| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology 45(6) © Oxford University Press 2004; all rights reserved
Pressure–Temperature Path Recorded in the Yangkou Garnet Peridotite, in Su-Lu Ultrahigh-pressure Metamorphic Belt, Eastern China
1 DEPARTMENT OF GEOLOGY AND MINERALOGY, GRADUATE SCHOOL OF SCIENCE, KYOTO UNIVERSITY, KYOTO 606-8502, JAPAN
2 DEPARTMENT OF EARTH SCIENCES, FACULTY OF SCIENCE, KANAZAWA UNIVERSITY, KANAZAWA 920-1192, JAPAN
RECEIVED OCTOBER 30, 2000; ACCEPTED NOVEMBER 25, 2003
 |
ABSTRACT |
|---|
Chemical variations along with changes in microstructure of the principal constituent minerals make it possible to identify at least four equilibrium stages in the evolution of the Yangkou garnet peridotite in the Su-Lu ultrahigh-pressure metamorphic belt, eastern China: Stage I—a primary garnet lherzolite stage represented by coarse-grained (a few millimeters size) porphyroclastic aluminous pyroxenes + chromian spinel ± garnet; Stage II—an ultrahigh-pressure (UHP) stage defined by fine-grained matrix phases (0·1–0·3 mm size) of garnet + extremely low-Al orthopyroxene + high-Na clinopyroxene + chromite; Stage III—a medium-pressure stage defined by fine-grained mineral aggregates (<0·1–0·2 mm size) mainly composed of aluminous spinel + high-Al orthopyroxene in the matrix; Stage IV—an amphibolite- to greenschist-facies stage defined by poikiloblastic amphibole. Orthopyroxene–clinopyroxene thermometry and an empirical spinel barometer give temperatures of around 800–830°C and pressures of 1·2–2·9 GPa for porphyroclasts of Stage I. Garnet–orthopyroxene, garnet–clinopyroxene and empirical spinel geothermobarometers give relatively uniform P–T conditions for the matrix garnet–orthopyroxene–clinopyroxene–chromite assemblage of Stage II (
730–760°C and 3·6–4·1 GPa). Aluminous spinel–olivine pairs in the aggregates give 650–680°C at 1·0–1·5 GPa for Stage III. The granulation of the studied rocks may have taken place during subduction (Stages I and II) of the host peridotite. The lack of kelyphitic rims around the matrix garnet and the preservation of incompatible assemblages in the matrix of the studied rocks is due to extremely low H2O activity during the early decompression stage. The peak temperature of associated eclogites in the Yangkou UHP unit is 700–800°C at 3·1–4·1 GPa. These observations suggest that the mantle fragment (garnet peridotite) and the crustal fragment (eclogite) in the Yangkou UHP unit both experienced a common UHP metamorphic event and that the exhumation path obtained by these reference points supports nearly isothermal decompression of the unit. The Yangkou UHP unit is considered to have been exhumed as part of a larger crustal mass (c. >10 km in size) along with the surrounding orthogneisses. Such isothermal decompression paths have also been reported from other areas of the Su-Lu belt and the Dabie Mountains. As the orthogneisses volumetrically account for >90% of these areas, buoyancy-controlled uplift probably played an important role in the exhumation of the UHP rocks, at least from the upper mantle to the lower crust. KEY WORDS: China; garnet peridotite; isothermal decompression path; Shandong; Su-Lu UHP belt
|
INTRODUCTION |
|---|
Peridotite massifs (alpine peridotites or orogenic peridotites) are volumetrically minor components in collisional orogenic belts; nevertheless, they have provided a wealth of information on the physico-chemical characteristics of the upper mantle and of orogenic root zones. Medaris (2000)
pointed out that spinel peridotite is predominant among such peridotite massifs, whereas garnet peridotite is common in ultrahigh-pressure (UHP) metamorphic belts. The distribution of peridotite massif in the Shandong Peninsula, eastern China, accords with this view. Spinel peridotite is characteristic of the Sino-Korean craton to the NW of the Yantai–Qingdao–Wulian Fault (YQWF), but garnet peridotite is the predominant ultramafic rock in the Su-Lu UHP belt to the SE of the fault (Fig. 1).
|
Garnet peridotite has been studied by many workers in the Su-Lu belt (e.g. Yang et al., 1993
; Medaris, 2000; Zhang et al., 2000). Medaris (2000) suggested that the garnet peridotite chemically resembles the Mg–Cr type of depleted mantle origin and contains a common UHP assemblage of olivine + low-Al2O3 (0·14–0·28 wt %) orthopyroxene + clinopyroxene + garnet ± phlogopite, giving P–T conditions ranging from 670°C and 4·1 GPa to 925°C and 7·0 GPa.
The Yangkou UHP unit is located near Qingdao City, central Su-Lu belt (Figs 1 and 2), and has been studied previously by many workers (e.g. Hirajima, 1996; Liou & Zhang, 1996; Jahn, 1998). The Yangkou UHP unit is characterized by the occurrence of a UHP metagranitoid preserving an original igneous texture, and by the presence of interstitial coesite in the matrix of eclogite (e.g. Hirajima et al., 1993; Ye et al., 1996), suggesting that fluid activity during the exhumation stage was limited. There is little evidence for post-eclogite stage deformation (e.g. Wallis et al., 1997). The mineralogy of the garnet peridotite in this unit has been reported briefly by Zhang et al. (2000). Our petrological study, however, shows that the Yangkou garnet peridotite preserves multiple, incompatible mineral assemblages, which indicate a nearly isothermal decompression path from the upper mantle to the lower crust. In this paper, we describe the detailed petrography and mineral chemistry of the Yangkou garnet peridotite and discuss its geological significance.
|
|
GEOLOGICAL SETTING |
|---|
The Yangkou UHP meta-igneous unit is mainly composed of coarse- and fine-grained eclogite, UHP metagranitoid, streaky orthogneiss and associated ultramafic rocks, and occurs as a small lens (100 m x 100 m) surrounded by amphibolite-facies regional orthogneiss of the Su-Lu unit (Fig. 2). The coarse-grained eclogite partly preserves a gabbroic texture, and some fine-grained eclogites contain coesite both as an inclusion phase in UHP minerals and as an interstitial matrix phase (e.g. Liou & Zhang, 1996
; Ye et al., 1996). Granitic textures indicate an igneous origin for the metagranitoids and their deformed equivalents in the streaky orthogneiss. The intrusive relationships between the metagranitoid and eclogites, and their common SHRIMP zircon ages (700–800 Ma, Hirajima & Fanning, 1999) imply that the metagranitoid and the eclogite were derived from an igneous unit and have shared a common metamorphic history.
