Journal of Asian Earth Sciences - Cugb · Paleotethys tectonics Tibetan Plateau abstract The...

Journal of Asian Earth Sciences 113 (2015) 748–765

Contents lists available at ScienceDirect

Journal of Asian Earth Sciences

journal homepage: www.elsevier .com/locate / jseaes

Structural style and metamorphic conditions of the Jinshajiangmetamorphic belt: Nature of the Paleo-Jinshajiang orogenic beltin the eastern Tibetan Plateau

http://dx.doi.org/10.1016/j.jseaes.2015.09.0031367-9120/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: The State Key Laboratory of Geological Processes andMineral Resources, China University of Geosciences, Beijing 100083, China

E-mail address: [email protected] (D.-P. Yan).1 Address: Department of Earth & Environmental Sciences, The University of Iowa,

Iowa City, IA 52242, USA.

Wentao Cao 1, Dan-Ping Yan ⇑, Liang Qiu, Yixi Zhang, Jingwei QiuThe State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Beijing 100083, China

a r t i c l e i n f o

Article history:Received 30 April 2015Received in revised form 14 August 2015Accepted 2 September 2015Available online 11 September 2015

Keywords:Jinshajiang sutureBarrovian metamorphismThermobarometryPaleotethys tectonicsTibetan Plateau

a b s t r a c t

The Jinshajiang metamorphic belt is a Barrovian sequence distributed within the Jinshajiang Suture Zone(JSZ), which was formed in the thickened Paleo-Jinshajiang orogenic belt after closure of the JinshajiangOcean. We have identified three metamorphic phases, M1, M2 and M3, corresponding to deformationstages D1, D2 and D3, in the Barrovian sequences. The metamorphic belt exhibits a metamorphic field gra-dient from chlorite to biotite, garnet, staurolite–kyanite and sillimanite grades. Inclusions in garnet andstaurolite (chlorite, mica, quartz, feldspar, ilmenite and graphite) indicate that M1 reflects greenschistfacies metamorphism. Pervasive M2 metamorphism formed a dominant S1 schistosity within theBarrovian sequence. Peak metamorphic conditions for metapelites of the garnet–staurolite and stauro-lite–kyanite grade were �580 �C and �0.65 GPa according to petrogenetic grids. Peak metamorphismwas in conditions of �635 �C and 0.50 GPa for the metapelites and �650 �C and �0.61 GPa for amphibo-lites in sillimanite grade. Greenschist facies retrograde metamorphism, M3, followed D2 deformation, atop-down-to-southeast shear in the JSZ. The D3 deformation is characterized by well-developed brittlefaults with fault gouge and breccia. Zircon grains from an amphibolite sample have cores with igneousoscillatory zoning and metamorphic rims. However, the metamorphic rims are too narrow to analyze.Laser ablation-inductively coupled plasma mass-spectrometry (LA-ICPMS) analyses of the igneous coresof zircons yielded a crystallization age of 242 Ma. 40Ar/39Ar dating of white mica from a garnet-schistgave a plateau age of 224 Ma. The peak metamorphism is thus limited to be between 242 and 224 Ma.We thus suggest a tectonic shift from collision to extension of for the Permian to Triassic Paleo-Jinshajiang orogenic belt.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

The Paleo-Jinshajiang Orogenic Belt (POB) includes the curvilinearJinshajiang Suture Zone (JSZ), which is an ophiolitic mélange zonecontaining peridotites, gabbros and basalts interlayered with vol-canoclastic sediments, and metamorphic complexes. The POBwas formed by closure of the Jinshajiang Ocean, a branch of thePaleotethys, and later collision between the Qiangtang Block andthe Songpan-Ganze terrane during Triassic (Fig. 1; Mo et al.,1993; Roger et al., 2003). However, experience multi-stage tectonicoverprinting, nature of the POB is not well-understood due to poorgeological constraining. For example, based on magmatic and local

metamorphic constraints, previous studies proposed various mod-els for the formation of the POB. The eastward subduction of theJinshajiang Oceanic plate and immediately collision produced thePOB was proposed based on occurrences of subduction-generatedplutonic arc complexes in eastern Qiangtang Block andblueschist-bearing metamorphic belt in the northern QiangtangBlock (e.g. Hou et al., 2003; Kapp et al., 2000), while westward sub-duction was suggested according to a west-vergent imbricatedthrust system (Reid et al., 2005a). The third school advocates adouble-vergent subduction of the Jinshajiang Oceanic plate onthe basis of buildup of accretionary prism and backarc volcanicsystem (Leeder et al., 1988). Therefore, re-establishing of the defor-mation–metamorphic sequence is the key to understand the nat-ure of the POB.

In the eastern Tibetan Plateau, a sequence of biotite-, garnet-,staurolite-, kyanite- and sillimanite-bearing metamorphic zonescrops out from the eastern JSZ. Previous workers termed these

http://crossmark.crossref.org/dialog/?doi=10.1016/j.jseaes.2015.09.003&domain=pdf

http://dx.doi.org/10.1016/j.jseaes.2015.09.003

mailto:[email protected]

http://dx.doi.org/10.1016/j.jseaes.2015.09.003

http://www.sciencedirect.com/science/journal/13679120

http://www.elsevier.com/locate/jseaes

Fig. 1. Regional tectonic map showing locations of terranes and suture zones adjacent to the JSZ (Modified after Reid et al., 2005a and Yan et al., 2012).

W. Cao et al. / Journal of Asian Earth Sciences 113 (2015) 748–765 749

metamorphic units the Eaqing complex as a basement unit withinthe JSZ based on a pre-Devonian age of 423 ± 40 Ma by Rb/Sr datingof a sillimanite-bearing mica schist (Wang et al., 2000a,b). How-ever, muscovite from a mica schist in the metamorphic complexwas dated at 225–215 Ma using the 40Ar/39Ar thermochronometry,which indicates a Triassic cooling age of the metamorphic units(Reid et al., 2005b). Additionally, no detailed research on metamor-phic evolution of the complex was conducted. The lack of consen-sus on the age of the metamorphic units has resulted in amisinterpretation in the relationship between tectonic activityand metamorphism in this Paleotethys domain, and a poor under-standing of the nature of this part of the Paleotethys orogenic belt.

In this contribution, we conducted petrographic observationand thermobarometrical estimation on six representative metape-lites and one amphibolite to study metamorphic evolution of thecomplex. U–Pb dating of zircon from an amphibolite, and 40Ar/39Arthermochronometry of muscovite from a metapelite wereemployed to constrain timing of the metamorphism. A P–T–t pathfor the Barrovian belt is delineated combining conventionalthermobarometry, U/Pb geochronometry and 40Ar/39Ar ther-mochronometry. In combination with previously published data,we present a new tectonic model from building of the Paleo-Jinshajiang orogenic belt, to a shift to collapse, and to laterexhumation of the metamorphic complex.

2. Geological background

2.1. Tectonic framework

The Jinshajiang suture is inferred to extend from the Pamirseastward to either the Ailaoshan suture (Jian et al., 2008;Metcalfe, 2006; Roger et al., 2003; Wang et al., 2000b) or theGanze-Litang suture (Yang et al., 2012), where the divide locates

in the Yushu region (Fig. 1). In the eastern Tibetan Plateau, theeastern segment of the JSZ, which is to the southeast of the Yushuregion and equals to the south segment of the JSZ in Yang et al.(2012), separates the Qiangtang Block to the west from the Yidunarc to the east (Fig. 1). The Qiangtang Block is separated from theLhasa Block by the Bangong-Nujiang suture further west, whereasthe Yidun arc is separated from the Songpan-Ganze terrane by theGanze-Litang suture in the east. To the south, the JSZ, Yidun Arc,Ganze-Litang suture and Songpan-Ganze terrane merge into west-ern margin of the Yangtze Block (Wang et al., 2011, 2013) (Fig. 1).

The Qiangtang Block is located to the south and west of the JSZand to the north and east of the Bangong-Nujiang suture (Fig. 1).Internal division of the Qiangtang Block is still under debate. Thegeology of the Qiangtang Block varies from the southeast to thenorthwest. In the southeastern part, the Qiangtang Block is mainlycomposed of weakly metamorphosed Paleozoic carbonates andclastic rocks unconformably overlain by Permian and Triassicshale, sandstone and volcanic strata (Reid et al., 2005a). TheJiangda-Weixi volcanic arc in the eastern margin of the QiangtangBlock consist of Triassic tholeiitic, calc-alkaline and shoshoniticvolcanic rocks followed by Late Triassic intra-plate volcanic rocks,including calc-alkaline basalt, andesite, dacite, and rhyolite, in theinner part of the Block. Occurrence of the arc setting, along withdetermined ages, was argued to be generated by westward sub-duction of the Jinshajiang Oceanic block (Hou et al., 2003; Moet al., 1994). In the northwestern part, ultrahigh-pressure (UHP)metamorphic rocks, which include garnet-bearing blueschist andeclogite (Li et al., 2006), and Late Paleozoic shallow marine stratacrops out the Qiangtang Block (Yin and Harrison, 2000 and refer-ences therein). The metamorphic rocks are regarded as an exten-sional core complex from the tectonic mélange underthrustbeneath the Qiangtang Block during southward/westward flat sub-duction of the Jinshajiang oceanic crust (Kapp et al., 2000, 2003;

750 W. Cao et al. / Journal of Asian Earth Sciences 113 (2015) 748–765

Yin and Harrison, 2000 and references therein). These metamor-phic rocks were exhumed by crustal-scale normal faults duringthe Late Triassic to the earliest Jurassic (Kapp et al., 2000, 2003;Yin and Harrison, 2000 and references therein). However, theoccurrence of the metabasites in the Qiangtang Block is arguedto form along a Paleotethyan suture named Longmu Co-Shuanghu suture (e.g. Li et al., 2006; Zhai et al., 2007). This sutureis argued to separate the Qiangtang block into the Northern Qiang-tang and the Southern Qiangtang blocks (e.g. Zhang et al., 2006).

TheYidunArc liesbetween the JSZ to thewestand theGanze-Litangsuture to the east (Fig. 2A). The arc is divided into eastern andwesternsegments on the basis of stratigraphy and lithology (Chen et al., 1987;Reid et al., 2005a; Wang et al., 2011, 2013). The eastern Yidun Arc iscomposedofmudstone, volcanic rocks and intrusionswhichweregen-erated by west-directed subduction of Ganze-Litang oceanic block(Hou et al., 2003;Mo et al., 1994; Reid et al., 2005a). The volcanic rocksare mostly calc-alkaline andesites, locally accompanied by basalt-rhyolite and shoshonitic assemblages (Mo et al., 1994). A sequenceof plutons with U–Pb zircon ages of 215–225Ma (Reid et al., 2007;Weislogel, 2008) lies along the western side of the Ganze-Litangsuture. Further to the west, Cretaceous plutons with U–Pb zircon agesof 94–104 Ma, are interpreted as products of intra-continental exten-sion, intrude the eastern Yidun Arc (Reid et al., 2007). The westernYidunArc, termed theZhongzaBlock, is composedofPaleozoic carbon-ates, intercalated mafic volcanic rocks and later intrusions (Reid et al.,2005a). The Permian basalts exposed in this part of the arc are compo-sitionally similar to the Emeishanbasalts,whichwas argued to form inan extensional setting by the same mantle plume (Xiao et al., 2008).Previous studies suggest that the Zhongza Block is a micro-continental fragment rifted from the Yangtze Block during formationof the Ganze-Litang ocean (Chang, 1997; Chen et al., 1987; Zhanget al., 1998), closure of which formed the current Ganze-Litang suture(Fig. 1). No age data are available for the later intrusions in thewesternYidun Arc, but they are inferred to be Triassic or younger according tosimilar lithology and age trend (Reid et al., 2007).

2.2. Stratigraphic units of the JSZ

The JSZ is an arcuate feature striking east–west in the westernpart, but bending to the NS-striking in the eastern part(Figs. 1 and 2A). The eastern part of the JSZ separates the YidunArc to the east from the Qiangtang Block to the west. Previousresearch divided the eastern segment of JSZ into four differenttectono-stratigraphic units according to their stratigraphic rela-tions and ages: the Eaqing Metamorphic Complex, the JinshajiangMélange, and the Gajinxueshan and Zhongxinrong Groups (Wanget al., 2000a,b; Xiao et al., 2008). The Gajinxueshan Group was sep-arated into two formations, Yangla and Jiaren in Wang et al.(2000b). Ages of the four units are still controversial, especiallyfor the Eaqing Complex; it was assigned as either Archean (Wanget al., 2000a) or Proterozoic (Wang et al., 2000b) (Fig. 3).

