Journal of Petrology Advance Access originally published online on November 3, 2006
Journal of Petrology 2007 48(1):113-139; doi:10.1093/petrology/egl056
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Clinopyroxene REE Geochemistry of the Red Hills Peridotite, New Zealand: Interpretation of Magmatic Processes in the Upper Mantle and in the Moho Transition Zone
1Earth Science Laboratory, Faculty of Education, Ehime University, Matsuyama 790-8577, Japan
2Department of Geoscience, Faculty of Science and Engineering, Shimane University, Matsue 690-8504, Japan
RECEIVED JUNE 23, 2005; ACCEPTED SEPTEMBER 5, 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL BACKGROUNDANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The Red Hills peridotite in the Dun Mountain ophiolite of South Island, New Zealand, is assumed to have been produced in a paleo-mid-ocean ridge tectonic setting. The peridotite is composed mostly of harzburgite and dunite, which represent residual mantle and the Moho transition zone (MTZ), respectively. Dunite channels within harzburgite blocks of various scales represent the MTZ component. Plagioclase- and clinopyroxene-bearing dunites occur sporadically within common dunites. These dunites represent products of melt–wall-rock interaction. Chondrite-normalized rare earth element (REE) patterns of MTZ clinopyroxenes show a wide compositional range. Clinopyroxenes in plagioclase dunites are extremely depleted in light REE (LREE) ([Lu/La]N >100), and are comparable with clinopyroxenes in abyssal peridotites from normal mid-ocean ridges. Interstitial clinopyroxenes in the common dunite have flatter patterns ([Lu/La]N
2) comparable with those for dunite in the Oman ophiolite. Clinopyroxenes in the lower part of the residual mantle harzburgites are even more strongly depleted in LREE ([Lu/La]N = 100–1000) than are mid-ocean ridge peridotites, and rival the most depleted abyssal clinopyroxenes reported from the Bouvet hotspot. In contrast, those in the uppermost residual mantle harzburgite and harzburgite blocks in the MTZ are less LREE depleted ([Lu/La]N = 10–100), and are similar to those in plagioclase dunite. Clinopyroxenes in the clinopyroxene dunite in the MTZ are similar to those reported from mid-ocean ridge basalt (MORB) cumulates, and clinopyroxenes in the gabbroic rocks have compositions similar to those reported from MORB. Strong LREE and middle REE (MREE) depletion in clinopyroxenes in the harzburgite suggests that the harzburgites are residues of two-stage fractional melting, which operated initially in the garnet field, and subsequently continued in the spinel lherzolite field. The early stage melting produced the depleted harzburgite. The later stage melting was responsible for the gabbroic rocks and dunite. Strongly LREE–MREE-depleted clinopyroxene in the lower harzburgite and HREE-enriched clinopyroxene in the upper harzburgite and plagioclase dunite were formed by later reactive melt migration occurring in the harzburgite. KEY WORDS: clinopyroxene REE geochemistry; Dun Mountain ophiolite; Moho transition zone; orogenic peridotite; Red Hills
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Complex magma behavior in the upper mantle and at the Moho transition zone (MTZ) has been investigated in many ophiolitic, orogenic, and oceanic peridotites (e.g. Kelemen et al., 1995
; Dijkstra et al., 2003). Partial melting processes in the upper mantle region are generally thought to be multi-stage near-fractional melting from the garnet to spinel stability fields (Klein & Langmuir, 1987; Johnson et al., 1990; Takazawa et al., 1999; Hellebrand et al., 2002a). However, the chemical processes that take place while melts move to the surface through the uppermost mantle and MTZ are still a matter of debate. For example, primary mid-ocean ridge basalt (MORB) generally cannot equilibrate with residual upper mantle harzburgite at low-pressure conditions (Stolper, 1980; Elthon & Scarfe, 1984). In contrast, dunite at the MTZ can be equilibrated with primary MORB (Kelemen et al., 1995). This implies that MORB was present at a shallow level, such as the MTZ. Recent studies have established the existence of ultra-depleted melts, which are trapped in olivine melt inclusions in MORB or MOR (mid-ocean ridge) cumulates (Ross & Elthon, 1993; Sobolev & Shimizu, 1993). Evidence of chemical modification of partial melts in the upper mantle by reaction with wall-rocks has also been documented (Takazawa et al., 1992; Garrido & Bodinier, 1999; Godard et al., 2000; Lenoir et al., 2001; Dijkstra et al., 2003). Such studies combine to shed light on the complex history of basaltic magma generation.
Large discordant dunite bodies in the MTZ are key rock types for direct observation of melt modification process in the shallow mantle. The origin of large discordant dunites within surrounding harzburgites has been discussed extensively (e.g. Quick, 1981a, 1981b; Kelemen, 1990; Kelemen et al., 1990, 2000; Suhr et al., 2003). Kelemen et al. (1995) examined melt behavior in the uppermost mantle and MTZ in the Oman ophiolite, and concluded that the Oman dunites represented conduits generated by focused melt flow, and that the dunites had equilibrated with MORB. The Red Hills ultramafic body in the Dun Mountain ophiolite belt of New Zealand is another good example of an association of upper mantle harzburgites and MTZ discordant dunites (Sano, 1991). We report here the geology, petrology, and clinopyroxene REE geochemistry of the Red Hills ultramafic body. We also debate the presence of melt fractions with diverse rare earth element (REE) chemistry in the MTZ dunites and upper mantle harzburgites, and possible processes that generate MORB.
|
GEOLOGICAL BACKGROUND |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The Red Hills ultramafic body is the largest body in the Dun Mountain ophiolite belt (Coombs et al., 1976
) of South Island, New Zealand (Fig. 1), and was largely unaffected by deformation during its emplacement. The Red Hills ultramafic body is exposed over an area of about 120 km2 in the northern part of the South Island, and is about 15 km in length and 8 km in width. Red Hills is separated from the Red Mountain ultramafic body to the SW (Sinton, 1977, 1980) by movement along the Alpine Fault. The western border of the Red Hills body is a fault contact with basalts and gabbros of the Lee River Group (Waterhouse, 1964), which constitutes the crustal section of the ophiolite sequence. The eastern part of the Lee River Group is mainly composed of a sheeted dike complex of diabase or microgabbro, and the western part is dominated by both massive and pillow basalts. At the western end of the Lee River Group, the pillow basalts are unconformably overlain by tuffaceous siltstone of the Permian Maitai Group (Johnston, 1981, 1982). At the western border of the Red Hills ultramafic body, strongly sheared and altered coarse-grained gabbroic rocks crop out in a serpentine mélange. The eastern border of the body is also in fault contact with metamorphosed sandstones and mudstones of the Permian to lower Mesozoic Caples Group (Bishop et al., 1976; Turnbull, 1979). The age of the igneous stage of the Dun Mountain ophiolite is about 280 Ma, as determined by zircon U–Pb dates for plagiogranites associated with the Red Hills ultramafic body (Kimbrough et al., 1992).
|
The geology of the Red Hills ultramafic body has been well described by Walcott (1969
). It consists of two zones that can be distinguished in terms of both the composition and structure of the rocks. The lower part of the body consists of massive harzburgite with a strongly developed porphyroclastic texture, whereas the upper part is composed mainly of dunite with an equigranular texture (Walcott, 1969).
