2. Geology & Geochemistry Group, Department of Earth Sciences, VU Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands;
3. College of Earth Sciences, Guilin University of Technology, Guilin 541004, China;
4. Faculty of Earth Resources, Key Laboratory of Tectonics and Petroleum Resources, Ministry of Education, China University of Geosciences, Wuhan 430074, China
The Sulu-Dabie ultrahigh-pressure (UHP) metamorphic belt in east-central China is possibly the largest UHP terrane in the world. It marks the Triassic collision zone between the Sino- Korean Plate and the Yangtze Plate (Fig. 1a). Identification of coesite and diamond in eclogites from the Sulu-Dabie Orogen demonstrates that supracrustal materials were subducted to mantle depths of > 100 km and experienced UHP metamorphism (for a summary see Liou et al., 2009; Ye et al., 2000). Based on the assumption that subducted continental crust was detached from its mantle lithosphere and exhumed to crustal levels as a single unit, various subduction and exhumation models have been proposed, such as an intracontinental thrusting and concomitant erosion model (Okay and Şengör, 1992), a buoyancy-driven model (Ernst, 2001; Ernst et al., 1997), a multi-stage exhumation model (Faure et al., 1999), an orogen-parallel extrusion and layer-parallel thinning model (Hacker et al., 2000) and a multi- slice exhumation model (Liu et al., 2009; Li et al., 2005, 2003).
More recently, different HP and UHP slices with contrasting subduction and exhumation histories within the Sulu Orogen were postulated (Liu et al., 2009; Xu et al., 2006), dividing the Sulu Orogen into three belts: one HP belt (Slice Ⅰ) and two UHP belts (slices Ⅱ and Ⅲ) from southeast to northwest (Fig. 1a). The slices feature increasing peaks and retrograde metamorphic P-T conditions and decreasing ages for peak metamorphism and subsequent retrogression (Table 1).
An apparent granulite-facies overprint in coesite-bearing eclogites was reported in the UHP belt of Slice Ⅲ (Banno et al., 2000; Nakamura and Hirajima, 2000; Yao et al., 2000; Zhang et al., 1995; Wang et al., 1993), revealing that some eclogites have undergone pervasive decompression reactions at a deep crustal level, but such phenomena have not been identified in Slice Ⅱ.
Reconstruction of P-T paths by phase equilibria modeling in metamorphic rocks is an essential tool in constraining models of the tectonic evolution of mountain belts. Refinements in internally consistent thermodynamic data sets, multicomponent mineral solution models, phase diagram calculation and pseudosection analysis techniques (Connolly and Petrini, 2002; Holland and Powell, 1998) have enabled realistic estimation of P-T paths of eclogite facies metamorphic rocks (e.g., Wei et al., 2003; Carson et al., 1999).
This study presents the petrography and mineral chemistry of eclogites from the two UHP belts within the Sulu Orogen. Pseudosection analysis is used to characterize the metamorphic P-T paths followed by eclogites from Donghai Town (in Slice Ⅱ and Slice Ⅲ), Jiangsu Province, and Rongcheng City (in Slice Ⅲ), Shandong Province, China (Fig. 1a). Thermal-dynamic structure models are compared with the calculated P-T paths of the two slices to obtain constraints on their subduction and exhumation processes.1 GEOLOGICAL SETTING
The Sulu Orogen constitutes the eastern part of the Sulu- Dabie Orogen in eastern China, which marks the Triassic collision zone between the Sino-Korean Plate and the Yangtze Plate. The Sulu metamorphic terrane is bounded by the Yantai- Qingdao-Wulian fault to the northwest, the Tanlu fault to the west and the Jiashan-Xiangshui fault to the southeast (Fig. 1a). From southeast to northwest the terrane is made up of a HP crustal slice and two UHP crustal slices that underwent different subduction and exhumation histories (Liu et al., 2009). The HP and UHP slices were subjected to an amphibolite facies retrograde overprint and separated by a thrust zone consisting of fault breccia and strongly deformed rocks. The HP belt (Slice Ⅰ) is composed of gneisses, quartz-mica schists, chloritoid-kyanite-quartz-mica schists, greenschists, piemontite-lepidolite-quartz schists, kyanite quartzites, and rare blueschists and marbles, whilst the UHP slices (Ⅱ and Ⅲ) mainly consist of gneisses with amphibolite layers and minor marbles and quartzites. Coesite-bearing eclogites and garnet peridotites occur as lenses, discontinuous layers, nodules, and boudins in the gneisses (Zhang et al., 2002). Both the HP and UHP metamorphic belts are unconformably overlain by Jurassic clastic sedimentary rocks and Cretaceous volcaniclastic deposits, and were intruded by post-collisional Mesozoic granites (Liu et al., 2004).
The Rongcheng area in the northeastern part of the Sulu UHP metamorphic belt (Slice Ⅲ) is underlain mainly by TTG gneisses and granitic gneisses with a series of schists. Minor marble and peridotite pods crop out in the gneisses (Fig. 1b). Eclogites are contained as lens-shaped blocks in the gneisses with migmatization developed locally in the contact zone (Jahn et al., 1996; Wang et al., 1993). Many Mesozoic magmatic intrusions developed in the gneisses.
The Donghai area in the south of the Sulu UHP metamorphic belt straddling the border between slices Ⅱ and Ⅲ is underlain mainly by quartz-feldspathic gneisses with a Proterozoic protolith age. The gneisses are intruded by Mesozoic granites and no Paleozoic strata occur (Fig. 1c). Eclogites commonly occur as pods or layers in the quartz-feldspathic gneisses. These pods and layers range from < 1 m to hundreds of meters in length. Most eclogite pods are aligned parallel to the regional schistosity of the gneisses, and some eclogite layers are folded concordantly with the gneisses (Zhang et al., 2000).2 PETROGRAPHY 2.1 Samples from Rongcheng
Two samples for the present study of Slice Ⅲ of the Sulu UHP metamorphic belt were collected from lenses of eclogites at Datuan Town, Rongcheng City, in the easternmost part of the Sulu UHP metamorphic terrane (Figs. 1a and 1b). Most of the eclogite bodies show mineralogical banding parallel to the foliation in the country rock gneisses. They are mainly composed of garnet and omphacite, with minor amounts of rutile, quartz and apatite. Many eclogitic blocks have experienced retrograde metamorphism and display a gradual transition of normal eclogite-retrograde eclogite- amphibolite eclogite-amphibolites from core to edge. A normal eclogite (RC23) and a retrograde eclogite (RC21) from the west of Datuan (Fig. 1b) were selected to reconstruct the exhumation path of the UHP rocks of Slice Ⅲ.