The Eaqing Complex, focus of our study, refers to metamorphicrocks exposed along the Jinshajiang River that exhibit an east towest gradient from chlorite grade, to biotite, garnet, staurolite–kyanite and sillimanite grades (Fig. 2; Wang et al., 2000b). Theserocks are metapelites, amphibolites and marbles (Wang et al.,2000b), ranging in grade from greenschist to upper amphibolitefacies. The Eaqing complex was attributed to be a basement unitbase on Rb–Sr whole rock dating (Wang et al., 2000a,b). Possibleblueschist location was mentioned in the geological report forRegional Geological Maps of Mangkang (1:200,000 scale); how-ever, the possible high pressure belt was not found during our fieldtrip in the area.

The other three stratigraphic units are located to the south ofour research area (Wang et al., 2000a,b). Following we briefly sum-marize their characteristics. The Jinshajiang Mélange consists of

scattered outcrops along the suture in Xiaruo-Tuoding, Shusong-Gongka and Xumai-Xuedui areas (Wang et al., 2000b). It is mostlycomposed of serpentinized peridotites, gabbros, and mafic volcanicrocks with intercalated limestone and radiolarian cherts (Zhanget al., 1994; Zi et al., 2012). U–Pb dating of zircon from gabbro-anorthosite, gabbro and trondhjemite yielded ages ranging from320 to 347 Ma (Jian et al., 2008, 2009a; Zi et al., 2012) which areinterpreted as the formation age of oceanic crust of the paleo-Jinshajiang Ocean. Plagiogranite, tonalite and plutonic arc com-plexes from the mélange yielded ages of 260–300 Ma (Jian et al.,2008; Wang et al., 2000b; Zi et al., 2012), and these are taken asthe age of subduction. On the basis of geochemical similarity, theJinshajiang ophiolite is thought to correlate with ophiolitic rocksin the Ailaoshan zone (Jian et al., 2009a,b; Metcalfe, 2006; Wanget al., 2000b). The Zhongxinrong Group consists of Late Permianto Middle Triassic weakly metamorphosed sandstone and siltstoneand carbonaceous slate with interbeds of limestone and volcanicrocks (Wang et al., 2000b). Whole-rock Rb–Sr dating of granitesthat intruded in the Group yield ages of 227 ± 2 Ma and255 ± 8 Ma, and thus are interpreted as syn-collisional plutons(Wang et al., 2000b). The Gajinxueshan Group is composed of car-bonates, clastic sediments and basic volcanic rocks (Wang et al.,2000b; Zhu et al., 2011). No reliable geochronologic data haveyet been obtained for these rocks, but they are thought to be LateDevonian to Carboniferous, perhaps even early Permian accordingto fossil evidence of corals and conodonts (Wang et al., 2000b; Zhuet al., 2011).

A series of Triassic granitic to granodioritic plutons intruded inthe fore-mentioned units along the JSZ. The Suwalong Pluton(Fig. 2) were dated to be 239–253 Ma by zircon U/Pb dating, andwas interpreted to be syn- and post-collisional plutons (Reidet al., 2007). Other plutons that intruded in the JSZ, such as Batangand Yarigong plutons (Fig. 2) have not been studied, but they wereinterpreted to be syntectonic as the Suwalong Pluton due to similaroccurrence and petrography.

3. Structural geology of the Barrovian belt

The Barrovian belt has experiences three stages of deformation,D1, D2 and D3, according to textural relationships from both fieldand microscopic scales. Among these three stages, D1 and D2

involved ductile deformation, whereas D3 recorded brittle defor-mation. The stage before deformation is defined as pre-D1, as indi-cated by foliation S0 (Fig. 4A).

Deformation D1 is defined by foliation S1 and lineation L1. On anoutcrop scale, S1 foliation is a penetrative schistosity, which is thecleavage of rootless folds within schist, because S1 totally replacesS0 in most locations. Locally, S0 lithons are preserved with foldedquartz veins, whereas S1 is typically perpendicular to S0 (Fig. 4A).Lineation L1 is defined by alignment of metamorphic minerals,including muscovite, biotite and locally hornblende, along the S1foliation. The S1 foliation is defined by oriented muscovite and/orbiotite, folded quartz, and inclusion trails of quartz and micawithin garnet and staurolite porphyroblasts. Rotated porphyrob-lasts, S-C fabrics, mica fish, grain-shape preferred orientation, root-less fold indicators and the orientation of foliation and lineation(Fig. 4A and B) consistently indicate dextral oblique top-to-eastthrusting (Fig. 4A and E).

Deformation D2 is defined by foliation S2 and lineation L2. S2foliation is a spaced crenulation cleavage, which is recorded byfolded S1 with horizontal hinges on different scales(Fig. 4B and C) and by bending of the main schistosity (Fig. 4Fand G). At most places, foliation S2 is at a low angle to foliationS1 or sub-parallel to it. At some areas, the S2 foliation is associatedwith chloritization (Fig. 4H), indicating a retrograde stage. In theSuwalong area, sheath folds are preserved in quartz-mylonites

Fig. 2. (A) Geological map of the east-JSZ (data from our field work, Reid et al., 2005a and Regional Geological Maps of Yidun, Bomi, Mangkang and Xiongsong with scale of1:200,000). Themetamorphic zones frombiotite to sillimanite grade are coveredbyblue, lighter blue, orange–red and red in sequence. Abbreviations for theplutons are SP for theSuwalong Pluton, BP for the Batang Pluton, and YP for the Yarigong Pluton. Detailed sample localities with GPS positions are in Table A2. Intersects (B) and (C) are lowerhemisphere equal areaprojections showing theD1 andD2deformational foliations (S1 andS2) and lineations (L1 and L2).Measurements of the orientation for the twodeformationstages are fromoutcrops of BT17 andGB12, respectively, locations ofwhich are shown in (A). (D) The cross section across the east-JSZwith Barrovianmetamorphic zones followsthe A–A0 line shown in the geological map. Glaucophane position is inferred according to Regional Geological Maps of Mangkang with a scale of 1:200,000.


Fig. 3. Stratigraphic divisions for units within the Jinshajiang suture zone, with different correlation between Wang et al. (2000a) and Wang et al. (2000b). Both researchattributed the Eaqing complex as a basement unit with either Proterozoic or Archean age. The Pt and Ar are not too scale.


(Fig. 4C), which were formed during deformation D2, indicatinglocal development of high-strain ductile shear during D2. Kine-matic marks, including overturned S1 folds, rotated porphyroblastsand S-C fabrics, indicate top-down-to-the-southeast obliquesinistral-normal sense of shear (Fig. 2A and C). Therefore, the D2

deformation stage represents an extensional setting.Localized brittle faulting and related fault gouge and/or fault

breccia characterize the deformation D3. Faults along the JSZmainly strike NW to NE and dip SE to NW or NE to SW (Fig. 4D).The major fault that separates the Jinshajiang suture with theQiangtang Block is a thrust fault that thrusted the Permian stratawithin the mélange onto the younger strata in Qiangtang Block.Fault striations, directional breccias and offset strata consistentlyindicate sinistral strike-slip and top-up-to-NE/NW thrusting(Fig. 4D).

4. Analytical procedures

Typical metamorphic minerals in the Barrovian zones were ana-lyzed with a JEOL JXA-8800R electron microprobe (EMP) equippedwith four wavelength dispersive spectrometer at the ElectronProbe Laboratory, Chinese Academy of Geological Sciences, Beijing.

The analytical conditions were set to 15 kV accelerating voltage,with 10 nA beam current, and 1 lm beam diameter for all phases.Natural minerals were used as standards. Sample GB15-6 and JS11-1was analyzed using a JEOL 8900 electron microprobe at the Univer-sity of Nevada in Las Vegas. That microprobe is equipped with fourwavelength dispersive detectors and an Oxford energy dispersive(EDS) detector. Mineral analyses on this machine were performedunder the conditions of 20 kV accelerating voltage and 10 nA beamcurrent, with a beam size of 10 lm and 30 s peak counting time.Below we summarize the mineral chemistry of major mineralsfor P–T estimation. Representative mineral composition is listedin Table 1 for the staurolite–kyanite grade and Table 2 for silliman-ite grade minerals.

We dated a sample of amphibolite (BT08-1), from the silliman-ite grade using the zircon U–Pb geochronology. The sample wascrushed, and conventional heavy liquid and magnetic techniqueswere used to separate zircon grains from the sample. Selectedzircon grains were mounted together for imaging by backscatteredelectron (BSE) and cathodoluminescence (CL) detectors. The grainswere then analyzed by LA-ICPMS at the State Key Laboratory ofGeological Processes and Mineral Resources, China University ofGeosciences, Wuhan. The LA-ICPMS is equipped with an Agilent7500a system for acquiring ion-signal intensities and GeoLas

Fig. 4. Pictures showing structural style and metamorphic mineral assemblages of the three deformation stages within the JSZ. (A) a remnant fold indicates that cleavage S1replaced a Late Permian S0 pelite bed. Note quartz vein along S0. S1 is a crenulation cleavage that cross-cuts the folded quartz vein. (B) Perpendicular relationship between S1 andS2 inhingepart of a S1 fold inmica-schist, viewing toward theeast. See a rockhammerat theupper right corner for scale. (C) Sheath fold in the Jinshajiangductile shear zone showsanear-parallel relationship betweenS1 and S2, viewing toward the east. (D) and (A) Steep-dippingNW-striking faultwith fault gouge andbreccia, viewing toward east. Scratchonthe fault surface dips to the south, showing northwestward transpression. (E) Garnet–quartz–mica-schist with banded quartz and mica and rotated garnet porphyroblast.Rectangle quartz grains occur along preferred orientation in themica band indicates penetrative S1 foliation, and inclusionswithin the garnet indicate schistosity S1with dextralstrike-slip movement. (F) Penetrative S1 schistosity is replaced by spaced S2 crenulation cleavage in staurolite–garnet–quartz mica schist with a rotated porphyroblast. Therotated garnet indicates sinistral strike-slipmovement for D2. (G) Photomicrograph ofmica–quartz-schist, showing S2 spaced cleavage sub-perpendicular to S1 foliation. (H) S2 issub-parallel to S1 in chlorite-schist. Note S1 foliation is defined by muscovite, biotite and quartz. S2 foliation is defined by retrograde chlorite.



2005 as a laser sampling system. Helium was employed as a carriergas, and argon was used as a make-up gas. Zircon 91500 was usedas an external standard for U–Pb dating. To correct time-dependent drift of sensitivity and mass discrimination, every 5sample analyses were followed by two analyses of the 91,500 stan-dard. Analytical and data-reduction procedures are similar to thosedescribed by Liu et al. (2008).

A garnet mica-schist (JS07-8) from east boarder of the stauro-lite–kyanite grade is selected for muscovite 40Ar/39Ar dating. Thesample JS07-8 is a coarse-grained schist, containing coarse mus-covite which appears to be in equilibrium with minerals garnet,biotite and feldspar. Following standard procedures, the samplewas crushed, sieved, separated and then hand-picked for whitemicas. The sample, GA-1550 biotite standards (98.50 Ma; Spelland McDougall, 2003), and K-glass and CaF2 fragments were irradi-ated at U.S. Geological Survey TRIGA Reactor, Denver, CO, and ana-lyzed at the Nevada Isotope Geochronology Laboratory (NIGL) atthe University of Nevada, Las Vegas. Correction factors for interfer-ing neutron reactions on K and Ca were determined by repeatedanalysis of K-glass and CaF2 fragments. J factors were determinedby fusion of 4–8 individual crystals of GA-1550 biotite neutron flu-ence monitors. Irradiated GA-1550 biotite standards together withCaF2 and K-glass fragments were placed in a Cu sample tray in ahigh vacuum extraction line and were fused using a 20W CO2

laser. The sample was analyzed by the furnace step heatingmethod using a double vacuum resistance furnace similar to the(Staudacher et al., 1978) design. Mass spectrometer discriminationand sensitivity was monitored by repeated analysis of atmosphericargon aliquots from an on-line pipette system. Line blanks werealso measured during the process. Discrimination, sensitivity, andblanks were relatively constant over the period of data collection.Computer automated operation of the sample stage, laser, extrac-tion line and mass spectrometer as well as final data reductionand age calculations were conducted using LabSPEC software byB. Idleman (Lehigh University).