The geological relationship between the mantle-derived ultramafic rocks and the UHP rocks derived from crustal materials in the Yangkou unit must originally have been a faulted contact, but any such relationship between them has been modified by pervasive deformation during the early decompression stage (D2 of Wallis et al., 1997; Fig. 2). We collected more than 20 fresh samples of ultramafic rock from intertidal outcrops and beach boulders. Most of the outcrops and boulders are severely serpentinized, but some of them still preserve multiply equilibrated mineral assemblages, which formed prior to any serpentinization.
|
PETROGRAPHY AND MINERAL CHEMISTRY |
|---|
The studied peridotites are weakly foliated and dark bluish green in hand specimen. Some peridotites contain light-colored, garnet clinopyroxenite layers of a few millimeters to 1 cm thickness, which are almost parallel to the foliation of the host peridotite. The garnet in the garnet clinopyroxenite layers is easily identified as reddish spots, up to a few millimeters in diameter. However, it is hard to identify garnet in the host peridotite with the naked eye, mainly because of the exceptionally fine-grained nature of the matrix-forming phases (0·2–0·3 mm in diameter). The following discussion concentrates on the detailed petrography of the host peridotite.
The garnet peridotite consists mainly of olivine (Ol), garnet (Grt), orthopyroxene (Opx), clinopyroxene (Cpx), amphibole (Amp), chromian spinel (CrSpl) and aluminous spinel (AlSpl). Chlorite (Chl), talc (Tc), tremolite (Tr), serpentine and magnetite are also recognized as retrograde phases. Tremolite layers are well developed between the garnet clinopyroxenite layers and the host peridotite. The main constituent minerals are classified into the following three categories on the basis of their size and texture: (1) porphyroclastic phases; (2) matrix phases; (3) poikiloblastic phases. The characteristic textures and chemical compositions of minerals of each category are described below.
Tables 1–5 show representative chemical compositions of the main constituent minerals, analyzed by scanning electron microscope (Hitachi S-550) with an energy dispersive X-ray (EDX) spectrometer (Kevex Quantum detector) at Kyoto University. We took special care with the lower-energy tail correction of the EDX detector. Details of the instrument and the analytical procedure have been given by Mori & Kanehira (1984) and Hirajima & Banno (1991). To increase the accuracy of the determined Al2O3 contents in Opx, we adopted a 1000 s counting time, four times that of routine analyses. The validity of this method has been shown by crosschecked analyses using a wavelength-dispersive X-ray spectrometer (see Hiramatsu et al., 1995). X-ray element distribution maps were obtained by Kevex Advanced Image software with 128 x 256 pixels and counting times between 1·0 and 1·5 s per point. Figure 3 shows the mineral assemblages in the representative peridotite samples along with the numbers of studied thin sections. The mineral assemblage of the peridotites varies from one sample to another.
|
|
|
|
|
|
Porphyroclastic phases
Porphyroclastic Opx and Cpx occur in some thin sections (Fig. 3). They are commonly oval in shape and are more than 1·0 mm in length (Fig. 4a). Some porphyroclastic grains show wavy extinction whereas fine-grained minerals in the matrix do not. The Opx porphyroclasts are rimmed by very fine-grained Ol, Tc and Tr. This Opx shows a distinct chemical zoning in Al2O3 content. The core is richer in Al2O3 (up to 3 wt %) than the rim (0·8 wt %; see Table 1). An elongated aggregate mainly composed of Ol sub-grains (c. 0·2 mm in width and 1·5 mm in length) was found in a 3 mm long Opx porphyroclast, but its forsterite content is identical to that of olivine in the matrix. Porphyroclastic Cpx contains up to 3·5 wt % Al2O3 (Table 2). It is commonly replaced by tremolitic amphibole at the margin.
|
A few porphyroclastic chromian spinel (CrSpl) grains were observed in Samples N and O, which also contain porphyroclastic Opx and Cpx (Figs 3 and 5). The CrSpl porphyroclasts (about 1 mm in diameter, significantly coarser than the matrix-forming phases) were found in the middle of a Chl–Spl composite pool (about 2 mm in diameter: Figs 4d and 5a). Many platy spinels (0·01–0·05 mm in width and 0·1–0·5 mm in length) are also intercalated with the basal plane of the Chl. The CrSpl porphyroclast and the platy CrSpl are mostly opaque, but the CrSpl porphyroclast is more transparent (brown) in its core. The brown core has an intermediate composition between the spinel and chromite end-members {Cr-number [= 100 x Cr/(Cr + Al + Fe3+)] = 40–45, and Mg-number [= 100 x Mg/(Mg + Fe2+)] = 40–50} and the opaque rim has a composition of approximately Cr-number = 55–60, and Mg-number = 17–25 (Table 3 and Fig. 6). Fe3+-number [= 100 x Fe3+/(Cr + Al + Fe3+)] is low in the brown core (5–12) but higher in the opaque rim (20–26). A magnetite outermost rim is partially developed at the margin of the porphyroclastic CrSpl (Table 3).
|
in 
in
|
A few Grt porphyroclasts, free from kelyphitic rims, were observed in Sample A (Figs 3 and 4b). The largest Grt porphyroclast is about 3 mm in diameter, and is accompanied by fine-grained (0·1–0·2 mm) Grt neoblasts on its rim. The largest Grt porphyroclast shows a slight chemical zoning: the core is slightly richer in Mg (c. prp69alm17grs14) than the rim (c. prp65alm22grs11sps2). The spessartine component shows a bowl-shaped chemical zoning (0·3–0·4 wt % of MnO in the core and 0·7–0·8 wt % at the rim; Fig. 7). The chemical variation of the Grt porphyroclast is mainly controlled by Mg–(Fe + Mn) exchange. Small Grt neoblasts have compositions that are similar to the rim of the Grt porphyroclasts. Although the Grt porphyroclasts are almost free from Cr2O3 (
0·1 wt %), the Grt neoblasts contain a significant amount of Cr2O3 (1·0–2·5 wt %; Table 4).
|
Matrix phases
Ol, Opx, Cpx, Grt and Amp form a fine-grained granoblastic texture in the matrix. Most grains are around 0·2 mm in size (Fig. 4). Most Spl in the matrix is smaller (<0·1 mm) than the other matrix minerals. Ol is the most common phase in the matrix and has a constant Mg-number of around 92–93.
Most of the matrix Opx is surrounded by a reaction corona of Tc and Tr (Fig. 4c), but only a few matrix Opx grains are in direct contact with other matrix phases such as Grt. The outline of the reaction corona shows a straight boundary against neighboring crystals, implying that it represents the original grain boundary of the Opx. The matrix Opx is homogeneous with Mg-number of 92–93 and 0·10–0·45 wt % Al2O3 (Table 1). It is almost free from Ol inclusions but rarely contains tiny inclusions of Grt and CrSpl/chromite. Another type of Opx is rarely observed in the matrix, closely associated with Ol, Cpx, AlSpl and Grt (Fig. 8). This Opx occurs as grains in contact with each other, appearing to form a domain (c. <0·1–0·2 mm in diameter) composed of very small subhedral to anhedral grains (0·05 mm in diameter). Small Amp grains are also developed along grain boundaries in this domain. This texture may suggest that the fine-grained domain is a product of garnet breakdown, although some Grt grains are in contact with AlSpl with a sharp boundary. The anhedral very fine-grained Opx has a composition of Mg-number 91 and contains about 1–2 wt % of Al2O3 (Table 1).
