petrogenesis of the early permian volcanic rocks in the chinese … · 2017-05-09 · petrogenesis...

Lithos 228–229 (2015) 23–42

Contents lists available at ScienceDirect

Lithos

j ourna l homepage: www.e lsev ie r .com/ locate / l i thos

Petrogenesis of the Early Permian volcanic rocks in the Chinese SouthTianshan: Implications for crustal growth in the Central AsianOrogenic Belt

He Huang a,b,c, Zhaochong Zhang a,⁎, M. Santosh a, Dongyang Zhang a, Tao Wang b,c

a State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Beijing 100083, Chinab Institute of Geology, Chinese Academy of Geological Sciences, Beijing 100037, Chinac 3D Geology Research Center, China Geological Survey, Beijing 100037, China

⁎ Corresponding author. Tel.: +86 10 82322195; fax: +E-mail address: [email protected] (Z. Zhang).

http://dx.doi.org/10.1016/j.lithos.2015.04.0170024-4937/© 2015 Elsevier B.V. All rights reserved.

a b s t r a c t
a r t i c l e i n f o
Article history:Received 30 September 2014Accepted 24 April 2015Available online 2 May 2015

Keywords:Volcanic rocksSouth TianshanCentral Asian Orogenic BeltTarim mantle plumeContinental growth

The Paleozoic and Early Mesozoic magmatic suites in the Central Asian Orogenic Belt (CAOB) provide importantinsights on the crustal growth and reworking process associatedwith the construction of the largest Phanerozoicorogen on the Earth. Among the tectonic blocks of the CAOB, the South Tianshan Terrane (STT) occupies thesouthwestern margin and is located adjacent to the Tarim Craton. Here we investigate the Early PermianXiaotikanlike Formation in the central part of the Chinese STT in Xinjiang in Northwest China. The formation iscomposed of a series of terrestrial volcanic lava flows and volcanic breccia, interbedded with siltstones, sand-stones and sandy conglomerates. Zircon U–Pb and Lu–Hf isotopic analysis, whole-rock major oxide, trace ele-ment and Sr–Nd isotopic data are presented for the volcanic lava flows of the Xiaotikanlike Formation exposedin the Boziguo'er, Laohutai andWensu regions. The new zircon ages fromour study, togetherwith those reportedin previous investigations on the rhyolitic lavaflow from theWensu region, suggest that the volcanic rocks of theXiaotikanlike Formation simultaneously erupted at ca. 285 Ma. The lavas of the formation show a wide range ofSiO2 (49.88 to 78.56wt.%). The basaltic rocks showSiO2 from49.88 to 53.78wt.%,MgO from3.73 to 7.01wt.% andMg# from 41 to 61. They possess slightly enriched Sr–Nd isotope signature [(87Sr/86Sr)t= 0.70495–0.70624 andεNd(t)=−0.5 to+0.6], and have trace and rare earth element patterns similar to those of oceanic island basalts(OIBs). Petrographic andwhole-rock chemical characteristics indicate that the basaltic lavaflows are dominantlytholeiitic, and were likely derived from a spinel-dominated peridotite asthenospheric mantle source. The felsiclavas of the Xiaotikanlike Formation show SiO2 in the range of 60.71 to 78.56 wt.% and display overall similarimmobile element pattern characterized by notable troughs at Nb–Ta, P and Ti and gently sloping REEs. ZirconLu–Hf analysis yields εHf(t) values of−8.7 to −0.3 for the felsic lavas from the Boziguo'er region. Geochemicaland isotopic data suggest that the felsic lava flows were likely derived from an ancient crustal source(s). Ourstudy suggests that the Xiaotikanlike volcanic lava flows erupted after the collision between the Tarim Cratonand the Kazakhstan–Yili–Central Tianshan Terrane, and that the South Tianshan Terrane was not affected byPermian Tarim mantle plume activities. Furthermore, our data also suggest that the lithosphere under the east-ern part of the STT at ~285Mahad been considerably thinned, whereas the lithosphere under thewestern part ofSTTwas still thick enough to allow the presence of a garnet-bearing source in the lower lithosphericmantle. Suchdistinction in the lithospheric thickness may be attributed to the oblique collisional tectonics. The widespreadLate Paleozoic igneous rocks from the Western Tianshan were genetically associated with regional-scale post-collisional extension. These rocks provide robust evidence for net vertical continental growth at the final stageof evolution of the CAOB.

© 2015 Elsevier B.V. All rights reserved.

1. Introduction

The Central Asian Orogenic Belt (CAOB), sandwiched between theSiberian and Europe Cratons to the north and the Tarim and NorthChina Cratons to the south, has been considered as a geologicalmuseum

86 10 82322176.

for largest Phanerozoic continental growth on earth. The crustal evolu-tion in the CAOB has been one of the focal themes in the recent years(e.g., Wilhem et al., 2012; Windley et al., 2007; Xiao et al., 2013, 2014;Zhou and Wilde, 2013; Kröner et al., 2014; Xiao and Santosh, 2014,and references therein). Some authors (e.g., Jahn et al., 2000), estimatethe proportion of juvenile crust that formed during the Phanerozoic as70% or more. The Late Neoproterozoic and Paleozoic crustal growth inCentral Asia is considered to have involved: 1) accretion of oceanic

http://crossmark.crossref.org/dialog/?doi=10.1016/j.lithos.2015.04.017&domain=pdf

http://dx.doi.org/10.1016/j.lithos.2015.04.017

mailto:[email protected]

http://www.sciencedirect.com/science/journal/00244937


24 H. Huang et al. / Lithos 228–229 (2015) 23–42

plateaus and islands together with magmatic underplating of mantle-derived materials and subduction of oceanic crust in island arc or/andactive continental margin settings (Safonova and Santosh, 2014, andreferences therein), and 2) mass transfer from the upper mantle to thecrust during post-collisional magmatism associated with orogenic col-lapse (Han et al., 1997; Long et al., 2011a, and references therein).Both mechanisms are related to the accretion-collision tectonics closelyassociated with the evolution of the Paleo-Asian domain followed bythe collisions of several terranes.

However, based on the recent identification of Permian mantleplume activities in the neighboring regions (such as the Siberian Cratonin the north as well as in the Tarim Craton in the south, see Zhou et al.,2009; Zhang et al., 2008a, 2010a; Zhang and Zou, 2013; Sobolev et al.,2011; Holt et al., 2012), some geologists suggest that crustal growth ofthe CAOB took place dominantly in anorogenic intraplate settings, unre-lated to subduction and collision (e.g., Kröner et al., 2014). For instance,in a recent article reviewing Nd–Hf isotopic studies on magmatic rocksfrom several tectonic units of the CAOB, Kröner et al. (2014) arguedthat the production of the newly formed crust during the orogenic evo-lution has been “grossly overestimated”. Alternatively, Kröner et al.(2014) suggest that much of the vertical juvenile growth in CentralAsia, which was previously believed to be related to the post-collisional extension (see Long et al., 2011a, and references therein), oc-curred after the completion of the CAOB evolution, and perhaps wasrelated to major plume activities. If it is the case, the significance of oro-genic vertical crustal accretion in theCAOB, and perhaps even in someofthe other orogenic belts in the world, needs to be re-evaluated.

The widespread Paleozoic and Early Mesozoic igneous rocks are di-rectly related to the Phanerozoic crustal growth/reworking of theCAOB, and may serve as compositional and thermal probes of theirsources and indicators of tectonic regimes (Jahn et al., 2000; Kröneret al., 2014; Long et al., 2011a; Seltmann et al., 2011). The majority ofrocks in these suites are of Late Carboniferous to Permian age (Longet al., 2011a; Seltmann et al., 2011). Previous studies have shown thatthe Siberianplumeactivity had a peak at ~251Ma, and that only the Lat-est Permian and Early Mesozoic magmatism in some parts of the CAOBmay be related to a far-field effect of the Siberian plume (e.g., Holt et al.,2012; Sobolev et al., 2011). However, as revealed by geochronologicalresults of kimberlitic intrusions, OIB-like mafic intrusions and basaltsand A1-type syenites and granites exposed in the Tarim Craton(e.g., Zhou et al., 2009; Zhang et al., 2008a, 2010a; Zhang and Zou,2013; Huang et al., 2012b; Zhang et al., 2013a, and references therein),the Permian Tarim large igneous province (Tarim LIP) has been widelyproposed by many authors, and the Tarim plume-related magmatismis constrained as ~300 Ma to ~270 Ma, which significantly overlapswith the age span of the major phase of magmatism in the adjacent re-gions of the CAOB. Thus, the Late Carboniferous and Early Permianmag-matic rocks in the CAOB, in particular those exposed in regions close tothe Tarim Craton, are important in evaluating the genetic link to plume,anorogenic crustal growth or orogenic magmatism.

Among the tectonic terranes of the CAOB, the South TianshanTerrane (STT) occupies the southwestern margin and is located im-mediately adjacent to the northern margin of Tarim Craton. If thenon-subduction continental growth of the STT took place in aplume-related setting, the magmatic suites in this terrane, especial-ly the mafic–ultramafic bodies, must bear isotopic and chemical fea-tures similar to those belonging to large igneous provinces (LIPs)generated by plume activities. The Early Permian Xiaotikanlike For-mation exposed in the central part of the Chinese STT contains basal-tic to rhyolitic volcanic rocks, and thus offer excellent proxies to testthe influence of a plume activity, and its constraints on the spatialdistribution of the Tarim LIP.

In this paper, we present new LA-ICP-MS U–Pb zircon ages, MC-ICP-MS zircon Hf isotopic data, bulk-rock major and trace element and Sr–Nd isotope results of representative samples collected from theXiaotikanlike basaltic to rhyolitic lava flows. Based on the results, we

discuss the nature of source composition, and the mechanisms of LatePaleozoic vertical crustal growth of the southwestern CAOB.

2. Geological outline of Chinese Southern Tianshan Terrane

As suggested in previous studies (e.g., Şengör and Natal'in, 1996;Sengör et al., 1993; Xiao et al., 2013), the CAOB can be subdivided intoeastern and western parts. The western CAOB is composed of severaltectonic terraneswhich have different origins and evolutionary historiesand are presently preserved in Kazakhstan, Kyrgyzstan, Uzbekistan,Tajikistan, southern Russia, NW China (Northern Xinjiang), and south-western Mongolia (Fig. 1a). The Northern Xinjiang is located in thesouthern segment of the western CAOB, and consists of, from north tosouth, the Chinese Altai, Junggar, Chinese Tianshan andnorthernmarginof Tarim Craton (Fig. 1b). Among these, the Chinese Tianshan is tradi-tionally subdivided into eastern and western parts broadly separatedby Tuokexun–Kumishi High Road. Tectonically, the Western Tianshancomprises the North Tianshan Terrane (NTT), the Kazakhstan–YiliBlock (KYB), the Central Tianshan Terrane (CTT), the South TianshanTerrane (STT) and the northern margin of Tarim Craton.

The South Tianshan Terrane investigated in this study includes vari-ous rocks exposed in the ribbon region bounded by the Northern TarimFault to the south and the Southern-Central Tianshan Suture to thenorth. As shown in Fig. 1c, the STT consists of two segments dividedby the Talas–Fergana diagonal dextral strike-slip fault, and the Chinesepart of STT is mostly located in the eastern segment. The nature and or-igin of the STT remain debated. Someworkers suggest that the STTdom-inantly represents an accretionary complex formed by the northwardsubduction of the South Tianshan Ocean, whereas others regard it as aPaleozoic passive margin of the Tarim Craton with minor accretionarycomponents (Gao et al., 2011). In our understanding from the natureand overall distribution of the rock types in the region, the STT is a com-posite terrane composedof high/ultrahigh pressuremetamorphic rocks,ophiolitic components, Early Paleozoic arc-type magmatic rocks, pas-sive margin sequence from the Tarim Craton and the underlyingTarim basement.

Some HP/UHP metamorphic rocks have been identified within Pa-leozoic ophiolites/ophioliticmélanges along the South-Central TianshanSuture. These rocks include blueschist-, eclogite- and greenschist-faciesmeta-sedimentary rocks and some mafic metavolcanic rocks with N-MORB, E-MORB, OIB and arc basalt affinities (Gao et al., 1998, 2011;Zhang et al., 2013b). The HP/UHP metamorphism was likely associatedwith the subduction of the South Tianshan Ocean. Ophiolitic mélangesare randomly distributed in the STT and occur generally parallel to thestrike of the STT. According to published dataset, the oldest age ofmafic rocks in these mélanges are ~600 Ma, and majority of these pos-sess Late Ordovician to Middle Devonian crystallization ages (see Jianget al., 2014). In addition, Middle Devonian to Early Carboniferous micro-fossils, i.e., radiolarians and conodonts, arewell-preservedwithin the sed-imentary rocks of mélanges. The basement rocks of this terrane, whichform part of the Tarim basement, are represented by PaleoproterozoicXingditagh and Muzha'erte formations and Mesoproterozoic Akesu For-mation, all exposed in the central part of the STT (Yang and Zhou,2009). Paleozoic strata consist predominantly of Lower Cambrian blackshales and phosphoric silicates, Cambrian–Carboniferous marine/non-marine carbonates, clastic rocks, cherts and interlayered volcanics (Allenet al., 1992; Jiang et al., 2014). Based on the assumption of a Late Paleozoicnorthward subduction model, Xiao et al. (2013) postulated that the latePaleozoic sequence should be a part of the accretionary complex initiallyaccreted onto the Central Tianshan Terrane. As shown in Fig. 1c, Permianstrata are generally sparse in the STT, and mainly occur in the middlesection of the Chinese part of the STT.

The outcrops of igneous rocks, most of which are granitoids, com-prise ~5% of the total area of the STT. As shown in Figs. 1c and d, thePaleozoic magmatism seems to have occurred mainly during the LateSilurian and Early Permian. Furthermore, available geochronological

KYRGHIZSTAN

KAZAKHSTANCHINA

South Tianshan Terrane

South Tianshan Belt

Central Tianshan Block

Central Tianshan Terrane

Kazakhstan-Yili Block

Tarim Craton

11

1

1

2

2

2

3

3

3

4

o40 N

o42 N

o40 N

o42 N

o75 E o78 E o81 E o84 E

o84 Eo81 Eo78 Eo75 E

Precambrian LowerPalaeozoic

Devonian to Carboniferousstrata

Permianstrata

OphioliteGranite Dioriticintrusion

Mesozoic &CenozoicstrataTectonicBoundary

StateBorder

0 20 40 60 80 100 km

Major tectonic boundaries:

1 Nikolaev Line-North Nalati Fault South-Central Tianshan Suture 2

3 North Tarim Fault Talas-Fergana dextral strike-slip fault4

Akesu

Kuche Korla

Aheqi

Baicheng

TekesZhaosu

EuropeanSilberian

Late Proterozoic Orogenic Belt

Orogenic Belt

Central Asian

N. ChinaTarim

60° N 60° N20° E 100° E

40° N 40° N

a)

Fig. 1bN. Xinjiang

Mongolia

Kazakhstan

Kyrgyzstan

c)

(1) 286.4±2.5 Maε (t)=-2.3 to -1.4Nd

(4) 278±3 Maε (t)=-0.3 to -0.2Nd

(3) 272.7±1.1 Maε (t)=-0.9 to 0Nd

(5) ε (t)=+0.2 to +0.6Nd

(2) 278±3 Maε (t)=-2.8 to -2.6Nd

(6) 272.4±1.1 Ma(7) 279.7±2.0 Ma

285.9±2.6 Maε (t)=+3.4 to +4.3Nd

(8) 273.7±1.5 Maε (t)=-0.3 to -0.2Nd

(9) 291.0±2.6 Maε (t)=-6.1 to -5.0Nd

(10) 290.1±1.4 Maε (t)=-4.4 to -3.1Nd

Fig. 2a

(11) 296.9±5.4 Ma

(12) 299±4 Ma(13)295.3±4.4 Ma(14) 288.6±6.3 Ma

(16) 279±3 Maε (t)=-3.6Nd

(15) 281±2 Maε (t)=-5.9Nd

(17) 296±4 Maε (t)=-5.3 and -6.9Nd

(18) 279±8 Maε (t)=-4.0 and -3.6Nd

(19) 292±3 Maε (t)=-4.8Nd

(20) 419.9±2.1 Maε (t)=-6.8 and -6.6Nd

(21) 382.1±6.2 Ma

(22) 420.6±2.3 Ma421.7±2.8 Ma419.8±3.3 Ma (23) 423±16 Ma

ε (t)=-7.3 to -3.7Nd

(24) 421±3 Ma

(25) 273±2 Ma

(26) 284±1 Ma(27) 280±3 Ma

(28) 281±2 Ma

300

290

280

270

11

10914

13

19

12

17

18

625

1

1516

2827

26

3

2, 4 7

8

Xiaotikanlike volcanic lava flows

o75 E o78 E o81 EAheqi Akesu Baicheng Kuche

0 20 40 60 80 100 km

Plutonic rocks from theSouth Tianshan TerranePlutonic rocks from theTarim Craton

Age

/Ma

d)