Sample RC23 shows an equigranular texture and is mainly composed of omphacite (45%), garnet (40%), quartz (10%), zoisite (3%) and traces of rutile and kyanite (2%). Most grains show irregular oblong shapes with uneven edges and developed cracks (Fig. 2a). Garnet is smaller than omphacite with a grain size < 1.5 mm. Some bigger garnet grains contain inclusions of omphacite+kyanite+rutile+quartz. Besides the inclusions, other two types of omphacite can be distinguished: one is medium- grained and usually contains inclusions of garnet+kyanite+ rutile+quartz (Fig. 2b), the other is scarce and occurs as rims around quartz and zoisite (Fig. 2a). The inclusion assemblages of garnet+omphacite+kyanite+rutile+quartz in the first type may represent the stable mineral assemblage during prograde or peak metamorphism. The occurrence of zoisite and omphacite rim in the second type suggests the hydrodynamic interaction during cooling process.
Sample RC21 shows a granoblastic texture and contains omphacite (45%), garnet (45%), quartz (5%), rutile (4%) and minor amphibole (1%). Garnet grains are 0.5–1.5 mm in size and show uneven colors, usually pink in the core with a colorless rim. The core is brighter in BSE images than the rim. Most garnet crystals are fresh with regular edges and are concentrated in centimeter-scale bands (Fig. 2c), some have irregular shapes and are distributed adjacent to omphacite, suggestive of reaction textures (Fig. 2d). Most omphacite grains are aggregates and have been resorbed to some extent, with edges and fractures being replaced by symplectites of albite+diopside (Fig. 2d), suggesting the retrogression of Grt+Omp+H2O→Amp+Ab±Cpx or Omp+ Qz→Ab+Cpx (all mineral abbreviations from Whitney and Evans, 2010). The minor amphibole grains are distributed in garnet bands, some of them showing clear boundaries with garnet, while some are surrounded with relict omphacite between garnet crystals (Fig. 2d). Some mica occurs in symplectite.2.2 Samples from Donghai
Two samples were collected from eclogite lenses at Donghai Town, Lianyungang City, one (DH09) is located in Sulu UHP metamorphic belt Slice Ⅲ, the other (DH03) in Slice Ⅱ (Figs. 1a and 1c). Most of the eclogites are banded with significantly different proportions of minerals and variable degrees of retrogression. Garnet, jadeite-rich clinopyroxene and rutile are present in all eclogites. Additional minerals, such as phengite, kyanite, epidote/zoisite, talc and amphibole, occur in many eclogites from the Donghai area. Mineral assemblages vary between areas and from one layer to another in banded eclogites. Two retrograde eclogites were selected to unravel the exhumation process, one (DH09) was from Jianchang in Slice Ⅲ, with granoblastic texture and the other one (DH03) was from Qinglongshan in Slice Ⅱ with abundant porphyroblastic epidote (Fig. 1c).
Sample DH09 is mainly composed of omphacite (70%), garnet (20%) and quartz (5%), with minor amounts of rutile+ ilmenite (3%) and glaucophane+paragonite+phengite (2%). Most omphacite and garnet grains are 1–4 mm in size, whereas individual omphacite may be as big as 6–7 mm. Both minerals show extensive fracturing and usually contain inclusions of quartz+ albite+graphite+rutile (Fig. 2e). Very fine symplectitic coronas of amphibole+albite+ilmenite have developed around garnet (Fig. 2f), suggesting the retrogression of Omp/Jd+Rt+fluid→Amp+ Ab+Ilm. Omphacite is rimmed by a fine border of albite (Fig. 2f), suggesting the reaction Jd+Qz→Ab. Many tiny ilmenite+ amphibole+quartz are distributed radially in the albite border, perpendicular to the edge of omphacite, with increasing size with distance from border (Fig. 2f), suggesting protracted retrogression.
Sample DH03 shows a porphyroblastic texture and has a preferred orientation in the matrix (Fig. 2g). The epidote porphyroblasts (15%) are usually bigger than 10 mm in size. The matrix consists of medium-grained garnet (40%), omphacite (30%), phengite (10%), quartz (3%), rutile (1%) and traces of barroisite+ kyanite+apatite+talc (1%); together, all matrix minerals define the foliation. Porphyroblastic epidote cuts the foliation and contains inclusions of omphacite+rutile/ilmenite+quartz+apatite± garnet with the same orientation as matrix (Fig. 2h), suggesting the formation of this porphyroblastic hydrous phase occurred during exhumation.3 MINERAL CHEMISTRY
Minerals were analyzed with the JEOL8800M electron microprobe at the Department of Earth Sciences, VU University Amsterdam, the Netherlands, using operating conditions of 15 kV accelerating voltage, ~24 nA beam current, peak counting time 10 s and background counting time 5 s. The following standards and crystals were used: amphibole TAP (Si, Al, Mg), PET (Ca, Ti), albite TAP (Na), or those PET (K), fayalite LiF (Fe, Mn). Representative mineral compositions are listed in Tables 2 to 4.
Garnet in Sample RC23 has a composition of Alm20–23 Grs36–40Prp37–40Sps~1Adr~1. Most garnet crystals show some Ca, Fe and Mg variation. They decrease in Ca and increase in Fe, Mg from core to rim, yielding the growing condition of increasing temperature and reducing pressure. Garnet in Sample RC21 has a composition of Alm52–53Grs19–24Prp18–19Sps~1Adr~5. Most grains have a bright core and a dark rim in BSE images. They have constant Mg and Ca content with decreasing Fe from core to rim, especially Fe3+, determined by charge balance, which maybe the reason of inhomogeneous brightness.