|
Most of the matrix Cpx is subhedral to anhedral. It is in contact with other matrix grains, such as Ol, Grt, Spl, or the reaction corona of Opx. The matrix Cpx shows a wide range of chemical variation (Table 2, Figs 9b and 10): Na2O varies from 0·0 to 2·8 wt % and shows a negative correlation with CaO, which ranges from 21 to 25 wt %. The matrix Cpx is less aluminous (0·0–1·0 wt % Al2O3) than the Cpx porphyroclasts (about 2·5 wt % Al2O3 on average). The rim of the matrix Cpx in contact with AlSpl is distinctly lower in sodic-pyroxene component [XNaCpx = Na/(Na + Ca) = 0·00–0·03] than the core (XNaCpx = 0·10–0·14: Fig. 10). One Cpx inclusion was found in a matrix garnet core. This Cpx inclusion (Table 2) has a composition (XNaCpx = 0·08) midway between that of the porphyroclastic and the matrix Cpx.
|
|
None of the matrix Grt grains have kelyphitic rims. Relatively large Grt grains (
0·3 mm) rarely show chemical zoning. Their compositions are c. prp61–62 alm22–24grs13uvr3 in the core and c. prp66alm22–24grs5–6 uvr5–7 at the rim. The Fe/Mg zoning profile of the matrix Grt is opposite to that of the Grt porphyroclasts. Judging from an X-ray mapping image (Fig. 9a), there is a reverse relationship between Ca–Al and Mg–Cr contents in Grt; namely, the Ca- and Al-rich core is less chromian (1·0 wt % of Cr2O3), whereas the Ca- and Al-poor rim is more chromian (up to 2·5 wt % of Cr2O3). Relatively small Grt grains in the matrix and small Grt inclusions in Amp are homogeneous, and they have a composition similar to the rim of the relatively large grains, i.e. c. prp63–68alm22–24grs5–7uvr4–7. Most of the matrix Amp is generally euhedral to subhedral and forms a granoblastic texture with other matrix phases. It is colorless pargasite. The granoblastic Amp rarely contains tiny Cpx inclusions with moderate sodic-pyroxene component. Other types of the Amp are also observed in the matrix. Some poikiloblastic Amp, up to 1 mm in size, includes other matrix grains, such as Grt, Opx or Ol. Anhedral, very fine-grained Amp is a member of the reaction domain, which contains AlSpl and high-Al Opx (Fig. 8).
Spl grains in the matrix are variable in color. Most of them are brown in the core and are oblique, pale yellow or colorless at the rim. Their Cr-number varies from grain to grain ranging from 5 to 60 (Table 3 and Fig. 6). The Cr-number varies also in single grains, generally decreasing from the core to the rim. The brown core is rich in Cr (Cr-number 50–60), whereas the pale yellow to colorless rim is poor in Cr (Cr-number 5–20). Spinel inclusions in the Grt or other matrix phases occur as tiny idiomorphic or subidiomorphic grains (Fig. 5c). Each grain is homogeneous in Cr-number, ranging from 55 to 68 (Fig. 6). Among them the idiomorphic inclusions in Ol and low-Al Opx have the highest Cr-number, around 63–68 (Fig. 6). Relatively homogeneous AlSpl grains are mainly found in the reaction domain, which contains Ol, Grt and high-Al Opx (Figs 5d and 8). Half of the studied specimens contain both CrSpl and AlSpl in the matrix (Fig. 3).
Poikiloblastic phases
In some Chl-bearing samples (Fig. 3), poikiloblastic Amp and large Chl grains are well developed (Fig. 4d). Chl grains are larger than the common matrix-forming phases and show a slight wavy extinction. The Chl commonly includes both porphyroclastic and platy Spl (Fig. 4d). In the Chl-bearing samples, Grt is recognized only as an inclusion phase, mainly in poikiloblastic Amp and large Chl, but not as a matrix phase. The poikilitic Amp is commonly zoned and is composed of a pargasitic core and an actinolitic hornblende rim (Fig. 4d and Table 5). The core composition is similar to those of the matrix granoblastic Amp.
Other retrograde opaque phases are Fe–Ni sulfides and magnetite. Magnetite occurs as a thin film surrounding the matrix Spl and is more abundant than Fe–Ni sulfides. Very thin carbonate veins are rarely observed. Most of the grain boundaries and olivine cracks are filled with serpentine.
|
EQUILIBRIUM STAGE DETERMINATION |
|---|
 TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY AND MINERAL... EQUILIBRIUM STAGE DETERMINATION CHEMOGRAPHICAL RELATIONSHIP OF...P-T ESTIMATIONDISCUSSIONAPPENDIX: CHEMOGRAPHICAL...REFERENCES |
|---|
We define the following four equilibrium stages on the basis of texture and chemical zoning of the constituent minerals described above.
Stage I
Aluminous porphyroclastic pyroxenes show wavy extinction, but fine-grained matrix pyroxenes do not (Fig. 4a). Therefore, the porphyroclastic pyroxene is interpreted as a relic mineral that formed before the matrix-forming stage. The porphyroclastic pyroxenes are termed Opx1 and Cpx1. The garnet porphyroclasts (Grt1) and the spinel porphyroclasts (Spl1) are likely also to be members of the pre-matrix stage, because of their larger grain size (Figs 4 and 5). Porphyroclastic Ol is not recognized in the studied specimens, but Ol may have been present at this stage, because some Ol is intercalated with the porphyroclastic Opx. Therefore, we tentatively infer an equilibrium assemblage of Opx1 + Cpx1 + Ol + Spl1, which was observed in Samples N and O (Fig. 3), for Stage I. The possible stability relationships between Grt1 and other porphyroclastic phases will be discussed later.
Stages II–III
According to the chemical compositions of the spinel and pyroxenes and the grain size of the reaction domain, we define the following two equilibrium stages in the matrix: (1) an earlier stage defined by Grt + low-Al Opx + high-Na Cpx + Ol, with grain size of 0·2–0·3 mm diameter. The Cr-rich core (Cr-number >50) of the matrix Spl and the chromite inclusions in the matrix phase are considered to have formed at this stage. (2) A later stage defined by an aggregate of aluminous Spl, high-Al Opx and Ol, which developed in a limited reaction domain less than 0·1–0·2 mm in diameter (Fig. 8).
Relatively large Grt grains (0·3 mm) show a chemical zonation from high-Ca and low-Cr cores to low-Ca and high-Cr rims. The rim composition is almost identical to the homogeneous matrix Grt, which is in direct contact with Ol, reaction coronas of the low-Al Opx and high-Na Cpx. Therefore, we consider that the rim of the matrix garnet was equilibrated along with low-Al Opx and high-Na Cpx in the matrix. We do not consider that the CaO-rich Grt core formed during Stage I, mainly for the following reasons: (1) the CaO-rich Grt core includes Cpx that has a distinctly lower Al2O3 content than that of Cpx1; (2) the chemical composition of the CaO-rich core is distinctly higher in CaO and Cr2O3 than the rim of Grt1. Cr contents of Grt coexisting with two pyroxenes, Ol and Spl increase with increasing metamorphic pressure (e.g. Webb & Wood, 1986). The idiomorphic chromite inclusions in low-Al Opx, Ol and Grt are the most chromian among the analyzed spinels in this study. Therefore, we consider that the Cr-rich Grt rim, low-Al (0·10–0·35 wt %) Opx and the inclusion chromite define the highest-pressure stage of this peridotite. The low-Cr Grt core may record the mid-stage of the compression stage of this rock.