Fig. 2c

ChineseAltay

E. Junggar

W. Junggar

JunggarKazakhstan-Yili-Central Tianshan Block

Beishan TerraneTarim Craton

O79 E O83 E O91 E O95 E

O46 N

O42 N

North Tianshan enTerraerrT ana nh esnaiThtuoS

b)Fig . 1c

1

Fig. 1. (a) Tectonic sketch map of the Central Asia Orogenic Belt showing the location of South Tianshan Terrane and the northern margin of the Tarim Craton (modified after Gao et al.,2011). (b) Tectonic units of the northern Xinjiang, NW China (modified after Xie et al., 2012). (c) Geological map of the South Tianshan Terrane and northernmargin of the Tarim Craton.(d) Simplified schematic diagram showing ages and locations of post-collisional plutonic rocks from the South Tianshan Terrane and northern margin of Tarim Craton. Data sources forages and εNd(t) values of felsic intrusive rocks: (1) (Huang et al., 2012a); (2), (3), (4) and (5) (Huang et al., 2012b; Zhang et al., 2010a); (6) (Zhang and Zou, 2013); (7) (Sun et al.,2008; Wei and Xu, 2011); (8) (Zhang et al., 2008a); (9) (Huang et al., 2011); (10) (Huang et al., 2014); (11) (Zhu et al., 2008a); (12), (13) and (14) (Konopelko et al., 2009); (15),(16), (17), (18) and (19) (Konopelko et al., 2007); (20) (Huang et al., 2013); (21) (Zhu et al., 2008b); (22) (Ge et al., 2012); (23) (Guo et al., 2013); (24) (Wang et al., 2009); (25)(Wang et al., 2007); (26), (27) and (28) (Biske et al., 2013).

25H. Huang et al. / Lithos 228–229 (2015) 23–42

data illustrate an irregularmigration of Early Permianmagmatism fromEast to West (Fig. 1d).

3. Geology and petrography of the Xiaotikanlike Formation

3.1. Geology of study area

The volcanic rocks, including lava flows and pyroclastic rocks, inter-bedded with siltstones, sandstones and sandy conglomerates, havebeen grouped as the Xiaotikanlike Formation by local geologists andare mainly distributed in the middle section of the Chinese part of theSouth Tianshan Terrane. We carried out field investigations of theBoziguo'er and Laohutai sections of the Xiaotikanlike formation in thenorth of the Baicheng County (Figs. 1c, 2 and 3). A brief summary of

the geologic and petrographic features of the rocks exposed in thesetwo regions is given below.

The stratigraphic successions in the Boziguo'er region contain theUpper Silurian Keketiekedaban Formation (Gao et al., 2014), theLower Carboniferous Gancaohu and Yeyungou formations (Wanget al., 2005), the Lower Permian Xiaotikanlike Formation andthe Lower Triassic Ehuobulake Formation (Liu et al., 2013). TheXiaotikanlike Formation tectonically overlies the KeketiekedabanFormation, uncomfortably overlies the Gancaohu and Yeyungouformations, and is uncomfortably overlain by Lower TriassicEhuobulake Formation (Fig. 2a).

The exposed strata of the Keketiekedaban Formation show a thick-ness of ~2600 m and include two stratigraphic units (units III and IVby local geologists). The unit III, predominately composed of lowgrade metamorphic rocks including biotite–quartz schist, biotite–

Fig. 2. (a) Regional geological sketch map of the Boziguo'er region (modified from Geological Map at the scale of 1:50,000, Xinjiang Bureau of Geology and Resources, unpublished).(b) Geological section of the Xiaotikanlike Formation. The sampling locations (arrow) are shown. (c) Regional geological sketch map of the Laohutai region (modified from GeologicalMap at the scale of 1:50,000, Xinjiang Bureau of Geology and Resources, unpublished).


muscovite–quartz schist, and biotite–muscovite–calcite schist, is tec-tonically overlain by the unit IV that consists mainly of gray, thick-bedded marbles interlayered with biotite–muscovite–quartz schist.Besides, the unit IV is intruded by a highly differentiated granitic plutonwith a LA-ICP-MS age of ~290 Ma (Huang et al., 2014). The Lower Car-boniferous Gancaohu and Yeyungou formations, both of which tectoni-cally overlie the Upper Silurian Keketiekedaban Formation, are stronglyfolded. The former is mainly composed of ~4000 m-thick, gray andceladon siltstones, and the latter refers to a series of ~450 m-thickgray, medium- to thick-bedded limestone interlayered with thin- tomedium-bedded siltstone, sandstone and minor conglomerate. TheXiaotikanlike Formation from the Boziguo'er region is ~3400 m thick,and contains large amounts of pyroclastic rocks, such as dacitic

agglomerates, breccias, and breccia-bearing tuffs, and relatively smallamount of lavas and sandstones. The Lower Triassic Ehuobulake Forma-tion, unconformably overlying the Xiaotikanlike Formation, has a thick-ness of ~350 m and consists dominantly of purple-red, thick-beddedconglomerate and grayish-green, medium- to thick-bedded lithicsandstone.

In the Laohutai region, the stratigraphic successions include theMesoproterozoic Akesu Formation (Yang and Zhou, 2009), the UpperCarboniferous Kangkelin Formation, the Lower Permian XiaotikanlikeFormation and the Lower Triassic Ehuobulake Formation. TheXiaotikanlike Formation unconformably overlies the Kangkelin Forma-tion and is uncomfortably overlain by the Ehuobulake Formation(Fig. 2c).

Fig. 3. Geology and petrography of the Xiaotikanlike Formation. (a) The Xiaotikanlike Formation unconformably lying over the Lower Carboniferous Gancaohu Formation;(b) Xiaotikanlike basalts showing non-porphyritic, holocrystalline texture (cross-polarized light); (c) strongly altered Xiaotikanlike felsic rocks with a felsophyric texture (cross-polarizedlight); (d) weakly altered Laohutai rhyolitic rocks showing porphyritic texture, with phenocrysts mainly composed of alkali-feldspar and quartz.


The Akesu Formation in the Laohutai and adjacent regions is morethan 3300 m thick. It is dominantly composed of granulite facies meta-morphic basement rocks. The strongly folded Upper CarboniferousKangkelin Formation, with a thickness of ~800 m, uncomfortably over-lies the Akesu Formation, and consists of a series of gray, thick-beddedlimestone interlayered with thin- to medium-bedded marble, siltstoneand sandstone. Our field observations show that the Lower PermianXiaotikanlike Formation in the Laohutai region is dominated by~400m-thick rhyolitic lava flow. The Lower Triassic Ehuobulake Forma-tion, with a thickness of ~470 m, unconformably overlies theXiaotikanlike Formation and is composed mainly of purple-red, thick-bedded conglomerate and grayish-green, medium- to thick-beddedlithic sandstone.

3.2. Petrography of studied rocks

The studied section from the Boziguo'er region is composed of~400 m-thick basaltic flow in the lower part and ~500 m-thick felsicflow close to the top. The two lava flows are separated by sandstonesand pyroclastic rocks. Both lava flows show steep southerly dip. Therock types in the basaltic flow show similar textures andmineral assem-blages from the bottom upward. They are massive, and have a non-porphyritic, intergranular texture, in which the angular interstices be-tween medium-grained, euhedral plagioclase grains are occupied bygrains of hypidiomorphic–anhedral pyroxene and anhedral iron titani-um oxides (Fig. 3b). Apatite, titanite and zircon are present as accessoryminerals. The basalts have experienced low to medium degree of low-temperature post-magmatic alteration, represented by clay formationas observed in some thin sections. The felsic flow is gray to pale red incolor,fine-grained andmassive, with a felsophyric texture. As suggestedby thewidespread occurrence of secondaryminerals in our samples, thefelsic rocks are moderately to severely altered, and in thin sections therock-forming minerals have been almost completely replaced by clay,zeolite, quartz, and calcite (Fig. 3c). The initial mineral assemblage is

possibly made up of quartz and feldspar, with subordinate plagioclaseand minor Fe–Ti oxides.

The samples of the rhyolitic lava flow from the Laothutai region aremoderately porphyritic in texture, with up to 15–20% phenocrysts com-prising alkali-feldspar, quartz, and minor hornblende. The groundmassis usually a fine-grained assemblage of feldspar, quartz and minor Fe–Ti oxides. The low- to medium-degree of post-magmatic alteration,mainly represented by kaolinization of feldspars, is present in mostsamples (Fig. 3d).

Petrographic and geochemical features of rhyolites of XiaotikanlikeFormation exposed in the Wensu region, ~200 km southwest of theLaohutai region, have been recently reported by Luo et al. (2008), andare used here for comparison. According to their description, the petro-graphic features of those rocks are similar to that of rhyolites from theLaohutai region, and the alkali feldspars in the Wensu rhyolites arelargely altered. Detailed petrographic features are not reproduced here.

4. Analytical methods

4.1. Zircon U–Pb and Lu–Hf isotopic data

Two representative samples, one (XKT-01) from the basaltic flowand the other (TLK-01) from the felsic flow within the studied sectionin the Boziguo'er region, were selected for zircon separation and U–Pbdating by laser-ablation-inductively coupled plasmamass spectrometry(LA-ICP-MS) in this study. Zircon grains were separated throughconventional magnetic and density techniques to concentrate non-magnetic, heavy fractions. The grains were hand-picked under a binoc-ular microscope. Internal structures of the zircon grains were examinedusing transmitted electron, backscattered electron (BSE) and cathodeluminescence (CL) prior to U–Pb isotopic analyses.

Zircon U–Pb dating for the basaltic sample XKT-01 was performedon single zircons using a Thermo Finnigan Element 2 multi-collectorICP-MS with a New Wave UP213 laser ablation system housed at the


National Research Center for Geoanalysis, Beijing, China. Helium wasused as the carrier gas within the ablation cell inside the ablation celland was mixed with argon (makeup gas) before entering the ICP tomaintain stable and optimum excitation conditions. A spot size of40 μm with a 10 Hz repetition rate and a laser power of 16–17 J/cm2

were used for all analyses. All U–Th–Pb isotope measurements wereperformed using zircon GJ-1 as an external standard and Plesovice zir-con was used as the reference material. Tow mean ages of 601.3 ±4.7 Ma and 335 ± 4.3 Ma were obtained for zircons GJ-1 and Plesovice,respectively. Isotopic ratios were calculated using GLITER 4.0 (vanAchterbergh et al., 2001) while common lead correction was carriedout using the EXCEL program ComPbCorr# 151 (Andersen, 2002).

Zircon U–Pb dating for the felsic sample TLK-01 was carried out atthe Key Laboratory of Continental Collision and Plateau Uplift, ChineseAcademy of Sciences, Beijing, China, using an Elan 6100 DRC ICP-MSequippedwith 193nmExcimer lasers. U–Th–Pb ratioswere determinedrelative to the Plesovice standard zircon, and the absolute abundancesof U, Th and rare earth elements (REEs) were determined using theNIST 612 standard glass. A mean age of 338.2 ± 1.5 Ma was obtainedfor the Plesovice zircon standard. Most analyses were carried outusing a beam with a 40 μm diameter. Corrections for common-Pbwere made using themethod of Andersen (2002). Data were processedvia theGLITTER and ISOPLOT (Ludwig, 2003) EXCEL programs. Errors onindividual analyses by LA-ICPMS are quoted at the 95% (1σ) confidencelevel. Details of the analytical procedures can be found in Yuan et al.(2004).

Given the post-magmatic alteration witnessed by the felsic lavaflow, the whole-rock Rb–Sr and Sm–Nd system are expected to havebeen perturbed (Poitrasson et al., 1995), and therefore felsic rocksfrom the Boziguo'er regionwere not analyzed forwhole-rock Sr–Nd iso-tope. In contrast,magmatic zircons can survivemultiple episodes of par-tial melting, magmatic differentiation as well as post-magmatic processincluding metamorphism, hydrothermal alteration and weathering(Rubatto andHermann, 2003). Therefore, they are capable of preservingthe Hf isotopic composition of the magma. In-situ zircon Hf isotopeanalysis of zircon grain from the sample TLK-1 was conducted using aNewwave UP213 laser-ablation microprobe attached to a Neptunemulti-collector ICP-MS, at Institute of Mineral Resources, Chinese Acad-emy of Geological Sciences, Beijing. Guided by CL images, Lu–Hf spotswere selected to be partly overlapping with or close to those of U–Pbdating. Instrumental conditions and data acquisition were comprehen-sively described by Hou et al. (2007) andWu et al. (2006). A stationaryspot was used for the analyses, with a beamdiameter of either 40 μmor55 μm depending on the size of ablated domains. Helium was used ascarrier gas to transport the ablated sample from the laser-ablation cellto the ICP-MS torch via a mixing chamber mixed with Argon. In orderto correct the isobaric interferences of 176Lu and 176Yb on 176Hf,176Lu/175Lu = 0.02658 and 176Yb/173Yb = 0.796218 ratios were deter-mined (Chu et al., 2002). For instrumental mass bias correction Ybisotope ratios were normalized to 172Yb/173Yb of 1.35274 (Chu et al.,2002) and Hf isotope ratios to 179Hf/177Hf of 0.7325 using an exponen-tial law. The mass bias behavior of Lu was assumed to follow that ofYb. Zircon GJ1 was used as the reference standard, with a weightedmean 176Hf/177Hf ratio of 0.282013 ± 0.00008 (2σ, n = 10) or0.282013 ± 0.000024 (2σ, n = 10) during our routine analyses. It isnot distinguishable from a weighted mean 176Hf/177Hf ratio of0.282013 ± 19 (2σ) from in-situ analysis by Elhlou et al. (2006).

4.2. Electron microprobe analyses

Mineral chemistry of representative clinopyroxene, plagioclase andtitanium iron oxides was analyzed using a JEOL JXA-8320 Superprobeat the EMPA Laboratory of China University of Geosciences, Beijing.Operating conditions were set at 15 kV at 1 × 10−8 A bean current.The beamsize for EMPA analyses is 2 μm.Naturalminerals and syntheticpure oxides from SPI Company of America were used as standards.

4.3. Whole-rock geochemistry

After careful petrographic examination, a representative suite oftwenty-two relatively fresh samples collected from the Xiaotikanlikelava flows (ten samples from the basaltic lava flow and six from thefelsic lava flow within the studied section in the Boziguo'er region,and another six samples from the Laohutai rhyolitic lava flow) werecrushed and powdered in an agate mill for geochemical analysis.Major elements were determined in the Key Laboratory of OrogenicBelts and Crustal Evolution, Ministry of Education, School of Earth andSpace Sciences, Peking University. Major element compositions weremeasured by a scanning wave length dispersion X-ray fluorescence(XRF) spectrometer (THERMO ARL ADVANT XP+) on fused glassdisks. The analytical precision, as determined on the Chinese Nationalstandard GSR-1 and GSR-3, was better than 1%. Trace (including rareearth) element analyses were determined by inductively coupled plas-mamass spectrometry (ICP-MS) hosted in the National Research Centerfor Geoanalysis, Chinese Academy of Geological Sciences (CAGS) in Bei-jing, China. For trace element determination, about 50 mg of powder ofeach sample was dissolved for about 7 days at ca. 100 °C usingHF-HNO3

(10:1) mixtures in screw-top Teflon beakers, followed by heating to70 °C for 12 h (overnight). Subsequently, the material was dissolvedin 7 N HNO3 and taken to incipient dryness again, and then was re-dissolved in 2% HNO3 to a sample/solution weight ratio of 1:1000. Theanalytical errors vary from 3% to 7% depending on the concentrationof any given element. An internal standard was used for monitoringdrift during analysis; further details have been given by Gao et al.(2008).