Challis (1965) argued that the Red Hills ultramafic rocks formed by crystal accumulation beneath a Permian island arc, based especially on the well-developed layering between dunite and harzburgite of the upper zone. Blake & Landis (1973) and Coombs et al. (1976) considered that the Red Hills ultramafic body and volcanic rocks in the Lee River Group collectively represent the ophiolite member of the Permian Dun Mountain ophiolite belt. Davis et al. (1980) concluded that both harzburgite and dunite in the Red Hills ultramafic body could be interpreted as residual mantle that formed the floor of a magma chamber, and that the basalts and gabbros of the Lee River Group were the differentiated products of that chamber. They also concluded that the basalts in the Lee River Group originated at a mid-ocean ridge, based on their chemical and isotopic characteristics. Sano (1991) reported that the 87Sr/86Sr isotopic ratios of clinopyroxene in the Red Hills ultramafic body are very low, ranging from 0·7019 to 0·7027. Such low ratios have never been reported from island-arc or back-arc basin rocks. The Sr isotope ratios of clinopyroxenes in the Red Hills ultramafic body are similar to those from MORB in the East Pacific Rise, the Gorda Ridge, and the Juan de Fuca Ridge (Sano, 1991). On the other hand, many late-stage basalt and dolerite dikes in the dunite dominant zone exhibit thermal aureoles against the surrounding ultramafic rocks. Sano et al. (1997) suggested that the dike rocks have island arc characteristics, based on their isotope and trace element chemistry. The available data suggest that the Dun Mountain ophiolite formed at a mid-ocean ridge, and was transported and accreted to the continental margin (see Davis et al., 1980).
|
ANALYTICAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Mineral analyses were carried out by electron microprobe at the University of Otago, using mineral standards and Bence–Albee data correction. Whole-rock major elements were determined by X-ray fluorescence (XRF) at the University of Otago, and rare earth element (REE) abundances in clinopyroxene were determined by laser ablation–inductively coupled plasma mass spectrometry (LA-ICP-MS) at Fukushima University. Polished surfaces of clinopyroxenes in rock slabs were directly ablated using a 1063 µm IR Nd–YAG laser probe. The cores of the crystals were analysed. The crater size induced by the laser ablation for each analysed point was 120 µm. NIST SRM612 glass was used for calibration. 29Si was used for internal standardization between SRM612 and the samples. Lower limits of detection are 1 ppb for La, Ce, Pr, Tb, Ho, Tm and Lu, 2 ppb for Eu, Dy, Er and Yb, 5 ppb for Nd and Gd, and 13 ppb for Sm. Analytical accuracy is better than 15% (1
) for all REE when the element concentration is greater than 50 ppb. Analytical techniques follow those of Kimura et al. (1997, 2000). The REE standard values for NIST612 used were from Pearce et al. (1996). |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Field observations
The Red Hills ultramafic body is divided into two parts, a dunite-dominant unit in the western part and a harzburgite-dominant unit in the eastern part (Walcott, 1969
; Johnston, 1982; Fig. 1). As the compositional layering trends north–south and the dip is almost vertical, we constructed a 1/500 scale measured columnar section to crosscut the banding from west to east. Based on our field observations, we concluded that there is no significant tectonic gap in the section. A schematic section through the Red Hills ultramafic body and a sample location map are given in Figs 2 and 3, respectively. The thickness, including the harzburgite-rich parts, of the dunite-dominant unit (Upper Unit) is about 3 km. The Upper Unit is predominantly dunite that occurs as a matrix to the other lithologies. Sparse small-scale plagiogranite dikes (>5 cm width) and coarse-grained gabbro layers or sills occur only in the uppermost part of the Upper Unit. Late-stage diabase or microgabbro dikes often occur throughout the Upper Unit, and these show chilled margins against the surrounding dunite. Clinopyroxene-bearing lithologies (olivine clinopyroxenite and wehrlite) and plagioclase-bearing dunite are often observed in the Upper Unit (Fig. 4a and d). Harzburgite is observed as interbands or blocks in a dunite matrix. Field relationships between the harzburgite and dunite are illustrated in Fig. 5. The field relationship between harzburgite and dunite in Fig. 5a is similar to that of the dunite in the Oman ophiolite, which has been interpreted as a replacive origin (Kelemen et al., 1995).
|
|
|
|
The harzburgite-dominant unit (Lower Unit: Fig. 2) is composed of spinel harzburgite, plagioclase-bearing spinel harzburgite and a small amount of dunite. The uppermost part of the Lower Unit is characterized by the appearance of a plagioclase-bearing lithology (Fig. 4b and c). As shown in Fig. 4b, the alignment of plagioclase is oblique to the harzburgite foliation. Large irregular-shaped crystals of clinopyroxene are also observed in the uppermost part. The transition from the lowermost part of the Upper Unit to the uppermost part of the Lower Unit is indistinct. Small dunite bodies also occur rarely in the Lower Unit. In the field, transitions between the dunite and surrounding harzburgite are gradual. The harzburgite adjacent to dunite shows the most depleted characteristics, with low modal orthopyroxene contents. The grain size of olivine in the dunite can be up to a few centimeters.
Modal compositions
Modal compositions of the rocks (Fig. 6; Table 1) were determined by point counting, based on over 2000 points in each sample. Rock types are clearly divided into a dunite–olivine clinopyroxenite series and a dunite–harzburgite series.
|
|
Upper Unit
The Upper Unit is composed of olivine clinopyroxenite, wehrlite, troctolite, harzburgite and dunite. We use the terms ‘clinopyroxene-bearing dunite’ for olivine clinopyroxenite, wehrlite and dunite containing minor clinopyroxene, and ‘plagioclase-bearing dunite’ for troctolite and dunite containing minor plagioclase, because these lithologies are heterogeneous and it is difficult to apply the standard nomenclature for ultramafic rocks. Plagioclase dunite is almost free of clinopyroxene. The range of modal plagioclase is 4–24%. Harzburgite in the Upper Unit lies in the modal compositional range of oceanic peridotite (Fig. 6). The modal abundance of clinopyroxene in harzburgite is up to 2%, less than that of harzburgite in the Lower Unit.
Lower Unit
Samples from the Lower Unit are scattered throughout the harzburgite field, and fall within the oceanic peridotite field (Fig. 6). Modal clinopyroxene is up to 5%. Both plagioclase-bearing and plagioclase-free harzburgites plot in the same field.
Microscopic observations
Upper Unit
Rocks in the Upper Unit are generally characterized by coarse-grained, equigranular, recrystallized textures (Fig. 7a–d). Olivine crystals are polygonal in shape. Many 120° triple grain junctions are observed in dunites. Kink bands in coarse and equigranular olivines are observed in all dunites and harzburgite (Fig. 7a–d). Clinopyroxene in clinopyroxene dunite is coarse-grained and subhedral, and sometimes shows magmatic twinning (Fig. 7a). Plastic deformation in clinopyroxene is not observed. Plagioclase crystals surrounded by pargasite rims occupy melt-pocket shaped domains (Fig. 7c), and show no evidence of plastic deformation. Harzburgites in the dunite matrix in the Upper Unit (Fig. 7d) display high-temperature, low-stress porphyroclastic textures (Nicolas, 1986), which are interpreted as statically recrystallized textures. The harzburgites exhibit recrystallized textures, characterized by curved crystal boundaries and spherical spinel aggregations (Mercier & Nicolas, 1975).
|
Lower Unit
Harzburgites in the Lower Unit (Fig. 7e–h) display porphyroclastic textures (Ceuleneer et al., 1988
). The harzburgites are tectonites, and kink band structures are strongly developed in olivine and orthopyroxene porphyroclasts. Plagioclase-bearing harzburgite occurs mostly in the uppermost part of the Lower Unit (Fig. 7f and g). The plagioclase crystals occupy melt-pocket shaped domains similar to those in the plagioclase dunites in the Upper Unit and in the plagioclase harzburgites in the Lower Unit. As shown in Fig. 7g, plagioclase appears along the grain boundaries of olivine porphyroclasts. The harzburgite fabric and the orientation of the plagioclase is clearly discordant. Irregular megacrysts of clinopyroxene (c. 5–6 mm) containing distinct lamellae of orthopyroxene are observed in the uppermost part of the Lower Unit (Fig. 7e). Dunite in the Lower Unit has a coarse-grained porphyroclastic texture with sutured crystal boundaries. Distinct kink band structures are developed in olivine. Small rectangular spinels are scattered within the olivine grains, which have a grain size of about 2 cm. The texture of the Lower Unit dunite is distinctly different from that of the Upper Unit. The dunite contains small amounts of orthopyroxene and interstitial clinopyroxene.