The Cpx inclusions in the core of the matrix Grt and the matrix Cpx with high sodic-pyroxene component have similar Al2O3 contents (1·0–1·5 wt %). Therefore, they are named Cpx2. Most of the matrix Opx grains surrounded by Tc and Tr (Fig. 4c) are homogeneous, and are characterized by extremely low Al2O3. Some Opx grains in UHP garnet peridotites in the Su-Lu belt show a bowl-shaped Al2O3 zoning (e.g. Yang et al., 1993; Hiramatsu et al., 1995) and their core is characterized by extremely low Al2O3 contents (0·15–0·25 wt %) similar to those studied here. The matrix Opx with low Al2O3 (Opx2) in this study may predate the very fine-grained Opx with moderate Al2O3 (Opx3) in view of the observed Al2O3 zoning pattern of the Opx in other garnet peridotites of the Su-Lu belt (e.g. Yang et al., 1993; Hiramatsu et al., 1995). Some Opx2 grains contain chromite inclusions with Cr-number 70. The brown Spl core with Cr-number around 50–60 (Spl2, as shown in Fig. 6) predates the colorless Spl rim with Cr-number around 5–10 (Spl3) as judged from the spinel zoning pattern. Therefore, low-Al Opx2, high-Na Cpx2, high-Cr Grt2 and high-Cr Spl2 along with matrix Ol form Stage II.
Many pargasitic Amp grains form a granoblastic texture along with the other matrix phases. Some granoblastic Amp grains contain tiny Cpx inclusions, of which the composition is similar to that of Cpx2. Therefore, the granoblastic Amp was also formed in Stage II or later than that. In this paper, we tentatively interpret pargasitic Amp as Amp3.
Half of the studied peridotite samples contain transparent aluminous Spl3 with a smaller grain size (up to 0·1 mm long) than the other matrix Cr-rich spinels (Spl2). The aluminous Spl3 commonly accompanies high-Al Opx3, Ol and pargasitic Amp3. Some Grt2 grains are in contact with Spl3 with a sharp boundary but in other cases Grt2 seems to be decomposed to form Spl3 and high-Al Opx3. We consider that the decomposition of Grt2 may supply Al to form Spl3 in a limited reaction domain at Stage III. Cpx characterized by less sodic-pyroxene component, which is in contact with Spl3 (Figs 8 and 10), may form during Stage III, but its composition is variable from grain to grain (Table 2). Neoblastic Na2O-free Cpx3 is rarely found, but Cpx3 commonly forms the rim of Cpx2, probably controlled by volume diffusion in precursor Cpx2.
Stage IV
The poikilitic Amp and the large Chl are closely associated with each other. Chl never occurs in the matrix when Grt2 occurs as a matrix phase. On the contrary, in the Chl-bearing sample, Grt2 grains occur only as inclusion phases in Amp but do not exist in the matrix. This suggests that the Chl and Amp are of a later stage than Grt2. The poikilitic Amp composition (pargasitic core and actinolitic hornblende rim) suggests they were formed under amphibolite- to greenschist-facies conditions. We define this later equilibrium stage as Stage IV and tentatively name the poikilitic Amp and the large Chl as Amp4 and Chl4. Tc and tremolite replacing the rim of Opx2, magnetite outermost rim of the matrix Spl and serpentine may also belong to this stage.
|
CHEMOGRAPHICAL RELATIONSHIP OF EACH EQUILIBRIUM STAGE |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY AND MINERAL...EQUILIBRIUM STAGE DETERMINATIONCHEMOGRAPHICAL RELATIONSHIP OF...P-T ESTIMATIONDISCUSSIONAPPENDIX: CHEMOGRAPHICAL...REFERENCES |
|---|
To visualize the minor chemical variations in Opx, Cpx and other relevant phases, the mineral compositions of each stage were plotted in an Sp (spinel, A2S–0·5), Di (diopside, C1S1·5) and En (enstatite, S1) triangular diagram projected from olivine and the jadeite component of clinopyroxene (Fig. 11), where A = [Al2O3] + [Cr2O3] + [Fe2O3] – [Na2O], C = [CaO], and S = [SiO2] –
[MgO + FeO] – 2[Na2O]. Detailed projection formulation is shown in the Appendix. Figure 12 shows schematic plots of the mineral assemblages for each stage in a tetrahedron of Al = [Al2O3] + [Fe2O3] – [Na2O], Cr = [Cr2O3], C = [CaO], and S = [SiO2] – [MgO + FeO] – 2[Na2O].
|
|
During Stages I and II, Opx + Cpx + Ol + CrSpl ± Grt are stable. The compositional change of these phases from Stage I to Stage II should be controlled by a continuous reaction, which mainly supplies Cr from Spl1 to Spl2 and Grt2, and Al from Spl, Opx1 and Opx2 to Grt2:
 |
During Stage III, aluminous Spl3 and pargasitic Amp3 were newly formed. Aluminous Opx3 is always associated with Spl3. To produce these phases, the reactant should supply Al and Ca to the product. One of the plausible reactions is the decomposition of Grt2 enhanced by infiltrated H2O:
 | (2) |
):
 | (3) |
|
P–T ESTIMATION |
|---|
Stage I
The heterogeneous composition of porphyroclastic phases, the possible chemical modification in later stages and the lack of a direct association of Grt1 and Opx1/Cpx1 make it difficult to evaluate the P–T conditions of Stage I. We tentatively choose the core composition of the Grt1 with the highest Mg-number, the most calcic Opx1, the least calcic Cpx1 and the core composition of the Spl1 to estimate the P–T conditions of Stage I (see Tables 1–4). For these pairs, the Grt–Opx geobarometer of Harley (1984b)
gives c. 1·2 GPa at 800°C (GO-P in Fig. 13). The Opx–Cpx geothermometer (Bertrand & Mercier, 1985: 2Px-T), Grt–Cpx geothermometer (Powell, 1985: GC-T) and Grt–Opx geothermometer (Harley, 1984a: GO-T) give c. 800–830°C at 1·2 GPa (Fig. 13). The pressure obtained using Opx–Grt geothermobarometry (1·2 GPa) is slightly lower than the pressure of typical garnet lherzolite facies; a minimum of 1·5 GPa at 800°C in the CMAS system (Fig. 13). The empirical Spl barometer (O'Neill, 1981) gives 2·8–2·9 GPa at 800°C assuming (Cr + Fe3+)/R3+ in Spl1 as 0·50–0·55 and Mg-number of Ol as 92. This discrepancy in the pressure estimation is probably caused by the difficulty of defining both the equilibrium composition of relevant phases and the equilibrium assemblage at Stage I.
|
Stage II
The Grt–Opx geothermobarometer (Harley, 1984a
, 1984b) gives c. 730–760°C and 3·6–4·1 GPa for Stage II. The Grt–Cpx geothermometer (Powell, 1985), after estimating the Fe3+/Fe2+ ratio by the approximation of Fe3+ = Na – Al – Cr3+, gives a similar temperature (Fig. 13). The empirical Spl barometer of O'Neill (1981) gives 3·7 GPa at 750°C assuming (Cr + Fe3+)/R3+ in Spl2 as 0·15 and Mg-number of Ol as 92. The pyroxene thermometer of Bertrand & Mercier (1985) gives a lower temperature (c. 630°C at 4 GPa, Fig. 13). We do not attach importance to this last result, because the pyroxene thermometer is considered to be less reliable at such low temperatures. Therefore, we adopt the P–T conditions of 745 ± 15°C and 3·8 ± 0·3 GPa as the equilibrated conditions of Stage II.