Garnet in Sample DH09 has a composition of Alm58–60Grs8–10 Prp28–29Sps~2Adr~2. Some grains decrease in Ca, Mg and increase in Fe from core to rim, which is presumably related to retrograde metamorphism. The symplectitic corona around garnet supports this. Garnet in Sample DH03 has a composition of Alm43–44 Grs17–18Prp35–36Sps~2Adr~1 with very little zoning, only displaying slightly decrease in the rim. Garnet inclusions in epidote have similar compositions to the bigger grains, suggesting they formed during the same stage of metamorphism.3.2 Pyroxene
Pyroxene in Sample RC23 is classified as Ca-Mg-Fe pyroxene, and others are Ca-Na pyroxenes (Table 3). The compositions of Ca-Na pyroxene in the eclogites are plotted in the ternary diagram of Morimoto et al. (1988) (Fig. 3b).
In Sample RC23, pyroxene belongs to the diopside species with a composition of Aug72–76Jd24–27Aeg~0. Most grains show clear zoning with increasing augite and decreasing jadeite from core to rim. There are two generations of clinopyroxene in Sample RC21. The first is distributed in the interstitial spaces between garnet grains and plots in the omphacite field with a homogeneous composition of Aug60–61Jd27–28Aeg11–13. The second generation is symplectitic clinopyroxene and plots in the aegirine-augite field with a composition of Aug75Jd7Aeg18, as the aegirine content is low, this clinopyroxene is augite.
Pyroxenes in Sample DH09 plot in the jadeite field with a composition of Aug10–11Jd65–66Ae23–24. There are no differences in composition between jadeite inclusions in garnet and larger matrix jadeite grains. Pyroxenes in Sample DH03 plot in the omphacite field with a composition of Aug48–50Jd34–35Ae15–17. Most grains are homogeneous in compositions. Omphacite inclusions in garnet have lower jadeite and higher aegirine than omphacite in the matrix.3.3 White Mica
There is scarce white mica in samples RC23 and RC21 and a little more in samples DH09 and DH03 (Table 4). White mica in Sample DH03 is phengite with an average Si content of 3.49 per formula unit based on 11 oxygens. In Sample DH09 they are paragonite with average Si value of 3.18 per formula unit. The composition for each grain is homogeneous. White mica in samples RC21 and RC23 was not analyzed.3.4 Amphibole
Two types of amphibole occur in Sample RC21, both calcic amphiboles (Fig. 4a). Larger grains mimic the grain shape of omphacite with worm-like omphacite relics surrounded by albite in the core (Fig. 2c), suggesting amphibole+albite replaced omphacite. The other type occurs as symplectites with albite around omphacite. Symplectitic amphibole has higher FeO and CaO, but lower Al2O3 and Na2O than the coarser grained amphibole. The symplectitic amphibole is a Mg-Hs-hornblende and the larger amphibole is magnesio-hastingsite or Mg-Hs-hornblende (Fig. 4b).
There are two types of amphiboles in Sample DH09, and one has big grain size and belongs to sodic amphibole (Fig. 4a) with compositions in the glaucophane field (Fig. 4c). The Fe3+ content indicates that the rims are more oxidized than the core with Fe3+/(Fe3++AlVI) increasing and Fe2+/(Fe2++Mg) remaining constant from core to rim. The second set of amphibole grains occurs as symplectites or coronas (Fig. 2f) and belongs to the sodic- calcic amphiboles (Fig. 4a) with a high content of Al2O3 and low SiO2. They are classified as magnesio-taramite. Amphibole in Sample DH03 is sodic-calcic amphibole and plots in barroisite field with homogeneous compositions (Figs. 4a and 4d).3.5 Epidote/Zoisite
Epidote occurs as porphyroblasts in Sample DH03 and displays compositional zoning with a slight decrease from a core value of Ps22 to a rim with Ps20 (Ps=100×Fe3+/(Fe3++Al)), implying the transition from epidote towards clinozoisite from core to rim. Zoisite in Sample RC23 has a homogeneous composition with a Ps-value of 2 (Table 4).4 P-T PSEUDOSECTION CALCULATIONS
To decipher the evolution of mineral assemblages with changing metamorphic P-T conditions, P-T pseudosections were calculated for samples RC23 and DH03 in the MnC(K)NFMASHO system using the Perple_X v. 6.7.7 software (Connolly and Petrini, 2002) and the updated version of the internally consistent thermodynamic database of Holland and Powell(1998, 1991). The solution models of White et al. (2000) for garnet, Green et al. (2007) for omphacite, Holland et al. (1996) for phengite, Holland et al. (1998) for chlorite, Tajčámanova et al. (2009) for biotite, Holland and Powell (2003) for feldspar and Dale et al.(2005, 2000) for amphibole were used. Quartz is abundant in the matrix of eclogite samples RC23 and DH03, thus SiO2 is considered to be available in excess. The H2O was used as a saturated pure fluid phase since both two samples contain abundant hydrous phases. The bulk chemical compositions of the eclogites were determined by X-ray fluorescence (XRF) using a Rigaku RIX 2000 spectrometry (analytical precision in within 0.5% for major and minor elements) at the Department of Earth Sciences, VU Amsterdam, the Netherlands. The analytical results are listed in Table 5.
For Sample RC23, a model system MnNCFMASHO was chosen to calculate the pseudosection in the P-T range of 1.3–4.0 GPa and 500–1 000 ℃. As is shown in Fig. 5a, lawsonite and talc are stable in HP fields with relatively low temperature (T < 800 ℃, P > 1.7 GPa), and zoisite occurs in a wide range of < 2.6 GPa. Kyanite-bearing mineral assemblages are stable in HT and HP fields. As a result of none lawsonite and talc according to thin section observation, it can be preliminarily inferred that the peak assemblage would be located in the field of Grt+Omp+Ky+ Rt+Coe/Qz with a P-T range of P > 2.6 GPa, T > 700 ℃.