Mineral chemistry
Olivine in the Red Hills ultramafic rocks has a wide compositional variation, ranging from Fo87 in Upper Unit clinopyroxene dunite to Fo93 in Lower Unit dunite (Fig. 8). Olivine in the Upper Unit dunite exhibits a bimodal compositional variation of Fo88·5–90·5 and Fo91·5–92·5. The high-Fo olivine composition is similar to that of olivine in the Lower Unit dunite, whereas the low-Fo olivine composition is similar to that of olivine in the plagioclase dunite and clinopyroxene dunite of the Upper Unit. Olivine in the Lower Unit harzburgites has a relatively uniform composition (Fo91), whereas that in the dunite exhibits the most refractory nature (Fo91·5–93·0).
|
The compositional range of cr-number [=Cr/(Cr + Al)) and mg-number [=Mg/(Mg + Fe2+)] in spinel lies within the field of ocean-floor peridotites (Dick & Bullen, 1984
; Hellebrand et al., 2002b) except for the most refractory dunite of the Lower Unit (Fig. 9). The composition of the high cr-number spinel in the dunite from the Lower Unit resembles that from the Western Pacific Margin (Ishii, 1985) or that in troctolite from the Hess Deep (Dick & Natland, 1996). Spinels in both the Upper Unit and the Lower Unit have almost the same cr-number, although the mg-number values of the spinels of the Upper Unit are lower than those of the Lower Unit. TiO2 contents of spinels in the Upper Unit vary considerably, but plagioclase dunites and common dunites are characterized by high TiO2.
|
A negative correlation between the modal olivine and Al2O3 contents of coexisting orthopyroxene is recognized in Lower Unit rocks (Fig. 10). This agrees with the residual harzburgite trend of oceanic peridotites (Dick et al., 1984
).
|
Whole-rock chemistry
Whole-rock major element compositions are reported in Table 2. All harzburgites in the Lower Unit have high MgO contents, and low Al2O3 and CaO contents (Fig. 11). Major element compositions of the Lower Unit harzburgites (filled symbols in Fig. 11) are homogeneous, and the compositional ranges are similar to those of oceanic, ophiolitic and alpine peridotites. Plagioclase-bearing harzburgites have less refractory compositions than plagioclase-free harzburgites, characterized by lower MgO and higher SiO2, Al2O3 and CaO contents. In contrast, rocks from the Upper Unit show wide compositional ranges (open symbols in Fig. 11). The chemical compositions of clinopyroxene dunite are controlled by the modal abundance of clinopyroxene, and have higher CaO and Al2O3 contents than harzburgites from the Lower Unit. Chemical variation within the plagioclase dunites is controlled by plagioclase modal abundances, and the rocks have high Al2O3 contents. The chemical compositions of the Upper Unit harzburgites differ from those of the Lower Unit. Blocky harzburgites in the Upper Unit are characterized by lower SiO2, Al2O3 and CaO contents and higher FeO* than the Lower Unit harzburgites at a given MgO. Chemical variations in the Upper Unit harzburgites thus have different trends from those of the Lower Unit harzburgites (Fig. 11b and d). These whole-rock chemical characteristics of the harzburgite in the Upper Unit are consistent with the modal compositions (Fig. 6), in which modal clinopyroxene in the Upper Unit harzburgite is lower than that in the Lower Unit.
|
|
Clinopyroxene REE chemistry
REE abundances in clinopyroxenes in the Red Hills ultramafic rocks are listed in Table 3. Chondrite-normalized REE patterns of clinopyroxenes from the Lower Unit harzburgite have extremely steep slopes in the light REE–middle REE (LREE–MREE) region (Fig. 12f). Although some clinopyroxenes are similar to those of oceanic peridotites from normal ridge segments (Johnson et al., 1990
; Johnson & Dick, 1992; Seyler & Bonatti, 1997), the extent of LREE depletion is much higher, and is comparable with or even greater than that observed in the Bouvet hotspot (Johnson et al., 1990; Johnson & Dick, 1992). HREE abundances range between one and nine times chondrite; one of the most depleted samples is comparable with the most depleted of the Hess Deep peridotites (Dick & Natland, 1996).
|
|
Clinopyroxenes in harzburgite from the uppermost part of the Lower Unit and in blocky harzburgite in the Upper Unit exhibit different REE patterns from those in the Lower Unit (Fig. 12e). Although two samples have very depleted HREE, LREE abundances are clearly an order of magnitude greater than those in the Lower Unit harzburgite.
Clinopyroxenes in common dunite, clinopyroxene dunite and plagioclase dunite from the Upper Unit show REE distinct patterns. Clinopyroxenes interstitial to olivine in common dunite have convex-upward patterns with Tb apices (Fig. 12b).
Clinopyroxenes from clinopyroxene dunite and plagioclase dunite also display strongly LREE-depleted patterns, whereas the MREE–HREE (heavy REE) abundances in the clinopyroxene from the plagioclase dunite are significantly greater than those in clinopyroxene dunite (Fig. 12c and d). The REE abundances and patterns of clinopyroxenes in the clinopyroxene dunite are in the range of those exhibited by mid-ocean ridge cumulates (Ross & Elthon, 1993), whereas those in plagioclase dunite are very similar to those in normal ridge abyssal peridotite (Johnson et al., 1990; Johnson & Dick, 1992).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Field and microscopic observations, modal compositions and geochemical characteristics of minerals and whole-rocks indicate that the Red Hills ultramafic body resembles the MTZ and residual mantle sections seen in many ophiolites (Quick, 1981a
, 1981b; Benn et al., 1988; Boudier & Nicolas, 1995). In the case of the Oman Ophiolite, the MTZ includes the melt-impregnated part of the residual mantle section. Our interpretation is that the MTZ in the Red Hills body is composed of the dunites of the Upper Unit and plagioclase harzburgite zone of the Lower Unit, whereas the plagioclase-free harzburgite in the Lower Unit corresponds to the residual mantle section. In the following section we discuss the origin of the Red Hills peridotites using clinopyroxene REE chemistry. Fractional melting, refertilization, and reactive melt flow play key roles in the formation of these peridotites.
Formation of residual harzburgite by two-stage fractional melting
Clinopyroxenes in residual harzburgite from the mantle section show extreme LREE depletion (Fig. 12). Although the REE patterns of clinopyroxenes in the residual peridotites of normal mid-ocean ridges can be formed by fractional melting of a MORB-source mantle in the spinel field (Johnson et al., 1990), the ultra-depleted LREE patterns observed here (Fig. 12) cannot be explained by simple spinel field processes. Johnson et al. (1990) also reported extremely MREE–LREE-depleted clinopyroxenes from the Bouvet and Discovery II fracture zones of the Southwest Indian Ridge. These fracture zones are thought to have been influenced by hotspot magma activity. Johnson et al. (1990) explained the characteristic REE patterns by dynamic melting that occurred in the source mantle in the garnet field, prior to melting in the spinel field. The ultra-depletion of LREE in the Red Hills mantle section harzburgites may also be explained by such polybaric melting processes.