Stage III
As Opx3 and Cpx3 vary in their compositions from grain to grain, we were unable to apply the two-pyroxene thermometer for evaluation of the P–T conditions of Stage III. The rim of Spl3, however, has a relatively constant composition around Mg-number 80 and Cr-number 5. Therefore, we used the Ol–Spl thermometer for temperature estimation for Stage III, i.e. 650–680°C by Fabriès (1979) and 590–620°C at 1·5 GPa by O'Neill & Wall (1987). Some workers (e.g. Melcher et al., 1997) have pointed out that the Ol–Spl thermometer of O'Neill & Wall (1987) gives a lower temperature (80–100°C) than other thermometers. Furthermore, it is considered that the Ol–Spl thermometer is easily affected by diffusion during cooling after the formation of the primary assemblage. Therefore, we assume 650–680°C or a slightly higher value as the Stage III temperature. The empirical Spl barometer of O'Neill (1981) may give the maximum pressure of Stage III, around 1·5 GPa at 700°C, assuming that Spl3 with Cr-number 5 was equilibrated with Opx3, Cpx3 and Ol. As no plagioclase occurs in the studied peridotite, the pressure conditions of Stage III should be around 1·0 GPa or higher. Our best-fit conditions for Stage III, therefore, are 650–680°C and 1·0–1·5 GPa.
|
DISCUSSION |
|---|
Why do incompatible assemblages exist in the matrix?
The Yangkou garnet peridotites preserve multi-stage equilibrium assemblages in the matrix, similar to the Caledonian peridotites (e.g. Carswell, 1986
), but have some distinctive features in comparison with other UHP garnet peridotites worldwide: (1) there are no kelyphitic rims around garnet, which are commonly found in Alpine-type garnet peridotites (e.g. Obata, 1980); (2) the very fine grain size (0·2–0·3 mm) of the main constituent minerals (e.g. 2–3 mm in Chijiadian garnet peridotite; Hiramatsu et al., 1995).
What is the main reason for the multi-stage equilibrium assemblages remaining in the matrix? Low H2O activity during the exhumation stage may have played an important role. This idea is supported by the lack of kelyphitic rims around garnet. Schmidt & Poli (1998) suggested that the Chl + Opx + Cpx + Ol assemblage, instead of Grt + Opx + Cpx + Ol, is stable in hydrated peridotite under the P–T conditions of Stage II for the Yangkou garnet peridotite. Therefore, we conclude that the Yangkou peridotite was dehydrated or almost dry during Stage II.
Synthetic studies of equilibrium between antigorite, enstatite, forsterite, Tc and H2O (e.g. Mysen et al., 1998, fig. 12) suggest that the assemblage forsterite + Tc is stable instead of enstatite at the P–T conditions of Stage III, and that the assemblage forsterite + Tc decomposes to antigorite at less than 620°C at 1 GPa, if sufficient H2O is supplied to the reaction system. At Yangkou Opx2 is always rimmed by a Tc–Tr corona, but not completely replaced by them. The very fine-grained Opx3 is not decomposed to Tc, but is in direct contact with the product aluminous Spl3 and the reactant Grt2 (Fig. 8). These textures suggest that sufficient H2O was not supplied to the matrix of the garnet peridotite during Stage III, and hence apparently incompatible assemblages can occur together in the matrix, probably through the development of local chemical equilibrium controlled by H2O availability. Such dry conditions are not unlikely in the Yangkou UHP unit, because low H2O activity during the exhumation stage is also supported by the existence of interstitial coesite in the eclogite (e.g. Hacker & Peacock, 1995; Liou & Zhang, 1996; Ye et al., 1996). The preservation of the multi-stage equilibrium assemblages is, therefore, a reflection of slow kinetics and limited fluid access.
The degree of local chemical equilibrium is probably related to the deformation of the rocks, as inferred from the granulation of the peridotite. Granulation of the Cpx has been proposed for the Nonsberg lherzolite, northern Italy, by Morten & Obata (1983) and Obata & Morten (1987), and for some peridotites in the Su-Lu belt by Yang et al. (1993) and Hiramatsu & Hirajima (1995). The granulation of the primary phases is also observed in the Yangkou peridotite (Fig. 4a and b). Such granulation of the peridotite may reflect shear deformation during subduction and/or early exhumation stages.
According to Poirier (1985), the recrystallized grain size depends only on the applied stress, and decreases as the stress increases. Accordingly, small grain size should reflect the last high-temperature, major stress episode. For the Yangkou peridotite, the Opx porphyroclasts show distinct chemical zoning of Al2O3, decreasing from the core (up to 3 wt %) to the rim (0·8 wt %), approaching the Al2O3 content of Opx2, and the chemical composition of the rim of the porphyroclastic Grt is similar to the surrounding neoblasts (Grt2). These observations suggest that granulation of the Yangkou peridotite took place mainly during Stage II (Fig. 13).
Assessment of the estimated PT conditions
To evaluate the equilibrium conditions between the lherzolite minerals, the distribution of minor elements such as Al, Ca, Ti and Cr can provide useful information. Mori & Green (1978) pointed out that the contents of TiO2 and Cr2O3 in the minerals of garnet lherzolites decrease in the order Grt > Cpx > Opx and that the Cr concentration ratio on a six-oxygen basis is roughly 0·5 for Cpx/Grt and 0·2 for Opx/Grt. Cr concentration ratios of the Stage II assemblage (0·50 for Cpx2/Grt2 and 0·12 for Opx2/Grt2) are comparable with the values suggested above. However, the Cr concentration of Grt1 is significantly lower than those of Opx1 and Cpx1 (Tables 1, 2 and 4). These data suggest that the assemblage for Stage II can be considered as equilibrated on the basis of the Cr distribution, but that this cannot be assumed for minerals of Stage I. Therefore, P–T conditions estimated using the element distribution between Grt and pyroxenes for Stage I are for reference only, as equilibration cannot be assumed.
The Al2O3 and CaO contents of porphyroclastic Opx1 are higher in Opx1 cores, and decrease rimwards (Fig. 11 and Table 1). In contrast, Opx2 is characterized by extremely low Al2O3 contents (0·1 wt %) comparable with Opx in other UHP garnet peridotites in the Su-Lu belt (e.g. Ye & Xu, 1992; Yang et al., 1993; Zhang et al., 1994, 2000; Hiramatsu et al., 1995). The Cr contents of inclusion Spl2 and cores of the matrix Spl2 are significantly higher than the Cr contents of the cores of porphyroclastic Spl1. Taken together, these Opx1 and Spl1 compositional features suggest that the P–T conditions of Stage I were higher T and/or lower P than those of Stage II.