4.4. Bulk-rock Rb–Sr and Sm–Nd isotope

Four samples of basaltic lavas from the Boziguo'er region wereanalyzed for bulk-rock Rb–Sr and Sm–Nd isotopic compositions at theInstitute of Geology, CAGS. The Rb–Sr isotopic compositions were mea-sured by isotope dilution on a Finnigan MAT-262 mass spectrometer,and Sm–Nd isotopic compositions were acquired by a Nu Plasma HRMC-ICP-MS (Nu Instruments). The procedures followed those describedby He et al. (2007). The 143Nd/144Nd values for JMC Nd2O3 standardwere 0.511126 ± 10 (2σ), and 87Sr/86Sr for SRM 987 SrCO3 standardwere 0.710247 ± 12 (2σ). Total procedural blanks were b100 pg forSr and b50 pg for Nd, and the estimated analytical uncertainties of147Sm/144Nd and 87Rb/86Sr ratios are b0.5%. The 143Nd/144Nd and87Sr/86Sr were corrected for mass fractionation by normalization to146Nd/142Nd = 0.7219 and 86Sr/88Sr = 0.1194, respectively.

5. Results

5.1. Zircon U–Pb geochronology

Representative CL images of the zircon grains from the rocks ana-lyzed in this study, together with spot ages, are shown in Fig. 4. Zirconsfrom the basaltic sample (XKT-01) are 60–120 μm long and are mostlysubhedral–euhedral prismatic crystals. In addition, a small amount ofzircons display oscillatory zoning and re-crystallized rims. Sixty-twograins were analyzed and the results are listed in SupplementaryTable 1. Among these, twenty-six spots are discordant, probablyreflecting post-magmatic lead loss, and the other 36 data are concor-dant. The data show Th/U ratios from 0.22 to 1.96, and thirty-oneamong these have Th/U ratios N 0.4. The high Th/U ratios, togetherwith morphological characteristics, indicate magmatic crystallizationof the zircons (Hoskin and Black, 2000). The 36 analyses do not yield acoherent mean 206U/238Pb age, and instead define a range of agesbetween 287 ± 5 Ma and 2566 ± 37 Ma (Figs. 5a and b). Obviously,the majority, if not all, of zircons from the basaltic sample arexenocrystic grains captured duringmagma ascend. Two zircons yielding206U/238Pb ages of 287 ± 5 and 288 ± 6 Ma define the youngest age

Fig. 4. Cathodoluminescence (CL) images of typical zircon populations. Solid circles show the domains for zircon U–Pb analysis, and dashed circles show those for zircon Lu–Hf analysis.


population, and provide the maximum possible age for the eruption. Insummary, the data indicate that the lower basaltic lava erupted no ear-lier than ~290 Ma.

Zircons separated from the felsic lava sample (TLK-01) are prismatic,transparent and colorless or pale yellow, with lengths varying from 100to 250 μm and width/length ratios generally ranging from 1:1.5 to 1:4.Thirty-two grains were analyzed and the results are listed in Supple-mentary Table 1. Thirteen of the 32 zircon grains have discordant ages(Figs. 5c and d); the remaining 19 grains show Th/U values in therange of 0.39 to 1.15 suggesting magmatic origin. Fourteen of the 19grains are clustered around 206Pb/238Pb ages in the range of279–289 Ma, with a weighted mean age of 285.0 ± 2.3 Ma (n = 14,MSWD = 0.90, Fig. 5e). We interpret that these zircons crystallizedfrom the parental magma of the upper felsic lava flow, and thereforethe weighted mean age represents the eruption age of the magma.Based on the Ti concentrations in those youngest zircon grains(10.27–36.09 ppm), the temperature of zircon crystallization is estimat-ed to be between 755 and 882 °C (Watson et al., 2006). The other fivezircons with concordant isotopic ages exhibit 206Pb/238Pb ages of296± 3, 298±5, 300±5, 609±6 and 730±22Ma, respectively, sug-gesting that they are inherited zircons.

The eruption age of the felsic lava flow is nearly contemporaneouswith the maximum possible eruption age of the basaltic flow withinerror. Combined with the stratigraphic location of the felsic flowsabove the basaltic lava flows, it is inferred that both basaltic and felsiclava flows within the section were simultaneously erupted duringEarly Permian.

5.2. Zircon Lu–Hf isotope

Zircon Lu–Hf isotopic compositions were measured on 19 zircongrains, all of which have been previously dated for U–Pb isotopeand show a well-defined concordia in 207Pb/235Pb vs. 206Pb/238Pb di-agram, from the felsic lava sample TLK-01. Guided by CL images(Fig. 4), the analytical spots for Lu–Hf isotope largely overlap or areclose to those for U–Pb. The decay constant of 176Lu 1.867 × 10−11

per year (Scherer et al., 2001) was adopted in the calculation. Thesingle stage model ages (TDM) were calculated based on the chon-drite model with 176Hf/177Hf = 0.282772 and 176Lu/177Hf = 0.0332(Blichert-Toft and Albarède, 1997) and the depleted mantle modelwith present-day 176Hf/177Hf = 0.28325 and 176Lu/177Hf = 0.0384(Griffin et al., 2000), respectively. The two-stage continental modelage (TDM2) was calculated based on the assumption that the initial176Hf/177Hf of zircon back to the depleted mantle growth curve andusing 176Lu/177Hf = 0.015 for the average continental crust (Griffinet al., 2000). The results are listed in Table 1.

The zircons yield a range of 176Lu/177Hf ratios from 0.000666 to0.002269 indicating very low radiogenic Hf. Among the 19 grains, 14ones with 206Pb/238U ages from ~289 to ~279 Ma show 176Hf/177Hf ra-tios between 0.282414 and 0.282592, corresponding εHf(t) values of−6.6 to−0.4 (calculated at 285 Ma, the eruption age of the felsic lavaflow, Fig. 6a). The two-stage continental model ages (TDM2) for the 14youngest zircons vary from 1.3 to 1.7 Ga. Given the chemical and phys-ical resistance of zircon that allow survival from even high crustalthermotectonic conditions, theHf isotopic compositions of zircon grains

Fig. 5. (a) LA-ICP-MS U–Pb zircon concordia diagram and (b) frequency distribution histogram for zircon separated from the lower basaltic lava flows; (c and d) LA-ICP-MS U–Pb zirconconcordia diagram and (e) weighted mean ages of zircons from the upper felsic lava flows.


in felsic lavasmight not been altered by the post-magmatic alteration asindicated by petrographic observations and high LOI contents (seebelow). Thus, the zirconHf isotope composition is taken to record infor-mation on the source region (Scherer et al., 2007).

5.3. Mineral chemistry

The electron microprobe analyses of representative clinopyroxene,plagioclase and titanium-iron oxides of basaltic rocks from theBoziguo'er region are listed in Supplementary Tables 2–4. The analyzed

Table 1Zircon Hf isotope compositions of the upper felsic flow (TLK-01).

Sample 206Pb/238Pb age 176Yb/177Hf 2σ 176Lu/177Hf 2σ

01 296 0.055222 0.000663 0.000919 0.008 300 0.046548 0.000302 0.000761 0.009 289 0.067187 0.002037 0.001084 0.010 287 0.111682 0.000954 0.002269 0.014 279 0.088201 0.004062 0.001715 0.015 298 0.040316 0.000639 0.000675 0.017 288 0.046211 0.000412 0.000755 0.018 730 0.085190 0.000725 0.001895 0.019 287 0.075594 0.000533 0.001299 0.020 281 0.068198 0.001051 0.001261 0.022 288 0.040123 0.000409 0.000666 0.023 283 0.060946 0.000229 0.001008 0.025 285 0.090905 0.000258 0.001501 0.026 282 0.044249 0.000384 0.000729 0.028 288 0.075891 0.000729 0.001359 0.029 284 0.042359 0.000232 0.000813 0.030 609 0.076138 0.000370 0.001373 0.031 281 0.040559 0.000169 0.000734 0.033 287 0.040735 0.000173 0.000744 0.0

clinopyroxenes are predominately augite (Fig. 7a), with two exceptionsthat show hedenbergite and diopside composition. The augite grainsdisplay a moderate compositional range (Wo37–42En36–46Fs15–22) withlow TiO2 (0.09 to 1.05 wt.%), Al2O3 (0.34 to 2.30 wt.%) and Na2O (0.17to 0.32 wt.%) (Supplementary Table 2). Analysis of two augite grainsshows Mg-, Ti-, Al-, Ca- and Na-rich cores together with more Si- andFe-rich rims. Based on the clinopyroxene thermometer proposed byPutirka et al. (2003), 6 results represent clinopyroxenes that were inequilibrium with the melts, and thus yield crystallization temperaturesbetween 1309 °C to 1072 °C. Except for some plausible alteration-

176Hf/177Hf 2σ εHf(t) TDM1 (Ga) TDM2 (Ga)

00008 0.282499 0.000019 −3.5 1.1 1.500003 0.282472 0.000024 −4.5 1.1 1.600039 0.282512 0.000022 −3.1 1.0 1.500018 0.282580 0.000034 −0.9 1.0 1.400085 0.282491 0.000026 −3.9 1.1 1.600007 0.282352 0.000022 −8.7 1.3 1.900004 0.282483 0.000029 −4.0 1.1 1.600028 0.281864 0.000027 −17.0 2.0 2.700012 0.282539 0.000024 −2.2 1.0 1.400021 0.282414 0.000024 −6.6 1.2 1.700005 0.282491 0.000022 −3.8 1.1 1.500006 0.282508 0.000023 −3.2 1.1 1.500004 0.282504 0.000023 −3.5 1.1 1.500005 0.282515 0.000022 −2.9 1.0 1.500022 0.282592 0.000022 −0.3 0.9 1.300006 0.282512 0.000025 −3.1 1.0 1.500002 0.282352 0.000023 −2.0 1.3 1.700001 0.282440 0.000029 −5.6 1.1 1.700004 0.282499 0.000023 −3.5 1.1 1.5

Fig. 6. (a) Histogram showing the distribution of εHf(t) values and (b) age vs.εHf(t) diagram for analyzed zircons from the sample TLK-1 and from other felsic intrusiverocks in the South Tianshan Terrane. Evolution line of the early Paleoproterozoic crust iscited from Long et al. (2011b).

Fig. 7. (a) En–Wo–Fs composition diagram for pyroxenes and (b) An–Ab–Or compositiondiagram for feldspars from the Xiaotikanlike basalts.


related albite and sanidine, the analyzed plagioclase grains frombasalticrocks mostly show a compositional range from labradorite to andesine(Ab36–57An42–62Or1–4) (Fig. 7b). The data also show that titanium–ironoxides in the basaltic lava samples mainly consist of ilmenite, withTiO2 from 47.84 to 50.36 wt.% and FeOT from 47.07 to 50.09 wt.%.

5.4. Major and trace elements

5.4.1. Major elementsMajor and trace element compositions of the representative samples

of the Xiaotikanlike lava flows exposed in the Boziguo'er and Laohutairegions are listed in Table 2. Petrographic observations have revealedthat most of studied samples have been subjected to varying degreesof alteration, which is consistent with their relatively high LOI contentsranging from 0.30 to 3.76 wt.% for basaltic lava samples, from 3.24 to9.67 wt.% for felsic lava samples from the Boziguo'er region, and from1.54 to 1.88 from the rhyolitic lava flow in the Laohutai region. On aLOI-free basis, the samples of the Xiaotikanlike Formation investigatedin the present study and those from previous data show SiO2 from49.88 to 78.56 wt.%.

The basaltic rocks sampled from the Boziguo'er region in this studywhen normalized to 100 wt.% anhydrous, show SiO2 from 49.88 to53.78, Al2O3 from 13.39 to 16.18 wt.%, and MgO from 3.73 to 7.01 wt.%,Fe2O3T from 10.12 to 12.66 wt.%, TiO2 from 1.68 to 2.48 wt.%, and P2O5

from 0.26 to 0.41 wt.%. The Mg# [atomic Mg/(Mg + Fe)] displays arange from 41 to 61, suggesting that the basaltic rocks were derivedfrom a relatively evolved magma. They are characterized by enrichmentin Na2O (2.69 to 4.55 wt.%) relative to K2O (0.98 to 1.84 wt.%), withNa2O/K2O ratios of 1.78–3.63. On the SiO2 vs. Na2O + K2O and Zr/TiO2vs. Nb/Y diagrams (Fig. 8), they generally plot in the boundary between

alkaline and subalkaline series. Previous data on the mafic–ultramaficigneous rocks from the northern margin of Tarim Craton are shown inFig. 8, where they generally plot in the field of alkaline series (Fig. 8).

The felsic lavas from the Boziguo'er, Laohutai and Wensu regionsshow 60.71 to 78.56 wt.% SiO2 and 6.17 to 8.79 wt.% total alkali(K2O + Na2O), and mostly fall in the subalkaline series in the plot ofSiO2 vs. Na2O+K2O (Fig. 8a). The felsic lavas from the Boziguo'er regionhave lower SiO2 (60.71 to 70.65 wt.%) and higher Al2O3 (11.58 to14.26 wt.%) than those from the two other regions. Probably due tothe high-degree post-magmatic alteration (LOI = 3.24–9.67 wt.%),felsic lavas from the Boziguo'er region show a Na-rich (Na2O =4.16–8.01 wt.%) and K-poor (0.05–2.83 wt.%) characteristic, coupledwith variable CaO contents. The above features indicate that the felsicrocks have possibly undergone intense Na-metasomatism by seawater. In theNb/Y vs. Zr/TiO2 diagramwhich iswidely used as an immo-bile element proxy for the TAS diagram, the Boziguo'er felsic lavas gen-erally plot in the boundary between subalkaline and alkaline series(Fig. 8b). The rhyolitic rocks from the Laohutai and Wensu regionsexhibit similar major element features characterized by relatively highSiO2 (73.22 to 78.56wt.%) and lowAl2O3 (10.98 to 13.27wt.%) contents.Besides, they range from 0.16 to 0.37 wt.% for TiO2, 1.06 to 4.21 wt.% forFe2O3T and 0.08 to 0.41 wt.% for MgO. They are characterized by rela-tively high K2O concentrations, from 4.69 to 8.60 wt.%, in comparisonwith low Na2O contents from 0.18 to 3.37 wt.%. As shown in SiO2 vs.Na2O + K2O and Nb/Y vs. Zr/TiO2 diagrams (Fig. 8), the Laohutai andWensu rhyolites plot mostly in the field of rhyolite.

5.4.2. Trace elementsIn general, the basalts from the Boziguo'er region have moderate to

low contents of compatible elements, such as Ni (31.0–188.3 ppm)and Cr (45.9–439.7 ppm), and relatively high contents of incompatibleelements. In primitive mantle-normalized trace element spidergrams(Fig. 9a), the basalts show significant enrichment in large-ion lithophileelements (LILEs, e.g., Rb, Ba and K) and high-field strength elements

Table 2Major (wt.%) and trace (in ppm) element data for the Xiaotikanlike basaltic and felsic lava flows.