To model the melting process, we assume fractional melting, because neither batch nor critical melting (Langmuir et al., 1977; Sobolev & Shimizu, 1992) models can generate ultra-depleted LREE patterns. We used a MORB-source mantle (Salters & Stracke, 2004) for the model calculation. The distribution coefficients between minerals and melt, starting solid mode and melting mode parameters used are listed in Tables 4 and 5. The calculation scheme follows Shaw (1970) and Johnson et al. (1990). If a MORB-source mantle melts fractionally in the garnet field and garnet remains as a residual solid, the clinopyroxene in the residue shows REE patterns that are humped in the segment Nd to Er (Fig. 13b). Even after 15% melting in either the garnet or spinel fields, clinopyroxene REE patterns in the residual solid similar to those in the harzburgite cannot be achieved by the models (Fig. 13a and b).
|
|
We also tested a two-stage fractional melting model, with initial melting in the garnet field followed by melting in the spinel field. Four cases were calculated, for 1%, 5%, 10% and 15% melting in the garnet field. The residue compositions at each degree of melting are listed in Table 6. The garnet-bearing residual mantle changes in accordance with the subsolidus decompression reaction from olivine + garnet to 1·25 orthopyroxene + spinel + 0·75 clinopyroxene (Johnson et al., 1990
). The changes in modal abundances of the mantle source from the garnet field to the spinel field following this reaction are shown in Table 5. For example, the modal composition of the residue after 10% melting in the garnet field is olivine: orthopyroxene:clinopyroxene:garnet = 0·60:0·24:0·08:0·08 (Table 5). From the subsolidus decompression reaction above, the modal composition becomes olivine: orthopyroxene:clinopyroxene:spinel = 0·49:0·32:0·13:0·07 (Table 5). Accordingly, the clinopyroxene REE composition would change to preserve the appropriate interphase partitioning, as diffusion is probably sufficiently rapid at relatively high temperatures. The clinopyroxene REE pattern in the residue after 1% melting in the garnet field shown in Fig. 13b would change to that indicated by the bold dashed line in Fig. 13c. Similarly, the clinopyroxene REE patterns in the residue after 5%, 10% and 15% melting in the garnet field (Fig. 13b) change to form the patterns displayed by the bold dashed lines in Fig. 13d–f, respectively. With the modal abundances newly equilibrated to the spinel field, melting calculations were performed for each residual mantle source composition to estimate the residual clinopyroxene REE patterns after 1–25% melting in the spinel field (Fig. 13c–f). As a result, the ultra-depleted LREE patterns in clinopyroxenes can be explained by two-stage melting, with 10% fractional melting of a MORB-source mantle in the garnet field, and 12% further fractional melting in the spinel field. This corresponds to total melting of 10–22% through the garnet to spinel fields.
|
|
Johnson et al. (1990
), Hellebrand et al. (2002a), and Barth et al. (2003) have previously pointed out that LREE-depleted clinopyroxenes in harzburgites cannot be explained by simple batch or fractional melting processes in the spinel field, and that multiple-stage fractional melting in the garnet and spinel stability fields is required. Barth et al. (2003) estimated that 4% near-fractional melting in the garnet field and an additional 12–20% near-fractional melting in the spinel field was required to model the clinopyroxene REE pattern from the Othris ophiolite. Hellebrand et al. (2001) developed a method for estimating the degree of melting in MOR peridotites based on spinel cr-number. The range of total melting estimated from the cr-number of the Red Hills harzburgites is about 10–18% (Fig. 9). This agrees well with the estimation from the clinopyroxene REE patterns (Fig. 13). Olivine and spinel compositions from the harzburgites in the Lower Unit plot on the residual mantle array (Fig. 14). The wide variation of spinel cr-number from 0·25 to 0·55 in a single mantle massif appears to indicate variable degrees of partial melt extraction. In contrast, the Red Hills MTZ dunites of all types plot well away from the mantle melting array with lower Fo composition in olivine (Fig. 14), suggesting re-equilibration with MORB melts or a cumulate origin.
|
Refertilization in the uppermost part of the mantle section
The petrological characteristics of the plagioclase-bearing harzburgites in the uppermost part of the Lower Unit are similar to those of oceanic plagioclase peridotites (Seyler & Bonatti, 1997
; Tartarotti et al., 2002). Plagioclase peridotites have also been reported from ophiolites and peridotite massifs. For example, the Othris peridotite massif contains a large volume of plagioclase peridotite (Menzies, 1973; Dijkstra et al., 2001; Barth et al., 2003). Oceanic, ophiolitic and massif plagioclase peridotites are thought to have formed by melt impregnation of a depleted peridotite protolith (Rampone et al., 1997; Seyler & Bonatti, 1997; Dijkstra et al., 2001; Tartarotti et al., 2002; Barth et al., 2003). Crystallization of plagioclase can be regarded as evidence of refertilization of depleted peridotite by melt impregnation (Rampone et al., 1997; Müntener et al., 2004). Plagioclase lenses are often observed in the harzburgite in the uppermost part of the mantle section at Red Hills (Fig. 4). The occurrence of the plagioclase is similar to that in the Othris massif. Plagioclase lenses are discordant to the foliation of the harzburgite in both cases. Under the microscope plagioclase crystals display rounded droplet- or bleb-like shapes (Fig. 7c, f and g). The plagioclase blebs are found along grain boundaries between mafic minerals. The field occurrence and microstructures of plagioclase in the harzburgite (Fig. 4b and c) support plagioclase crystallization by melt impregnation of the harzburgite.
Irregular aggregations of large crystals of clinopyroxene are observed in the harzburgite of the uppermost part of the mantle section (Fig. 7e). The occurrence of these clinopyroxene aggregations is similar to that in peridotites at the recrystallization front of the Ronda massif (Lenoir et al., 2001). Lenoir et al. (2001) suggested that coarse-granular peridotite containing clinopyroxene aggregations was formed by partial solidification of percolating melt, leading to peridotite refertilization. Harzburgite blocks in the dunite matrix in the Red Hills MTZ are also characterized by identical recrystallization textures (Fig. 7d), which look like the statically recrystallized harzburgites in the Ronda peridotite, suggesting that melt impregnation and recrystallization took place in the uppermost residual mantle section and in the blocky harzburgite in the MTZ.
Clinopyroxenes in the uppermost harzburgite in the mantle section and those in the blocky harzburgite in the MTZ (Upper Unit) have LREE-depleted REE patterns; however, the degree of depletion is less marked than in the Lower Unit harzburgites (Fig. 12e). The chondrite-normalized La abundance in the clinopyroxenes is (La)N = 0·05–0·1, whereas that in the normal harzburgite in the mantle section is much lower [(La)N <0·01; Fig. 12f]. Relatively high LREE abundances in clinopyroxene in the uppermost harzburgite and in the blocky harzburgite may be caused by circulation of a small melt fraction. Godard et al. (2000) examined the REE patterns of peridotites in the residual mantle of the Oman ophiolite, and concluded that cpx-rich harzburgites showing ‘spoon-shape’ REE patterns were the products of a cpx-forming melt–rock reaction, indicating refertilization. The degree of selective LREE enrichment is probably related to the circulation of a small melt fraction through the peridotite (Bodinier & Godard, 2003). Spoon-shaped REE patterns or LREE enrichment of clinopyroxene have also been regarded as the products of chromatographic reaction between melt and wall-rock (Navon & Stolper, 1987; Takazawa et al., 1992). The Red Hills clinopyroxenes with relatively high LREE abundances in the uppermost mantle section and in the blocky harzburgites could presumably be generated by refertilization of the depleted mantle by melt percolation. The varied lithologies that formed in the uppermost part of the residual mantle, ranging from plagioclase-bearing harzburgite and dunite to clinopyroxene-bearing dunite, can be explained by this secondary process.
Reactive melt flow model for harzburgite and dunite formation
An alternative model for mantle melting is a ‘chromatographic’ or a ‘reactive melt percolation’ process (e.g. Navon & Stolper, 1987; Vernières et al., 1997). Vernières et al. (1997) proposed a plate model for the geochemical simulation of melt reaction and migration in the mantle. We applied REE plate model calculations to simulate the reactive melt migration in the Red Hills harzburgites and dunites, to examine the role of percolated melts in depletion and refertilization in the upper mantle section.