Mori (1977) suggested that the Al2O3 content of Opx in spinel lherzolite could not achieve equilibrium in laboratory conditions at less than 1000°C and warned against assuming complete redistribution of Al in ultramafic rocks. On the other hand, Mori & Green (1978) pointed out that Opx approaches equilibrium compositions most quickly among the garnet lherzolite minerals under experimental conditions of 1100–1450°C and 2·7–4·4 GPa. Yang et al. (1993) and Hiramatsu et al. (1995) showed that the Opx in their garnet lherzolite preserves a 0·1–0·4 mm diameter homogeneous core with low Al2O3 (0·15 wt %) and a slightly more Al2O3-rich rim (0·2–0·5 wt %). They proposed that the core of the Opx once equilibrated under UHP conditions. As the majority of the matrix Opx in this study (Opx2) is homogeneous and characterized by similar low Al2O3 contents, we consider that Opx2 was once equilibrated with the higher-Cr and lower-Ca Grt2 under UHP conditions. Mørk (1985) reported heterogeneous Opx coronas showing Al2O3 zoning from 0·1 to 1·3 wt % from metagabbro recrystallized in a quartz eclogite facies condition in the Western Gneiss Region of Norway. She concluded that the low Al2O3 content of Opx reflects the slow diffusion rate of Al in Opx. The relevant metagabbro was not strongly deformed during high-P metamorphism and the high-P metamorphic minerals form domains of local equilibrium. However, this is not the case for our samples from the Yangkou UHP unit, because the host peridotite suffered pervasive deformation and granulation during Stage II.
Ye et al. (2000) estimated a maximum pressure for the Yangkou eclogite as >7 GPa using a ‘subsilicic calculated original garnet composition’ derived from the compositions of the host garnet and its inclusion phases. We have not observed similar inclusion-rich porphyroclastic garnet in either eclogite or ultramafic rocks at Yangkou and so are not able to confirm this extreme pressure.
Both aluminous Spl3 and Amp3, which are thought to be produced through the decomposition of Grt, are observed as small grains in the matrix instead of kelyphitic rims (Fig. 8). Opx3 occurs only in the reaction domain closely associated with Spl3. Stage III is clearly distinguishable on the basis of texture, even though geothermobarometry results are widely scattered.
Significance of isothermal decompression path from Stage II to Stage III
A precisely determined P–T path for the UHP rocks can provide an important constraint on their mechanism of exhumation. Petrological evidence has established the following two types of exhumation path so far:
- a nearly isothermal decompression path from the upper mantle to the lower crust; for example, in the Bohemian Massif and northeastern part of the Su-Lu belt (e.g. Schmädicke et al., 1992; Banno et al., 2000; Nakamura & Hirajima, 2000);
- decompression associated with a significant cooling; for example, in the Dora Maira Massif, Western Alps (e.g. Chopin et al., 1991; Schertl et al., 1991; Hirajima & Compagnoni, 1993).
This difference should arise from the shapes, volumes, and exhumation rates of the ascending UHP bodies. These factors may control the decompression P–T path (Hacker & Peacock, 1995). For the type (2) path of the Dora Maira Massif, the ascending UHP body may have been cooled by newly subducted materials above or below. Radiometric studies by Gebauer et al. (1997) showed a very rapid exhumation rate (20–24 mm/year) for the Dora Maira UHP rocks, based on SHRIMP dating. The thinness of the thrust nappe of the UHP unit in the Dora Maira Massif (1–0·5 km) may have contributed to its significant cooling during rapid exhumation.
To produce an isothermal decompression path, the ascending UHP body requires either thermal isolation from (or even heating by) the wall rocks. To create such a situation, either rapid exhumation or large volumes of the ascending hot materials are necessary. Nakamura & Hirajima (2000) estimated that the size of the exhumed mass required to prevent the loss of heat to or addition of heat from the wall rock should be >10 km on the basis of petrological studies in the northeastern part of the Su-Lu belt and with an assumed exhumation rate of 20 mm/year. This result requires that the country rock gneiss or at least materials surrounding the peridotite and/or eclogite at UHP depths should have been exhumed together with the small peridotite and/or eclogite bodies (commonly <1 km).
The Grt–Opx geothermobarometer (Harley, 1984a, 1984b), the empirical Spl barometer (O'Neill, 1981) and the Spl–Ol thermometer (Fabriès, 1979; O'Neill & Wall, 1987) suggest that the Yangkou garnet peridotite experienced an isothermal decompression (ITD) path or a decompression with only a slight temperature drop from Stage II (730–760°C at 3·6–4·1 GPa) to Stage III (>650–680°C at 1·0–1·5 GPa) (Fig. 13). Temperature conditions of Stage II for the Yangkou garnet peridotite are similar to those of the UHP stage of the associated coesite eclogites (Fig. 13), i.e. 700–800°C at 3·1–4·1 GPa using the Grt–Cpx thermometer of Powell (1985) or Krogh (1988), a Grt–Cpx–phengite barometry of Waters & Martin (1996), and a Grt–Cpx–kyanite–coesite (or quartz) barometry of Nakamura & Banno (1997) (see Hirajima & Nakamura, 2003, fig. 12). Therefore, the Yangkou garnet peridotite and the other UHP members in the Yangkou UHP unit could have shared the same exhumation path. The size of the Yangkou UHP unit (c. 100 m x 100 m, Fig. 2), however, is too small to prevent the loss or addition of heat by conduction to or from the wall rocks during exhumation. The acidic orthogneiss is the predominant lithotype around the Yangkou area, as well as in the other areas of the Su-Lu UHP belt. Therefore, the Yangkou UHP unit and surrounding orthogneiss are considered to have been exhumed together as a large mass.
ITD paths have been reported from the Rongcheng area (Nakamura & Hirajima, 2000) and the Weihai area (Banno et al., 2000) in the Su-Lu UHP metamorphic belt (Fig. 1) and from the Dabie Mountains, the western extension of the Su-Lu belt (Castelli et al., 1998; Compagnoni et al., 2001). The present study establishes an ITD path in the middle part of the Su-Lu belt, suggesting that such a path is characteristic for exhumation of the Dabie–Su-Lu belt, which is mainly composed of acidic country gneisses.
ITD paths have been reported from other UHP belts mainly dominated by acidic country rocks (e.g. the Bohemian Massif, Schmädicke et al., 1992; Kotková et al., 1997). The volume of UHP eclogite and garnet peridotite is commonly <10% in these areas. Therefore, the average density of the inferred exhumed mass (>90 vol. % acidic rocks and minor basic and ultrabasic UHP rocks) is less than that of mantle peridotite (e.g. Irifune, 1994). This geological setting supports a buoyancy uplift model from the mantle to the lower crust as proposed by Ernst et al. (1996).
|
APPENDIX: CHEMOGRAPHICAL TREATMENT IN ACS COMPONENT SYSTEM |
|---|
The mineral assemblages in the studied peridotites should be discussed in the K–Na–Ca–Mg–Fe2+–Al–Cr–Fe3+–Si–O–H system. To see the compositional change of the main phases in each equilibrium stage, the relevant phases are plotted in an ACS diagram (Fig. 11). To decrease the number of components in the system, components A, C and S stand mainly for Al3+, Ca2+ and Si4+, respectively, according to following assumptions:
- Al3+, Cr3+ and Fe3+ are regarded as one component;
- Mg2+ and Fe2+ are regarded as one component;
- olivine (
) is regarded as an excess phase;
- H2O is regarded as a perfectly mobile component;
- Na2O is mainly contained in Cpx and amphibole; Na2O is omitted as Jd (NaR3+Si2O6) component from the system;
- the other minor components are negligible.