Isopleths of the XMg (Mg/(Fe2++Mn+Mg+Ca)) and XCa (Ca/(Fe2++Mn+Mg+Ca)) contents in garnet and Mg# (Mg/(Fe2++Mg)) and Na contents in omphacite are shown in Fig. 5b. In the P-T condition range of P > 26 kbars, T > 700 ℃, the XMg isopleths have moderately positive slopes increase with rising temperature. The XCa isopleths have relatively flat slopes and decrease as pressure declines, being preferable pressure indicators. Whereas in the field of Grt+Omp+Ky+Zo+Rt+Qz, the XMg and XCa isopleths in garnet have almost the same flat slopes, by which it is not persuasive to constrain the pressure condition. Howerver, the Na content isopleths in this field have steep slopes increase with rising temperature, being good temperature indicators. The measured XMg (0.37–0.39) and XCa (0.40–0.36) isopleths in garnet from core to rim, together with the Mg# isopleth (0.91) in the core of omphacite and the Na content isopleth (0.25) in the rim of omphacite indicate a P-T path from 3.2 GPa and ~840 ℃ to 2.4 GPa and 910 ℃, where the maximum pressure (stage A) was recorded by the core composition in garnet and omphacite, and the maximum temperature (stage B) was recorded by the rim composition in garnet and the Na content in the rim of omphacite.4.2 P-T Pseudosection for Eclogite Sample DH03 from Slice Ⅱ
Pseudosection of Sample DH03 was calculated in a MnNCKFMASHO system in the P-T range of 1.3–4.0 GPa and 450–800 ℃ (Fig. 6a). Isopleths of the XMg and XCa contents in garnet and Si contents in phengite are shown in Fig. 6b. Contours of XMg and XCa in garnet have steep slopes, sensitive to temperature changes, while Si isopleths in phengite have moderately positive slopes, which are more appropriate to reflect pressure variation. The measured XMg (0.36) and XCa (0.18) content in the core of garnet and the Si content (3.49) in phengite constain the P-T range of 3.1–3.3 GPa and 660–690 ℃, located in the stability field of Mica+Grt+Omp+Law+Rt+Coe, which is irreconcilable to the present mineral assemblage. Mineral chemical characteristics indicate that garnet has a composition zoning because of chemical difussion, reflected in the XFe, XMg and XCa content slightly decreasing from core to rim, which manifests that the eclogites were subjected to a certain extent reform. Therefore, the pressure and temperature constrained by the measured mineral content in the core of garnet and phengite probably represents the peak P-T condition (stage A), and the present mineral assemblage can not represent the peak mineral assemblage. The final thermodynamic equilibrium assemblage according to petrographic observation and mineral chemisty corresponds to the mineral assemblage of Ep+Amp+Ms+Grt+Omp+Ky+Zo+Rt+Qz in the pseudosection, of which the P-T is about 2.2 GPa, 600 ℃ (stage B). Consequently, combined with the Na content of omphacite from core to rim (Na=0.50 wt.%–0.52 wt.%), it is speculated that the eclogite recorded a cooling decompressive process.5 DISCUSSION 5.1 Metamorphic P-T Evolution for Eclogite Sample RC23 from Slice Ⅲ
By P-T pseudosection and petrographic observation, it is speculated that the prograde P-T path may pass the stability field of Grt+Omp+Ky+Rt+Qz (Fig. 5b). According to the model, the peak P-T condition is located in the stability field of Grt+Omp+ Ky+Rt+Coe, matching well with the petrographic observation except Coe, which is difficult to preserve probably owing to the heating decompressive process, thus turning into quartz. From stage A to stage B, the peak temperature recorded lagged behind the peak pressure, implying that Sample RC23 experienced persistent overheating process, which overlaid an apparent granulite-facies metamorphism during exhumation. With temperature rising and pressure declining, coesite transformed to quartz and the composition of garnet changed obviously embodied in the increasing content of XMg and the decreasing content of XCa from core to rim. The H2O content pseudosection shows that the P-T path evolves along the MH2O increasing direction (Fig. 7a), indicating that there are few enthetic fluids joining in this process. Garnet with fluid reacted to zoisite, finally achieving thermodynamic equilibrium and stabilizing in the field of Grt+Omp+Ky+Rt+Qz+Zo, which is consistent with the petrographic observation.
Compared to the other two samples from Slice Ⅲ, the HP minerals such as garnet and omphacite can be preserved without apparent reform during exhumation, probably because the eclogite was always in a relatively dry state, which is also the reason why little muscovite can be found in the rock. Besides, though the eclogite experienced a stage of thermal relaxation above 700 ℃, the garnet still retained compositional zoning and was not homogenous. Experimental researches show that under the temperature of 650–750 ℃, growth compositional zonation of garnet will be homogenous because of persistent heating and the differences in mineral grains (Carlson, 2002; Carlson and Schwarze, 1997; Spear, 1991). Previous study has reported that garnet of UHP eclogite in Sulu Orogen could not only form growth compositional zoning, but also record peak temperature above 900 ℃, for which the interpretation is that the dwelling time of the UHP eclogite in peak P-T condition was very short, and the eclogite exhumated to the Earth's surface at a rapid rate (Zhang et al., 2005). Therefore, a short thermal relaxation stage and rapid exhumation rate should be another reason why the UHP minerals could retain without obvious retrograde reform.5.2 Metamorphic P-T Evolution for Sample DH03 from Slice Ⅱ
Pseudosection modeling of Sample DH03 indicates that the peak P-T condition is located in the stability field of lawsonite bearing assemblage, above the transformed line of coesite to quartz, which is consistent with discovery of coesite in previous studies in Sulu-Dabie orogenic belt. But the calculated peak P-T condition is lower than those calculated by geothermobarometers of other eclogites in previous researches (Liu et al., 2008; Liu and Xue, 2007; Liu and Xu, 2004; Zhang et al., 2002, 1995; Yao et al., 2000; Wang et al., 1993; Hirajima et al., 1990). Wei et al. (2013) considered that in decompressive process of exhumation, like thermal driving dehydration reaction, decompressive driving dehydration reaction will balance metamorphic mineral assemblages, inducing transformation of eclogite-facies peak mineral assemblages. Therefore, there is a certain uncertainty in defining ecologite peak P-T condition by geothermobarometers, while pseudosection calculation based on bulk compositions can well define peak P-T conditions. Furthermore, it is confusing that no lawsonites even their pseudomorphs in present mineral assemblages were found according to petrographic observation. Previous studies of experimental petrology and pseudosection modeling manifests that lawsonite has a wide stability region and extensively consists in cold subduction zones, but there are still few lawsonite bearing eclogites exposed on the Earth's surface around the world, particularly those formed in continent-continent collision zones. Lawsonite bearing eclogites in China are mainly exposed in HP-LT metamorphic belt, such as the North Qilian, the North Altyn-Tagh and the Southwest Tianshan. Sulu-Dabie orogenic belt is deemed to have once experienced lawsonite bearing eclogite-facies peak metamorphism (Wei et al., 2010; Li et al., 2005), suggesting that the continental subduction can bring the water of water-bearing mineral such as lawsonite into the deep mantle. Whitney and Davis (2006) considered that the conservation of lawsonite bearing eclogite is relevant to rapid exhumation. Tsujimori et al. (2006) stated that under the control of slab subduction and exhumation mechanism, typically exhumation process is adverse to the conservation of lawsonite bearing eclogites, especially those experiencing early staged dehydration reaction. As consequences of being subjected to late alteration, that there are no lawsonites or lawsonite pseudomorphs is not contradict against the pseudosection modeling result of Sample DH03 that the peak P-T condition is located in lawsonite stability field.