For the harzburgite model, we used the bulk REE composition of a Red Hills harzburgite as the starting composition (Table 7). The Red Hills harzburgite itself is much more depleted than any mantle reservoir, such as primitive mantle (e.g. Sun & McDonough, 1989) or depleted MORB-source mantle (e.g. Salters & Stracke, 2004). To generate such marked depletion, near fractional melting in the garnet or spinel fields with about 10–18% melting is required, as discussed above. The melts generated under these conditions should have a MORB composition (Salters & Stracke, 2004). Consequently, the strongly depleted harzburgite composition observed at Red Hills is regarded as residual after melt extraction.
|
We assumed the initial mineralogy to be olivine:orthopyroxene:clinopyroxene in the proportions 0·70:0·20:0·10 (Table 7). This assumption is based on the sub-solidus mineralogy at 1 GPa of a depleted mantle source (Workman & Hart, 2005
) after 10% MORB extraction. This was obtained from calculations using pMELTS (Ghiorso et al., 2002) associated with the adiabat_1ph program (Smith & Asimow, 2005). We set 20 vertical cells for the reaction model, using the following melting parameters: (1) 5% compaction limit; (2) melting mineralogy olivine:orthopyroxene:clinopyroxene = 0:0·01:0·99 in a cell at 5% melting; (3) no melt influx at the bottom cell. This model simulates melting occurring in the spinel field at the base of the column. The melts generated migrate upwards when degrees of partial melting exceed 5% in each cell [parameters from Vernières et al. (1997)]. The harzburgite mineralogy produced after 20 steps is designed to correspond to the observed harzburgites (modal clinopyroxene <5%). The clinopyroxene REE compositions in each cell were calculated from the interstitial melt compositions (Fig. 15).
|
The calculated results from the lower five cells reproduce the ultra LREE-depleted patterns of the clinopyroxenes in the Lower Unit harzburgites fairly well (Fig. 15a). The clinopyroxenes in the upper cells (cells 10–20) are less depleted in LREE and are enriched in HREE, similar to those from the plagioclase-bearing upper harzburgite and the plagioclase dunite in the MTZ (Fig. 15b). Moreover, the interstitial melt compositions in the upper cells (cells 10–20) are similar to the LREE-depleted melts reported by Shimizu (1998
) from olivine-hosted melt inclusions in MORB from the FAMOUS area (Fig. 15a–c).
The model calculations suggest that the clinopyroxenes in the lower harzburgite and plagioclase harzburgite and plagioclase dunite with variable compositions can be generated by reactive melt flow in the spinel field, if upward reactive melt percolation occurred in the system. The advantages of this model are two-fold: (1) the vertical spatial variation in the clinopyroxene REE chemistry is reasonably explained by upward melt migration; (2) melt compositions generated at the top of the harzburgite column are similar to those actually observed in MOR. This model is also consistent with the lower Fo contents of olivine in the plagioclase dunite and upper plagioclase-bearing harzburgite (Fig. 14). A discrepancy occurs in the LREE compositions of some clinopyroxenes from harzburgites, which have slightly elevated LREE and ‘spoon-shaped’ REE patterns as described above. This feature could be generated by influx of less depleted melts in the percolation column (e.g. Vernières et al., 1997). It is possible that the harzburgite had both relatively focused and diffuse melt flow regimes, which allowed complex 3D variations in the melt migration pathways.
The REE pattern of the melt estimated from the Red Hills gabbros is not fractionated, and is similar to that of the MORB glass composition reported by Niu & Hekinian (1997) (Fig. 15). Therefore, it is probable that the gabbroic rock is a product from normal mid-ocean ridge basalt (N-MORB). The clinopyroxenes in clinopyroxene dunite and some clinopyroxenes in harzburgite in the Upper Unit have different REE patterns with flatter and lower HREE regions (Fig. 15c). The patterns are similar to those in MOR cumulates reported elsewhere (Ross & Elthon, 1993), suggesting derivation from pooled MORB melts. This view is consistent with the observation that the modal clinopyroxene content in the clinopyroxene dunite exceeds 45% and may reach 83% (clinopyroxenite; see Table 1 and Fig. 6). The clinopyroxene chemistries of the gabbroic rocks and the clinopyroxene dunite may be explained by precipitation from MORB melts, rather than by the reactive flow model.
The origin of the MREE-enriched clinopyroxenes in the common dunite is still problematic. However, a plate model using the conditions of (1) source dunite with its REE composition and modal mineralogy similar to those in the Red Hills, (2) slightly LREE-enriched MORB melt infiltration, and (3) fully fractional reaction (Table 7) can generate MREE-humped clinopyroxenes in the bottom cells (cells 1 and 2, Fig. 15d). In this case, the trace amount of interstitial clinopyroxenes present in the common dunite would be the result of secondary impregnation by incipient melt infiltration after formation of the dunite body. An alternative model is simpler. If the host dunite had an extremely depleted composition, the MREE-enriched patterns could also be generated by simple precipitation of the clinopyroxene from residual interstitial MORB melt. The lowermost cells (cells 1 and 2) in the reactive flow model correspond to this condition.
Origin of common dunite in the Lower Unit and MTZ
A problem with the ‘reactive melt flow after depletion’ model is the formation mechanism of the massive common dunite and clinopyroxene dunite in the MTZ. As melts percolate from the bottom to the top of the reaction column, incremental melting would produce dunite in the lower column rather than in the upper column. This model may explain the Lower Unit dunite, which has a more depleted nature as shown by high cr-number = 0·7–0·8 in spinel (Fig. 14). Although focused melt flow can generate dunite conduits by the same mechanism, the formation of massive common dunite in the MTZ still seems difficult.
Common dunite in the MTZ could represent a MORB cumulate. In this scenario, the common dunite would have been refertilized by an incipient MORB melt to form the MREE-humped clinopyroxene during the later stage. However, it is difficult to form the more than 2 km thickness of common dunite in the MTZ by this mechanism. An alternative explanation is that intensive melt extraction by reactive flow occurred during the early melting stage. If this is the case, dunite can be formed by channeled flow and at the same time as depleted harzburgite by porous flow (Kelemen, 1990). MREE-enriched clinopyroxenes in the dunite can also be formed by the residual interstitial MORB melt in an extremely depleted dunite host, as discussed above. This model seems reasonable, because MORB melts extracted in the early stage could also form the Red Hills gabbroic rocks and clinopyroxene dunite. Generation of a huge volume of MORB is also a prerequisite for the MOR system. With this model, compositionally varied clinopyroxenes in the harzburgite were generated in the later melting stage after a massive volume of MORB components had been extracted.
Mantle melting regime beneath mid-ocean ridge systems
Trace element geochemistry suggests that MORB do not equilibrate with residual abyssal peridotites (Johnson et al., 1990; Johnson & Dick, 1992). This discrepancy between MORB and residual peridotite compositions at shallow levels suggests that the melt must be transported to the surface without complete re-equilibration with the surrounding peridotite (Kelemen et al., 1995). Diffuse porous flow melt transport should cause intensive chemical re-equilibration and reaction, and therefore focused flow in spatially restricted conduits is needed for melt aggregation and transport, the remnant of which might be represented by dunite channel networks in the MTZ (Kelemen et al., 1995). Reaction of melts against host peridotite during transport also causes significant modification and variations in melt REE composition (Vernières et al., 1997). The same relationship between residual harzburgites and dunites is observed in the Red Hills mantle–MTZ section.
Our study of the major and REE element compositions of clinopyroxenes from the Red Hills residual mantle–MTZ section suggests the following important conclusions for the mantle melts in the MOR system.
- Clinopyroxene compositions in the residual harzburgite in both the Lower Unit and the MTZ can be explained by two-stage melting in the garnet and spinel fields with reactive melt flow migration taking place in the spinel field. The harzburgite body would essentially have been formed in the early stage as a result of MORB extraction.
- Plagioclase dunite in the MTZ formed by impregnation of the dunite by melts extremely enriched in HREE. Such melts can be generated by reactive melt flow in the spinel field. LREE-depleted MORB melt (Shimizu, 1998) is generated by this process.
- Clinopyroxene dunite in the MTZ formed by impregnation by a larger amount of MORB melt.
- The thick common dunite in the MTZ formed by melt reaction during MORB extraction in the early stage. MREE-enriched clinopyroxene in common dunite formed from residual N-MORB.
- Clinopyroxenes in the gabbroic rock indicate precipitation from a typical MORB melt.