 |
 | (A1) |
|
ACKNOWLEDGEMENTS |
|---|
The studied samples were collected as part of a co-operative project between Kyoto University and the Academia Sinica supported by the Japan Society for the Promotion of Science and the Chinese National Natural Science Foundation, and we would like to express thanks to the members of the project. We are grateful to G. Medaris, P. Robinson, S. Harley, M. Obata, T. Mori, D. Nakamura and an anonymous reviewer for their critical and constructive comments, and encouragement.
|
FOOTNOTES |
|---|
Corresponding author. Fax: 81-(0)75-753-4189. E-mail: hirajima{at}kueps.kyoto-u.ac.jp
* Present address: IBM Systems Engineering, Japan. 
|
REFERENCES |
|---|
Banno, S., Enami, M., Hirajima, T., Ishiwatari, A. & Wang, Q. C. (2000). Decompression P–T path of coesite eclogite to granulite from Weihai, eastern China. Lithos 52, 97–108.[CrossRef][Web of Science]
Bertrand, P. & Mercier, J. C. C. (1985). The mutual solubility of coexisting ortho- and clinopyroxene: toward an absolute geothermometer for the natural system? Earth and Planetary Science Letters 76, 109–122.[CrossRef][Web of Science]
Carswell, D. A. (1986). The metamorphic evolution of Mg–Cr type Norwegian garnet peridotites. Lithos 19, 279–297.[CrossRef][Web of Science]
Castelli, D., Rolfo, F. & Compagnoni, R. (1998). Metamorphic veins with kyanite, zoisite and quartz in the Zhu-Jia-Chong eclogite, Dabie Mountains, China. The Island Arc 7, 159–173.
Chopin, C., Henry, C. & Michard, A. (1991). Geology and petrology of the coesite-bearing terrain, Dora Maira massif, Western Alps. European Journal of Mineralogy 3, 263–291.[Web of Science]
Compagnoni, R., Rolfo, F., Wu, W. & Xu, S. (2001). A coherent lithostratigraphic unit in the coesite-eclogite complex of Dabie Shan, China: geologic and petrologic evidence. In: Abstracts of Sixth International Eclogite Conference, Niihama, Japan, pp. 24–25.
Ernst, W. G., Maruyama, S. & Wallis, S. R. (1996). Buoyancy and rapid exhumation of thin slabs of ultrahigh-pressure metamorphosed continental crust. Proceedings of the National Academy of Sciences of the USA 94, 9532–9537.
Fabriès, J. (1979). Spinel–olivine geothermometry in peridotites from ultramafic complexes. Contributions to Mineralogy and Petrology 69, 329–336.[CrossRef][Web of Science]
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.[CrossRef][Web of Science]
Gebauer, D., Schertl, H. P., Brix, M. & Schreyer, W. (1997). 35 Ma old ultrahigh-pressure metamorphism and evidence for very rapid exhumation in the Dora Maira Massif, Western Alps. Lithos 41, 5–24.[CrossRef][Web of Science]
Hacker, B. R. & Peacock, S. M. (1995). Creation, preservation, and exhumation of UHPM rocks. In: Coleman, R. G. & Wang, X. (eds) Ultra High Pressure Metamorphism. Cambridge: Cambridge University Press, pp. 159–181.
Harley, S. L. (1984a). An experimental study of the partitioning of Fe and Mg between garnet and orthopyroxene. Contributions to Mineralogy and Petrology 86, 359–373.[CrossRef][Web of Science]
Harley, S. L. (1984b). The solubility of alumina in orthopyroxene coexisting with garnet in FeO–MgO–Al2O3–SiO2 and CaO–FeO–MgO–Al2O3–SiO2. Journal of Petrology 25, 665–696.
Hirajima, T. (1996). Effect of jadeite-content on the garnet–clinopyroxene geothermometer for an ultrahigh-pressure eclogite complex. Proceedings of the Japan Academy, Series B 72, 249–254.
Hirajima, T. & Banno, S. (1991). Electron-microprobe analysis of rock forming minerals with Kevex-delta IV (Quantum detector). Hitachi Scientific Instrument News 34, 1118–1134 (J).
Hirajima, T. & Compagnoni, R. (1993). Petrology of a jadeite–quartz/coesite–almandine–phengite fels with retrograde ferro-nybøite from the Dora Maira Massif, Western Alps. European Journal of Mineralogy 5, 943–955.
Hirajima, T. & Fanning, M. (1999). U–Pb age of the Yangkou UHP metamorphic complex in the Su-Lu belt, China. In: Abstracts of Japanese Earth and Planetary Science Joint Meeting, Ge-004.
Hirajima, T. & Nakamura, D. (2003). The Dabie Shan–Sulu orogen. In: Carswell, D. A. & Compagnoni, R. (eds) Ultrahigh Pressure Metamorphism. European Mineralogical Union Notes in Mineralogy 5, 105–144.
Hirajima, T., Wallis, S. R., Zhai, M. & Ye, K. (1993). Eclogitized metagranitoid from the Su-Lu ultra-high pressure (UHP) province, eastern China. Proceedings of the Japan Academy, Series B 69, 249–254.
Hiramatsu, N. & Hirajima, T. (1995). Petrology of Hujialin garnet clinopyroxenite in the Su-Lu ultrahigh-pressure province, eastern China. The Island Arc 4, 310–323.[CrossRef]
Hiramatsu, N., Banno, S., Hirajima, T. & Cong, B. (1995). Ultrahigh-pressure garnet peridotite from Chijiadian, Rongcheng County, in the Su-Lu region of eastern China. The Island Arc 4, 324–333.[CrossRef]
Irifune, T. (1994). Phase transformation in pyrolite and subducted crust compositions down to a depth of 800 km in the lower mantle. Mineralogical Magazine 58, 444–445.
Jahn, B.-M. (1998). Geochemical and isotopic characteristics of UHP eclogites and ultramafic rocks of the Dabie orogen. Implications for continental subduction and collision tectonics. In: Hacker, B. & Liou, J. G. (eds) When Continents Collide. Geodynamics and Geochemistry of Ultrahigh-Pressure Rocks. Dordrecht: Kluwer, pp. 203–239.
Kotková, J. Harley, S. L. & Fisera, M. (1997). A vestige of very high-pressure (ca. 28 kbar) metamorphism in the Variscan Bohemian Massif, Czech Republic. European Journal of Mineralogy 9, 1017–1033.
Krogh, E. J. (1988). The garnet–clinopyroxene Fe–Mg geothermometer—a reinterpretation of existing experimental data. Contributions to Mineralogy and Petrology 99, 44–48.[CrossRef][Web of Science]
Liou, J. G. & Zhang, R. Y. (1996). Occurrences of intergranular coesite in ultrahigh-P rocks from the Sulu region, eastern China: implications for lack of fluid during exhumation. American Mineralogist 81, 1217–1221.[Abstract]
Medaris, L. G., Jr (2000). Garnet peridotites in Eurasian high-pressure and ultrahigh-pressure terranes: a discovery of origins and thermal histories. In: Ernst, W. G. & Liou, J. G. (eds) UHP Metamorphism and Geodynamics in Collision Type Orogenic Belts. Boulder, CO: Geological Society of America, pp. 57–73.