Basalts from the Boziguo'er region Felsic rocks from the Boziguo’er region Rhyolites from the Laohutai region

BZXW-1 BZXW-2 BZXW-4 BZXW-5 BZXW-6 BZXW-7 BZXW-8 BZXW-9 XKT-1(1) XKT-1(2) TLK-01 BZLW-1 BZLW-2 BZLW-3 BZLW-4 BZLW-11 SLHT-17 SLHT-18 SLHT-19 SLHT-20 SLHT-21 SLHT-22

SiO2 51.97 49.11 48.00 50.29 50.90 50.74 52.29 50.04 52.88 53 64.09 62.22 54.81 59.41 68.13 59.82 76.31 71.84 74.9 75 74.8 72.59TiO2 1.88 1.69 1.63 1.69 1.68 1.64 2.10 1.94 2.44 2.43 1.26 0.43 0.29 0.43 0.43 0.36 0.18 0.2 0.22 0.16 0.24 0.36Al2O3 14.31 14.98 15.18 15.21 15.86 15.82 13.76 14.86 13.27 13.2 13.79 10.67 11.46 11.78 13.45 11.74 10.81 12.71 12.01 11.67 12.37 12.68Fe2O3T 10.09 10.87 11.25 11.38 10.90 10.91 11.77 10.93 12.3 12.48 7.21 1.71 4.55 3.23 3.15 2.37 3.56 3.23 3.40 3.01 1.04 2.93MnO 0.17 0.16 0.14 0.15 0.15 0.15 0.15 0.17 0.16 0.15 0.11 0.04 0.12 0.06 0.07 0.06 0.02 0.01 0.01 0.01 0.02 0.06MgO 6.89 6.36 6.84 5.64 5.20 5.07 4.14 6.29 3.91 3.68 0.66 0.22 0.27 0.17 0.29 1.12 0.18 0.12 0.27 0.22 0.15 0.4CaO 9.98 9.46 7.84 8.90 8.61 7.11 8.47 6.64 7.29 7.22 2.43 9.93 11.58 9.91 2.69 8.43 0.57 1.11 0.49 0.58 1.43 1.28Na2O 3.04 3.13 3.20 2.89 3.27 3.86 2.63 4.45 3.22 3.27 4.21 6.09 6.54 6.70 7.73 3.78 0.6 0.18 0.18 0.66 2.22 3.31K2O 0.98 1.02 1.63 1.09 1.34 1.67 1.25 1.23 1.81 1.73 2.68 0.07 0.10 0.05 0.06 2.57 5.97 8.44 7.05 6.79 6.03 4.61P2O5 0.30 0.26 0.25 0.31 0.28 0.27 0.34 0.35 0.38 0.4 0.29 0.08 0.07 0.11 0.10 0.09 0.04 0.04 0.03 0.04 0.04 0.08LOI 0.30 2.82 3.76 1.74 1.69 2.21 2.41 2.13 1.41 1.43 3.24 7.78 9.67 7.96 3.56 9.06 1.54 1.88 1.56 1.64 1.7 1.76Sc 33.07 32.84 33.99 33.25 33.78 32.07 32.82 31.48 27.7 .623 6.32 5.12 8.91 7.95 5.56 4.68 5.58 6.26 6.73 6.25 4.03 4.85V 230.60 221.20 226.80 227.60 230.40 230.80 260.10 242.10 261 243 61.47 28.92 56.26 54.27 25.75 38.28 9.52 12.1 7.38 9.28 4.66 10.3Cr 344.35 439.65 103.65 265.76 150.22 191.33 61.17 84.99 40.9 25.9 17.77 45.60 55.33 56.16 34.94 42.56 5.17 8.25 3.13 3.67 3.15 4.02Co 51.49 46.47 65.04 49.35 42.79 40.91 50.18 33.36 32.4 28.1 7.56 19.34 7.59 34.27 1.59 2.40 1.89 1.26 1.98 1.54 1.02 2.67Ni 119.09 188.34 55.22 91.18 50.13 76.41 38.41 41.58 40.7 31 7.94 11.95 20.24 21.95 13.26 16.48 3.81 3.53 2.71 2.69 2.13 3.03Mn 1520 1203 1339 1297 1194 1253 1456 1553 697 401 788 672 492 413Ga 18.59 17.32 18.59 18.42 19.46 20.06 20.61 20.09 20.8 18.8 24.98 14.66 15.26 15.94 17.64 14.10 29.9 31.9 29.4 28.1 21.2 22.2Cs 8.18 6.14 0.96 0.95 0.85 0.50 1.35 0.74 4.77 0.08 0.09 0.09 0.10 4.52Rb 30.41 29.36 48.58 25.99 34.63 41.99 29.13 33.62 54.6 43 41.41 1.25 1.82 0.81 0.14 103.20 201 243 269 219 187 87.1Ba 247.60 253.30 322.60 282.70 360.50 561.10 345.30 442.40 457 416 566.70 157.40 108.10 107.50 112.30 519.50 393 397 602 510 1189 1460Th 3.56 2.43 2.75 3.41 3.76 3.54 4.91 5.24 5.94 5.63 8.51 9.92 15.41 12.19 15.36 15.25 21.7 24.7 25.8 26.3 19.6 14.7U 1.11 0.62 0.76 0.79 0.88 0.84 1.13 1.20 1.17 1.16 3.54 1.93 5.02 1.65 5.99 6.55 2.58 2.43 1.65 2.02 4.72 3.28Nb 16.47 11.89 15.05 13.76 13.76 14.25 20.57 18.75 20.5 18.4 33.02 11.56 10.59 15.82 18.80 14.96 30.4 34.9 25.6 30.4 28.5 37.7Ta 1.19 0.94 1.39 1.10 1.10 1.17 1.43 1.39 1.28 1.2 1.35 1.04 1.20 1.19 1.35 1.43 1.97 2.34 2.35 2.3 2.06 2.15La 18.08 15.47 16.06 19.07 19.87 19.55 26.12 26.59 31.1 30.1 28.23 23.01 26.97 29.13 25.09 25.39 107 136 83.2 118 77.8 65.5Ce 40.73 34.59 36.28 41.87 43.44 43.34 56.80 57.53 67.5 64.5 96.36 44.28 54.26 59.78 72.89 54.40 188 240 163 217 107 123Pb 16.43 3.35 0.87 3.72 5.12 3.60 5.89 9.59 6.33 5.31 37.60 12.22 24.69 23.09 9.00 38.28 21.8 22.3 21.9 21.1 16.2 13.6Pr 5.17 4.42 4.65 5.28 5.68 5.44 7.11 7.18 9.08 8.68 8.10 4.94 6.13 6.77 6.13 6.00 23.8 29.7 18.7 26.6 15.9 14.7Sr 230.30 220.10 217.20 219.20 261.40 306.20 224.20 277.40 261 236 91.84 171.90 187.30 195.70 79.88 248.70 29 37.3 29.2 34 100 91.1Nd 23.45 20.41 21.46 23.66 25.04 24.46 31.42 31.76 40.4 39.7 33.53 18.58 24.22 27.01 24.42 23.33 91.2 112 71.6 105 60 54.6Zr 164.40 133.20 137.50 166.80 176.40 170.40 228.10 232.60 247 236 420.60 109.20 117.60 120.60 382.90 140.00 329 361 394 332 335 367Hf 4.52 3.69 3.85 4.57 4.85 4.71 6.24 6.27 6.45 6.29 11.46 3.65 4.34 7.27 16.17 4.35 10 11.8 12 10.3 9.87 9.56Sm 5.03 4.46 4.62 5.13 5.38 5.33 6.63 6.82 10.1 9.85 6.82 3.14 4.86 5.01 4.49 4.38 16.4 19.6 14 20.3 10.8 9.79Eu 1.68 1.50 1.51 1.73 1.84 1.79 2.12 2.04 2.73 2.54 1.47 0.64 0.90 0.86 0.62 0.78 0.42 0.37 0.4 0.26 1.24 1.77Gd 5.44 4.90 5.13 5.69 5.91 5.88 7.23 7.39 9.03 8.75 7.31 3.11 4.99 5.02 4.81 4.35 14.4 15.7 13.5 19.2 8.97 9.13Tb 0.85 0.77 0.81 0.90 0.93 0.92 1.13 1.15 1.38 1.32 1.17 0.45 0.77 0.73 0.73 0.65 2.3 2.49 2.08 3.1 1.43 1.42Dy 4.89 4.47 4.70 5.16 5.37 5.35 6.48 6.67 7.89 7.5 6.78 2.58 4.33 4.06 4.19 3.63 13 13.9 12.3 17.9 8.21 7.98Y 23.61 21.87 22.75 25.58 26.63 26.67 31.89 33.02 38.8 36 35.67 14.49 24.13 21.58 25.14 20.34 60.4 63.5 64.5 79.6 44.1 41.1Ho 0.96 0.87 0.93 1.02 1.06 1.06 1.29 1.32 1.52 1.46 1.37 0.53 0.87 0.80 0.86 0.74 2.46 2.72 2.43 3.32 1.67 1.57Er 2.59 2.35 2.51 2.75 2.89 2.85 3.52 3.57 4.16 3.99 3.89 1.54 2.48 2.24 2.58 2.12 6.8 7.88 7.11 8.6 4.96 4.65Tm 0.40 0.36 0.39 0.42 0.44 0.43 0.53 0.54 0.59 0.57 0.62 0.25 0.40 0.37 0.44 0.35 0.92 1.06 0.96 1.05 0.65 0.62Yb 2.54 2.23 2.43 2.66 2.75 2.73 3.35 3.39 3.65 3.47 4.17 1.60 2.62 2.44 3.04 2.32 6.07 7.03 6.37 6.58 4.05 4.19Lu 0.38 0.33 0.36 0.39 0.41 0.40 0.49 0.50 0.53 0.51 0.64 0.24 0.39 0.36 0.49 0.35 0.84 1.03 0.86 0.91 0.57 0.59

32H.H

uangetal./Lithos

228–229

(2015)23

–42

Fig. 8. (a) Total alkalis vs. SiO2 diagram (Le Bas et al., 1986) for classification of theXiaotikanlike lava rocks. All the major element data have been recalculated to 100%on a LOI-free basis. Pc, picrobasalt; B, basalt; O1, basaltic andesite; O2, andesite; O3,dacite; R, rhyolite; S1, trachybasalt; S2, basalt trachyte; S3, trachyandesite; T, tra-chyte; U1, basanite and tephrite; U2, phonotephrite; U3, tephritic phonolite; Ph,phonolite; (b) Nb/Y vs. Zr/TiO2 diagram (Winchester and Floyd, 1977). Data sourcesfor Early Permian mafic–ultramafic rocks from the northern margin of Tarim Craton:Keping basalts (Jiang et al., 2004; Li et al., 2012; Zhang et al., 2010a; Zhang et al.,2012a; Zhou et al., 2009); Piqiang mafic–ultramafic complex (Zhang et al., 2010a);Xiaohaizi mafic–ultramafic complex (Zhou et al., 2009); Bachu mafic–ultramaficdykes (Zhang et al., 2010a).


(HFSE, e.g., Nb, Ta, Zr, Hf, P, HREE and Th). Except for slightly negativeNb–Ta, Sr and P anomalies, the patterns of incompatible elements inour basalts are fairly similar to those of many ocean island basalts(OIBs, Fig. 9a). On the chondrite-normalized REE diagram (Fig. 10a),all the samples possess essentially parallel REE patterns characterizedby the relative enrichment in light rare earth elements (LREEs), withlow and restricted (La/Yb)N ratios from 4.74 to 6.22. Besides, theyshownil to slightly negative Eu anomalies, and have Eu/Eu* values vary-ing from 0.84 to 1.0. Compared to the Keping basalts exposed in thenorthern margin of Tarim Craton, which are widely accepted as thebest exposures of Tarim Large Igneous Province (TLIP) basalts, our sam-ples are generally more enriched in Ba, Sr and HREE but more depletedin HREE (Figs. 9a and 10a).

As expected from the intense post-magmatic alteration, the mo-bile element (e.g., Rb–K, Sr and Ba) concentrations in the felsicrocks from the Boziguo'er region are variable (Fig. 9b). The immobileelements, on the contrary, exhibit overall similar patterns. For in-stance, they are mostly enriched in Th–U, with moderate contents

of Nd and Zr–Hf. Besides, in primitive mantle-normalized trace ele-ment spidergrams they show sizeable troughs at Nb–Ta, P and Ti.On the chondrite-normalized REE plot (Fig. 10b), the studied felsicsamples are characterized by gently sloping REEs with (La/Yb)N of4.86 to 10.34, and moderate negative Eu anomalies (Eu/Eu* =0.41–0.64). Obvious positive Ce anomalies are observed in two sam-ples that have Ce/Ce* values [Ce/SQRT (La × Pr)] of 1.56 and 1.44, re-spectively. Despite the variable concentrations in mobile elements,the patterns of immobile element for the felsic lavas from theBoziguo'er region are fairly comparable to that for the rhyolitesfrom Laohutai and Wensu regions (see below).

The rhyolites from the Laohutai and Wensu regions possess essen-tially similar trace element patterns (Fig. 9c) characterized by high con-centrations of Rb–K, Ba, Th–U, moderate Zr–Hf contents, and strongnegative Nb–Ta, Sr, P and Ti anomalies. In terms of rare earth elements(Fig. 10c), they show slightly fractionated REE patterns [(La/Yb)N =5.4–13.9], with slight fractionation of the HREEs [(Gd/Yb)N = 1.1–2.4]and negative Eu anomalies (Eu/Eu* = 0.04–0.67).

5.5. Whole-rock Sr–Nd isotopic compositions

Rb–Sr and Sm–Nd isotopic compositions for four basaltic samplesfrom the Boziguo'er region are listed in Table 3 and the plots of(87Sr/86Sr)t versus εNd(t) are shown in Fig. 11. Some of the publishedisotopic data for Paleozoic igneous rocks exposed in the South TianshanTerrane and the Northern Tarim Craton are also shown for comparison.The basaltic samples display a small range in Sr–Nd isotopic composi-tions, with initial 87Sr/86Sr ratios from 0.70495 to 0.70624 and age-corrected εNd(t) values clustered around −0.5 to +0.6. In Fig. 11, ourdata lie within the OIB field.

6. Discussion

6.1. Petrogenesis of basaltic lava flow

6.1.1. Volcanic seriesThe volcanic series of basaltic flows can provide important clues on

the nature of the mantle sources, conditions for partial melting andmagma differentiation and tectonic regimes. Given the alteration affect,the volcanic series of the basaltic lavas is discriminated basedmainly onmineralogical characteristics and immobile element compositions.

The basaltic samples in our study are dominantly composed ofeuhedral plagioclase in association with subordinate hypidiomorphicpyroxenes (Fig. 3b), suggesting that the feldspars formed earlier thanpyroxene. Furthermore, the samples have low-Ti augites (with TiO2 of0.05–1.30wt.%), moderate to high An contents (An= 42–62) of plagio-clase, and absence of olivine. The above features are consistent withthose of tholeiitic series (McBirney, 1993; Wilson, 1989; Zhang et al.,2012a). Based on the CIPW normative composition (SupplementaryTable 5), five of the basalt samples are quartz-, diopside- andhypersthene-normative and the other five are olivine-, diopside- andhypersthene-normative, suggesting magma with compositions varyingfrom silica-saturated to silica-undersaturated. The Nb/Y ratios of 0.51 to0.70 in the basaltic samples generally fall on the boundary betweensubalkaline and alkaline series, but mostly in the field of subalkaline se-ries (Fig. 8b). In general, petrographic andwhole-rock chemical charac-teristics suggest that the lower basaltic lava flows are mainly tholeiitic,although they straddle a transitional composition between alkalineand tholeiite series.

6.1.2. Crustal contaminationConsidering the relatively low MgO (3.73–7.10 wt.%) and Mg#

(41–61) as well as low contents of compatible element including Ni(31–188ppm) and Cr (26–440 ppm),we infer that the basalts representevolved magmas rather than primitive magmas directly generated bypartial melting of mantle sources.

Fig. 9. Primitivemantle normalized incompatible element patterns of (a) lower basaltic lavaflows (present study), (b) upper felsic lavaflows (present study) and (c)Wensu rhyolites (Luoet al., 2008) from the Xiaotikanlike Formation. Values of primitive mantle, ocean island basalts, N-MORB and E-MORB are from Sun and McDonough (1989). Data sources: mafic rocks ofthe Chuanwulu complex (Huang et al., 2012a); Keping basalts (Jiang et al., 2004; Li et al., 2012; Zhang et al., 2010a; Zhang et al., 2012b; Zhou et al., 2009).