These observations indicate that the melts beneath the Red Hills MOR system can be classified into two types: (1) depleted-MORB to N-MORB type melts, which may be related to first stage depletion of the MORB-source mantle peridotite in the garnet or spinel fields; (2) extremely LREE-depleted and HREE-enriched melts, both of which are related to intra-harzburgite reactive melt flow. The MORB-type melts occur only in the MTZ, and hence are always related to regions in which the presence of massive volumes of melt can be expected. The melts with extreme compositions occur largely in or near the harzburgite bodies with depleted melt in the Lower Unit and enriched melt in the upper MTZ.
The above observations lead us to conclude that a consecutive two-step melting model operated, with the processes of (1) early stage intensive MORB melt extraction by both focused and diffused flow systems, generating the dunites and the depleted harzburgites, respectively, and (2) late-stage reactive melt flow in the depleted harzburgites. The former process generated the gabbroic rocks, clinopyroxene dunite and depleted harzburgite. The latter process generated the ultra LREE-depleted clinopyroxenes in the harzburgite and the HREE-enriched clinopyroxenes in the upper plagioclase-bearing harzburgite and plagioclase dunite in the Red Hills peridotite. The MOR melt extraction system consists of a combination of diffuse and focused flow regimes (e.g. Iwamori, 1993; Kelemen et al., 1995), and the Red Hills case study described here provides a good example. The composite melt transport mechanism could have existed either in the early melting stage or in the later melting stage. The model presented here appears to best explain the variation of melts present in the Red Hills mantle–MTZ section, and provides the missing link between the MORB and MOR mantle residual melts.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The Red Hills peridotite body of the Dun Mountain ophiolite belt can be divided into the MTZ and residual mantle. Harzburgite in the mantle is the residue after multi-stage fractional melting, which operated initially in the garnet stability field, and continued in the spinel field. In contrast, discordant dunites in the MTZ record a complex magma migration process with melt influx and reaction under varying conditions. Formation of the massive dunite in the MTZ and harzburgite in the residual mantle may have been related to an early MORB extraction stage with both focused melt flow and diffuse melt flow regimes, respectively. A later melting stage in the depleted harzburgite generated extreme melt compositions by reactive melt flow. Evidence for the two-stage melting is recorded in the clinopyroxene REE chemistry. Traces of N-MORB and depleted-MORB melt compositions are found in the clinopyroxenes in the gabbroic rocks and the possible MOR cumulate clinopyroxenes in the clinopyroxene dunite in the MTZ. Extreme melt compositions are exhibited by clinopyroxenes in depleted harzburgite and in the refertilized plagioclase dunite in the MTZ. The formation processes of the melts with extreme compositions were successfully simulated by upward migration of reactive melt in the harzburgite in the spinel field. The combination of early melt depletion followed by later reactive melt migration successfully explains the variations in REE chemistry of the clinopyroxenes from the Red Hills peridotite.
|
ACKNOWLEDGEMENTS |
|---|
We extend our sincere thanks to D. S. Coombs and Y. Kawachi of Otago University for their discussion and suggestions while S.S. visited Otago. We also thank B. P. Roser of Shimane University for comments on the draft. Constructive comments from reviewers Drs A. Djikstra, E. Rampone and J.-L. Bodinier considerably improved this paper. Drs G. Suhr and E. Takazawa also improved an early version of the manuscript. Dr J.-L. Bodinier also kindly provided the code of a Plate Model calculation program to J.-I.K. S.S. received financial support from D. S. Coombs for field survey of the Red Hills area. This work was partially supported by grants-in-aid for scientific research from the Japan Society for the Promotion of Science (JSPS) to S.S. (Nos 11640454, 14540429) and to J.-I.K (Nos 12874058, 10304038).
*Corresponding author. Telephone: +81-89-927-9443. Fax: +81-89-927-9396. E-mail: sano{at}ed.ehime-u.ac.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUNDANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Arai S. (1994) Characterization of spinel peridotites by olivine–spinel compositional relationships: review and interpretation. Chemical Geology 113:191–204.[CrossRef][Web of Science]
Anders E and Grevesse N. (1989) Abundances of the elements: meteoric and solar. Geochimica et Cosmochimica Acta 53:197–214.[CrossRef][Web of Science]
Baker MB and Stolper EM. (1994) Determining the composition of high-pressure mantle melts using diamond aggregates. Geochimica et Cosmochimica Acta 58:2811–2827.[CrossRef][Web of Science]
Barth MG, Mason PRD, Davies GR, Dijkstra AH, Drury MR. (2003) Geochemistry of the Othris Ophiolite, Greece: evidence for refertilization? Journal of Petrology 44:1759–1785.
Benn K, Nicolas A, Reuber I. (1988) Mantle–crust transition zone and origin of wehrlitic magmas: evidence from the Oman ophiolite. Tectonophysics 151:75–85.[CrossRef][Web of Science]
Bishop DG, Bradshaw JD, Landis CA, Turnbull IM. (1976) Lithostratigraphy and structure of the Caples terrane of the Humboldt Mountains, New Zealand. New Zealand Journal of Geology and Geophysics 19:827–848.
Blake MC Jr and Landis CA. (1973) The Dun Mountain ultramafic belt—Permian oceanic crust and upper mantle in New Zealand. US Geological Survey Journal of Research 1:529–534.
Bodinier J-L. (1988) Geochemistry and petrogenesis of the Lanzo peridotite body, western Alps. Tectonophysics 149:67–88.[CrossRef][Web of Science]
Bodinier J-L and Godard M. (2003) Orogenic, ophiolitic, and abyssal peridotites. In Carlson RW, Holland HD, Turekian KK (Eds.). The Mantle and Core. Volume 2 of Treatise on Geochemistry(Elsevier–Pergamon, Oxford, Oxford) pp. 103–170.
Bodinier J-L, Dupuy C, Dostal J. (1988) Geochemistry and petrogenesis of Eastern Pyrenean peridotites. Geochimica et Cosmochimica Acta 52:2893–2907.[CrossRef][Web of Science]
Boudier F and Nicolas A. (1995) Nature of the Moho Transition Zone in the Oman Ophiolite. Journal of Petrology 36:777–796.
Ceuleneer G, Nicolas A, Boudier F. (1988) Mantle flow patterns at an oceanic spreading center: the Oman peridotites record. Tectonophysics 151:1–26.[CrossRef][Web of Science]
Challis GA. (1965) The origin of New Zealand ultramafic intrusion. Journal of Petrology 6:322–364.
Constantin M, Hékinian R, Ackermand D, Stoffers P. (1995) Mafic and ultramafic intrusions into upper mantle peridotites from fast spreading centers of the Easter microplate (South east Pacific). In Vissers RLM and Nicolas A (Eds.). Mantle and Lower Crust Exposed in Oceanic Ridges and in Ophiolites(Kluwer Academic, Dordrecht) pp. 71–120.
Coombs DS, Landis CA, Norris RJ, Sinton JM, Borns DJ, Nakamura Y, Craw D. (1976) The Dun Mountain ophiolite belt, New Zealand, its tectonic setting, constitution, and origin, with special reference to the southern portion. American Journal of Science 276:561–603.
Davis TE, Johnston MR, Rankin PC, Stull RJ. (1980) The Dun Mountain ophiolite belt in east Nelson, New Zealand. Proceedings of International Ophiolite Symposium1979Cyprus(Geological Survey DepartmentIn Panayiotiou A (Ed.). , Nicosia) pp. 480–496.
Dick HJB. (1989) Abyssal peridotites, very slow spreading ridges and ocean ridge magmatism. In Saunders AD and Norry MJ (Eds.). Magmatism in the Ocean Basins. Geological Society, London, Special Publications 42: pp. 71–105.