5. Petrology and mineral chemistry

5.1. Petrology

The Barrovian sequence mainly exposes medium metamorphicgrade pelitic schists and amphibolites. In the metapelites, typicalBarrovian grades, chlorite, biotite, garnet, staurolite–kyanite andsillimanite occurs from east to west. Amphibolites were found inthe sillimanite grade zone. All the Barrovian isograd belts strikeessentially N–S and are parallel with each other. We studied sam-ples from the garnet–staurolite grade, staurolite–kyanite and silli-manite grade in order to constrain metamorphic evolution of theBarrovian sequence. Detailed sample localities are marked inFig. 2A, with their GPS positions in Table A1. All samples collectedfrom these grades contain ilmenite, rutile and graphite, in additionto major phases. Abbreviations for minerals follow Whitney andEvans (2010).

5.1.1. Garnet to staurolite gradeThe garnet grade takes a large part of the Barrovian sequence. It

exposes diagnostic garnet-bearing schists in some localities of thearea. Grain sizes of garnet in northern part of this grade, less than1 mm, is smaller than those in the southern part, up to 1 cm. Otherfeatures of the garnet schists between northern and southern partof the grade are similar.

Sample JS11-1 (Fig. 5A) is a garnet schist collected around theboundary between garnet and staurolite grade, and thus we attri-bute it to be a garnet–staurolite metamorphic grade. The samplecontains 10% garnet, 20% muscovite, 15% biotite, 40% quartz, 5%

feldspar, 5% chlorite, and 5% rutile and opaque minerals. The garnetis subhedral and rounded in shape, and contains inclusions ofplagioclase, quartz, mica and opaque minerals. These mineralsdefine a previous greenschist facies metamorphism. Biotite,muscovite and quartz in the matrix define the major foliation.

5.1.2. Staurolite to kyanite gradeThe staurolite–kyanite grade is a narrower area comparing with

the other Barrovian zones. There is no clear boundary betweenstaurolite and kyanite occurrences in the field, thus we use thestaurolite–kyanite grade rather than separating them. Typical min-eral assemblage of St + Grt + Bt + Ms + Pl + Qz is mostly developedin metapelites. Stable coexistence of garnet and kyanite is uncom-mon in this grade.

Sample GB16-2 (Fig. 5B) is a staurolite schist, and contains 30%quartz, 20% muscovite, 15% plagioclase, 15% biotite, 10% garnet, 5%staurolite, 2% chlorite and small amounts of opaque minerals.Garnet is anhedral in shape, and up to 1 cm in diameter. Garnetcontains abundant inclusions, such as quartz, feldspar biotite, mus-covite, chlorite and opaque minerals (e.g. graphite). The inclusionsare more abundant in garnet cores, and less abundant in the rims.Garnet rims are usually irregular, showing evidence of later resorp-tion. Staurolite is subhedral to anhedral in shape, and up to 3 cm. Itis usually poikiloblastic in shape, with inclusions of quartz, feld-spar and mica. Two groups of white mica exist in the thin section.Group 1 is elongate, and defines the major foliation (S1). Group 2 isplate shaped, crosscutting S1, and is usually associated withbiotite. Biotite is dark brown in color, and is usually big in size,up to 5 mm. The quartz grains are usually anhedral in shape.Quartz grains display subgrain rotation recrystallization. Somestaurolite grains have a ring of sericite grains at the rims.

Stable coexistence of kyanite and garnet occurs in rare circum-stance. Sample BT2009K was collected in the staurolite–kyanitegrade. The sample contains intergrown biotite and kyanite grains(Fig. 5C). This texture indicates that it may have experienced thefollowing mineral reaction: staurolite + chlorite = biotite + kyan-ite + H2O. The sample was subjected to retrograde metamorphism,with garnet grains retrogressed to sericite and chlorite.

5.1.3. Sillimanite grade (GB14-7, BT13-8, BT17-1-1, and BT18-2)The sillimanite grade is at the western end of the Barrovian

sequence, and is adjacent to the Batang and Suwalong plutons(Fig. 2A). Sillimanite-bearing schist is a common rock within thisgrade; a few localities in southern part of the sillimanite gradeexpose tourmaline-bearing schists. Four samples were selected tostudy the sillimanite grade metamorphic conditions, with GB14-7collected at southern part of the Barrovian zone, and BT13-8,BT17-1-1 and BT18-2 collected at northern part of the Barrovianzone (Fig. 2A).

GB14-7 is a sillimanite schist collected around Suwalong area(Fig. 2A). The sample contains 40% biotite, 25% plagioclase, 10%potassium feldspar, 10% sillimanite, 5% garnet, 5% quartz and asmall amount of opaque minerals. Biotite in the sample is largein size, up to 5 mm, and dark brown in color. K-feldspar iscommon, showing cross-hatched twinning in the sample. Mus-covite is much finer grained than biotite, and its presence showsthat metamorphic conditions were still below the muscovite dehy-dration reaction conditions. In Fig. 5D, the texture of the sillimanite–garnet–mica-schist shows a potential reaction as: garnet +muscovite = biotite + sillimanite. Quartz grains commonly showundulose extinction indicating grain boundary migrationrecrystallization.

Sample BT13-8, BT17-1-1 and BT18-2 were collected aroundthe town of Batang in northern part of the sillimanite grade. Thesample BT13-8 has 15% garnet, 25% plagioclase, 35% quartz, 10%biotite, 10% potassium feldspar, 3% sillimanite and 2% opaque

Table 1Representative analyses of minerals from the garnet–staurolite grade and the staurolite–kyanite grade.

Garnet–staurolite grade Staurolite–kyanite gradeSample no. JS11-1 GB16-2

Grt-core Grt-rim Bt Pl Ms Grt-core Grt-rim Bt Pl Ms StSpot no. 47 19 142 94 mtx1-13 37 3 24 mtx1-2 mtx1-15 5

SiO2 35.77 36.49 35.07 67.70 46.94 36.33 36.87 35.01 66.17 46.16 29.22TiO2 0.11 0.00 1.62 0.03 0.25 0.14 0.03 1.61 0.00 0.17 0.71Al2O3 21.20 21.52 19.91 19.23 35.57 21.23 21.73 19.67 20.08 36.26 54.26FeO 29.74 38.56 22.63 0.12 0.85 26.24 37.99 22.22 0.06 0.88 11.95MnO 8.97 1.24 0.03 0.00 0.00 11.71 0.50 0.07 0.00 0.00 0.31MgO 0.80 2.44 7.73 0.00 0.54 0.91 2.58 8.33 0.01 0.47 1.71CaO 3.53 0.48 0.02 0.40 0.00 4.60 1.51 0.02 0.81 0.01 0.00Na2O 0.04 0.06 0.07 11.38 1.65 0.00 0.06 0.12 11.24 1.83 0.00K2O 0.00 0.00 8.46 0.02 8.68 0.00 0.01 8.75 0.07 8.88 0.00Cr2O3 0.04 0.01 0.03 0.01 0.04 0.03 0.02 0.05 0.00 0.00 0.00P2O5 0.00 0.09 0.01 0.00 0.00 0.00 0.02 0.02 0.15 0.00F 0.28 0.00 0.04 0.30 0.16 0.00Cl 0.03 0.00 0.00 0.02 0.00 0.01Y2O3 0.26 0.00 0.00 0.00Total 100.44 100.88 95.58 98.88 94.55 101.18 101.31 96.18 98.75 94.69 98.16

Si 2.92 2.95 5.33 2.99 6.23 2.93 2.95 5.31 2.94 6.13 7.99Ti 0.01 0.00 0.19 0.00 0.03 0.01 0.00 0.18 0.00 0.02 0.15Al 2.04 2.05 3.57 1.00 5.56 2.02 2.05 3.52 1.05 5.68 17.47Fe2+ 2.01 2.61 2.88 0.00 0.09 1.74 2.55 2.82 0.00 0.10 2.73Fe3+ 0.02 0.00 0.00 0.03 0.00 0.00Mn 0.62 0.08 0.00 0.00 0.00 0.80 0.03 0.01 0.00 0.00 0.07Mg 0.10 0.29 1.75 0.00 0.11 0.11 0.31 1.88 0.00 0.09 0.70Ca 0.31 0.04 0.00 0.02 0.00 0.40 0.13 0.00 0.04 0.00 0.00Na 0.00 0.01 0.02 0.97 0.43 0.00 0.01 0.04 0.97 0.47 0.00K 0.00 0.00 1.64 0.00 1.47 0.00 0.00 1.69 0.00 1.50 0.00Cr 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00P 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00F 0.13 0.00 0.02 0.14 0.01 0.00Cl 0.01 0.00 0.00 0.00 0.00 0.00Y 0.01 0.00 0.00 0.00T (CAT) 8.05 8.05 15.53 4.99 13.93 8.04 8.03 15.61 5.03 14.00 29.10

Fe2+/(Fe2+ + Mg) 0.95 0.90 0.94 0.89 0.80Mg/(Fe2+ + Mg) 0.38 0.53 0.40 0.49K/(Na + Ca + K) 0.78 0.76


minerals. The garnet is anhedral and corroded forming a poikilo-blast shape. The quartz grains display grain boundary diffusionrecrystallization. However, there is no preferred foliation definedby minerals. BT17-1-1 and BT18-2 display similar texture. Garnetin the two samples is subhedral in shape, and is poikiloblastic inshape. Muscovite, biotite, quartz and feldspar are other majorphases in the sample; but in contrast with sample BT13-8, theyare aligned with the major foliation S1.

5.1.4. Amphibolites (GB13-1)An amphibolite (GB13-1) was collected from the sillimanite

grade around Suwalong area, southern part of the Barrovian zone.The sample intercalated with other units in the sillimanite grade. Itcontains typical mineral assemblages of hornblende, plagioclase,biotite, potassium feldspar, and quartz. Hornblende and plagio-clase are subhedral, showing no preferred orientation.

5.2. Mineral chemistry

Biotite is widely distributed in the Barrovian zones. The matrixbiotite grains change in color from dark greenish to dark brownfrom the garnet grade to the sillimanite grade. Biotite from thematrix of the garnet–staurolite grade sample has a Ti content from0.08 to 0.09, and XMg (Mg/(Fe + Mg)) values from 0.36 to 0.41. Inthe staurolite–kyanite grade, matrix biotite has Ti contents andXMg values in the range of 0.16–0.22 p.f.u. (per formula unit) and0.36–0.42, respectively. Some biotite grains in the sillimanite gradecontain oriented sagenitic rutile. In sillimanite grade, biotite has

XMg values ranging from 0.31 to 0.44, and Ti contents from 0.21to 0.30 p.f.u.

Garnet occurs in the garnet, kyanite-staurolite and sillimanitegrades. Garnets in samples JS11-1 from the garnet–staurolite gradeand GB16-2 from the kyanite–staurolite grade have distinctivecompositional zoning (Fig. 6; Table 1). Mole fractions (X) for ana-lyzed garnet grains in JS11-1 are 0.66–0.86 for almandine compo-nent, 0.03–0.10 for pyrope component, 0.09–0.01 for grossularcomponent and 0.20–0.03 for spessartine component (Fig. 6B).Mole fractions for garnet grains in GB16-2 are 0.57–0.84 for alman-dine, 0.04–0.10 for pyrope, 0.26–0.01for spessartine and 0.15–0.05for grossular (Fig. 6C). The increase in almandine and pyrope anddecrease in spessartine from core to rim (‘bell’ shape distributionfor Mn) in these garnets indicates that prograde zoning patternswere preserved and that thresholds for conditions for cation diffu-sion were not reached (Spear, 1993). Therefore, garnet rim compo-sitions are used for calculating peak P–T conditions. Garnet rims inother samples fall within the compositional ranges of JS11-1 andGB16-2.