Crustal contamination generally seems inevitable when continentalbasaltic magmas rise from their sources in the mantle through the con-tinental crust (Watson, 1982). Despite the variable MgO and Mg#, theXiaotikanlike basalts exhibit fairly restricted initial 87Sr/86Sr ratios of0.70495 to 0.70624 and εNd(t) values clustered of −0.5 to +0.6(Fig. 11), suggesting that they were derived from a common source re-gion. The slightly negative Nb-Ta and Ti anomalies are present in prim-itivemantle-normalized trace element spidergrams (Fig. 9a), which canbe ascribed to either the nature of mantle source (i.e., lithospheric man-tle previously metasomatized by slab-derived fluid (Baker and Wyllie,1992; Ringwood et al., 1992)) or the contamination by crustalmaterials.However, compared with those in Xiaotikanlike basalts, the mafic rocksfrom the Chuanwulu complex in the western part of the STT displaymore “arc/crust-like” compositions characterized by significant enrich-ment in K and incompatible elements as well as more pronouncedNb–Ta and Ti anomalies (Fig. 9a), which have been considered to bethe distinct signature of lithospheric mantle (Huang et al., 2012a).Therefore, the basaltic lavaflowof theXiaotikanlike Formationwas like-ly to have been derived from a different mantle source.

On the other hand, Precambrian xenocrystic zircons with207Pb/206Pb ages up to 2.6 Ga are abundant in the rock, implying the in-volvement of Tarim basement rocks. In addition, xenocrystic zircons

yield one major age population of ca. 390–440 Ma, with two majorpeak 206Pb/238U ages at 392 Ma and 432 Ma (Fig. 5b), suggesting thatMiddle Silurian to Middle Devonianmagmatic rocks were also involvedduring magma ascent. Some of the earlier studies have demonstratedthat such Early Paleozoic magmatism occurred in a continental arc set-ting associated with the subduction of Paleozoic South Tianshan Ocean(Ge et al., 2012; Huang et al., 2013; Zhang et al., 2007), and that theserocks are characterized by depletion of high-field-strength elements(e.g., Nb, Ta, and Ti) relative to large-ion lithophile elements (e.g., Rb,Ba, and Th). In addition, the Early Paleozoic arc-related felsic rocksfrom the STT also bear “ancient”whole-rock Sr–Nd and zircon Hf isoto-pic signatures (Fig. 11). Correspondingly, we speculate that the involve-ment of Tarim basement and Early Paleozoic continental arc-type rockshas elevated the Th but lowered the Nb–Ta–Ti contents in parentalmagmas, and led to more “ancient” isotopic compositions.

The above interpretation is reinforced by the trace elemental data.All of the studied basaltic samples plot close to the boundary betweenthe MORB-OIB array and the field of active continental margins in theNb/Yb vs. Th/Yb diagram (Fig. 12a), implying the assimilation of conti-nental arc materials. Some incompatible elements, such as Nb, Th andLa, have similar bulk distribution coefficients in most mantle rocktypes, and therefore ratios of Nb/La and Th/Nb are expected to be

Fig. 10. Chondrite-normalized rare earth element patterns of (a) lower basaltic lava flows (present study), (b) upper felsic lava flows (present study) and (c) Wensu rhyolites (Luo et al.,2008) from the Xiaotikanlike Formation. Values of Chondrite, ocean island basalts, N-MORB and E-MORB are from Sun and McDonough (1989). Data sources: mafic rocks of theChuanwulu complex (Huang et al., 2012a); Keping basalts (Jiang et al., 2004; Li et al., 2012; Zhang et al., 2010a; Zhang et al., 2012b; Zhou et al., 2009).


insignificantly modified by partial melting or fractional crystallizationbut very sensitive to the addition of crustal contaminant (Li et al.,2012; Mahoney et al., 2008). Modeling by (Nb/La)N vs. (Th/Nb)N ratios

Table 3Whole-rock Sr and Nd isotopic data for basaltic rocks.

BZXW-4 BZXW-6 XKT-1 XKT-2

(87Rb/86Sr)m 0.64751 0.383527 0.4765 0.3275(87Sr/86Sr)m 0.70809 0.707194 0.706846 0.707548(147Sm/144Nd)m 0.130032 0.129773 0.1401 0.1382(143Nd/144Nd)m 0.512544 0.512494 0.512514 0.512503t(Ma) 286 286 286 286(87Sr/86Sr)t 0.705511 0.705666 0.704948 0.706243(143Nd/144Nd)t 0.512301 0.512251 0.512252 0.512244263f(Sm/Nd)s −0.3 −0.3 −0.3 −0.3εNd(0) −1.8 −2.8 −2.4 −2.6εNd(t) 0.6 −0.4 −0.4 −0.5TDM(Ma) 1105 1192 1318 1307T2DM(Ma) 1006 1085 1084 1096

Fig. 11. (87Sr/86Sr)t vs. εNd(t) plot for the Xiaotikanlike basalts and some other magmaticrocks. Data sources: OIB (Nowell et al., 1998; Pearce et al., 1999; Woodhead et al., 1998).

Fig. 12. (a) Nb/Yb vs. Th/Yb diagram for the Xiaotikanlike basalts and mafic–ultramaficrocks from the northern Tarim Craton. The field of MORB-OIB array is from Pearce(2008). (b) (Nb/La)N vs. (Th/Nb)N diagram for those rocks. Themixing curves are betweenthe OIB and three possible assimilants. One of these is represented by Late Silurian conti-nental arc-type rock composition (blue circle) carrying 17.3 ppm Th, 18.3 ppm Nb and74.9 ppm La (Huang et al., 2013). Tarim basement compositions are represented bythose of two samples. One sample (blue square, Qi et al., 2011) contains 9 ppm Th,6 ppm Nb and 130 ppm La, within the range of granitic or adakitic gneiss compositions.The other (orange circle, Long et al., 2010) has 20 ppm Th, 10 ppm Nb and 90 ppm La,within the range ofmica schist compositions,within the range ofmica schist compositions(Li et al., 2012). Data sources: Xiaotikanlike basalts (present study); Keping basalts (Jianget al., 2004; Li et al., 2012; Zhang et al., 2010a; Zhang et al., 2012b; Zhou et al., 2009);Piqiangmafic–ultramafic complex (Zhang et al., 2010a); Xiaohaizi mafic–ultramafic com-plex (Zhou et al., 2009); Bachu mafic-ultramafic dykes (Zhang et al., 2010a).


indicates that the Xiaotikanlike basaltic samples closely follow themixing curves between the OIB composition and the appropriateassimilant compositions represented by one monzonite sample fromEarly Paleozoic Tie'reke pluton (pluton 20 in Fig. 1c) and by one samplefrom the Precambrian Akesu Formation (Fig. 12b), suggesting the mod-erate degree of crustal contamination.

6.1.3. Fractional crystallizationAs mentioned above, augite grains in samples show Al-, Ca- and Na-

rich cores with more Fe-rich rims, suggesting that the residual magmasevolved to Al-, Ca- and Na-poor compositions. In compliance withmass

balance, the coexisting fractional phase(s) was likely Al-, Ca- and Na-rich. Coupled with the low Sr contents (217–306 ppm) and slightlynegative Eu anomalies (Eu/Eu* = 0.84–1.00), above inference indicatesthe fractionation of plagioclase. More importantly, petrographic obser-vation shows that the plagioclase formed earlier than the other rock-forming minerals in their paragenetic sequence, comparable to afractionation process that was dominated by plagioclase separation.

The Fe2O3T shows increase with decreasing MgO and Mg# in our

samples (figures are not shown), implying the progressive enrichmentof Fe in residual melts. Petrographic observations show that anhedralFe–Ti oxides are present as an interstitialmineral in basaltic samples, in-dicating that they crystallized at a late magmatic stage. As pointed outbymany authors (e.g., Toplis and Carroll, 1995; Veklser, 2009), the frac-tional crystallization of the magma is largely controlled by oxygen fu-gacity. The above features are indicative of reducing conditions duringthe evolution of the basaltic magma, since only low oxygen fugacity(fO2) would prevent the crystallization of magnetite at early stagesand thereby lead to the increasing Fe concentrations in residual liquids(Jang et al., 2001; Toplis and Carroll, 1995). The almost completeabsence of magnetite grains in our samples further suggests a low fO2

condition in the late magmatic stage.In summary, it can be inferred that, in addition to the crustal con-

tamination, the Xiaotikanlike basaltic lavas have undergone fractionalcrystallization dominated by the fractionation of plagioclases underlow oxygen fugacity conditions.

6.1.4. Mantle sourcesWith the exception of slight Nb–Ta–Ti and Sr negative anomalies,

the Xiaotikanlike basalts have trace element and REE patterns broadlyresembling those in oceanic island basalts (OIBs). As discussed above,Nb–Ta–Ti troughswere likely produced bymoderate amounts of crustalcontamination, and Sr negative anomalies can be attributed to fraction-ation of plagioclase in basaltic magmas. Thus, the geochemistry of oursamples is most likely indicative of an OIB-like mantle source.

In most cases, the OIB-like sources were genetically linked to thepartial melting of mantle plume (e.g., Safonova and Santosh, 2014;Zhou et al., 2009) or asthenospheric mantle (Hofmann et al., 1997;Jahn et al., 1999; Kou et al., 2012; Zindler et al., 1984). According tothe estimated clinopyroxene crystallization temperatures of 1309 °Cto 1072 °C, the mantle potential temperature appears normal, and notmarkedly higher than normal asthenospheric mantle as plume-derived magma (Mckenzie and Bickle, 1988; Putirka, 2005; Zhanget al., 2012b), precluding the possibility of a plume-head origin. Alterna-tively, a normal asthenospheric mantle derivation seems much morelikely.

The absence of hydrous minerals in basaltic rocks suggests thatthe primitive magmas were generated by partial melting of anhy-drous mantle peridotite. The involvement of the asthenosphericmantle source is also supported by REE modeling. Partial meltingtaking place in mantle sources with different relative amounts ofgarnet and spinel could extract melts with markedly different trajec-tories in plots of La/Sm vs. Sm/Yb and (Tb/Yb)N vs. (Yb/Sm)N(Fig. 13). Except for two samples (XKT-1-1 and XKT-1-2) that fallclose to the curve of garnet–lherzolite, the Xiaotikanlike basalticsamples mostly plot below the melt path of garnet- and spinel–lherzolite (Fig. 13a). Likewise, with the same exceptions, on theplot of (Tb/Yb)N vs. (Yb/Sm)N (Fig. 13b), majority of Xiaotikanlikebasalts lie closer to a melt path for spinel–peridotite than to one forgarnet–peridotite. Therefore, we propose that the primary magmasof Xiaotikanlike basalts were derived from the partial melting of gar-net–spinel (garnet-bearing, spinel-dominated) peridotitic mantle.

However, the above estimates acquired from REE modeling shouldbe approached with caution, because the positions of the curves inany type of REE modeling can vary with a different choice of meltingmodels, distribution coefficients, source composition, and/or mineralproportions (Zhang et al., 2008b, 2012b). In spite of this drawback,

Fig. 13. (a) La/Sm vs. Sm/Yb diagram showing themelting curves for Xiaotikanlike basaltsandmafic–ultramafic rocks from the northern TarimCraton. The lines are non-modal frac-tionalmelting curves for garnet lherzolite and spinel lherzolite. Numbers on the lines referto percentages of melt. DM= depleted mantle, PM= primitive mantle, N-MORB= nor-mal mid-ocean ridge basalt, Sp = spinel, Gt = garnet. (b) (Tb/Yb)P vs. (Yb/Sm)P for thethose rocks. The grid indicates the range of model melt compositions produced by 1%,5%, 10%, and 15% of aggregated fractional melting (Shaw, 1970) of peridotite in whichthe amount of melting that occurs in the presence of garnet varies from 0% to 100%. Thelight lines indicate the percentage of melt contribution from garnet-facies mantle (Gar);e.g. Gar 0 corresponds to melt from spinel peridotite. Curves of constant melt fractionare shown by bold lines. The curves are for a source consisting of a 1:1 mix of estimatedaverage depleted mantle (Workman and Hart, 2005) and model enriched mantle perido-tite (Ito and Mahoney, 2005). Partition coefficients are taken or interpolated from Saltersand Stracke (2004). The unmelted peridotite is assumed to be 53% olivine, 30%orthopyroxene, 10% clinopyroxene, and 7% garnet or spinel, andmelting of thesemineralsis assumed to occur in proportions of 10%, 10%, 40%, and 40%, respectively, after Janneyet al. (2000). Data sources as in Fig. 12.


REE models provide a first order estimation of the mantle source forXiaotikanlike basalts and suggest that the sources likely containedsmaller amounts of garnet as compared to those for themafic–ultramaf-ic rocks from the northernmargin of Tarim Craton. The latter rockswerepossibly derived from deeper magma source(s) (Fig. 13). This inferenceis in good agreement with the alkali affinity of mafic–ultramafic rocksfrom the northern margin of Tarim Craton (Fig. 8; Zhou et al., 2009;Zhang et al., 2008a, 2010a; Li et al., 2012).

6.2. Genesis of felsic lava flows

The stratigraphic location of the felsic flows above the basaltic lavaflows in the Boziguo'er region suggests the possibility that the formerwas derived by high-degree fractional crystallization of basaltic

magmas or by remelting of earlier-solidified basaltic rocks duringthe same volcanic episode. However, considering the negativeεHf(t) values (from −6.6 to −0.4 with an average of −3.4) of the~290 to ~279 Ma zircons, the felsic rocks do not appear to beco-magmatic with the basalts. Alternatively, ancient crustal rocks,or mixed sources in which the presence of crustal materials issubstantial are more plausible. The variation in ~289 to ~279 Mazircon Hf isotopic compositions necessitates a process that juxta-poses magmas with variable Hf isotopic compositions. This leadsto the possibility that the partial melting of a crustal source waslikely triggered by underplating of mantle-derived magmas thatwere possibly the equivalents to parental magmas of basalts. Thisinterpretation is consistent with the high magma temperatures(755 to 882 °C) calculated based on zircon Ti concentrations.

The Laohutai rhyolitic lavas exhibit trace element patterns largelycomparable to those from the Wensu region, implying that the rhyoliticrocks occurring in the two regions were dominantly derived from thepartial melting of a common source, or compositionally similar protoliths.The high SiO2 concentrations of Laotutai (SiO2 = 73.22–77.51 wt.%, LOI-free, similarly hereinafter) andWensu (SiO2=74.01–78.56wt.%) rhyolit-ic lava flows, coupled with the absence of basaltic to andesitic volcaniclava flows or their plutonic counterparts in the two regions, can be con-sidered as diagnostic criteria for crustal source(s).

Except for three samples, most of Laohutai and Wensu rhyoliticlavas show a strongly peraluminous affinity (ASI N 1.1). However,given the low- to moderate-degree of kaolinization of feldspars presentin the Laohutai rhyolitic lavas and secondary replacement of feldsparsreported from theWensu rocks (Luo et al., 2008), the post-magmatic al-teration may be responsible for the elevated ASI values. Thus, the com-position of their parental magmas is expected to be of metaluminous toweakly peraluminous. If with the above inference is valid, the source re-gionswere predominately composed of igneous or meta-igneous rocks.This interpretation is further supported by the high zircon saturationtemperatures varying from 851 to 906 °C for Laotutai rhyolites andfrom 887 to 933 °C for Wensu rhyolites, respectively. Given the high-K characteristic, we postulate the dehydration melting involving thebreakdown of hydrous minerals, such as amphibole and/or zoisite, inthe genesis of the initial melts (Beard and Lofgren, 1991; Rapp andWatson, 1995). Furthermore, the rhyolitic samples from Laohutai andWensu regions have low Yb/Lu (6.45–7.41), Dy/Yb (1.39–2.72) and(Ho/Yb)N (0.93–1.52) ratios, suggesting that amphibole was the pre-dominant phase in their source region (Moyen, 2009). The low Sr/Y(0.43–2.27 for Laohutai rhyolites and 0.96–12.3 forWensu rhyolites, re-spectively) and Sr/Yb (4.58–24.69 for Laohutai rhyolites and 10.1–100.7for Wensu rhyolites, respectively) ratios, as well as relatively flat HREEpatterns in Laohutai and Wensu rhyolites argue for a garnet-free resi-due (Drummond andDefant, 1990). Negative Sr and Eu anomalies com-monly present in the above samples are indicative of a significant role ofresidual plagioclase during partial melting.

In general, the felsic lavas of the Xiaotikanlike Formation were likelyderived from the partial melting of crustal sources, probably triggeredby underplating of mantle-derived magmas. Specifically, the rhyoliticlava flows from Laohutai and Wensu regions were likely derived fromthe partial remelting of K-rich metabasaltic rocks (amphibolite) withinthe stability field of plagioclase, resulting in a garnet-free residue.