Although the eclogite only recorded the exhumation process, on the basis of petrographic observation, it can be supposed that the prograde P-T path would pass the stability field of Tal-bearing assemblage because some traces of Tal can be found in the rock (Fig. 6b). From stage A to stage B, with temperature and pressure declining, the composition of garnet and omphacite changed slightly, which might result from the rapid exhumation of the rock. Therefore, the rock had already achieved thermodynamic equilibrium when the garnet and omphacite were not able to record the changes entirely. The P-T path incises the H2O content isopleths and points out to the decreasing direction of H2O content (Fig. 7b), which means after eclogite-facies peak P-T contition, driven by decompression, hydrous mineral lawsonite started to break down and released H2O, leading to the dehydration reaction of Law→Grs+Ky+Coe+H2O. The involvement of fluid is beneficial to the evolution of metamorphic assemblage, so the eclogite is difficult to record the authentic peak P-T condition, which should be higher than that calculated by present minerals. When the pressure declined under the transformed line of coesite to quartz, omphacite started to break down to amphibole with the XJd and XAeg content slightly reducing. As the pressure further declined, lawsonite broke down to epidote, kyanite, quartz and fluid, and the rock finally stabilized in the field of Ep+Amp+Mica+Grt+ Omp+Ky+Zo+Rt+Qz, which is consistent with the compositional zoning of epidote that displays a slight decrease from a core value of Ps22 to a rim value with Ps20, implying the transition from epidote towards zoisite from core to rim.5.3 Geological Implications
The formation and exhumation of the Sulu-Dabie UHP rocks have been extensively studied in recent years (Suo et al., 2012; Li et al., 2011; Liou et al., 2009; Zheng et al., 2003). Various subduction and exhumation models have been proposed. Textural relations and mineral paragenesis indicate that the Sulu eclogites investigated here have experienced multistage metamorphism. The P-T paths derived for the rocks are illustrated in Fig. 8. The eclogites experienced a clockwise P-T evolution. The peak P-T conditions of samples RC23 and DH03 indicate that they were subducted beneath the Sino-Korean Plate to ~100–105 km depth. Sample RC23 was subducted to almost the same depth as Sample DH03, but still experienced a high- temperature overprint. This suggests differences in the metamorphic evolution of the two crustal slices and that the UHP rocks in Slice Ⅲ did not necessarily undergo significant cooling during early exhumation.
The uplift of Slice Ⅲ may have been temporarily obstructed by the overlying slices. Liu et al. (2009) suggested that HP crustal Slice Ⅰ was first detached from the underlying subducted continental crust, and then exhumed due to buoyancy (Fig. 9). When HP Slice Ⅰ returned to upper crustal levels, the underlying continental crust Slice Ⅱ continued to subduct to 100–120 km depth and experienced UHP metamorphism that is evidenced by the widespread occurrence of coesite in eclogites and their host rocks (Liu et al., 2009). After that, the UHP Slice Ⅱ was detached from the subducting slab, and exhumed causing overprinting of eclogites by retrograde assemblages, while the subduction of the underthrusted Slice Ⅲ continued to a depth of > 100–120 km. After it experienced UHP metamorphism this slice also detached, and was exhumed and retrogressed (Fig. 9a).
On the basis of this tectonic scenario, it is concluded that the two UHP crustal slices in the Sulu terrane have a similar geodynamic evolution, but UHP metamorphism and subsequent exhumation of Slice Ⅲ most likely took place later than Slice Ⅱ. After the detachment of Slice Ⅱ from the subducting slab, buoyancy-driven crustal advection through the subduction channel may be one of the principal mechanisms for exhuming the UHP eclogites, which resulted in earlier exhumation of Sample DH03 as part of Slice Ⅱ. After the ascent to crustal levels of Slice Ⅱ, the UHP Slice Ⅲ was detached from the subducting slab and rose buoyantly (Fig. 9a). The presence of the overlying Slice Ⅱ reduced the buoyancy force and the exhumation rate of Slice Ⅲ (Fig. 9b), which may not last long, thus resulting in slower exhumation and the granulite-facies overprint of the eclogite facies assemblages. As a whole, Slice Ⅲ still exhumed at a rapid rate. To summarize, the crustal slices in the Sulu UHP metamorphic terrane appear to have experienced differential subduction and exhumation processes.
A collisional orogenic belt was formed in two stages, i.e., orogeny during oceanic crust subduction before the closure of the oceanic basin (oceanic-type subduction zone) and the continent- continent collisional orogeny after its disappearance (continental- type subduction zone). The Sulu UHP metamorphic belt mainly consists of granitic and pelitic gneisses intercalated with blocks of eclogite and varying amounts of ultramafic rocks, especially garnet peridotite (Liou et al., 2009; Zhang et al., 2002). The rock assemblages suggest that this belt is typical of a continental-type subduction zone (Song et al., 2014). Therefore, the Sulu orogenic belt represents a complete evolutionary sequence from oceanic subduction to continental collision. Its subducted continental crust is made up of different crustal slices that underwent different subduction and exhumation histories, rather than a single unit.6 CONCLUSIONS
(1) Eclogite samples from the UHP belt of Slice Ⅲ of the Sulu orogenic belt mainly consist of omphacite, garnet and quartz, with minor rutile, ilmenite, amphibole and phengite. Eclogite samples from Slice Ⅱ show a porphyroblastic texture with epidote porphyroblasts and garnet, omphacite, phengite, quartz and rutile in matrix.