Melcher, F., Grum, A., Simon, G., Thalhammer, T. V. & Stumpfl, E. F. (1997). Petrogenesis of the ophiolitic giant chromite deposits of Kempirasi, Kazakhstan: a study of solid and fluid inclusions in chromite. Journal of Petrology 38, 1419–1458.[CrossRef][Web of Science]
Mori, T. (1977). Geothermometry of spinel peridotites. Contributions to Mineralogy and Petrology 59, 261–279.[CrossRef][Web of Science]
Mori, T. & Green, D. H. (1978). Laboratory duplication of phase equilibria observed in natural garnet lherzolites. Journal of Geology 86, 83–97.[Web of Science]
Mori, T. & Kanehira, K. (1984). X-ray energy spectrometry for electron-probe analysis. Journal of Geological Society of Japan 90, 271–285.
Mørk, M. B. E. (1985). A gabbro to eclogite transition on Flemsøy, Sunnmøre, western Norway. Chemical Geology 50, 283–310.[CrossRef][Web of Science]
Morten, L. & Obata, M. (1983). Possible high-temperature origin of pyroxenite lenses within garnet peridotite, northern Italy. Bulletin de Minéralogie 106, 775–780.
Mysen, B. O., Ulmer, P., Konzett, J. & Schmidt, M. W. (1998). The upper mantle near convergent plate boundaries. In: Hemley, R. J. (ed.) Mineralogical Society of America, Reviews in Mineralogy 37, 97–138.
Nakamura, D. & Banno, S. (1997). Thermodynamic modelling of sodic pyroxene solid-solution and its application in a garnet–omphacite–kyanite–coesite geothermobarometer to UHP metamorphic rocks. Contributions to Mineralogy and Petrology 130, 93–102.[CrossRef][Web of Science]
Nakamura, D. & Hirajima, T. (2000). Granulite-facies overprint on UHP rocks in the northeastern part of the Su-Lu region, eastern China. Journal of Petrology 41, 563–582.
Obata, M. (1980). The Ronda peridotite: garnet-, spinel-, and plagioclase peridotite facies and the P–T trajectories of a high-temperature mantle intrusion. Journal of Petrology 21, 533–572.
Obata, M. & Morten, L. (1987). Transformation of spinel lherzolite to garnet lherzolite in ultramafic lenses of the Austridic crystalline complex, northern Italy. Journal of Petrology 28, 599–623.
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.[CrossRef][Web of Science]
O'Neill, H. St. C. & Wall, V. J. (1987). The olivine–orthopyroxene–spinel oxygen geobarometer, the nickel precipitation curve, and the oxygen fugacity of Earth's upper mantle. Journal of Petrology 28, 1169–1191.
Poirier, J. P. (1985). Creep of Crystals: High-temperature Deformation Processes in Metals, Ceramics and Minerals. Cambridge: Cambridge University Press.
Powell, R. (1985). Regression diagnostics and robust regression in geothermometer/geobarometer calibration: the garnet–clinopyroxene geothermometer revised. Journal of Metamorphic Geology 3, 231–243.[Web of Science]
Schertl, H. P., Schreyer, W. & Chopin, C. (1991). The pyrope–coesite rocks and their country rocks at Parigi, Dora Maira Massif, Western Alps: detailed petrography, mineral chemistry and PT-path. Contributions to Mineralogy and Petrology 108, 1–21.[CrossRef][Web of Science]
Schmädicke, E. (2000). Phase relations in peridotitic and pyroxenitic rocks in the model systems CMASH and NCMASH. Journal of Petrology 41, 69–86.
Schmädicke, E., Okrusch, M. & Schmidt, W. (1992). Eclogite-facies rocks in the Saxonian Erzgebirge, Germany: high pressure metamorphism under contrasting P–T conditions. Contributions to Mineralogy and Petrology 110, 226–241.[CrossRef][Web of Science]
Schmidt, M. W. & Poli, S. (1998). What causes the position of the volcanic front? Experimentally based water budget for dehydrating slabs and consequences for arc magma generation. Earth and Planetary Science Letters 163, 367–378.
Wallis, S. R., Ishiwatari, A., Hirajima, T., Ye, K., Guo, J., Nakamura, D., Kato, T., Zhai, M., Enami, M., Cong, B. & Banno, S. (1997). Occurrence and field relationships of ultrahigh-pressure metagranitoid and coesite eclogite in the Su-Lu terrane, eastern China. Journal of the Geological Society, London 153, 45–54.
Waters, D. J. & Martin, H. N. (1996). The garnet–cpx–phengite barometer. http://www.earth.ox.ac.uk/~davewa/research/ecbarcal.html.
Webb, S. A. C. & Wood, B. J. (1986). Spinel–pyroxene–garnet relationships and their dependence on Cr/Al ratio. Contributions to Mineralogy and Petrology 92, 471–480.[CrossRef][Web of Science]
Yang, J.-J., Godard, G., Kienast, J. R., Lu, Y. & Sun, J. (1993). Ultrahigh-pressure (60 kbar) magnesite-bearing garnet peridotites from northeastern Jiangsu, China. Journal of Geology 101, 541–554.[Web of Science]
Ye, K. & Xu, P. (1992). Petrogenesis of eclogites and related peridotitic rocks in Datuan area of Rongcheng county, Shandong province. Acta Petrologica Sinica 8, 27–39 (in Chinese with English abstract).
Ye, K., Hirajima, T., Ishiwatari, A., Guo, J. & Zai, M. (1996). Finding of an intergranular coesite from Yangkou eclogite, Qingdao, and its significance. Chinese Science Bulletin 41, 1407–1408 (in Chinese).
Ye, K., Cong, B. & Ye, D. (2000). The possible subduction of continental material to depths greater than 200 km. Nature 407, 734–736.[CrossRef][Medline]
Zhang, R. Y., Liou, J. G. & Cong, B. (1994). Petrogenesis of garnet-bearing ultramafic rocks and associated eclogites in the Su-Lu ultrahigh-P metamorphic terrane, eastern China. Journal of Metamorphic Geology 12, 169–186.[Web of Science]
Zhang, R. Y., Liou, J. G., Yang, J. S. & Yui, T.-F. (2000). Petrochemical constraints for dual origin of garnet peridotites from the Dabie–Sulu UHP terrane, eastern–central China. Journal of Metamorphic Geology 18, 149–166.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Ikuta, N. Kawame, S. Banno, T. Hirajima, K. Ito, J. F. Rakovan, R. T. Downs, and O. Tamada First in situ X-ray identification of coesite and retrograde quartz on a glass thin section of an ultrahigh-pressure metamorphic rock and their crystal structure details American Mineralogist, January 1, 2007; 92(1): 57 - 63. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Y. Zhang, J. G. Liou, J.-P. Zheng, W. L. Griffin, T.-F. Yui, and S. Y. O'Reilly Petrogenesis of the Yangkou layered garnet-peridotite complex, Sulu UHP terrane, China American Mineralogist, May 1, 2005; 90(5-6): 801 - 813. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||