Staurolite grains are essentially homogeneous, with no signifi-cant zoning. Small variations in FeO (11.74–14.49 wt%) and MgO(0.54–2.20 wt%) exist between grains. XFe (Fe/(Fe + Mg)) valuesrange from 0.92 to 0.94, and Al/(Al + Si) ratios are from 0.68 to0.70. TiO2 contents are usually 0.48–0.67 wt%.

EMP analytical results indicate that kyanite is almost purealuminosilicate, with small amounts of FeO (0.1–1.12 wt%) and CaO(<0.03 wt%). Every analyzed sillimanite grain has small amountsof FeO, from 0.28 to 1.51 wt%. Other chemical components, such

Table 2Representative analyses of minerals from metapelites and amphibolite within the sillimanite grade.

Sample no. GB14-7 BT13-8 BT17-1-1 BT18-2 GB13-1

Grt Bt Pl Sill Grt Bt Pl Grt Bt Pl Grt Bt Pl Amp PlSpot no. grt1-2 mtx-1 mtx-2 11 grt2-9 mtx2-10 mtx7-4 grt5-2 mtx4-2 mtx2-3 grt4-3 mtx4-1 mtx4-4 9 10

SiO2 37.82 35.18 62.69 38.69 37.97 36.09 58.32 36.36 32.83 59.57 36.11 34.29 61.76 48.37 57.89TiO2 0.01 2.16 0.00 0.00 0.05 2.66 0.00 0.00 2.52 0.00 0.13 2.41 0.09 0.92 0.00Al2O3 20.46 19.27 23.23 60.20 22.51 18.83 26.19 21.4 18.95 25.52 20.56 19.33 22.31 7.98 26.56FeO 30.25 20.82 0.21 0.65 28.37 19.58 0.34 32.12 24.32 0.04 31.53 20.95 0.21 12.61 0.06MnO 7.24 0.19 0.00 0.00 6.19 0.33 0.00 7.25 0.43 0.00 7.53 0.21 0.13 0.23 0.04MgO 2.64 8.34 0.00 0.21 2.43 8.63 0.00 1.84 6.64 0.09 2.56 9.08 0.02 13.17 0.00CaO 1.64 0.02 4.96 0.00 2.38 0.11 6.64 1.6 0.07 5.78 1.13 0.08 4.35 11.68 9.61Na2O 0.00 0.15 8.60 0.04 0.10 0.02 7.88 0.11 0.18 8.08 0.14 0.22 9.61 0.85 5.63K2O 0.00 9.55 0.05 0.27 0.07 9.27 0.13 0.00 9.34 0.17 0.03 8.93 0.64 0.60 0.09Cr2O3 0.02 0.00 0.00 0.00 0.01 0.00 0.00 0.13 0.00 0.00 0.17 0.00 0.00 0.07 0.00NiO 0.01 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.05 0.03Total 100.08 95.67 99.74 100.06 100.09 95.52 99.5 100.81 95.28 99.25 99.89 95.5 99.12 96.53 99.9

Si 3.05 5.38 2.78 1.05 2.99 5.47 2.62 2.94 5.17 2.67 2.95 5.25 2.78 7.09 2.59Ti 0.00 0.25 0.00 0.00 0.00 0.30 0.00 0.00 0.30 0.00 0.01 0.28 0.00 0.10 0.00Al 1.94 3.47 1.21 1.93 2.11 3.37 1.39 2.11 3.52 1.35 1.98 3.49 1.18 1.38 1.40

2.03 2.66 0.01 0.00 1.99 2.48 0.01 2.17 3.20 0.00 2.10 2.68 0.01 1.41 0.000.01 0.00 0.01 0.00 0.00 0.00 0.00 0.05 0.00 0.13

Mn 0.49 0.02 0.00 0.00 0.41 0.04 0.00 0.50 0.06 0.00 0.52 0.03 0.00 0.03 0.00Mg 0.32 1.90 0.00 0.01 0.28 1.95 0.00 0.22 1.56 0.01 0.31 2.07 0.00 2.88 0.00Ca 0.14 0.00 0.24 0.00 0.20 0.02 0.32 0.14 0.01 0.28 0.10 0.01 0.21 1.84 0.46Na 0.00 0.04 0.74 0.00 0.02 0.01 0.69 0.02 0.05 0.7 0.02 0.07 0.84 0.24 0.49K 0.00 1.86 0.00 0.00 0.01 1.79 0.01 0.00 1.88 0.01 0.00 1.75 0.04 0.11 0.01Cr 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.01 0.00 0.00 0.01 0.00Ni 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00T (CAT) 7.98 15.59 4.98 3.00 8.02 15.44 5.03 8.11 15.74 5.01 8.06 15.63 5.06 15.23 4.95

Fe2+/(Fe2+ + Mg) 0.86 0.88 0.91 0.87Mg/(Fe2+ + Mg) 0.42 0.44 0.33 0.44 0.67K/(Na + Ca + K) 0.00 0.01 0.01 0.04 0.01

Fig. 5. Photomicrographs of the metamorphic zones showing metamorphic mineral assemblages. (A) Mineral assemblage of garnet–staurolite grade. The garnet is inclusion-rich with a bigger amount of inclusions in core than rim. Quartz, feldspar and mica define the major foliation. (B) Mineral assemblage of staurolite grade. The sample containsminerals of garnet, staurolite, mica, quartz, plagioclase, rutile and opaque minerals. Garnet is rounded in shape and is rich in inclusions; staurolite is stably coexisting withgarnet. (C) Photomicrograph of kyanite-schist showing mineral assemblage of garnet, biotite, plagioclase, tourmaline, chlorite, and rutile. (D) Photomicrograph of sillimanite–garnet–mica-schist under plane polarized light. Garnet, muscovite, biotite and sillimanite grains are intergrown with each other, showing a possible sillimanite-formingreaction from garnet.


Fig. 6. Element distribution maps and zoning patterns of a garnet grain in sample JS11-1. The BSE image (A) of garnet displays mineral inclusions with analyzed spots markedas a black dotted line. The compositional maps exhibit obvious compositional change in Mg (B) and Mn (C), whereas there is only a slight change in Ca (D). (E) Compositionalzoning of a garnet in JS11-1. Compositional zoning in garnet porphyroblasts shows increases in almandine and pyrope components and decreases in spessartine componentfrom core to rim, indicating prograde zoning. X coordinate shows analyzed position within the garnet grain, while Y coordinate shows ratios of compositions in a certain spot.


as TiO2, MnO, MgO, CaO, Na2O and K2O, are also present but makeup less than 0.3 wt% combined.

Amphibole from the amphibolites within the sillimanite grade iscalcic-amphibole. In the studied amphibolite sample, Ca on theM4site

ranges from1.78 to1.86p.f.u., total cationonAsite (Na + K + 2Ca) from0.06 to 0.29, and octahedral sites cations (Al + Fe3+ + 2Ti) from 0.29 to0.90.Using thenomenclatureof amphibolebyHawthorneet al. (2012),the amphiboles are tremolite andmagnesio-hornblende species of the

Fig. 7. Petrogenetic grid for pelites. Black solid lines are equilibria within thesystem of NKFMASH system and black dashed line is for the KMASH system (Spearet al., 1999); Black dotted lines are equilibria in the (Mn)KFMASH system and graydashed lines in the NCKASH system (Vannay and Grasemann, 1998). The browncolored cube is calculated by the Ti-in-biotite thermometer with pressureestimated at 0.50 and 0.60 GPa respectively; the red colored cube is estimated bythe GB-GBPQ thermobarometer; the blue colored cube is from the HPQ barometer.The P–T spots on the diagram contain absolute errors for each result. The thick blackarrow shows a P–T path for evolution of the Jinshajiang metamorphic complex. (Forinterpretation of the references to colour in this figure legend, the reader is referredto the web version of this article.)


Ca-subgroup. Geochemical compositions of typical plagioclases in themetapelites are shown in Tables 1 and 2. The plagioclase compositionsvary betweenmetamorphic zones. Plagioclase grains from staurolite–kyanite grade are almost pure albite (An0.01–0.04), whereas grains fromthe sillimanite grade have very slightly more CaO (An0.19–0.32). Plagio-clase grains in the amphibolites have composition of An0.43–0.61 witha very small orthoclase component (<1%) (Table 2).

Muscovite is widely distributed in all the Barrovian zones. TheXMg (Mg/(Fe + Mg)) values vary within the range of 0.23–0.58, andthe K/(K + Na + Ca) ratios range from 0.64 to 0.95. Si cations in allsamples are 5.94–6.55 per 22 oxygen atoms, with muscovite fromthe staurolite–kyanite grade having the highest values and thatfrom the sillimanite grade having the lower. The analyzed mus-covite grains have TiO2 contents up to 1.33 wt%. Small amounts ofK-feldspar exist in metapelites of the sillimanite grade and in theamphibolites. The analyzed grains have K/(K + Na + Ca) ratios rang-ing from 0.94 to 0.97 with one low value of 0.83. Green and flakygrains of chlorite occur as both prograde and retrograde mineralsin the metapelites and amphibolites. The Fe/(Fe + Mg) values ofthe chlorite show only very slight variations from 0.53 to 0.56.

6. P–T–t evolution of the Barrovian belt

6.1. Geothermobarometry

We estimated P–T conditions using exchange and net-transferreaction thermobarometers and the titanium in biotite thermome-ter. Representative samples of metapelites and amphibolites, asdescribed above, from high-grade metamorphic zones were usedfor this purpose. Due to lack of reliable thermobarometry, we didnot use the sample BT2009K to calculate peak metamorphic P–Tconditions. The calculated P–T results with absolute errors areplotted on a petrogenetic grid to show the metamorphic historyand field P–T gradient from the garnet to sillimanite grade(Fig. 7; Table 3).

Garnet-biotite (GB) thermometer (Holdaway, 2000) andgarnet–biotite–plagioclase–quartz (GBPQ) barometer (Wu et al.,2004) were utilized to calculate peak P–T conditions since mostsamples contain a mineral assemblage including garnet, biotiteand plagioclase. The GB thermometer utilizes Fe–Mg exchangebetween garnet and biotite to estimate metamorphic temperatures(Ferry and Spear, 1978). Holdaway (2000) developed a new calibra-tion of the thermometer using an average of three garnet Margulesactivity models: Berman and Aranovich (1996), Ganguly et al.(1996) and Mukhopadhyay et al. (1997). This version has provento be more accurate than others (Holdaway, 2000; Wu et al.,2004). The barometer GBPQ (Wu et al., 2004) use the garnet activ-ity model of Holdaway (2000) and has been re-calibrated to bemore accurate for samples with XAn > 17% and Xgrs > 3%. TheGB-GBPQ thermobarometer has an absolute error of ±50 �C,±0.12 GPa (Wu et al., 2004).

We also summarize metamorphic temperature conditions usingthe recently developed Ti-in-biotite thermometer (Henry, 2005)for samples with biotite to compare with results by the other ther-mometer. Biotite grains without later chloritization in ilmenite-bearing metapelites were selected for this purpose. The analyzedbiotite grains have XMg (Mg/(Mg + Fe)) values and Ti contentswithin the applicable ranges of 0.275–1.000 and 0.04–0.60 atomsp.f.u. respectively (Henry, 2005). Errors of the thermometer arededuced to be ±12 �C for temperatures higher than 700 �C and±24 �C for temperatures lower than 600 �C (Henry, 2005).

Geobarometrical calculations for amphibolites were performedusing the hornblende–plagioclase–quartz (HPQ) barometer(Bhadra and Bhattacharya, 2007). The HPQ barometer requiresspecific composition limits of both amphibole and plagioclase.For plagioclase, XAn needs to be in the range of 0.20–0.82;

amphibole requires compositional limits of Si from 6.09 to 7.27, Alfrom 0.90 to 2.65, Ti from 0.00 to 0.29, Mg from 1.79 to 4.32, Fefrom 0.76 to 2.63, Mn from 0.00 to 0.09, Ca from 1.38 to 1.88, Nafrom 0.23 to 0.77, K from 0.00 to 0.27, and total Ca, Na and K from1.73 to 2.67 p.f.u. for 23 oxygen atoms (Bhadra and Bhattacharya,2007). We carefully selected the amphibole and plagioclase thatlocate within the compositional limits for pressure estimation.The HPQ barometer has an uncertainty of 0.2 GPa (Bhadra andBhattacharya, 2007).