(2) Pseudosection modeling indicates that eclogites from Slice Ⅱ witness a peak metamorphism of eclogite-facies under conditions of 3.1–3.3 GPa and 660–690 ℃, and a retrograde cooling decompression process. The eclogites from Slice Ⅲ record a heating decompressive P-T path with a peak-P stage of 3.2 GPa and 840 ℃ and a peak-T stage of 2.4 GPa and 950 ℃, suggesting an apparent granulite-facies metamorphism overprint during exhumation. Both eclogites recorded clockwise P-T paths with peak P-T conditions suggesting a subduction beneath the Sino-Korean Plate to ~100–105 km depth.
(3) The two UHP crustal slices in the Sulu terrane have a similar geodynamic evolution. The UHP rocks in Slice Ⅱ exhumed after the eclogitic peak pressure conditions earlier than that of Slice Ⅱ. The existence of Slice Ⅱ diminished the buoyancy force on Slice Ⅲ, resulting in a granulite-facies overprint displayed by the UHP rocks of Slice Ⅲ. The Sulu orogenic belt is made up of different crustal slices that underwent different subduction and exhumation histories, rather than a single unit.ACKNOWLEDGMENTS
We are indebted to Dr. Rong-Guo Hu from the Department of Earth Sciences, VU Amsterdam for his help with sample preparation for XRF analysis. We gratefully acknowledge the careful and insightful comments of the three anonymous reviewers, which considerably improved the manuscript. This study was funded by the National Key R & D Program of China (No. 2016YFC0600403), the State Scholarship Fund of the China Scholarship Council (CSC) to Yilong Li, and the Fundamental Research Funds for the Central Universities, China University of Geosciences (Wuhan) (Nos. CUGL170404, CUG160232). Fieldwork and sample collection were supported by the KNAW- CAS China Exchange Program to Jan R. Wijbrans, Fraukje M. Brouwer, and Hua-Ning Qiu, and the NWO Meervoud Program to Fraukje M. Brouwer. The final publication is available at Springer via https://doi.org/10.1007/s12583-018-0845-x.
Banno, S., Enami, M., Hirajima, T., et al., 2000. Decompression P-T Path of Coesite Eclogite to Granulite from Weihai, Eastern China. Lithos, 52(1/2/3/4): 97-108. DOI:10.1016/s0024-4937(99)00086-9
Carlson, W. D., 2002. Scales of Disequilibrium and Rates of Equilibration during Metamorphism. American Mineralogist, 87(2/3): 185-204. DOI:10.2138/am-2002-2-301
Carlson, W. D., Schwarze, E. T., 1997. Petrological Significance of Prograde Homogenization of Growth Zoning in Garnet:An Example from the Llano Uplift. Journal of Metamorphic Geology, 15(5): 631-644. DOI:10.1111/j.1525-1314.1997.tb00640.x
Carson, C. J., Powell, R., Clarke, G. L., 1999. Calculated Mineral Equilibria for Eclogites in CaO-Na2O-FeO-MgO-Al2O3-SiO2-H2O:Application to the Pouébo Terrane, Pam Peninsula, New Caledonia. Journal of Metamorphic Geology, 17(1): 9-24. DOI:10.1046/j.1525-1314.1999.00177.x
Connolly, J. A. D., Petrini, K., 2002. An Automated Strategy for Calculation of Phase Diagram Sections and Retrieval of Rock Properties as a Function of Physical Conditions. Journal of Metamorphic Geology, 20(7): 697-708. DOI:10.1046/j.1525-1314.2002.00398.x
Dale, J., Holland, T. J. B., Powell, R., 2000. Hornblende-Garnet-Plagioclase Thermobarometry:A Natural Assemblage Calibration of the Thermodynamics of Hornblende. Contributions to Mineralogy and Petrology, 140: 353-362. DOI:10.1007/s004100000187
Dale, J., Powell, R., White, R. W., et al., 2005. A Thermodynamic Model for Ca-Na Clinoamphiboles in Na2O-CaO-FeO-MgO-Al2O3-SiO2-H2O-O for Petrological Calculations. Journal of Metamorphic Geology, 23(8): 771-791. DOI:10.1111/j.1525-1314.2005.00609.x
Ernst, W. G., 2001. Subduction, Ultrahigh-Pressure Metamorphism, and Regurgitation of Buoyant Crustal Slices-Implications for Arcs and Continental Growth. Physics of the Earth and Planetary Interiors, 127(1/2/3/4): 253-275. DOI:10.1016/s0031-9201(01)00231-x
Ernst, W. G., Maruyama, S., Wallis, S., 1997. Buoyancy-Driven, Rapid Exhumation of Ultrahigh-Pressure Metamorphosed Continental Crust. Proceedings of the National Academy of Sciences, 94(18): 9532-9537. DOI:10.1073/pnas.94.18.9532
Faure, M., Lin, W., Shu, L., et al., 1999. Tectonics of the Dabieshan (Eastern China) and Possible Exhumation Mechanism of Ultra High-Pressure Rocks. Terra Nova, 11(6): 251-258. DOI:10.1046/j.1365-3121.1999.00257.x
Green, E. C. R., Holland, T. J. B., Powell, R., 2007. An Order-Disorder Model for Omphacitic Pyroxenes in the System Jadeite-Diopside-Hedenbergite-Acmite, with Applications to Eclogitic Rocks. American Mineralogist, 92(7): 1181-1189. DOI:10.2138/am.2007.2401