6.3. Tectonic implication

6.3.1. Early Permian volcanism in the South Tianshan Terrane and NorthernTarim Craton

A previous study by Luo et al. (2008) reported an LA-ICP-MS age of289.4 ± 5.5 Ma for rhyolites in the Xiaotikanlike Formation from theWensu region. The results from present study assign a crystallizationage of 285.0 ± 2.3 Ma for upper felsic lava flows in the study area,which is broadly coeval with that of the Wensu rhyolites within errorranges. The two youngest 206Pb/238U ages of zircons from the lower


basaltic lava flows show 287 ± 5 and 288 ± 6 Ma. Taking into accountthe above results, we postulate that the volcanic rocks of Xiaotikanlikeformation formed within a short time in the Early Permian and arethereby likely to be related to the same tectono-thermal event.

Apart from the presently studied Xiaotikanlike volcanic rocks fromthe STT, some Permian volcanic rocks and volcanic-sedimentary succes-sions have been identified in the northern margin of Tarim Craton. Thebasaltic units in the Early Permian Kupukuziman and Kaipaizileike for-mations in the Keping area are considered as the best exposure of theLower Permian basalt sequence, and have attracted considerable atten-tion. Zhang et al. (2010b) reported 40Ar/39Ar plateau ages from282.90 ± 1.55 Ma to 271.93 ± 3.67 Ma for the Kupukuziman basalts,and, based on LA-ICP-MS U–Pb zircon dating, Zhang et al. (2012a) pro-posed that the Kaipaizileike basalts erupted at 291.9 ± 2.2 Ma. In addi-tion to above-mentioned exposures, substantial volume of volcanicsequence is considered to be present in the subsurface. Drill-hole datarevealed that the Early Permian basalts, together with related tuff andtuffaceous rocks in the Tarim Craton may cover an area of ca.250,000 km2 (Chen et al., 2006; Zhou et al., 2009),with an estimated av-erage thickness of ~ 300 m (Chen et al., 2006; Jiang et al., 2004; Zhanget al., 2012a).

6.3.2. Early Permian magmatism in the STT: subduction-, collision- orplume-related?

The magmatic rocks in the South Tianshan Terrane are dominantlyof Early Permian ages (Figs. 1c and d). Different hypotheses have beenproposed for the origin of these magmatic suites. These include:1) northward subduction of the Paleozoic South Tianshan Ocean (Xiaoet al., 2013, and references therein), 2) post-collisional extension fol-lowing the amalgamation of Tarim Craton and Kazakhstan–Yili–CentralTianshan Terrane (Long et al., 2011a; Han et al., 2011, and referencestherein; Gao et al., 2011, and references therein) or 3) Early PermianTarim mantle plume-related activities (Zhang et al., 2008a, 2010a;Zhou et al., 2009).

During the convergence of oceanic crust, subduction-relatedmagmatism occurs only along the active margin of the overridingplate and part of the oceanic crust can be decoupled and accreted ontothe margin of the overriding plate to form an accretionary wedge. Inthe case of the STT, although the southernmost boundary between theaccretionary complex and the Tarim passive margin are still not well-defined, the South-Central Tianshan Suture, i.e., the Atbashi–Inyl'chekFault in Kyrgyzstan and South Nalati Shan Fault and Qawabulak Faultin China (Gao et al., 2009; Xiao et al., 2013), is widely accepted as thenorthernmost boundary. Furthermore, given the widespread occur-rence of island-arc magmatic rocks exposed in the southern margin ofthe Kazakhstan–Yili–Central Tianshan Terrane, a region north of the su-ture, the Late Paleozoic evolution of the Paleozoic South Tianshan Oceanwas likely characterized by northward subduction until thefinal closure(Huang et al., 2013; Xiao et al., 2013, and references therein). Therefore,the South-Central Tianshan Suture appears to mark the site where thePaleozoic South Tianshan Ocean was consumed with northward sub-duction beneath the Kazakhstan–Yili–Central Tianshan Terrane. Thisscenario suggests that the Late Paleozoic strata in the study area belongto either the southernmargin of the accretionary complex or Tarim pas-sive margin sequence, since the area is located very far from the South-Central Tianshan Suture.

The Xiaotikanlike Formation rests unconformably on strongly foldedLower Carboniferous Gancaohu and Yeyungou formations. In fact, thedeformation structures, such as folding, in the STT were plausibly builtby the collisional event, and the ages of overlying volcanic rocksmay provide an upper limit of the suturing time. Regionally, the geo-chemical and isotopic features of other Late Carboniferous to EarlyPermian STT felsic rocks are predominately characterized by K-rich(K2O + 2 wt.% N Na2O), metaluminous to peraluminous affinities withnegative εNd(t) values (Huang et al., 2012a and references therein;Gao et al., 2011), generally pointing to Precambrian (Meso- to Neo-

Proterozoic) mature crustal protoliths. Considering that a large popula-tion of the Late Carboniferous to Early Permian magmatic rocks finallyintruded the country rocks carrying ophiolitic components (stitchinggranitic plutons, Han et al., 2011, and references therein), these rocksevidently postdated the collisional event during which the basementof Tarim crust was underthrust northward beneath the accretionarywedge (Konopelko et al., 2009). In general, the above features suggestthat the collision of the Tarim Craton and Kazakhstan–Yili–CentralTianshan Terrane likely took place during the Late Carboniferous. Thisinterpretation is further supported by several lines of geologic evidence.First, metamorphic zircon U–Pb ages of ~319 Ma (Su et al., 2010), rutileU–Pb age of 318 Ma (Li et al., 2011) and the Sm–Nd isochron age of319Ma (Hegner et al., 2010) for eclogite samples from thewestern seg-ment of the STT consistently indicate that high pressure peakmetamor-phism of subducted oceanic material occurred at the end of the EarlyCarboniferous (Han et al., 2011). Second, the youngest ages ofradiolarian and conodonts fossils from ophiolitic mélanges indicateEarly Carboniferous ages (see Han et al., 2011, and references therein;Huang et al., 2013), and Permian molasse sedimentation are wide-spread in the western part of the STT and adjacent tectonic domains(Shu et al., 2007).

Taking into account the above features, we infer that theXiaotikanlike volcanic lava flows formed after the collision of Tarim Cra-ton and Kazakhstan–Yili–Central Tianshan Terrane. Some workers pro-pose that the tectonomagmatic history of southwestern CAOB duringPermianwas partially or totally related to Tarimmantle plume tectonics(e.g., Pirajno, 2010; Xia et al., 2004, 2008; Zhang and Zou, 2013; Zhanget al., 2008a, 2010a; Zhou et al., 2009). The inferred Tarim Large IgneousProvince (TLIP) consists of mafic–ultramafic intrusive rocks and dykes,flood basalts, and A1-type syenites/granites from the Tarim Craton(Huang et al., 2012b; Zhang and Zou, 2013). However, debates continueover the regional extent of the TLIP, in particular whether the Tianshanregion was part of the TLIP or not. As neighbors to the TLIP, Permianmagmatic rocks in STT can be used to address this dispute. The estimat-ed clinopyroxene crystallization temperature of the studied basalts liesbetween 1309 °C to 1072 °C, representing the lower limit of the mantlepotential temperature (TP). The calculated TP is no higher than those ofnormal asthenospheric mantle, and excess heat supplied by an upwell-ing plume seems not required in the genesis of these basalts. On theregional scale, individual Permianplutons in the STT aremostly elongat-ed in the W–E direction in accordance with the general trend of theSouthern-Central Tianshan Suture (SCTS). Also, the Permian STT igne-ous belt generally extends parallel to the SCTS (Fig. 1c). Such lineardistribution of plutonic rocks does not favor a plume–related origin.

In summary, the Permian magmatism in the STT occurred in re-sponse to the lithospheric thinning in the post-collisional regime, andthat majority, if not all, of those rocks were genetically unrelated tothe Tarim mantle plume.

6.3.3. Implications for the lithosphere extension and crustal evolution in thesouthern CAOB

As discussed in the previous section, the magmatism seems domi-nantly associated with the post-collisional extension, without anyprominent plume-related signature. The breakoff of the subducted oce-anic slab or/and delamination of the thickened lithospheric root werelikely major processes that produced post-collisional magmatism inthe southwestern CAOB (e.g., Han et al., 2011; Long et al., 2011a). Weinfer that, despite evidence for slab breakoff, the delamination oflower lithosphere appears to be a more feasible model to explain thewidespread occurrence of Permianmagmatism in the STT that is locatedsouth of the present South-Central Tianshan Suture and represent thenorthern margin of the “lower” plate. Accompanying the collision ofthe Tarim Craton and Kazakhstan–Yili–Central Tianshan Terrane, theregion beneath the suture zone developed a thickened lithosphere. Sub-sequent to the collision, the thickened subcontinental mantle root andparts of lower crust might have been detached, replaced by hot


asthenospheric mantle. This upwelling mantle then partially melted toproduce the post-collisional magmatism and also induced a corre-sponding extensional tectonic regime.

The geochemical characteristics of the Xiaotikanlike basalts argue forthe presence of garnet-bearing, spinel-dominated peridotitic astheno-spheric source in the middle-eastern part of the STT at ~287 Ma. Ourprevious study (Huang et al., 2012a) on coeval (287.8 ± 4.3 Ma) maficintrusive rocks of the Chuanwulu complex, from the western part ofthe STT, suggested a garnet-bearing lithospheric mantle source. Differ-ences in source components plausibly reflect the variable thickness ofthe lithosphere from east towest.We infer that at ~285Ma, in thewest-ern part of the STT the thickness of the lithospherewas still high enoughto allow the presence of garnet-bearing source in the lithospheric man-tle; whereas considerable lithosphere-thinning occurred in the easternpart of the STT and resulted in the formation of a spinel-dominatedupper asthenosphere source (Fig. 14). In other words, the litho-sphere–asthenosphere boundary beneath the eastern part of the STTwas shallower than that beneath the western part.

Such discrepancy in the lithospheric thickness may be attributed tooblique collisional tectonics. It has been well documented that, along

Fig. 14. Cartoon model showing the tectonic evolution of the South Tianshan Terrane and adjaOcean is proposed as diachronous, and, accordingly, the time of amalgamation between Tarim Ccollisional tectonics, the post-collisional lithospheric extension and thinningwas initiated beneawest. This process resulted in a prominent distinction in the thickness of the lithospheric durinleading to the presence of a spinel-dominated upper asthenosphere source, whereas in the wpresence of garnet-bearing source in the lithospheric mantle. Meanwhile, the upwelling mantaffect the STT.

the length of the suture, the time of the closure of the Paleozoic SouthTianshan Ocean, as well as the amalgamation of Tarim Craton andKazakhstan–Yili–Central Tianshan Terrane, could have been differentand diachronous, with a general westward younging (Chen et al.,1999; de Jong et al., 2006; He et al., 2014; Xiao et al., 2013). As a resultof such scissors-like collisional process, post-collisional extension andthinning would have been initiated beneath the eastern part of theSTT, propagating from east to west progressively. The diachronousextension of lithosphere is also revealed by the age distribution of LatePaleozoic felsic plutons, which show an irregular migration of post-collisional magmatism of the STT from the East to the West (Fig. 1c).

The collision between Tarim Craton and the Kazakhstan–Yili–Cen-tral Tianshan terrane occurred nearly coevally with that between theKazakhstan–Yili–Central Tianshan and Junggar terranes along theNorthern Central Tianshan suture (Han et al., 2010, 2011). Our studydoes not support the presence of Permian plume-triggered magmatismin the South Tianshan Terrane. Consequently, other tectonic units of theWestern Tianshan, which are located farther away from the Tarim Cra-ton, were more unlikely overprinted by plume-related magmatism.Combined with the scenario that Late Carboniferous to Permian

cent regions during the Late Paleozoic. The final closure of the Paleozoic South Tianshanraton and Yili-Central Tianshan Terrane showswestward younging (a). Due to the obliqueth eastern part of the South Tianshan Terrane, and then occurredprogressively fromeast tog Early Permian. Significant lithosphere thinning occurred in the eastern part of the STT,

estern part of the STT, the thickness of the lithosphere was still high enough to allow thele plume beneath Tarim Craton triggered extensive magmatism, but did not significantly


plutonic rocks from theWestern Tianshan show linear disposition alongterrane boundaries, they might represent formation in an orogenic,post-collisional setting. Thus, the generation of majority, if not all, ofPermian igneous in the Western Tianshan still occurred within, ratherthan after, the timespan of the CAOB evolution. Such regional-scale,post-collisional magmatism must have been associated with the up-welling of mantle materials that supplied sufficient heat source. Thus,we believe that mantle materials were largely involved in the genesisof post-collisional rocks exposed in the present Western Tianshan, andthat the rocks themselves represent the net vertical continental growthat the final stage in the orogenic cycle.

7. Conclusions

(1) Based on LA-ICP-MS zircon U–Pb isotope analysis, theXiaotikanlike volcanic lava flows are considered to have eruptedduring the Early Permian (ca. 285 Ma).

(2) The basaltic lava flows show petrographic, mineralogical andgeochemical characteristics similar to those of tholeiitic basalts,and were derived from an asthenospheric mantle source domi-nantly composed of garnet–spinel (garnet-bearing, spinel-dominated) peridotite.

(3) The felsic lavas of the Xiaotikanlike Formation were not co-magmatic with basaltic lavas, and were derived from the partialmelting of crustal sources, probably triggered by underplating ofmantle-derived magmas.

(4) The eruption of Xiaotikanlike volcanic lava flows postdated thecollision of Tarim Craton and Kazakhstan–Yili–Central TianshanTerrane. Coupled with other geological evidence, the SouthTianshan Terrane, as well as other tectonic terranes of theChinese Western Tianshan, was not affected by Permian Tarimmantle plume activities.

(5) Around ~285Ma, the thickness of lithosphere in thewestern partof STT was still high enough to allow the presence of garnet-bearing source in the lower lithosphericmantle.Meanwhile, con-siderable lithosphere-thinning occurred in the eastern part of theSTT and resulted in the generation of a spinel-dominated upperasthenosphere source. Such distinction in the lithospheric thick-ness may be attributed to the oblique collisional tectonics.

(6) Thewidespread Late Paleozoicmagmatic rocks from theWesternTianshan are genetically associated with regional scale post-collisional extension. They represent net vertical continentalgrowth at the final stage of the CAOB evolution.

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.lithos.2015.04.017.

Acknowledgment

We thank Editor Prof. Nelson Eby and three referees from the Journalfor constructive comments. This study was financially supported byNational Nature Science Foundation of China (grant No. 41390442),National 305 Project (2011BAB06B02-04), the 111 Project (No. B07011)and projects of China Geological Survey (No. 12120113094000). Thiswork also contributes to the Talent Award to M. Santosh under the1000 Plan from the Chinese Government. We are grateful to XiuchunZhan and Yuejin Deng of National Research Center for Geoanalysis,Beijing, Ling Ding of Key Laboratory of Continental Collision and PlateauUplift, Chinese Academy of Sciences, Beijing. Zhenyu Chen and SuohanTang of Institute of Geology, Chinese Academy of Geological Sciences,and other workers for their assistance in geochemical, isotopic and LA-ICP-MS determinations.

References

Allen, M.B., Windley, B.F., Zhang, C., 1992. Palaeozoic collisional tectonics and magmatismof the Chinese Tien Shan, central Asia. Tectonophysics 220, 89–115.

Andersen, T., 2002. Correction of common lead in U–Pb analyses that do not report Pb-204. Chemical Geology 192, 59–79.

Baker, M.B., Wyllie, P.J., 1992. High-pressure apatite solubility in carbonate-rich liquids:implications for mantle metasomatism. Geochimica et Cosmochimica Acta 56,3409–3422.

Beard, J.S., Lofgren, G.E., 1991. Dehydration melting and water-saturatedmelting of basal-tic and andesitic greenstones and amphibolites at 1, 3, and 6.9 kb. Journal of Petrology32, 365–401.

Biske, Y.S., Konopelko, D.L., Seltmann, R., 2013. Geodynamics of late Paleozoic magmatismin the Tien Shan and its framework. Geotectonics 47, 291–309.