Application of these thermobarometers yielded a wide range ofP–T conditions for the samples from garnet–staurolite and stauro-lite–kyanite grades (Table 2). Peak metamorphic temperatures ofsamples JS11-1 and GB16-2 from the two grades are about583 �C using the GB thermometer (Holdaway, 2000). The GBPQbarometer by Wu et al. (2004) yields pressures about 1.3 GPa forthe same samples. However, the peak pressure conditions areprobably overestimated because plagioclase in these two samplesis nearly pure albite (XAn < 4%) which is far below the calibrationrange (XAn > 17%) of the barometer. Thus, we corrected the meta-morphic pressure according to petrogenetic grids from previousresearch of metapelitic equilibria (stated below). Peak tempera-tures for these samples are about 550 �C given by Ti-in-biotitethermometer, which is lower than the temperature estimated byGB thermometer. This is probably due to local reequilibration(Henry, 2005) and pressure effects on the thermometer.

For samples from the sillimanite grade, the thermobarometersyield similar P–T results. The metapelites from the sillimanitegrade yield P–T conditions of 635 ± 15 �C, 0.51 ± 0.11 GPa usingthe GB-GBPQ thermobarometer (Holdaway, 2000; Wu et al.,2004). Metamorphic temperatures for the same samples from thesillimanite grade are 638 ± 22 �C calculated using the Ti-in-biotitethermometer. Amphibolites from the sillimanite grade yield pres-sure at 0.61 ± 0.02 GPa by HPQ barometer (Bhadra andBhattacharya, 2007). The pressure estimates for the metapelitesand amphibolite are similar, within error ranges of the calculation

Table 3Summary of calculated P–T conditions in the Barrovian zones.

Sample Metamorphic grade GB-GBPQ Ti-in-biotite HP-HPQ

T (�C) P (GPa) T (�C) T (�C) P (GPa)

JS11-1 Garnet schist Grt-St 583 0.65a 549GB16-2 Staurolite schist St-Ky 583 0.65a 550GB14-7 Sillimanite schist Sill 651 0.56 616BT13-8 Sillimanite schist Sill 633 0.62 656BT17-1-1 Sillimanite schist Sill 635 0.39 638BT18-2 Sillimanite schist Sill 620 0.45 640GB13-1 Amphibolites Sill 0.60

a The pressures were overestimated by GBPQ barometer due to low XAn values, which is below calibration range of the barometer, but are estimated to their current valuesusing petrogenetic grids. See text for more details.


methods. Quartz from both grades experienced grain boundarymigration recrystallization, showing that metamorphic tempera-tures were above �500 �C (Stipp et al., 2002).

The petrogenetic grid for anatectic pelites built within the sys-tem of NKFMASH (Na2O–K2O–FeO–MgO–Al2O3–SiO2–H2O) (Spearet al., 1999) was employed to evaluate the calculated P–T resultswith phase equilibria built within the system of (Mn)KFMASH(MnO–K2O–FeO–MgO–Al2O3–SiO2–H2O) and NCKASH (Na2O–CaO–K2O–Al2O3–SiO2–H2O) system (Vannay and Grasemann,1998). The P–T results plotted on the petrogenetic grid are wellwithin reasonable limits expected for the pressures calculated usingthe GBPQ barometer. The overestimated pressure conditions werecorrected according to the P–T diagram. The mineral assemblagesfor the staurolite–kyanite grades, St + Grt + Bt + Ms + Pl + Qz + Chl +Il + C ± Tur and Ky + Grt + Bt + Ms + Pl + Qz + Chl + Il + C ± Tur,constrained the peak metamorphic pressure to be 0.50–0.80 GPa

Fig. 8. (A) Cathodoluminescence images of selected zircon grains from amphibolite samand metamorphic rims. The black circles show the positions of analyzed spots. Analysis nthe lower right corner. (B) A Th/U plot showing that the Th/U ratios of oscillatory zonindicating protolith age of the sample BT08-1. (D) REE pattern for the zircons from BT 0

on the diagram (Fig. 7).We thus estimated the pressure to have beenaround 0.65 GPa. The calculated P–T results for the sillimanite gradeplot within the sillimanite stability field, and thus are relativelyreliable.

6.2. Age of the metamorphism

The amphibolite sample (BT08-1) contains zircons with varioustextures and morphologies, although grains have rather simpleinternal structures, as shown in the CL images (Fig. 8A). Mostzircons display inherited cores with oscillatory-zoned mantlesand irregular rims (Fig. 8A). For example, the zircon grain 11(Fig. 8A) has a rhythmic zones from its core to the irregular shapedblack growth zone, and a near-rounded outermost overgrowth. Theoutmost rims are not wide enough for LA-ICPMS analysis, thus theoscillatory zones were analyzed instead.

ple BT08-1. The images show zircon grains with inherited cores, oscillatory zoningumber and ages of the zircon grains are also marked. A scale of 100 lm is shown ines of zircons in BT08-1 are close to 1. (C) Concordia plot 206Pb/238U vs. 207Pb/235U8-1. Detailed original datasets are in Table A2.


Zircons from the sample BT08-1 are 120–340 lm long withlength/width ratio of from 1:1 to 3:1 (Fig. 8A). A total of 30 analy-ses of zircons were obtained (Table 4 and Fig. 8A). Four analysesthat have significantly higher errors are excluded from the agecalculation. In the other 26 zircon grains, uranium contents varyfrom 391 to 869 ppm, and thorium contents are 225–658 ppm,yielding Th/U ratios of 0.48–0.79 (Fig. 8B). Treated as a singlegroup, the remaining 26 analyses yield a concordia U–Pb age of242.0 ± 3.8 Ma (MSWD = 0.92) (Fig. 8C). The zircon REE (rare earthelement) patterns are normalized to the chondrite values of Andersand Grevesse (1989). The REE patterns (Fig. 8D) show typicalcharacteristics of HREE (heavy rare earth element)-enrichment,Ce positive anomaly and Eu negative anomaly.

Muscovite in sample JS07-8 yielded a flat plateau age for steps4–13 out of 14, which represents 79.0% of the total 39Ar released(Fig. 9; Table 5). The three ages at lower temperatures and the laststep gave incompatible ages, and thus were discarded. Plateau ageof this sample is calculated to be 224.7 ± 1.1 Ma. Closure tempera-ture of muscovite 40Ar/39Ar dating is about 400 �C for ourmuscovitesize (e.g. Harrison et al., 2009), which is lower than peak metamor-phic temperatures. Thus the age represents a cooling age of the rock,and postdates the peak metamorphism.

6.3. Deformation, metamorphic sequences and P–T–t constraints forthe Barrovian zones

As described above, at least three deformational events, D1, D2

and D3 and three corresponding phases of metamorphism, M1,M2 and M3, are recognized. The relationship between deformationand metamorphism is shown in Table 6.

M1 metamorphism is recorded by mineral inclusions in porphy-roblastic garnet and poikiloblastic staurolite. The minerals assem-blage in the porphyroblasts and poikiloblasts includes biotite,muscovite, quartz, feldspar and ilmenite. This assemblage repre-sents a typical greenschist facies metamorphism. No reliable ther-mobarometer could be used to estimate the exact pressure andtemperature. Timing of the M1 metamorphism postdated D0. Therelationship between M1 metamorphism with later deformationis unclear, and we treat it as a pre-D1 event.

D1 is characterized by ductile deformation with eastwardthrusting. The deformational features include foliation S1 and min-eral lineation L1, S-C fabrics, tightly closed folds and inclusion trailsin garnet and staurolite. The inclusion trails are aligned with the S1foliation, which implies that the M2 metamorphismwas essentiallycoeval with the D1 deformation. Evidence for the M2 metamor-phism is widespread in the study area, as represented by the west-ward increase of metamorphic grade in the east-JSZ. Within themetamorphic zones, P–T conditions increase toward the west. Peakpressure conditions were achieved in the staurolite–kyanite grade,and then decreased in the sillimanite grade, whereas peak temper-atures were achieved in the sillimanite grade. Although the age ofthe M2 metamorphism along the east-JSZ is not well-constrained,our work suggests that it occurred in the Middle to Late Triassic.

Characterized by ductile deformation with eastward extension,the D2 deformation produced the S2 foliation and L2 mineral lin-eation, sheath folds, S-C fabrics, and a spaced crenulation cleavage.The D2 deformation could also be divided into two stages, D2a andD2b according to temperature differences. The D2a stage is repre-sented by the Jinshajiang shear zone with a top-down-to-the-southeast shear sense (Fig. 2C). Quartz grains within the shear zoneshow grain boundary migration and subgrain rotation recrystal-lization. These recrystallization mechanisms indicate deformationtemperatures exceeding 400 �C (Stipp et al., 2002). D2b stage issymbolized by chloritization along the S2 foliation. This supportsthat the M2, metamorphism produced a retrograde greenschistfacies assemblage of chlorite + ilmenite + sericite and that

chloritization happened synchronously with the D2 deformation.The D2 deformation represents an exhumation process withdecreasing pressure and temperature.

The timing of the D2 deformation and M3 metamorphism isdeduced to be Late Triassic. The Jinshajiang shear zone showsdeformation temperatures higher than 400 �C, which is close tothe closure temperature of muscovite for 40Ar/39Ar dating. Thus,the entire area was probably exhumed to a depth at which thetemperature reached about 400 �C by the Late Triassic.

D3 deformation is recorded by NNW–SSE striking brittle faults,fault gouge and/or breccia, which indicate sinistral transpression.As one of the major faults in the research area, the Jinshajiang faultis thought to have been reactivated in the Cenozoic due to thenortheastward growth of the Tibetan Plateau or extrusion tectonicsof the terrane. Southeast movement of the land mass has generateda series of medium- to small-scale transpressional faults.

7. Discussion

7.1. Origin of the Barrovian zones

The metamorphic sequence was previously assigned as theEaqing Complex mainly based on whole rock Rb–Sr geochronolog-ical dating of one sillimanite-schist (Wang et al., 2000a,b), whichyielded an age of 423 ± 40 Ma. The unit was thus attributed as abasement with the Early-Middle Proterozoic age. However, thevalidity of the age is debatable. This method may not be appropri-ate for constraining the metamorphic ages, because the separatelyanalyzed rock segments may experienced different histories, andtheir Rb/Sr systematics may behaved differently during the meta-morphic process.

In comparison with the Silurian age, our geochronological anal-yses on zircon from the amphibolite show a much younger age,242 Ma. The analyzed spots on the zircons are of igneous originaccording to their CL images, Th/U rations, and REE patterns(Flores et al., 2013). The CL images show that the zircon grains con-tain oscillatory-zoned core with irregularly shaped inner rim,which suggests either termination of zircon growth or reshapedby forces such as transportation. The analytical results byLA-ICPMS show that the inner part of the zircons has high Th/Uratios, and typical igneous signatures (positive Ce and negative Euanomalies). Thus we suggest that the zircon originated from anigneous source, with an age of 242 Myr. The outmost of the zirconrims grew on the irregularly-shaped inner texture, and form theoval to sub-rounded shaped zircon as shown from the CL images.This rim is too thin to analyze with the laser beam. From the geo-logical context, the rim is preferable to be of metamorphic origin.

The U/Pb concordia age of the igneous zircon cores, 242 Ma,predates the Barrovian metamorphism. The muscovite 40Ar/39Arage, 224 Ma, has provided lower limit of the metamorphic age.These data suggest that magmatic crystallization of zircons fromamphibolites took place at 242 Ma, synchronous with the syntec-tonic 240–241 Ma Suwalong Pluton (Reid et al., 2007). Metamor-phism of these rocks must have postdated the igneous activity at242 Ma and preceded the uplift and cooling of the rocks at224 Ma. The Triassic ages imply that the Eaqing Complex areincompatible with the interpretation of originating from basement.

Barrovian sequences are commonly interpreted to result fromcontinental collision and subsequent thickening of crusts(Thompson and England, 1984). The Jinshajiang Barrovian belt isexposed within the mélange zone of the Jinshajiang suture, whichis uncommon for existence of Barrovian zones. Although it is stillunclear with the subduction polarity of the Paleo-Jinshajiang ocea-nic plate, the subsequent collision and thickening of Jinshajiangorogenic belt could explain the formation of the sequence. Peak

Table 4Common lead corrected LA-ICP-MS zircon U–Pb analytical results for amphibolite sample BT08-1.