Hacker, B. R., Ratschbacher, L., Webb, L., et al., 2000. Exhumation of Ultrahigh-Pressure Continental Crust in East Central China:Late Triassic-Early Jurassic Tectonic Unroofing. Journal of Geophysical Research:Solid Earth, 105(B6): 13339-13364. DOI:10.1029/2000jb900039
Hirajima, T., Ishiwatari, A., Cong, B. L., et al., 1990. Coesite from Mengzhong Eclogite at Donghai County, Northeastern Jiangsu Province, China. Mineralogical Magazine, 54(377): 579-583. DOI:10.1180/minmag.1990.054.377.07
Holland, T. J. B., Babu, E. V. S. S. K., Waters, D. J., 1996. Phase Relations of Osumilite and Dehydration Melting in Pelitic Rocks:A Simple Thermodynamic Model for the KFMASH System. Contributions to Mineralogy and Petrology, 124(3/4): 383-394. DOI:10.1007/s004100050198
Holland, T. J. B., Baker, J., Powell, R., 1998. Mixing Properties and Activity-Composition Relationships of Chlorites in the System MgO-FeO-Al2O3-SiO2-H2O. European Journal of Mineralogy, 10(3): 395-406. DOI:10.1127/ejm/10/3/0395
Holland, T. J. B., Powell, R., 1998. An Internally Consistent Thermodynamic Data Set for Phases of Petrological Interest. Journal of Metamorphic Geology, 16(3): 309-343. DOI:10.1111/j.1525-1314.1998.00140.x
Holland, T. J. B., Powell, R., 2003. Activity-Composition Relations for Phases in Petrological Calculations:An Asymmetric Multicomponent Formulation. Contributions to Mineralogy and Petrology, 145(4): 492-501. DOI:10.1007/s00410-003-0464-z
Holland, T. J. B., Powell, R., 1991. A Compensated-Redlich-Kwong (CORK) Equation for Volumes and Fugacities of CO2 and H2O in the Range 1 bar to 50 kbar and 100-1 600℃. Contributions to Mineralogy and Petrology, 109(2): 265-273. DOI:10.1007/bf00306484
Jahn, B. M., Cornichet, J., Cong, B. L., et al., 1996. Ultrahigh-εNd Eclogites from an Ultrahigh-Pressure Metamorphic Terrane of China. Chemical Geology, 127(1/2/3): 61-79. DOI:10.1016/0009-2541(95)00108-5
Leake, B. E., Woolley, A. R., Arps, C. E. S., et al., 1997. Nomenclature of Amphiboles:Report of the Subcommittee on Amphiboles of the International Mineralogical Association, Commission on New Minerals and Mineral Names. The Canadian Mineralogist, 35: 219-246.
Li, S. G., Huang, F., Zhou, H. Y., et al., 2003. U-Pb Isotopic Compositions of the Ultrahigh Pressure Metamorphic (UHPM) Rocks from Shuanghe and Gneisses from Northern Dabie Zone in the Dabie Mountains, Central China:Constraint on the Exhumation Mechanism of UHPM Rocks. Science in China Series D:Earth Sciences, 46(3): 200-209. DOI:10.1360/03yd9019
Li, S. Z., Kusky, T. M., Zhao, G. C., et al., 2011. Thermochronological Constraints on Two-Stage Extrusion of HP/UHP Terranes in the Dabie-Sulu Orogen, East-Central China. Tectonophysics, 504(1/2/3/4): 25-42. DOI:10.1016/j.tecto.2011.01.017
Li, S. G., Li, Q. L., Hou, Z. H., et al., 2005. Cooling History and Exhumation Mechanism of the Ultrahigh-Pressure Metamorphic Rocks in the Dabie Mountains, Central China. Acta Petrologica Sinica, 21: 1117-1124.
Liou, J. G., Ernst, W. G., Zhang, R. Y., et al., 2009. Ultrahigh-Pressure Minerals and Metamorphic Terranes-The View from China. Journal of Asian Earth Sciences, 35(3/4): 199-231. DOI:10.1016/j.jseaes.2008.10.012
Liu, F. L., Gerdes, A., Xue, H. M., 2009. Differential Subduction and Exhumation of Crustal Slices in the Sulu HP-UHP Metamorphic Terrane:Insights from Mineral Inclusions, Trace Elements, U-Pb and Lu-Hf Isotope Analyses of Zircon in Orthogneiss. Journal of Metamorphic Geology, 27(9): 805-825. DOI:10.1111/j.1525-1314.2009.00833.x
Liu, F. L., Gerdes, A., Zeng, L. S., et al., 2008. SHRIMP U-Pb Dating, Trace Elements and the Lu-Hf Isotope System of Coesite-Bearing Zircon from Amphibolite in the SW Sulu UHP Terrane, Eastern China. Geochimica et Cosmochimica Acta, 72(12): 2973-3000. DOI:10.1016/j.gca.2008.04.007
Liu, F. L., Xu, Z. Q., 2004. Fluid Inclusions Hidden in Coesite-Bearing Zircons in Ultrahigh-Pressure Metamorphic Rocks from Southwestern Sulu Terrane in Eastern China. Chinese Science Bulletin, 49(4): 396-404. DOI:10.1007/bf02900324
Liu, F. L., Xu, Z. Q., Liou, J. G., 2004. Tracing the Boundary between UHP and HP Metamorphic Belts in the Southwestern Sulu Terrane, Eastern China:Evidence from Mineral Inclusions in Zircons from Metamorphic Rocks. International Geology Review, 46(5): 409-425. DOI:10.2747/0020-68220.127.116.119
Liu, F. L., Xue, H. M., 2007. Review and Prospect of SHRIMP U-Pb Dating on Zircons from Sulu-Dabie UHP Metamorphic Rocks. Acta Petrologica Sinica, 23: 2737-2756.
Morimoto, N., Ferguson, A. K., Ginzburg, I. V., et al., 1988. Nomenclature of Pyroxenes. American Mineralogist, 73: 1123-1133.