Dick HJB and Bullen T. (1984) Chromian spinel as a petrogenetic indicator in abyssal and alpine-type peridotites and spatially associated lavas. Contributions to Mineralogy and Petrology 86:54–76.[CrossRef][Web of Science]
Dick HJB, Fisher RL, Bryan WB. (1984) Mineralogic variability of the uppermost mantle along mid-ocean ridges. Earth and Planetary Science Letters 69:88–106.[CrossRef][Web of Science]
Dick HJB and Natland JH. (1996) Late-stage melt evolution and transport in the shallow mantle beneath the East Pacific Rise. In Mével C, Gillis KM, Allan JF, Meter PS (Eds.). Proceedings of the Ocean Drilling Program, Scientific Results, 147(Ocean Drilling Program, College Station, TX) pp. 103–134.
Dijkstra AH, Drury MR, Vissers RLM. (2001) Structural petrology of plagioclase-peridotites in the West Othris Mountains (Greece): melt impregnation in mantle lithosphere. Journal of Petrology 42:5–24.
Dijkstra AH, Barth MG, Drury MR, Mason PRD, Vissers LM. (2003) Diffuse porous melt flow and melt–rock reaction in the mantle lithosphere at a slow-spreading ridge: a structural petrology and LA-ICP-MS study of the Othris Peridotite Massif (Greece). Geochemistry, Geophysics, Geosystems 4: doi:10.1029/2001GC000278.
Elthon D and Scarfe CM. (1984) High-pressure phase equilibria of a high-magnesia basalt and the genesis of primary oceanic basalt. American Mineralogist 69:1–15.[Abstract]
Ernst WG. (1978) Petrochemical study of lherzolitic rocks from the Western Alps. Journal of Petrology 19:341–392.
Frey FA, Suen CJ, Stockman HW. (1985) The Ronda high temperature peridotite: geochemical and petrogenesis. Geochimica et Cosmochimica Acta 49:2469–2491.[CrossRef][Web of Science]
Garrido CJ and Bodinier J-L. (1999) Diversity of mafic rocks in the Ronda peridotite: evidence for pervasive melt rock reaction during heating of subcontinental lithosphere by upwelling asthenosphere. Journal of Petrology 40:729–754.[CrossRef][Web of Science]
Ghiorso MS, Hirschmann MM, Reiners PW, Kress VC. (2002) The pMELTS: a revision of MELTS for improved calculation of phase relations and major element partitioning related to partial melting of the mantle to 3 GPa. Geochemistry, Geophysics, Geosystems 3: paper number 1030, doi:10.1029/2001GC000217.
Godard M, Jousselin D, Bodinier JL. (2000) Relationships between geochemistry and structure beneath a paleo-spreading centre: a study of the mantle section in the Oman Ophiolite. Earth and Planetary Science Letters 180:133–148.[CrossRef][Web of Science]
Hamlyn PR and Bonatti E. (1980) Petrology of mantle-derived ultramafics from the Owen fracture zone, Northwest Indian Ocean: implications for the nature of the oceanic upper mantle. Earth and Planetary Science Letters 48:65–79.[CrossRef][Web of Science]
Hékinian R, Bideau D, Francheteau J, Cheminée JL, Armijo R, Lonsdale P, Blum N. (1993) Petrology of the East Pacific Rise crust and upper mantle exposed in Hess Deep (Eastern Equatorial Pacific). Journal of Geophysical Research 98:8069–8094.
Hellebrand E, Snow JE, Dick HJB, Hofmann AW. (2001) Coupled major and trace elements as indicators of the extent of melting in mid-ocean-ridge peridotites. Nature 410:677–681.[CrossRef]
Hellebrand E, Snow JE, Hoppe P, Hofmann AW. (2002a) Garnet-field melting and late-stage refertilization in ‘residual’ abyssal peridotites from the Central Indian Ridge. Journal of Petrology 43:2305–2338.
Hellebrand E, Snow JE, Mühe R. (2002b) Mantle melting beneath Gakkel Ridge (Arctic Ocean): abyssal peridotite spinel compositions. Chemical Geology 182:227–235.[CrossRef][Web of Science]
Ishii T. (1985) Dredged samples from the Ogasawara fore-arc seamount or ‘Ogasawara Paleoland’–‘fore-arc ophiolite’. In Nasu N, Kobayashi K, Uyeda S, Kushiro I, Kagami H (Eds.). Formation of Active Ocean Margins(Terra Scientific Publishing Company (Terrapub), Tokyo) pp. 307–342.
Iwamori H. (1993) Dynamic disequilibrium melting model with porous flow and diffusion-controlled chemical equilibration. Earth and Planetary Science Letters 114:301–313.[CrossRef][Web of Science]
Johnson KT, Dick HJB, Shimizu N. (1990) Melting in the oceanic upper mantle: an ion microprobe study of diopsides in abyssal peridotites. Journal of Geophysical Research 95:2661–2678.
Johnson KT and Dick HJB. (1992) Open system melting and temporal and spatial variation of peridotite and basalt at the Atlantis II Fracture Zone. Journal of Geophysical Research 97:9219–9241.
Johnston MR. (1981) Sheet O27 AC_Dun Mountain, 1st edn. Geological map of New Zealand, 1:50 000(Department of Scientific and Industrial Research, Wellington).
Johnston MR. (1982) Sheet N28 BD_Red Hills, 1st edn. Geological map of New Zealand, 1:50 000(Department of Scientific and Industrial Research, Wellington).
Kelemen PB. (1990) Reaction between ultramafic rocks and fractionating basaltic magma 1. Phase relations, the origin of calc-alkaline magma series, and the formation of discordant dunite. Journal of Petrology 31:51–98.
Kelemen PB, Joyce DB, Webster JD, Holloway JR. (1990) Reaction between ultramafic rock and fractionating basaltic magma 2. Experimental investigation of reaction between olivine tholeiite and harzburgite at 1150–1050°C and 5 kbar. Journal of Petrology 31:99–134.
Kelemen PB, Shimizu N, Salters VJM. (1995) Extraction of mid-ocean-ridge basalt from the upwelling mantle by focused flow of melt in dunite channels. Nature 375:747–753.[CrossRef]
Kelemen PB, Braun M, Hirth G. (2000) Spatial distribution of melt conduits in the mantle beneath oceanic spreading ridges: observations from the Ingalls and Oman ophiolites. Geochemistry, Geophysics, Geosystems 1: 1999GC000012.
Kimbrough DL, Mattinson JM, Coombs DS, Landis CA, Johnston MR. (1992) Uranium–lead ages from the Dun Mountain ophiolite belt and Brook Street terranes, South Island, New Zealand. Geological Society of America Bulletin 104:429–443.
Kimura J-I, Tsuchiya N, Sano S, Chuman N, Yoshida T. (1997) Quantitative trace element analysis using inductively coupled plasma-mass spectrometry (ICP-MS). Chikyukagaku (Geochemistry) 31:133–151.
Kimura J-I, Danhara T, Iwano H. (2000) Preliminary report on trace element determination in zircon and apatite crystals using excimer laser ablation-inductively coupled plasma mass spectrometry (ExLA-ICPMS). Fission Track News Letter 13:11–20.
Klein E and Langmuir CH. (1987) Global correlations of ocean ridge basalts with axial depth and crustal thickness. Journal of Geophysical Research 92:8089–8115.
Langmuir CH, Bender JF, Bence AE, Hanson GN, Taylor SR. (1977) Petrogenesis of basalts from the FAMOUS area: Mid-Atlantic Ridge. Earth and Planetary Science Letters 36:133–156.[CrossRef][Web of Science]
Lenoir X, Garrido CJ, Bodinier J-L, Dautria J-M, Gervilla F. (2001) The recrystallization front of the Ronda peridotite: evidence for melting and thermal erosion of subcontinental lithospheric mantle beneath the Alboran basin. Journal of Petrology 42:141–158.
Loney RA and Himmelberg GR. (1989) The Kanuti Ophiolite, Alaska. Journal of Geophysical Research 94:15869–15900.[CrossRef]
McDonough WF and Sun S-S. (1995) The composition of the Earth. Chemical Geology 120:223–253.[CrossRef][Web of Science]
McKenzie D and O’Nions RK. (1991) Partial melt distributions from inversion of rare earth element concentrations. Journal of Petrology 32:1021–1091.