Blichert-Toft, J., Albarède, F., 1997. The Lu–Hf isotope geochemistry of chondrites and theevolution of the mantle–crust system. Earth and Planetary Science Letters 148,243–258.

Chen, C.M., Lu, H.F., Jia, D., Cai, D.S., Wu, S.M., 1999. Closing history of the southernTianshan oceanic basin, western China: an oblique collisional orogeny.Tectonophysics 302, 23–40.

Chen, H.L., Yang, S.F., Wang, Q.H., Luo, J.C., Jia, C.Z., Wei, G.Q., Li, Z.L., He, G.Y., Hu, A.P.,2006. Sedimentary response to the Early–Mid Permian basaltic magmatism in theTarim plate. Geology in China 33 (3), 545–552 (in Chinese with English abstract).

Chu, N.C., Taylor, R.N., Chavagnac, V., Nesbitt, R.W., Boella, R.M., Milton, J.A., German, C.R.,Bayon, G., Burton, K., 2002. Hf isotope ratio analysis using multi-collector inductivelycoupled plasma mass spectrometry: an evaluation of isobaric interference correc-tions. Journal of Analytical Atomic Spectrometry 17, 1567–1574.

de Jong, K., Xiao, W.J., Windley, B.F., Masago, H., Lo, C.-H., 2006. Ordovician 40Ar/39Arphengite ages from the blueschist-facies Ondor Sum subduction–accretion complex(Inner Mongolia) and implications for the early Palaeozoic history of continentalblocks in China and adjacent areas. American Journal of Science 306, 799–845.

Drummond, M.S., Defant, M.J., 1990. A model for trondhjemite–tonalite–dacite genesisand crustal growth via slab melting: Archean to modern comparisons. Journal ofGeophysical Research 95, 21503–21521.

Elhlou, S., Belousova, E., Griffin, W.L., Pearson, N.J., O'Reilly, S.Y., 2006. Trace element andisotopic composition of GJ-red zircon standard by laser ablation. Geochimica etCosmochimica Acta 70 (18), A158.

Gao, J., Li, M.S., Xiao, X.C., Tang, Y., He, G.Q., 1998. Paleozoic tectonic evolution of theTianshan orogen, northwestern China. Tectonophysics 287, 213–231.

Gao, S., Rudnick, R.L., Xu, W.L., Yuan, H.L., Liu, Y.S., Walker, R.J., Puchtel, I.S., Liu, X.M.,Huang, H., Wang, X.R., Yang, J., 2008. Recycling deep cratonic lithosphere and gener-ation of intraplatemagmatism in the North China Craton. Earth and Planetary ScienceLetters 270 (1–2), 41–53.

Gao, J., Long, L., Klemd, R., Qian, Q., Liu, D., Xiong, X., Su, W., Liu, W., Wang, Y., Yang, F.,2009. Tectonic evolution of the South Tianshan orogen and adjacent regions, NWChina: geochemical and age constraints of granitoid rocks. International Journal ofEarth Sciences 98, 1221–1238.

Gao, J., Klemd, R., Qian, Q., Zhang, X., Li, J.L., Jiang, T., Yang, Y.Q., 2011. The collision be-tween the Yili and Tarim blocks of the Southwestern Altaids: geochemical and ageconstraints of a leucogranite dike crosscutting the HP–LT metamorphic belt in theChinese Tianshan Orogen. Tectonophysics 499, 118–131.

Gao, X.F., Lin, X., Xu, Y.D., Le, M.L., 2014. Evolution of sedimentary basins in SouthTianshan during Paleozoic–Mesozoic. Earth Science-Journal of China University ofGeosciences 8, 1119–1128 (in Chinese with English abstract).

Ge, R., Zhu, W., Wu, H., Zheng, B., Zhu, X., He, J., 2012. The Paleozoic northern margin ofthe Tarim Craton: passive or active? Lithos 142–143, 1–15.

Griffin, W., Pearson, N., Belousova, E., Jackson, S., Van Achterbergh, E., O'Reilly, S.Y., Shee,S., 2000. The Hf isotope composition of cratonic mantle: LAM-MC-ICPMS analysis ofzircon megacrysts in kimberlites. Geochimica et Cosmochimica Acta 64, 133–147.

Guo, R.Q., Nijiati, A., Qin, Q., Jia, X.L., Zhu, Z.X., Wang, K.Z., Li, Y.P., 2013. Geological charac-teristics and tectonic significance of Silurian granitic intrusions in the northern Tarimcraton, Xinjiang. Geological Bulletin of China 32 (2/3), 220–238 (in Chinese with En-glish abstract).

Han, B.F., Wang, S.G., Jahn, B.M., Hong, D.W., Kagami, H., Sun, Y.L., 1997. Depletedmantlemagma source for the Ulungur River A-type granites from north Xinjiang, China: geo-chemistry and Nd–Sr isotopic evidence, and implication for Phanerozoic crustalgrowth. Chemical Geology 138, 135–159.

Han, B.F., Guo, Z.J., Zhang, Z.C., Zheng, L., Chen, J.F., Song, B., 2010. Age, geochemistry, andtectonic implications of a late Paleozoic stitching pluton in the North Tian Shan suturezone, western China. Geological Society of America Bulletin 122, 627–640.

Han, B.-F., He, G.-Q., Wang, X.-C., Guo, Z.-J., 2011. Late Carboniferous collision between theTarim and Kazakhstan–Yili terranes in the western segment of the South Tian ShanOrogen, Central Asia, and implications for the Northern Xinjiang, western China.Earth-Science Reviews 109, 74–93.

He, X.X., Tang, S.H., Zhu, X.K., Wang, J.H., 2007. Precise measurement of Nd isotopic ratiosby means of multi-collector magnetic sector inductively coupled plasma mass spec-trometry. Acta Geoscientica Sinica 28, 405–410 (in Chinese with English abstract).

He, Z.-Y., Zhang, Z.-M., Zong, K.-Q., Xiang, H., Chen, X.-J., Xia, M.-J., 2014. Zircon U–Pb andHf isotopic studies of the Xingxingxia Complex from Eastern Tianshan (NW China):Significance to the reconstruction and tectonics of the southern Central Asian Oro-genic Belt. Lithos 190–191, 485–499.

Hegner, E., Klemd, R., Kröner, A., Corsini, M., Alexeiev, D.V., Iaccheri, L.M., Zack, T., Dulski,P., Xia, X., Windley, B.F., 2010. Mineralogy ages and P–T conditions of late Paleozoichigh-pressure eclogite and provenance of mélange sediments in the South TianshanOrogen of Kyrgyzstan. American Journal of Science 310, 916–950.



http://refhub.elsevier.com/S0024-4937(15)00153-X/rf0005
















































































Hofmann, C., Courtillot, V., Feraud, G., Rochette, P., Yirgu, G., Ketefo, E., Pik, R., 1997.Timing of the Ethiopian flood basalt event and implications for plume birth and glob-al change. Nature 389, 838–841.

Holt, P.J., van Hunen, J., Allen, M.B., 2012. Subsidence of theWest Siberian Basin: effects ofa mantle plume impact. Geology 40, 703–706.

Hoskin, P.W.O., Black, L.P., 2000. Metamorphic zircon formation by solid-state recrystalli-zation of protolith igneous zircon. Journal of Metamorphic Geology 18, 423–439.

Hou, K.J., Li, Y.H., Zou, T.R., Qu, X.M., Shi, Y.R., Xie, G.Q., 2007. Laser ablation-MC-ICP-MStechnique for Hf isotope microanalysis of zircon and its geological applications.Acta Petrologica Sinica 23, 2595–2604 (in Chinese with English abstract).

Huang, H., Zhang, Z.C., Zhang, D.Y., Du, H.X., Ma, L.T., Kang, J.L., Xue, C.J., 2011. Petrogenesisof Late Carboniferous to Early Permian Granitoid Plutons in the Chinese SouthTianshan: implications for Crustal Accretion. Acta Geologica Sinica 85 (8),1305–1333 (in Chinese with English abstract).

Huang, H., Zhang, Z.C., Kusky, T., Zhang, D.Y., Hou, T., Liu, J.L., Zhao, Z.D., 2012a. Geochro-nology and geochemistry of the Chuanwulu complex in the South Tianshan, westernXinjiang, NW China: Implications for petrogenesis and Phanerozoic continentalgrowth. Lithos 140–141, 65–84.

Huang, H., Zhang, Z.C., Kusky, T., Santosh, M., Zhang, S., Zhang, D.Y., Liu, J.L., Zhao, Z.D.,2012b. Continental vertical growth in the transitional zone between South Tianshanand Tarim, western Xinjiang, NW China: insight from the Permian Halajun A1-typegranitic magmatism. Lithos 155, 49-66.

Huang, H., Zhang, Z.C., Santosh, M., Zhang, D.Y., Zhao, Z.D., Liu, J.L., 2013. Early Paleozoictectonic evolution of the South Tianshan Collisional Belt: evidence from geochemistryand zircon U–Pb geochronology of the Tie'reke Monzonite Pluton, Northwest China.The Journal of Geology 121, 401–424.

Huang, H., Zhang, Z.C., Santosh, M., Zhang, D.Y., 2014. Geochronology, geochemistry andmetallogenic implications of the Boziguo'er rare metal-bearing peralkaline graniticintrusion in South Tianshan, NW China. Ore Geology Reviews 61, 157–174.

Ito, G., Mahoney, J.J., 2005. Flow and melting of a heterogeneous mantle: 1. Importance tothe geochemistry of ocean island and mid-ocean ridge basalts. Earth Planetary Sci-ence Letters 230, 29–46.

Jahn, B.M., Wu, F.Y., Lo, C.H., Tsai, C.H., 1999. Crust–mantle interaction induced by deepsubduction of the continental crust: geochemical and Sr\Nd isotopic evidence frompost-collisional mafic–ultramafic intrusions of the northern Dabie complex, centralChina. Chemical Geology 157, 119–146.

Jahn, B.M., Wu, F., Chen, B., 2000. Massive granitoid generation in Central Asian: Nd iso-tope evidence and implication for continental growth in the Phanerozoic. Episodes23 (2), 82–92.

Jang, Y.D., Naslund, H.R., McBirney, A.R., 2001. The differentiation trend of the Skaergaardintrusion and the timing ofmagnetite crystallisation: iron enrichment revisited. EarthPlanetary Science Letters 189, 189–196.

Janney, P.E., Macdougall, J.D., Natland, J.H., Lynch, M.A., 2000. Geochemical evidence fromthe Pukapuka volcanic ridge system for a shallow enriched mantle domain beneaththe South Pacific superswell. Earth Planetary Science Letters 181, 47–60.

Jiang, C.Y., Zhang, P.B., Lu, D.R., Bai, K.Y., Wang, Y.P., Tang, S.H., Wang, J.H., Yang, C., 2004.Petrology, geochemistry and petrogenesis of the Keping basalts and their Nd. Sr andPb isotopic compositions. Geological Review 50, 492–500 (in Chinese with Englishabstract).

Jiang, T., Gao, J., Klemd, R., Qian, Q., Zhang, X., Xiong, X., Wang, X., Tan, Z., Chen, B., 2014.Paleozoic ophiolitic mélanges from the South Tianshan Orogen, NW China: Geologi-cal, geochemical and geochronological implications for the geodynamic setting.Tectonophysics 612–613, 106–127.

Konopelko, D., Biske, G., Seltmann, R., Eklund, O., Belyatsky, B., 2007. Hercynian post-collisional A-type granites of the Kokshaal Range, Southern Tien Shan, Kyrgyzstan.Lithos 97, 140–160.

Konopelko, D., Seltmann, R., Biske, G., Lepekhina, E., Sergeev, S., 2009. Possible source di-chotomy of contemporaneous post-collisional barren I-type versus tin bearing A-typegranites, lying on opposite sides of the South Tien Shan suture. Ore Geology Reviews35, 206–216.

Kou, C.H., Zhang, Z.C., Santosh, M., Huang, H., Hou, T., Liao, B.L., Li, H.B., 2012. Picriticporphyrites generated in a slab-window setting: implications for the transitionfrom Paleo-Tethyan to Neo-Tethyan tectonics. Lithos 155, 375–391.

Kröner, A., Kovach, V., Belousova, E., Hegner, E., Armstrong, R., Dolgopolova, A., Seltmann,R., Alexeiev, D.V., Hoffmann, J.E., Wong, J., Sun, M., Cai, K., Wang, T., Tong, Y., Wilde,S.A., Degtyarev, K.E., Rytsk, E., 2014. Reassessment of continental growth during theaccretionary history of the Central Asian Orogenic Belt. Gondwana Research 25,103–125.

Le Bas, M.J., Le Maitre, R.W., Streckeisen, A., Zanettin, B., 1986. A chemical classification ofvolcanic rocks based on the total alkali-silica diagram. Journal of Petrology 27,745–750.

Li, Q.L., Lin, W., Su, W., Li, X.H., Shi, Y.H., Liu, Y., Tang, G.Q., 2011. SIMS U–Pb rutile age oflow-temperature eclogites from southwestern Chinese Tianshan, NW China. Lithos122, 76–86.

Li, Y.-Q., Li, Z.-L., Sun, Y.-L., Santosh, M., Langmuir, C.H., Chen, H.-L., Yang, S.-F., Chen,Z.-X., Yu, X., 2012. Platinum-group elements and geochemical characteristics ofthe Permian continental flood basalts in the Tarim Basin, northwest China: im-plications for the evolution of the Tarim Large Igneous Province. Chemical Geol-ogy 328, 278–289.

Liu, D., Jolivet, M., Yang, W., Zhang, Z., Cheng, F., Zhu, B., Guo, Z., 2013. Latest Paleo-zoic–Early Mesozoic basin–range interactions in South Tian Shan (northwestChina) and their tectonic significance: Constraints from detrital zircon U–Pbages. Tectonophysics 599, 197–213.

Long, X.-P., Yuan, C., Sun, M., Zhao, G.-C., Xiao, W.-J., Wang, Y.-J., Yang, Y.H., Hu, A.-Q.,2010. Archean crustal evolution of the northern Tarim craton, NW China: Zircon U–Pb and Hf isotopic constraints. Precambrian Research 180 (3–4), 272–284.

Long, L.L., Gao, J., Klemd, R., Beier, C., Qian, Q., Zhang, X., Wang, J.B., Jiang, T., 2011a. Geo-chemical and geochronological studies of granitoid rocks from theWestern TianshanOrogen: implications for continental growth in the southwestern Central AsianOrogenic Belt. Lithos 126 (3–4), 321–340.

Long, X.P., Yuan, C., Sun, M., Kröner, A., Zhao, G.C., Wilde, S., Hu, A.Q., 2011b. Reworking ofthe Tarim Craton by underplating of mantle plume-derived magmas: Evidence fromNeoproterozoic granitoids in the Kuluketage area, NW China. Precambrian Research187, 1–14.

Ludwig, K.R., 2003. User's Manual for Isoplot 3.00: A Geochronological Toolkit forMicrosoft Excel. In: Ludwig, Kenneth R. (Ed.).

Luo, J.H., Che, Z.C., Cao, Y.Z., Zhang, J.Y., 2008. Geochemical and geochronological charac-teristics and its tectonic significance of Early Permian acid volcanic rocks ofXiaotikanlike Formation in the southern margin of South Tianshan orogen, NWChina. Acta Petrologica Sinica 24, 2281–2288 (in Chinese with English abstract).

Mahoney, J.J., Saunders, A.D., Storey, M., Randriamanantenasoa, A., 2008. Geochemistry ofthe Volcan de l' Androy Basalt–Rhyolite Complex, Madagascar Cretaceous IgneousProvince. Journal of Petrology 49, 1069–1096.

McBirney, A.R., 1993. Differentiated rocks of the Galapagos hotspot. Geological Society,London, Special Publications 76, 61–69.

Mckenzie, D., Bickle, M.J., 1988. The volume and composition of melt generated by exten-sion of the lithosphere. Journal of Petrology 29, 625–679.

Moyen, J.F., 2009. High Sr/Y and La/Yb ratios: the meaning of the “adakitic signature”.Lithos 112, 556–574.