Nakamura, D., Hirajima, T., 2000. Granulite-Facies Overprinting of Ultrahigh-Pressure Metamorphic Rocks, Northeastern Su-Lu Region, Eastern China. Journal of Petrology, 41(4): 563-582. DOI:10.1093/petrology/41.4.563
Okay, A. I., Şengör, A. M. C., 1992. Evidence for Intracontinental Thrust-Related Exhumation of the Ultra-High-Pressure Rocks in China. Geology, 20(5): 411-414. DOI:10.1130/0091-7613(1992)020<0411:efitre>2.3.co;2
Song, S. G., Niu, Y. L., Su, L., et al., 2014. Continental Orogenesis from Ocean Subduction, Continent Collision/Subduction, to Orogen Collapse, and Orogen Recycling:The Example of the North Qaidam UHPM Belt, NW China. Earth-Science Reviews, 129: 59-84. DOI:10.1016/j.earscirev.2013.11.010
Spear, F. S., 1991. On the Interpretation of Peak Metamorphic Temperatures in Light of Garnet Diffusion during Cooling. Journal of Metamorphic Geology, 9(4): 379-388. DOI:10.1111/j.1525-1314.1991.tb00533.x
Tajčmanová, L., Connolly, J. A. D., Cesare, B., 2009. A Thermodynamic Model for Titanium and Ferric Iron Solution in Biotite. Journal of Metamorphic Geology, 27(2): 153-165. DOI:10.1111/j.1525-1314.2009.00812.x
Tsujimori, T., Sisson, V. B., Liou, J. G., et al., 2006. Very-Low-Temperature Record of the Subduction Process:A Review of Worldwide Lawsonite Eclogites. Lithos, 92(3/4): 609-624. DOI:10.1016/j.lithos.2006.03.054
Wang, Q. C., Ishiwatari, A., Zhao, Z. Y., et al., 1993. Coesite-Bearing Granulite Retrograded from Eclogite in Weihai, Eastern China. European Journal of Mineralogy, 5(1): 141-152. DOI:10.1127/ejm/5/1/0141
Wei, C. J., Li, Y. J., Yu, Y., et al., 2010. Phase Equilibria and Metamorphic Evolution of Glaucophane-Bearing UHP Eclogites from the Western Dabieshan Terrane, Central China. Journal of Metamorphic Geology, 28(6): 647-666. DOI:10.1111/j.1525-1314.2010.00884.x
Wei, C. J., Powell, R., Zhang, L. F., 2003. Eclogites from the South Tianshan, NW China:Petrological Characteristic and Calculated Mineral Equilibria in the Na2O-CaO-FeO-MgO-Al2O3-SiO2-H2O System. Journal of Metamorphic Geology, 21(2): 163-179. DOI:10.1046/j.1525-1314.2003.00435.x
Wei, C. J., Tian, Z. L., Zhang, L. F., 2013. Modelling of Peak Mineral Assemblages and P-T Conditions for High-Pressure and Ultrahigh-Pressure Eclogites. China Science Bulletin, 58(22): 2159-2164. DOI:10.1360/972013-605
White, R. W., Powell, R., Holland, T. J. B., et al., 2000. The Effect of TiO2 and Fe2O3 on Metapelitic Assemblages at Greenschist and Amphibolite Facies Conditions:Mineral Equilibria Calculations in the System K2O-FeO-MgO-Al2O3-SiO2-H2O-TiO2-Fe2O3. Journal of Metamorphic Geology, 18(5): 497-511. DOI:10.1046/j.1525-1314.2000.00269.x
Whitney, D. L., Davis, P. B., 2006. Why is Lawsonite Eclogite so Rare? Metamorphism and Preservation of Lawsonite Eclogite, Sivrihisar, Turkey. Geology, 34(6): 473. DOI:10.1130/g22259.1
Whitney, D. L., Evans, B. W., 2010. Abbreviations for Names of Rock-Forming Minerals. American Mineralogist, 95(1): 185-187. DOI:10.2138/am.2010.3371
Xu, Z. Q., Zeng, L. S., Liu, F. L., et al., 2006. Polyphase Subduction and Exhumation of the Sulu High-Pressure-Ultrahigh-Pressure Metamorphic Terrane. Geological Society of America Special Paper, 403: 93-113.
Yao, Y. P., Ye, K., Liu, J. B., et al., 2000. A Transitional Eclogite-to High Pressure Granulite-Facies Overprint on Coesite-Eclogite at Taohang in the Sulu Ultrahigh-Pressure Terrane, Eastern China. Lithos, 52(1/2/3/4): 109-120. DOI:10.1016/s0024-4937(99)00087-0
Ye, K., Cong, B. L., Ye, D. N., 2000. The Possible Subduction of Continental Material to Depths Greater than 200 km. Nature, 407(6805): 734-736. DOI:10.1038/35037566
Zhang, R. Y., Hirajima, T., Banno, S., et al., 1995. Petrology of Ultrahigh-Pressure Rocks from the Southern Su-Lu Region, Eastern China. Journal of Metamorphic Geology, 13(6): 659-675. DOI:10.1111/j.1525-1314.1995.tb00250.x
Zhang, R. Y., Liou, J. G., Shu, J. F., 2002. Hydroxyl-Rich Topaz in High-Pressure and Ultrahigh-Pressure Kyanite Quartzites, with Retrograde Woodhouseite, from the Sulu Terrane, Eastern China. American Mineralogist, 87(4): 445-453. DOI:10.2138/am-2002-0408
Zhang, R. Y., Liou, J. G., Yang, J. S., et al., 2000. Petrochemical Constraints for Dual Origin of Garnet Peridotites from the Dabie-Sulu UHP Terrane, Eastern-Central China. Journal of Metamorphic Geology, 18(2): 149-166. DOI:10.1046/j.1525-1314.2000.00248.x
Zhang, Z. M., Xiao, Y. L., Shen, K., et al., 2005. Garnet Growth Compositional Zonation and Metamorphic P-T Path of the Ultrahigh-Pressure Eclogites from the Sulu Orogenic Belt, Eastern Central China. Acta Petrologica Sinica, 21: 809-818.
Zheng, Y. F., Fu, B., Gong, B., et al., 2003. Stable Isotope Geochemistry of Ultrahigh Pressure Metamorphic Rocks from the Dabie-Sulu Orogen in China:Implications for Geodynamics and Fluid Regime. Earth-Science Reviews, 62(1/2): 105-161. DOI:10.1016/s0012-8252(02)00133-2