Sample no. U–Th–Pb ratios Ages (Ma)

Pb (ppm) Th (ppm) U (ppm) Th/U 207Pb/206Pb 207Pb/235U 206Pb/238U 208Pb/232Th 207Pb/206Pb 207Pb/235U 206Pb/238U 208Pb/232Th

Ratio 1r Ratio 1r Ratio 1r Ratio 1r Age (Ma) 1r Age (Ma) 1r Age (Ma) 1r Age (Ma) 1r

1 52.2 266 484 0.55 0.0507 0.0012 0.2685 0.0065 0.0383 0.0003 0.0124 0.0002 228 56 241 5 243 2 249 42 88 541 687 0.79 0.0588 0.0011 0.2946 0.0054 0.0363 0.0002 0.0110 0.0001 561 39 262 4 230 1 221 33 68.0 325 548 0.59 0.0558 0.0012 0.3275 0.0078 0.0424 0.0004 0.0144 0.0003 443 53 288 6 268 3 290 54 54.2 282 529 0.53 0.0514 0.0010 0.2754 0.0054 0.0389 0.0003 0.0123 0.0002 257 44 247 4 246 2 247 46 63.0 349 521 0.67 0.0518 0.0011 0.2766 0.0061 0.0387 0.0003 0.0126 0.0003 276 45 248 5 245 2 254 57 50.3 249 483 0.52 0.0526 0.0012 0.3054 0.0086 0.0421 0.0006 0.0150 0.0004 322 54 271 7 266 4 301 98 65.5 350 655 0.54 0.0532 0.0009 0.2788 0.0049 0.0380 0.0003 0.0121 0.0002 339 39 250 4 240 2 242 39 45.5 225 467 0.48 0.0510 0.0010 0.2717 0.0059 0.0386 0.0003 0.0127 0.0002 239 48 244 5 244 2 254 4

10 65.8 350 591 0.59 0.0508 0.0010 0.2715 0.0051 0.0388 0.0003 0.0124 0.0002 232 43 244 4 245 2 250 411 71 386 553 0.70 0.0489 0.0010 0.2653 0.0059 0.0392 0.0004 0.0135 0.0003 143 46 239 5 248 2 272 512 59.3 316 530 0.60 0.0509 0.0011 0.2701 0.0062 0.0384 0.0003 0.0125 0.0002 235 50 243 5 243 2 251 413 61.2 331 564 0.59 0.0506 0.0012 0.2593 0.0058 0.0372 0.0003 0.0120 0.0002 233 54 234 5 235 2 241 414 58.0 311 522 0.60 0.0501 0.0010 0.2657 0.0057 0.0386 0.0004 0.0120 0.0002 198 53 239 5 244 3 242 415 47.7 259 391 0.66 0.0504 0.0013 0.2669 0.0071 0.0384 0.0004 0.0128 0.0003 213 90 240 6 243 2 258 516 69 363 566 0.64 0.0505 0.0012 0.2661 0.0064 0.0382 0.0002 0.0126 0.0002 217 86 240 5 242 2 252 417 76 409 655 0.62 0.0485 0.0011 0.2603 0.0063 0.0389 0.0003 0.0123 0.0002 120 56 235 5 246 2 248 418 71 356 596 0.60 0.0497 0.0013 0.2762 0.0077 0.0402 0.0004 0.0129 0.0002 183 59 248 6 254 2 259 419 97 504 798 0.63 0.0499 0.0012 0.2656 0.0068 0.0387 0.0003 0.0127 0.0002 187 56 239 5 245 2 255 420 93 480 755 0.64 0.0545 0.0012 0.2939 0.0068 0.0392 0.0004 0.0127 0.0002 391 48 262 5 248 2 255 421 92 495 766 0.65 0.0524 0.0010 0.2752 0.0054 0.0381 0.0003 0.0121 0.0002 302 43 247 4 241 2 244 322 77 420 567 0.74 0.0530 0.0013 0.2770 0.0072 0.0379 0.0003 0.0124 0.0002 328 25 248 6 240 2 248 423 93 525 752 0.70 0.0511 0.0009 0.2602 0.0047 0.0369 0.0002 0.0116 0.0002 256 38 235 4 234 2 233 324 62.4 343 521 0.66 0.0514 0.0011 0.2697 0.0061 0.0380 0.0003 0.0119 0.0002 257 52 242 5 241 2 239 425 81 444 736 0.60 0.0523 0.0010 0.2675 0.0052 0.0370 0.0002 0.0115 0.0002 298 43 241 4 234 1 230 326 67.5 381 590 0.65 0.0511 0.0010 0.2738 0.0060 0.0388 0.0005 0.0127 0.0002 256 44 246 5 245 3 255 527 89 482 656 0.74 0.0524 0.0009 0.2840 0.0052 0.0392 0.0004 0.0123 0.0002 306 39 254 4 248 2 248 428 113 658 869 0.76 0.0516 0.0008 0.2669 0.0041 0.0375 0.0002 0.0114 0.0001 333 35 240 3 237 1 229 329 71 352 570 0.62 0.0590 0.0013 0.3296 0.0098 0.0399 0.0005 0.0140 0.0003 569 50 289 7 252 3 281 630 71 407 661 0.62 0.0515 0.0010 0.2585 0.0051 0.0363 0.0003 0.0112 0.0002 261 44 233 4 230 2 224 3

W.Cao

etal./Journal

ofAsian

EarthSciences

113(2015)

748–765

761

Fig. 9. Muscovite 40Ar/39Ar dating of sample JS07-8, a garnet-schist. Step-heatingstages from 4 to 13 yield a plateau age of 224.72 Ma.


metamorphic pressure for these zones is estimated to be 0.65 GPa,which is equivalent of �21 km in depth. Thus the Barroviansequence recorded evolution of the middle to lower crust of thePaleo-Jinshajiang orogenic belt, part of the Paleotethys orogen, asdiscussed later.

Heat sources of Barrovian sequences has long been discussed,either attributed to conductive thermal relaxation or synmetamor-phic magmatic heat (e.g. Ague and Carlson, 2013). This especiallyinvolves the heat source of sillimanite grade in Barroviansequences. In comparison with the typical Barrovian zone in Scot-land, where sillimanite grade has a sharp transition from kyanitegrade, the sillimanite grade in the Jinshajiang Barrovian sequencedoes not show textural evidence of sillimanite formation fromkyanite. The sillimanite-schists in JSZ generally do not show apervasive foliation, in contrast to the strongly foliated garnet tokyanite grades. The kyanite grade is restricted to a narrow area,with kyanite schists strongly foliated. Additionally, the sillimanitezones are usually surrounding the plutons, which are a possibleheat source. Thus, the sillimanite grade may be strongly influencedby, if not all from, magmatic heat; while the other Barrovian zonesmay have a preferential conductive heat source.

Table 5Muscovite 40Ar/39Ar analyses for the garnet-schist sample JS07-8.

JS07-8, muscovite, 4.58 mg, J = 0.005045 ± 0.37%4 amu discrimination = 1.0566 ± 0.48%, 40/39K = 0.0093 ± 74.72%, 36/37Ca = 0.000247 ±Step T (C) t (min.) 36Ar 37Ar 38Ar 39Ar 40Ar %40Ar⁄

1 720 12 3.416 0.187 1.781 53.656 2087.13 54.22 770 12 0.487 0.050 0.678 42.435 1124.18 88.63 820 12 0.666 0.054 1.775 121.548 3199.35 94.44 860 12 1.283 0.058 4.588 332.677 8997.72 96.15 900 12 0.809 0.038 3.670 268.017 7204.46 97.06 940 12 0.371 0.034 1.358 96.709 2599.11 96.47 980 12 0.243 0.029 0.716 50.097 1358.71 95.78 1020 12 0.185 0.027 0.488 34.827 949.052 95.59 1060 12 0.163 0.026 0.453 31.885 869.372 95.8

10 1110 12 0.143 0.029 0.392 27.874 762.728 96.011 1160 12 0.132 0.029 0.379 26.912 736.052 96.612 1210 12 0.114 0.020 0.194 13.511 383.678 94.713 1270 12 0.118 0.029 0.102 5.966 184.503 88.014 1400 12 0.413 0.021 0.243 12.774 454.737 76.8

Cumula

Total gas age = 221.4 ± 0.6, plateau age = 224.7 ± 1.1 (steps 4–13).Note: isotope beams in mV, rlsd = released, error in age includes J error, all errors 1 sigmcalculations).

7.2. Comparison of P–T results calculated from different methods

We used a variety of methods to estimate P–T conditions in dif-ferent metamorphic grades, such as the GB-GBPQ thermobarome-ter and Ti-in-biotite thermometer. These thermobarometersyielded rather different P–T results (Table 3). For samples fromthe garnet–staurolite and staurolite–kyanite grades, the GBPQthermobarometer yielded pressures about 1.30 GPa, which is toohigh for a Barrovian zone. This is probably due to the low grossularcontent in the metapelitic garnet (Xgrs < 0.05) (Holdaway, 2001)and the low anorthite content in the plagioclase (XAn < 0.17),ranges which are not appropriate for the GBPQ barometer (Wuet al., 2004). This limitation results from the application of solutionmodels for plagioclase and garnet (Holdaway, 2001). The activitycoefficient for anorthite is not well-constrained especially nearthe peristerite solvus (Ghent and Stout, 1981; Holdaway, 2001)and thus may generate large pressure errors.

For samples from the sillimanite grade, the thermobarometersapplied to both pelites and amphibolites yielded similar P–Tresults within a range of errors (Table 3). The metapelites havecompositions within the calibration ranges, because they containplagioclase with XAn values > 0.19, making the GB-GBPQ thermo-barometer applicable. The Ti-in-biotite thermometer is suitablefor temperature calculation for these rocks because the sampleshave a peraluminous composition with stable ilmenite and/orrutile. Without ilmenite and rutile, the thermometer may extrapo-late larger uncertainties. The GB-GBPQ thermobarometer andTi-in-biotite thermometer yielded similar metamorphic conditionsof �635 �C and 0.50 GPa for samples from the sillimanite grade;the HPQ barometer yielded slightly higher pressure at about 0.61GPa.

7.3. Tectonic model for the Paleo-Jinshajiang orogenic belt

Integrating the deformational and metamorphic sequencesalong the east-JSZ, we summarize a Late Permian to Early Jurassictectonic model in order to reconstruct the tectonic evolution of theeastern Paleo-Jinshajiang orogenic belt (POB), part of thePaleotethys orogen (Fig. 10).

In the Late Permian, the paleo-Jinshajiang oceanic plate startedeastward subduction (current positioning) and produced the LatePermian Yidun volcanic arc (Fig. 10A) with sequences of calc-alkaline volcanic and plutonic arc rocks.

10.38%, 39/37Ca = 0.000761 ± 12.08%% 39Ar rlsd Ca/K 40Ar⁄/39ArK Age (Ma) 1 s.d.

4.8 0.05985831 21.332356 184.40 1.623.8 0.02023677 23.679792 203.59 1.32

10.9 0.00763026 25.170456 215.67 1.3329.7 0.00299432 26.354111 225.20 1.3724.0 0.00243509 26.407892 225.64 1.378.6 0.00603817 26.199561 223.96 1.364.5 0.00994215 26.180280 223.81 1.373.1 0.01331503 26.196326 223.94 1.372.8 0.01400495 26.272840 224.55 1.372.5 0.01786873 26.374651 225.37 1.382.4 0.01850748 26.406896 225.63 1.381.2 0.0254237 26.518733 226.53 1.390.5 0.08348708 26.003353 222.38 1.421.1 0.02823508 27.104103 231.22 1.58

tive %39Ar rlsd= 100.0

a (36Ar through 40Ar are measured beam intensities, corrected for decay for the age

Table 6Correlation of deformation and metamorphism in the Jinshajiang metamorphic zone.