Menzies M. (1973) Mineralogy and partial melt textures within an ultramafic–mafic body, Greece. Contributions to Mineralogy and Petrology 42:273–285.[CrossRef][Web of Science]
Mercier J-CC and Nicolas A. (1975) Textures and fabrics of upper-mantle peridotites as illustrated by xenoliths from basalts. Journal of Petrology 16:454–487.
Michel PJ and Bonatti E. (1985) Peridotite composition from the North Atlantic: regional and tectonic variations and implications for partial melting. Earth and Planetary Science Letters 73:91–104.[CrossRef][Web of Science]
Müntener O, Pettke T, Desmurs L, Meier M, Schaltegger U. (2004) Refertilization of mantle peridotite in embryonic ocean basins: trace element and Nd isotopic evidence and implications for crust–mantle relationships. Earth Planetary Science Letters 221:293–308.
Navon O and Stolper E. (1987) Geochemical consequences of melt percolation: the upper mantle as a chromatographic column. Journal of Geology 95:285–307.[Web of Science]
Nicolas A. (1986) Structure and petrology of peridotites: clues to their geodynamic environment. Reviews of Geophysics 24:875–895.[Web of Science]
Niu Y. (1997) Mantle melting and melt extraction processes beneath ocean ridges: evidence from abyssal peridotites. Journal of Petrology 38:1047–1074.[CrossRef][Web of Science]
Niu Y and Hekinian R. (1997) Basaltic liquids and harzburgitic residues in the Garnet Transform: a case study at fast-spreading ridges. Earth and Planetary Science Letters 146:243–258.[CrossRef][Web of Science]
Pearce NJG, Perkins WT, Westgate JA, Gorton MP, Jackson SE, Neal CR, Chenery SP. (1996) A comparison of new and published major and trace element data for NIST SRM 610 and NIST SRM 612 glass reference materials. Geostandards Newsletter 21:115–144.
Prinz M, Keil K, Green A, Reid AM, Bonatti E, Honnorez J. (1976) Ultramafic and mafic dredge samples from the equatorial Mid-Atlantic Ridge and fracture zones. Journal of Geophysical Research 81:4087–4103.
Quick JE. (1981a) Petrology and petrogenesis of the Trinity peridotite, an upper mantle diapir in the eastern Klamath Mountains, northern California. Journal of Geophysical Research 86:11837–11863.[CrossRef]
Quick JE. (1981b) The origin and significance of large, tabular dunite bodies in the Trinity peridotite, northern California. Contributions to Mineralogy and Petrology 78:413–422.
Rampone E, Hofmann AW, Piccardo GB, Vannucci R, Bottazzi P, Ottolini L. (1995) Petrology, mineral and isotope geochemistry of the External Liguride peridotites (Northern Apennines, Italy). Journal of Petrology 36:81–105.
Rampone E, Piccardo GB, Vannucci R, Bottazzi P. (1997) Chemistry and origin of trapped melts in ophiolitic peridotites. Geochimica et Cosmochimica Acta 61:4557–4569.[CrossRef][Web of Science]
Ross K and Elthon D. (1993) Cumulates from strongly depleted mid-ocean-ridge basalt. Nature 365:826–829.[CrossRef]
Salters VJM and Stracke A. (2004) Composition of the depleted mantle. Geochemistry, Geophysics, Geosystems 5: Q05004, doi:10.1029/2003GC000597.
Sano S. (1991) Petrogenesis of the Red Hills Peridotite in the Dun Mountain Ophiolite Belt. Memoirs of Faculty of Education, Ehime University, Series III 12:1–118.
Sano S, Tazaki K, Koide Y, Nagao T, Watanabe T, Kawachi Y. (1997) Geochemistry of dike rocks in Dun Mountain Ophiolite, Nelson, New Zealand. New Zealand Journal of Geology and Geophysics 40:127–136.
Seyler M and Bonatti E. (1997) Regional-scale melt–rock interaction in lherzolitic mantle in the Romanche Fracture Zone (Atlantic Ocean). Earth and Planetary Science Letters 146:273–287.[CrossRef][Web of Science]
Shaw DM. (1970) Trace element fractionation during anatexis. Geochimica et Cosmochimica Acta 34:237–243.[CrossRef][Web of Science]
Shimizu N. (1998) The geochemistry of olivine-hosted melt inclusions in a FAMOUS basalt ALV519-4-1. Physics of the Earth and Planetary Interiors 107:183–201.[CrossRef][Web of Science]
Sinton JM. (1977) Equilibration history of the basal alpine-type peridotite, Red Mountain, New Zealand. Journal of Petrology 18:216–246.
Sinton JM. (1980) Petrology and evolution of the Red Mountain ophiolite complex, New Zealand. American Journal of Science 280-A:296–328.[Web of Science]
Smith PM and Asimow PD. (2005) Adiabat_1ph: a new public front-end to the MELTS, pMELTS, and pHMELTS models. Geochemistry, Geophysics, Geosystems 6: paper number Q02004, doi:10.1029/2004GC000816.
Sobolev AV and Shimizu N. (1992) Superdepleted melts and ocean mantle permeability. Doklady Rossiyskoy Akademii Nauk 326:354–360.
Sobolev AV and Shimizu N. (1993) Ultra-depleted primary melt included in an olivine from the Mid-Atlantic Ridge. Nature 363:151–154.[CrossRef]
Stolper E. (1980) A phase diagram for mid-ocean ridge basalts: preliminary results and implications for petrogenesis. Contributions to Mineralogy and Petrology 74:13–27.[CrossRef]
Suhr G, Hellebrand E, Snow JE, Seck HA, Hofmann AW. (2003) Significance of large, refractory dunite bodies in the upper mantle of the Bay of Islands Ophiolite. Geochemistry, Geophysics, Geosystems 4: doi: 10.1029/2001GC000277.
Sun SS and McDonough WM. (1989) Chemical and isotopic systematics of ocean basalts: implications for mantle composition and processes. In Saunders AD and Norry MJ (Eds.). Magmatism in the Ocean Basins. Geological Society, London, Special Publications 42: pp. 313–345.
Takazawa E, Frey FA, Shimizu N, Obata M, Bodinier JL. (1992) Geochemical evidence for melt migration and reaction in the upper mantle. Nature 359:55–58.[CrossRef]
Takazawa E, Fray FA, Shimizu N, Saal A, Obata M. (1999) Polybaric petrogenesis of mafic layers in the Horoman peridotite complex, Japan. Journal of Petrology 40:1827–1851.[CrossRef][Web of Science]
Tartarotti P, Susini S, Nimis P, Ottolini L. (2002) Melt migration in the upper mantle along the Romanche Fracture Zone (Equitorial Atlantic). Lithos 63:125–149.[CrossRef][Web of Science]
Turnbull IM. (1979) Stratigraphy and sedimentology of the Caples terrane of the Thompson Mountains, northern Southland, New Zealand. New Zealand Journal of Geology and Geophysics 22:555–574.
Vernières J, Godard M, Bodinier JL. (1997) A plate model for the simulation of trace element fractionation during partial melting and magma transport in the Earth's upper mantle. Journal of Geophysical Research 102:24771–24784.[CrossRef]
Walcott RI. (1969) Geology of the Red Hill Complex, Nelson, New Zealand. Transactions of Royal Society of New Zealand, Earth Science 7:57–88.
Walter MJ. (1998) Melting of garnet peridotite and the origin of komatiite and depleted lithosphere. Journal of Petrology 39:29–60.[CrossRef][Web of Science]
Waterhouse JB. (1964) Permian stratigraphy and faunas of New Zealand. New Zealand Geological Survey Bulletin 72:1–101.
Workman RK and Hart SR. (2005) Major and trace element composition of the depleted MORB mantle (DMM). Earth and Planetary Science Letters 231:53–72.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||