Nowell, G.M., Kempton, P.D., Noble, S.R., Fitton, J.G., Saunders, A.D., Mahoney, J.J., Taylor,R.N., 1998. High precision Hf isotopemeasurements of MORB and OIB by thermal ion-ization mass spectrometry: insights into the depleted mantle. Chemical Geology 149,211–233.

Pearce, J.A., 2008. Geochemical fingerprinting of oceanic basalts with applications toophiolite classification and the search for Archean oceanic crust. Lithos 100 (1–4),14–48.

Pearce, J.A., Kempton, P.D., Nowell, G.M., Noble, S.R., 1999. Hf–Nd element and isotopeperspective on the nature of provenance of mantle and subduction components inthe western Pacific arc–basin systems. Journal of Petrology 40, 1579–1611.

Pirajno, F., 2010. Intracontinental strike-slip faults, associated magmatism, mineral sys-tems and mantle dynamics: examples from NW China and Altay–Sayan (Siberia).Journal of Geodynamics 50, 325–346.

Poitrasson, F., Pin, C., Duthou, J.-L., 1995. Hydrothermal remobilization of rare earth ele-ments and its effect on Nd isotopes in rhyolite and granite. Earth and Planetary Sci-ence Letters 130, 1–11.

Putirka, K.D., 2005. Mantle potential temperatures at Hawaii, Iceland, and the mid-oceanridge system, as inferred from olivine phenocrysts: evidence for thermally drivenmantle plumes. Geochemistry, Geophysics, Geosystem 6, Q05L08. http://dx.doi.org/10.1029/2005GC000915.

Putirka, K.D., Mikaelian, H., Ryerson, F., Shaw, H., 2003. New clinopyroxene-liquidthermobarometers for mafic, evolved, and volatile-bearing lava compositions, withapplications to lavas from Tibet and the Snake River Plain, Idaho. American Mineral-ogist 88, 1542–1554.

Qi, Q.-J., Zhang, Z.-C., Dong, S.-Y., Zhang, D.-Y., Huang, H., Zhang, S., Ma, L.-T., 2011. Geo-chemical characteristics and tectonic setting of Mesoproterozoic metamorphicrocks in Aksu area, Southwestern Tianshan Mountains. Acta Petrologica etMineralogica 30 (2), 172–184 (in Chinese with English abstract).

Rapp, R.P., Watson, E.B., 1995. Dehydration melting of metabasalt at 8–32 kbar: implica-tions for continental growth and crust–mantle recycling. Journal of Petrology 36,891–931.

Ringwood, A., Kesson, S., Hibberson, W., Ware, N., 1992. Origin of kimberlites and relatedmagmas. Earth and Planetary Science Letters 113, 521–538.

Rubatto, D., Hermann, J., 2003. Zircon formation during fluid circulation in eclogites(Monviso, Western Alps): implications for Zr and Hf budget in subduction zones.Geochimica et Cosmochimica Acta 67, 2173–2187.

Safonova, I.Yu., Santosh, M., 2014. Accretionary complexes in the Asia-Pacific region: trac-ing archives of ocean plate stratigraphy and tracking mantle plumes. Gondwana Re-search 25, 126–158.

Salters, V.J.M., Stracke, A., 2004. Composition of the depleted mantle. Geochemistry, Geo-physics, Geosystems 5, Q05004. http://dx.doi.org/10.1029/02003GC000597.

Scherer, E., Münker, C., Mezger, K., 2001. Calibration of the lutetium-hafnium clock. Sci-ence 293, 683–687.

Scherer, E.E., Whitehouse, M.J., Münker, C., 2007. Zircon as a monitor of crustal growth.Elements 3, 19–24.

Seltmann, R., Konopelko, D., Biske, G., Divaev, F., Sergeev, S., 2011. Hercynian post-collisional magmatism in the context of Paleozoic magmatic evolution of the TienShan orogenic belt. Journal of Asian Earth Sciences 42, 821–838.

Şengör, A.M.C., Natal'in, B., 1996. Turkic-type orogeny and its role in the making ofthe continental crust. Annual Reviews of Earth and Planetary Sciences 24,263–337.

Sengör, A.M.C., Natal'in, B.A., Burtman, V.S., 1993. Evolution of the Altaid tectonic collageand Paleozoic crustal growth in Eurasia. Nature 364, 299–307.

Shaw, D.M., 1970. Trace element fractionation during anatexis. Geochimica etCosmochimica Acta 34, 237–243.

Shu, L.S., Wang, B., Zhu, W.B., 2007. Age and tectonic significance of radiolarian fossilsfrom the Heiyingshan ophiolitic mélange, southern Tianshan Belt. NW China. ActaGeologica Sinica 81, 1–8 (in Chinese with English abstract).

Sobolev, S.V., Sobolev, A.V., Kuzmin, D.V., Krivolutskaya, N.A., Petrunin, A.G., Arndt, N.T.,Radko, V.A., Vasiliev, Y.R., 2011. Linking mantle plumes, large igneous provincesand environmental catastrophes. Nature 477, 312–316.

Su, W., Gao, J., Klemd, R., Li, J.L., Zhang, X., Li, X.H., Chen, N.S., Zhang, L., 2010. U–Pb zircongeochronology of Tianshan eclogites in NW China: implication for the collision





















































































































http://dx.doi.org/10.1029/2005GC000915




















http://dx.doi.org/10.1029/02003GC000597























between the Yili and Tarimblocks of the southwestern Altaids. European Journal ofMineralogy 22, 473–478.

Sun, S.S., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts:implications for mantle composition and processes. In: Saunders, A.D., Norry, M.J.(Eds.), Magmatism in the Ocean Basins., Special Publications 42. Geological Society,London, pp. 313–345.

Sun, L.H., Wang, Y.J., Fan,W.M., Zi, J.W., 2008. A further discussion of the petrogenesis andtectonic implication of theMazhashan Syenites in the Bachu Area. Journal of Jilin Uni-versity (Earth Science Edition) 38 (1), 8–20 (in Chinese with English abstract).

Toplis, M.J., Carroll, M.R., 1995. An experimental study of the influence of oxygen fugacityon Fe–Ti oxide stability, phase relations and mineral-melt equilibria in ferro-basalticsystems. Journal of Petrology 36, 1137–1170.

van Achterbergh, E., Ryan, C., Jackson, S., Griffin,W., 2001. Data reduction software for LA-ICP-MS. Laser-Ablation-ICPMS in the earth sciences—principles and applications.Miner Assoc Can (short course series) 29, 239–243.

Veklser, I.V., 2009. Extreme iron enrichment and liquid immiscibility in mafic intrusions:experimental evidence revisited. Lithos 111, 72–82.

Wang, L.S., Xia, L.Q., Dong, Y.P., Xu, X.Y., Xia, Z.C., Li, X.M., Ma, Z.P., Yang, P.F., 2005. Geo-logical meaning of unconformility of dip between the Lower Carboniferous strataand its underlying strata. Northwestern Geology 38 (1), 1009–6248 (in Chinesewith English abstract).

Wang, C., Liu, L., Che, Z., Luo, J., Zhang, J., 2007. Geochronology, petrogenesis and signifi-cance of Baleigong mafic rocks in Kokshal segment, southwestern Tianshan Moun-tains. Geological review 53, 743–754 (in Chinese with English abstract).

Wang, C., Luo, J.H., Che, Z.C., Liu, L., Zhang, J.Y., 2009. Geochemical Characteristics and U–Pb LA-ICP-MS zircon dating of the Oxidaban pluton from Xinjiang, China: Implica-tions for a Paleozoic oceanic subduction process in Southwestern Tianshan. ActaGeologica Sinica 83 (2), 272–283 (in Chinese with English abstract).

Watson, E.B., 1982. Basalt contamination by continental crust: some experiments andmodels. Contributions to Mineralogy and Petrology 80, 73–87.

Watson, E.B., Wark, D.A., Thomas, J.B., 2006. Crystallization thermometers for zircon andrutile. Contributions to Mineralogy and Petrology 151, 413–433.

Wei, X., Xu, Y.G., 2011. Petrogenesis of Xiaohaizi syenite complex from Bachu area, Tarim.Acta Petrologica Sinica 27, 2984–3004 (in Chinese with English abstract).

Wilhem, C., Windley, B.F., Stampfli, G.M., 2012. The Altaids of Central Asia: a tectonic andevolutionary innovative review. Earth-Science Reviews 113, 303–341.

Wilson, B.M., 1989. Igneous Petrogenesis A Global Tectonic Approach. Springer.Winchester, J.A., Floyd, P.A., 1977. Geochemical discrimination of different magma series

and their differentiation products using immobile elements. Chemical Geology 20,325–343.

Windley, B.F., Alexeiev, D., Xiao, W., Kröner, A., Badarch, G., 2007. Tectonic models for ac-cretion of the Central Asian Orogenic Belt. Journal of the Geological Society 164,31–47.

Woodhead, J.D., Eggins, S.M., Johnson, R.W., 1998. Magma genesis in the New Britain is-land arc: further insights into melting and mass transfer processes. Journal of Petrol-ogy 39, 1641–1668.

Workman, R.K., Hart, S.R., 2005. Major and trace element composition of the depletedMORB mantle (DMM). Earth Planetary Science Letters 231, 53–72.

Wu, F.Y., Yang, Y.H., Xie, L.W., Yang, J.H., Xu, P., 2006. Hf isotopic compositions of the stan-dard zircons and baddeleyites used in U–Pb geochronology. Chemical Geology 234,105–126.

Xia, L.Q., Xu, X.Y., Xia, Z.C., Li, X.M., Ma, Z.P., Wang, L.S., 2004. Petrogenesis of Carbonifer-ous rift-related volcanic rocks in the Tianshan, northwestern China. Geological Socie-ty of America Bulletin 116, 419–433.

Xia, L.-Q., Xia, Z.-C., Xu, X.-Y., Li, X.-M., Ma, Z.-P., 2008. Relative contributions of crust andmantle to the generation of the Tianshan Carboniferous rift-related basic lavas, north-western China. Journal of Asian Earth Sciences 31, 357–378.

Xiao,W.J., Santosh, M., 2014. The western Central Asian Orogenic Belt: a window to accre-tionary orogenesis and continental growth. Gondwana Research 25, 1429–1444.

Xiao, W.J., Windley, B.F., Allen, M.B., Han, C.M., 2013. Paleozoic multiple accretionary andcollisional tectonics of the Chinese Tianshan orogenic collage. Gondwana Research23, 1316–1341.

Xiao, W.J., Han, C., Liu, W., Wan, B., Zhang, J., Ao, S., Zhang, Z., Song, D., Tian, X., Luo, J.,2014. How many sutures in the southern Central Asian Orogenic Belt: insights fromEast Xinjiang–West Gansu (NW China)? Geoscience Frontiers 5, 525–536.

Xie, W., Song, X.-Y., Deng, Y.-F., Wang, Y.-S., Ba, D.-H., Zheng, W.-Q., Li, X.-B., 2012. Geo-chemistry and petrogenetic implications of a Late Devonian mafic–ultramafic intru-sion at the southern margin of the Central Asian Orogenic Belt. Lithos 144–145,209–230.

Yang, S.H., Zhou, M.F., 2009. Geochemistry of the 430-Ma Jingbulake mafic–ultramafic in-trusion in Western Xinjiang, NW China: implications for subduction relatedmagmatism in the South Tianshan orogenic belt. Lithos 113, 259–273.

Yuan, H.L., Gao, S., Liu, X.M., Li, H.M., Günther, D., Wu, F.Y., 2004. Accurate U–Pb age andtrace element determinations of zircon by laser ablation inductively coupled plasma-mass spectrometry. Geostandards and Geoanalytical Research 28, 353–370.

Zhang, C.-L., Zou, H.-B., 2013. Permian A-type granites in Tarim and western part of Cen-tral Asian Orogenic Belt (CAOB): genetically related to a common Permian mantleplume? Lithos 172–173, 47–60.

Zhang, C.L., Zhou, D.W., Wang, J.L., Wang, R.S., 2007. Geochronology, geochemistry andSr–Nd isotopic composition and genesis implications of Huangjianshishan granite in-trusion in Kumishi area of southern Tianshan. Acta Petrologica Sinica 23, 1821–1829(in Chinese with English abstract).

Zhang, C.L., Li, X.H., Li, Z.X., Ye, H.M., Li, C.N., 2008a. A Permian layered intrusive complexin the western Tarim Block, northwestern China: product of a ca.275 Ma mantleplume? Journal of Geology 116, 112–128.

Zhang, Z.C., Xiao, X.C., Wang, J., Wang, Y., Kusky, T.M., 2008b. Post-collisional Plio-Pleistocene shoshonitic volcanism in thewestern KunlunMountains, NW China: geo-chemical constraints on mantle source characteristics and petrogenesis. Journal ofAsian Earth Sciences 31, 379–403.

Zhang, C.L., Xu, Y.G., Li, Z.X., Wang, H.Y., Ye, H.M., 2010a. Diverse Permian magmatism inthe Tarim Block, NW China: Genetically linked to the Permian Tarim mantle plume?Lithos 119, 537–552.

Zhang, Y., Liu, J., Guo, Z., 2010b. Permian basaltic rocks in the Tarim basin, NW China: im-plications for plume–lithosphere interaction. Gondwana Research 18, 596–610.

Zhang, D.Y., Zhou, T.F., Yuan, F., Jowitt, S.M., Fan, Y., Liu, S., 2012a. Source, evolution andemplacement of Permian Tarim Basalts: evidence from U–Pb dating, Sr–Nd–Pb–Hfisotope systematics and whole rock geochemistry of basalts from the Keping area,Xinjiang Uygur Autonomous region, northwest China. Journal of Asian Earth Sciences49, 175–190.

Zhang, Z.C., Kang, J.L., Kusky, T., Santosh, M., Huang, H., Zhang, D.Y., Zhu, J., 2012b. Geo-chronology, geochemistry and petrogenesis of Neoproterozoic basalts fromSugetbrak, northwest Tarim block, China: implications for the onset of Rodinia super-continent breakup. Precambrian Research 220–221, 158–176.

Zhang, D.Y., Zhang, Z.C., Santosh, M., Cheng, Z.G., He, H., Kang, J.L., 2013a. Perovskite andbaddeleyite from kimberlitic intrusions in the Tarim large igneous province signal theonset of an end-Carboniferous mantle plume. Earth and Planetary Science Letters361, 238–248.

Zhang, L.F., Du, J.X., Lü, Z., Yang, X., Gou, L.L., Xia, B., Chen, Z.Y., Wei, C.J., Song, S.G., 2013b.A huge oceanic-type UHP metamorphic belt in southwestern Tianshan, China: peakmetamorphic age and P–T path. Chinese Science Bulletin 58, 4378–4383.

Zhou, J.B., Wilde, S.A., 2013. The crustal accretion history and tectonic evolution of the NEChina segment of the Central Asian Orogenic Belt. Gondwana Research 23,1365–1377.

Zhou, M.F., Zhao, J.H., Jiang, C.Y., Gao, J.F., Wang, W., Yang, S.H., 2009. OIB-like, heteroge-neous mantle sources of Permian basaltic magmatism in the western Tarim Basin,NW China: implications for a possible Permian large igneous province. Lithos 113,583–594.

Zhu, Z.X., Li, J.Y., Dong, L.H.,Wang, K.Z., Liu, G.Z., Li, Y.P., Liu, Z.T., 2008a. Age determinationand geological significance of Devonian granitic intrusions in Seriyakeyilake region,northern margin of Tarim basin, Xinjiang. Acta Petrologica Sinica 24, 971–976 (inChinese with English abstract).

Zhu, Z.X., Li, J.Y., Dong, L.H., Zhang, X.F., Hu, J.W.,Wang, K.Z., 2008b. The age determinationof Late Carboniferous intrusions in Mangqisu region and its constraints to the closureof oceanic basin in South Tianshan, Xinjiang. Acta Petrologica Sinica 24 (12),2761–2766 (in Chinese with English abstract).

Zindler, A., Staudigel, H., Batiza, R., 1984. Isotope and trace element geochemistry ofyoung Pacific seamounts: implication foe the scale of upper mantle heterogeneity.Earth and Planetary Science Letters 70, 175–195.































































































































petrogenesis of the early permian volcanic rocks in the chinese … · 2017-05-09 · petrogenesis...

Documents