Deformation events Pre-D1 D1 D2 D3

D2a D2b

Metamorphic stages M1 M2 M3

Metamorphicgrades

Green-schistfacies

Amphibolite facies Amphibolite to green-schist facies Green-schistfacies

P/T Conditions 6.5 kbar/583 �C, 6.2 kbar/633 �C, 5.6 kbar/651 �C

5.0 bar/�500 �C �300 �C

Tectonic events Suturing? Orogenic wedge to orogenic plateau Crustal thinning HimalayanAge constraints 242–224 Ma 224–214 Ma Late

Cenozoic

Fig. 10. Schematic tectonic evolution model for the Jinshajiang suture zone and adjacent area under current E–W direction. The A to D show the formation and collapse of thePaleo-Jinshajiang Orogenic Belt.

Table A1Localities and lithology of the studied samples.

Sample no. GPS location Elevation (m) Lithology

JS07-8 29�34.9530N 99�03.6180E 2641 Garnet schistJS11-1 29�36.8020N 99�00.8380E 2431 Garnet schistGB13-1 29�28.5650N 99�03.7640E 2438 AmphiboliteGB14-7 29�29.8440N 99�03.0840E 2428 Sillimanite schistGB16-2 29�36.8280N 99�00.8620E 2444 Staurolite schistBT08-1 29�54.2000N 99�03.2300E 2494 AmphiboliteBT13-8 29�56.2170N 99�03.6600E 2491 Sillimanite schistBT17-1 29�57.2760N 99�03.8210E 2529 Sillimanite schistBT18-2 29�57.2790N 99�04.0370E 2529 Sillimanite schist


During the Early to early Middle Triassic, the paleo-Jinshajiangoceanic plate continued to subduct eastward underneath the YidunArc (Fig. 10B), adding to the volcanic and plutonic arc complex(Hou et al., 2003). At this time a number of major plutons wereintruded, e.g. the Suwalong granitic pluton (239–241 Ma; Reidet al., 2007), the Jiaren granitoid (231–234 Ma; Zhu et al., 2011),the Baimaxueshan granitoids (239–255 Ma; Reid et al., 2007) andthe Ludian granitoids. These plutons may have had mixed sourcesmagma sources including deep mantle, lower crust and upper crus-tal sedimentary rocks (Zhu et al., 2011). Additionally, subduction ofthe paleo-Jinshajiang oceanic block carried pelitic materials intodeep where they were metamorphosed (M1). In addition, contin-ued subduction may have rifted the western Yidun Arc from theYangtze Block to form the Ganze-Litang ocean.

During the late Middle to Late Triassic, the paleo-Jinshajiangoceanic plate finally sutured with the Qiangtang Plate, and started

thickening (Fig. 10C) to form the POB. The Ganze-Litang ocean mayhave started to subduct westward. Sequences of deeply subductedrocks were heated up to peak metamorphic conditions due to

Table A2Trace element datasets for zircon. All the analyses were performed on the LA-ICPMS at the State Key Laboratory of Geological Processes and Mineral Resources, China University ofGeosciences, Wuhan. Detailed procedures follow Liu et al. (2008).

Spot Ti49 Y Nb La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta

1 2.34 657 4.93 0.88 23.7 0.12 0.70 1.17 0.47 7.89 3.03 43.4 18.9 109 24.8 290 60.5 9977 1.263 12.6 807 4.61 9.37 46.0 3.06 13.8 4.55 1.07 14.1 4.54 59.1 23.1 128 27.3 306 60.3 9759 1.194 4.39 740 5.87 0.006 23.1 0.017 0.79 1.60 0.48 11.2 4.13 53.9 21.3 117 25.3 279 52.7 10241 2.496 4.96 596 4.39 2.53 27.7 0.50 2.37 1.82 0.68 9.07 3.34 44.1 17.2 94.3 20.2 223 44.0 10276 1.247 3.18 710 6.66 2.31 26.8 0.64 3.28 1.87 0.62 9.58 3.58 50.3 20.5 114 24.9 284 54.5 10346 1.658 2.89 610 4.58 3.58 27.9 0.72 2.98 1.75 0.58 8.08 2.76 39.1 16.6 96.7 22.5 266 55.9 10470 1.379 3.26 609 4.86 0.060 17.7 0.046 0.54 1.41 0.46 8.52 3.01 43.1 17.2 97.1 22.1 258 51.5 9909 1.83

10 2.29 542 3.50 0.011 20.6 0.033 0.53 1.23 0.51 7.61 2.80 38.2 15.3 85.7 19.0 220 44.1 10568 1.0211 3.93 848 4.03 0.39 25.5 0.22 1.67 2.38 0.84 12.7 4.49 61.2 23.9 133 29.2 335 67.3 10744 1.3012 2.39 732 3.92 0.003 20.7 0.042 0.87 1.68 0.68 11.2 3.99 54.0 21.1 117 25.4 297 60.9 10148 1.1613 4.57 611 3.78 0.15 22.4 0.045 0.54 1.57 0.58 9.21 3.40 45.7 17.8 98.0 21.0 232 45.0 10148 1.1514 2.89 450 2.71 0.15 20.2 0.087 0.56 1.14 0.41 6.09 2.19 31.1 12.7 72.3 16.1 190 39.5 10821 0.8015 3.35 963 5.46 0.29 29.7 0.085 1.13 2.54 0.85 16.1 5.62 74.1 28.7 156 32.2 350 66.3 10351 1.5016 3.82 463 2.80 0.45 20.0 0.19 1.03 1.14 0.45 7.42 2.50 33.8 13.4 74.9 16.2 184 36.5 10490 0.9617 3.73 793 5.94 1.44 30.7 0.42 2.32 2.29 0.70 13.3 4.55 60.8 23.3 125 26.6 287 52.5 10910 2.5319 2.60 1020 5.04 0.058 26.9 0.072 1.14 2.83 1.00 16.0 5.66 76.2 29.7 162 35.0 391 76.5 10097 1.3820 4.51 672 4.50 0.12 24.2 0.053 0.85 1.41 0.60 10.7 3.58 48.5 19.3 108 23.5 268 53.3 9937 1.3821 2.40 665 4.70 0.71 27.7 0.31 1.67 1.90 0.72 9.90 3.54 48.8 19.2 106 23.1 263 51.5 10585 1.5222 20.2 1182 7.39 2.80 34.8 0.47 3.32 3.37 1.18 19.6 6.88 94.3 36.1 191 39.4 419 76.9 9307 1.7723 4.03 524 3.39 0.65 29.0 0.26 1.66 1.54 0.57 7.35 2.65 36.0 15.0 84.4 19.0 226 47.7 11304 0.9924 2.61 389 2.36 13.2 37.6 1.61 5.62 1.51 0.44 5.74 1.96 26.8 10.6 61.4 14.1 172 35.8 10458 0.6625 2.55 699 5.24 0.030 26.6 0.058 0.56 1.93 0.60 10.1 3.82 51.4 19.9 113 24.6 280 53.5 10245 1.9126 2.75 593 3.02 0.21 22.7 0.18 1.53 1.62 0.78 8.56 2.96 40.9 16.5 95.7 22.4 269 57.1 10644 0.9127 4.92 735 5.18 6.89 38.6 0.90 3.64 2.14 0.72 11.1 3.94 53.1 21.3 120 26.3 301 60.0 10249 1.2728 3.61 932 6.82 1.06 34.9 0.39 2.38 2.85 0.88 14.9 5.44 72.6 28.1 150 31.9 345 64.9 9985 1.9330 3.60 544 3.43 0.54 21.4 0.15 1.25 1.20 0.45 8.18 2.87 39.4 15.7 87.9 19.6 230 46.3 10724 1.06


long-term thermal relaxation, forming peak metamorphism M2

within the Barrovian zones and deformation D1. This has led toformation of a short-period of higher grade metamorphism, asrecorded by the metamorphic rims of our zircons.

During the late Late Triassic, the thickened crust in the Jinsha-jiang orogenic belt started to collapse (Fig. 10D). Thinning of thecrust may have partially exhumed the Jinshajiang metamorphicrocks. At the same time deformation D2 led to formation of theJinshajiang shear zone. At shallow depths, pervasive chloritizationand retrogrademetamorphism occurred duringM3metamorphism.

From mid-Mesozoic to the Cenozoic, the metamorphic rockswere exhumed to the surface by D3, brittle faults formed as a resultof collision between the Indian and Eurasian plates, and the east-ward propagation of the plateau. These faults have complicatedthe styles and distribution of the faults.

8. Conclusions

A continuous Barrovian sequence of metapelites, marble andamphibolites is exposed in the east-Jinshajiang suture zone andhas experienced polyphase deformation and metamorphism. M1

metamorphism is confirmed to be greenschist facies with mineralassemblages of chlorite, white mica, biotite, quartz, feldspar, ilme-nite and graphite. M2 peak metamorphism formed Barroviansequences with dominant S1 schistosity correlated with D1 defor-mation. Peak P–T conditions for the sillimanite grade were about635 �C and 0.50 GPa for metapelites, and pressure were about0.61 GPa for amphibolites. For the staurolite–kyanite grade, peakpressure conditions are overestimated by conventional thermo-barometry, but corrected to be �0.65 GPa according to petroge-netic grids; peak temperature conditions are estimated to be�580 �C. The peak metamorphism lasted a short period and tookplace between 242 Ma and 224 Ma. During this time period, thePOB experienced a transition from compressional to extensionalsettings. Extensional D2 deformation produced top-down-to-southeast shear in retrograded metapelites from the Barrovianzone and mylonites from the shear zone. Timing of this stageis inferred to be late Triassic to early Jurassic. Deformation D3,

represented by brittle faults, formed as a consequence of Indian-Asian collision and later outward propagation of the plateau. Thesefaults have displaced the Barrovian sequences to form their currentconfiguration.

Acknowledgments

Critical and constructive reviews by Kennet E. Flores fromBrook-lyn College and an anonymous reviewer highly improved themanu-script. We appreciate Michael Wells and Paul Robinson for theircomments and suggestions on an earlier draft of the manuscript.Chanliang Wang, Xiangkun Meng, Xutuo Li and Xiangli Tang arethanked for their help with the research and field work. Thanksare also due to Jianxiong Zhou and Minghua Ren for their help withelectron microprobe analyses. This work was supported by theNational Natural Science Foundation of China (grant numbers.41172191 and 41372212), the 973 Program of Key Basic ResearchDevelopment Plan of China (grant number 2014CB440903), theSpecialized Research Fund for the Doctoral Program of HigherEducation (SRFDP 20120022110013), and the Oversea FamousProfessor Program to Nicholas Arndt (MS2011ZGDZ (BJ) 019).

Appendix A

See Tables A1 and A2.

References

Ague, J.J., Carlson, W.D., 2013. Metamorphism as garnet sees it: the kinetics ofnucleation and growth, equilibration, and diffusional relaxation. Elements 9,439–445.

Anders, E., Grevesse, N., 1989. Abundances of the elements: meteoritic and solar.Geochim. Cosmochim. Acta 53, 197–214.

Berman, R.G., Aranovich, L.Y., 1996. Optimized standard state and solutionproperties of minerals I. Model calibration for olivine, orthopyroxene,cordierite, garnet, and ilmenite in the system FeO–MgO–CaO–Al2O3–TiO2–SiO2. Contrib. Miner. Petrol. 126, 1–24.

Bhadra, S., Bhattacharya, A., 2007. The barometer tremolite + tschermakite + 2albite = 2 pargasite + 8 quartz: constraints from experimental data at unit silicaactivity, with application to garnet-free natural assemblages. Am. Mineral. 92,491–502.

http://refhub.elsevier.com/S1367-9120(15)30081-X/h0005


























Chang, C., 1997. Geology and Tectonics of Qinghai-Xizang Plateau. Science Press,Beijing.

Chen, B., Wang, K., Liu, W., Cai, Z., Zhang, Q., Peng, X., Qiu, Y., Zheng, Y., 1987.Geotectonics of the Nujiang–Lancangjiang–Jinshajiang Region. GeologicalPublishing House, Beijing.

Ferry, J.M., Spear, F.S., 1978. Experimental calibration of the partitioning of Fe andMg between biotite and garnet. Contrib. Miner. Petrol. 66, 113–117.

Flores, K.E., Martens, U.C., Harlow, G.E., Brueckner, H.K., Pearson, N.J., 2013. Jadeititeformed during subduction: in situ zircon geochronology constraints from twodifferent tectonic events within the Guatemala Suture Zone. Earth Planet. Sci.Lett. 371–372, 67–81.

Ganguly, J., Cheng, W., Tirone, M., 1996. Thermodynamics of aluminosilicate garnetsolid solution: new experimental data, an optimized model, and thermometricapplications. Contrib. Miner. Petrol. 126, 137–151.

Ghent, E.D., Stout, M.Z., 1981. Geobarometry and geothermometry of plagioclase-biotite-garnet-muscovite assemblages. Contrib. Miner. Petrol. 76, 92–97.

Harrison, T.M., Célérier, J., Aikman, A.B., Hermann, J., Heizler, M.T., 2009. Diffusion of40Ar in muscovite. Geochim. Cosmochim. Acta 73, 1039–1051.

Hawthorne, F.C., Oberti, R., Harlow, G.E., Maresch, W.V., Martin, R.F., Schumacher, J.C., Welch, M.D., 2012. Nomenclature of the amphibole supergroup. Am. Mineral.97, 2031–2048.

Henry, D.J., 2005. The Ti-saturation surface for low-to-medium pressure metapeliticbiotites: implications for geothermometry and Ti-substitution mechanisms.Am. Mineral. 90, 316–328.

Holdaway, M.J., 2000. Application of new experimental and garnet Margules data tothe garnet-biotite geothermometer. Am. Mineral. 85, 881–892.

Holdaway, M.J., 2001. Recalibration of the GASP geobarometer in light of recentgarnet and plagioclase activity models and versions of the garnet-biotitegeothermometer. Am. Mineral. 86, 1117–1129.

Hou, Z., Wang, L., Zaw, K., Mo, X., Wang, M., Li, D., Pan, G., 2003. Post-collisionalcrustal extension setting and VHMS mineralization in the Jinshajiang orogenicbelt, southwestern China. Ore Geol. Rev. 22, 177–199.

Jian, P., Liu, D., Kröner, A., Zhang, Q., Wang, Y., Sun, X., Zhang, W., 2009a. Devonian toPermian plate tectonic cycle of the Paleo-Tethys Orogen in southwest China (II):insights from zircon ages of ophiolites, arc/back-arc assemblages and within-plate igneous rocks and generation of the Emeishan CFB province. Lithos 113,767–784.

Jian, P., Liu, D., Kröner, A., Zhang, Q., Wang, Y., Sun, X., Zhang, W., 2009b. Devonian toPermian plate tectonic cycle of the Paleo-Tethys Orogen in southwest China (I):geochemistry of ophiolites, arc/back-arc assemblages and within-plate igneousrocks. Lithos 113, 748–766.

Jian, P., Liu, D., Sun, X., 2008. SHRIMP dating of the Permo-Carboniferous Jinshajiangophiolite, southwestern China: geochronological constraints for the evolutionof Paleo-Tethys. J. Asian Earth Sci. 32, 371–384.

Kapp, P., Yin, A., Manning, C.E., Harrison, T.M., Taylor, M.H., Ding, L., 2003. Tectonicevolution of the early Mesozoic blueschist-bearing Qiangtang metamorphicbelt, central Tibet. Tectonics 22, 1043.

Kapp, P., Yin, A., Manning, C.E., Murphy, M., Harrison, T.M., Spurlin, M., Lin, D., 2000.Blueschist-bearing metamorphic core complexes in the Qiangtang block revealdeep crustal structure of northern Tibet. Geology 28, 19–22.

Leeder, M.R., Smith, A.B., Yin, J., 1988. Sedimentology, palaeoecology andpalaeoenvironmental evolution of the 1985 Lhasa to Golmud Geotraverse.Philos. Trans. – Roy. Soc. Lond. 327, 107–143.

Li, C., Zhai, Q., Dong, Y., Huang, X., 2006. Discovery of eclogite and its geologicalsignificance in Qiangtang area, central Tibet. Chin. Sci. Bull. 51, 1095–1100.

Liu, Y., Hu, Z., Gao, S., Gunther, D., Xu, J., Gao, C., Chen, H., 2008. In situ analysis ofmajor and trace elements of anhydrous minerals by LA-ICP-MS withoutapplying an internal standard. Chem. Geol. 257, 34–43.

Metcalfe, I., 2006. Palaeozoic and Mesozoic tectonic evolution and palaeogeographyof East Asian crustal fragments: the Korean Peninsula in context. GondwanaRes. 9, 24–46.

Mo, X., Deng, J., Lu, F., 1994. Volcanism and the evolution of Tethys in Sanjiang area,southwestern China. J. SE Asian Earth Sci. 9, 325–333.

Mo, X.X., Lu, F.X., Sheng, S.Y., 1993. Sanjiang Tethyan Volcanism and RelatedMineralization. Geological Publishing House, Beijing.

Mukhopadhyay, B., Holdaway, M.J., Koziol, A.M., 1997. A statistical model ofthermodynamic mixing properties of Ca–Mg–Fe2+ garnets. Am. Mineral. 82,165–181.

Reid, A., Wilson, C.J.L., Shun, L., Pearson, N., Belousova, E., 2007. Mesozoic plutons ofthe Yidun Arc, SW China: U/Pb geochronology and Hf isotopic signature. OreGeol. Rev. 31, 88–106.

Reid, A.J., Wilson, C.J.L., Liu, S., 2005a. Structural evidence for the Permo-Triassictectonic evolution of the Yidun Arc, eastern Tibetan Plateau. J. Struct. Geol. 27,119–137.

Reid, A.J., Wilson, C.J.L., Phillips, D., Liu, S., 2005b. Mesozoic cooling across the YidunArc, central-eastern Tibetan Plateau: a reconnaissance 40Ar/39Ar study.Tectonophysics 398, 45–66.

Roger, F., Arnaud, N., Gilder, S., Tapponnier, P., Jolivet, M., Brunel, M., Malavieille, J.,Xu, Z., Yang, J., 2003. Geochronological and geochemical constraints onMesozoic suturing in east central Tibet. Tectonics 22, 1037.

Spear, F.S., 1993. Metamorphic Phase Equilibria and Pressure–Temperature–TimePaths. Mineralogical Society of America, Washington, D.C..

Spear, F.S., Kohn, M.J., Cheney, J.T., 1999. P-T paths from anatectic pelites. Contrib.Miner. Petrol. 134, 17–32.

Spell, T.L., McDougall, I., 2003. Characterization and calibration of 40Ar/39Ar datingstandards. Chem. Geol. 198, 189–211.

Staudacher, T., Jessberger, E.K., Dorflinger, D., Kiko, J., 1978. A refinedultrahigh-vacuum furnace for rare gas analysis. J. Phys. E: Sci. Instrum.11, 781–784.

Stipp, M., Stünitz, H., Heilbronner, R., Schmid, S.M., 2002. The eastern Tonalefault zone: a ‘‘natural laboratory” for crystal plastic deformation of quartzover a temperature range from 250 to 700 �C. J. Struct. Geol. 24, 1861–1884.

Thompson, B., England, P.C., 1984. Pressure–temperature–time paths of regionalmetamorphism. II. Their inference and interpretation using mineralassemblages in metamorphic rocks. J. Petrol. 25, 929–955.

Vannay, J.-C., Grasemann, B., 1998. Inverted metamorphism in the High Himalaya ofHimachal (NW India): phase equilibria versus thermobarometry. Schweiz.Mineral. Petrogr. Mitt. 78, 107–132.

Wang, B.-Q., Wang, W., Chen, W.T., Gao, J.-F., Zhao, X.-F., Yan, D.-P., Zhou, M.-F.,2013. Constraints of detrital zircon U–Pb ages and Hf isotopes on theprovenance of the Triassic Yidun Group and tectonic evolution of the YidunTerrane, Eastern Tibet. Sed. Geol. 289, 74–98.

Wang, B.-Q., Zhou, M.-F., Li, J.-W., Yan, D.-P., 2011. Late Triassic porphyriticintrusions and associated volcanic rocks from the Shangri-La region, Yidunterrane, Eastern Tibetan Plateau: adakitic magmatism and porphyry coppermineralization. Lithos 127, 24–38.

Wang, X., Metcalfe, I., Jian, P., He, L., Wang, C., 2000a. The Jinshajiang suture zone:tectono-stratigraphic subdivision and revision of age. Sci. China, Ser. D Earth Sci.43, 10–22.

Wang, X., Metcalfe, I., Jian, P., He, L., Wang, C., 2000b. The Jinshajiang–AilaoshanSuture Zone, China: tectonostratigraphy, age and evolution. J. Asian Earth Sci.18, 675–690.

Weislogel, A.L., 2008. Tectonostratigraphic and geochronologic constraints onevolution of the northeast Paleotethys from the Songpan-Ganzi complex,central China. Tectonophysics 451, 331–345.

Whitney, D.L., Evans, B.W., 2010. Abbreviations for names of rock-forming minerals.Am. Mineral. 95, 185–187.

Wu, C.-M., Zhang, J., Ren, L.-D., 2004. Empirical Garnet-Biotite-Plagioclase-Quartz(GBPQ) geobarometry in medium- to high-grade metapelites. J. Petrol. 45,1907–1921.

Xiao, L., He, Q., Pirajno, F., Ni, P., Du, J., Wei, Q., 2008. Possible correlation between amantle plume and the evolution of Paleo-Tethys Jinshajiang Ocean: evidencefrom a volcanic rifted margin in the Xiaru-Tuoding area, Yunnan, SW China.Lithos 100, 112–126.

Yan, D.-P., Zhou, M.-F., Robinson, P.T., Grujic, D., Malpas, J., Kennedy, A., Reynolds, P.H., 2012. Constraining the mid-crustal channel flow beneath the TibetanPlateau: Data from the Nielaxiongbo gneiss dome, SE Tibet. Int. Geol. Rev. 54,615–632.

Yang, T.-N., Hou, Z.-Q., Wang, Y., Zhang, H.-R., Wang, Z.-L., 2012. Late Paleozoic toEarly Mesozoic tectonic evolution of northeast Tibet: evidence from the Triassiccomposite western Jinsha-Garzê-Litang suture. Tectonics 31, TC4004.

Yin, A., Harrison, T.M., 2000. Geologic evolution of the Himalayan–Tibetan Orogen.Annu. Rev. Earth Planet. Sci. 28, 211–280.

Zhai, Q., Cai, L., Huang, X., 2007. The fragment of Paleo-Tethys ophiolite from centralQiangtang, Tibet: geochemical evidence of metabasites in Guoganjianian. Sci.China, Ser. D Earth Sci. 50, 1302–1309.

Zhang, K.-J., Zhang, Y.-X., Li, B., Zhu, Y.-T., Wei, R.-Z., 2006. The blueschist-bearingQiangtang metamorphic belt (northern Tibet, China) as an in situ suture zone:evidence from geochemical comparison with the Jinsa suture. Geology 34, 493–496.

Zhang, N., Cao, Y., Liao, Y., Zhang, H., Hu, D., Zhang, R., Wang, L., 1998. Geology andMetallogeny in the Ganzi-Litang Rift Zone. Geological Publishing House, Beijing.

Zhang, Q., Zhou, D., Zhao, D., Huang, Z., Han, S., Jia, X., Dong, J., 1994. Ophiolites ofthe Hengduan Mountains, China: characteristics and tectonic settings. J. SEAsian Earth Sci. 9, 335–344.

Zhu, J.-J., Hu, R.-Z., Bi, X.-W., Zhong, H., Chen, H., 2011. Zircon U–Pb ages, Hf–Oisotopes and whole-rock Sr–Nd–Pb isotopic geochemistry of granitoids in theJinshajiang suture zone, SW China: constraints on petrogenesis and tectonicevolution of the Paleo-Tethys Ocean. Lithos 126, 248–264.

Zi, J.-W., Cawood, P.A., Fan, W.-M., Wang, Y.-J., Tohver, E., 2012. Contrasting riftand subduction-related plagiogranites in the Jinshajiang ophiolitic mélange,southwest China, and implications for the Paleo-Tethys. Tectonics 31,TC2012.



































































































































































Journal of Asian Earth Sciences - Cugb · Paleotethys tectonics Tibetan Plateau abstract The...

Documents

Transcript of Journal of Asian Earth Sciences - Cugb · Paleotethys tectonics Tibetan Plateau abstract The...