Available online at www.sciencedirect.com

www.elsevier.com/locate/jaes

Journal of Asian Earth Sciences 32 (2008) 165–183

The association of mafic–ultramafic intrusions and A-type magmatismin the Tian Shan and Altay orogens, NW China:

Implications for geodynamic evolution and potentialfor the discovery of new ore deposits

Franco Pirajno a,*, Jingwen Mao b, Zhaochong Zhang c, Zuoheng Zhang b, Fengmei Chai c

a Geological Survey of Western Australia, 100 Plain Street, Perth, WA 6004, Australiab Institute of Mineral Resources, Chinese Academy of Geological Sciences, Beijing 100037, PR China

c State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Beijing 100083, PR China

Abstract

The NW China region is characterised by tectonic and lithostratigraphic domains, such as the Tian Shan and Altay orogens, theTarim, Junggar and Turpan Basins. The Tian Shan and Altay orogens are part of the Central Asian Orogenic Belt. The NW Chinaregion was affected by a series of thermal events that occurred between the Silurian and the Triassic, which resulted in the emplacementof numerous granitic plutons and mafic–ultramafic intrusions. A number of these granitic plutons are of A-type affiliation, which on thebasis of the positive eNd values are likely to have been derived from mantle sources. In addition, at least two large igneous provinces(LIPs) can be recognised in NW China, namely the 345–325 Ma Tian Shan LIP and the ca. 270–280 Ma Tarim LIP. Age and field datasuggest a spatial and temporal relationship between the mafic–ultramafic intrusions and A-type granites within the LIPs. In this paper wediscuss mafic–ultramafic intrusions that host magmatic Ni–Cu sulphide deposits (Kalatongke in the Altay, Huangshan and Poyi–Poshi)in the eastern Tian Shan. These intrusions are typically zoned, characterised by an envelope of early gabbroic rocks that enclose laterultramafic units. These zoned mafic–ultramafic intrusions have some features that are comparable with Alaskan-type complexes. Takinginto consideration the spatial–temporal relationship of the mafic, mafic–ultramafic rocks and A-type granites, we suggest that these mag-matic events occurred during an extensional regime, possibly related to a mantle superplume event that affected much of central Asiaduring the Permian, of which the Siberian Traps and the Emeishan continental flood basalts of SW China are part. If the A-type felsicmagmatism took place during a superplume event, we also suggest that these rocks may be conducive to host iron–oxide–copper–gold(IOCG) style mineralisation. We conclude with a model that attempts to explain the relationship between the zoned mafic–ultramaficintrusions and mantle plume activity in NW China during the Permian.� 2007 Elsevier Ltd. All rights reserved.

Keywords: Tian Shan and Altay orogens; Tarim; Junggar; Mafic–ultramafic intrusions; Large igneous provinces; Mantle plume

1. Introduction

The NW China region encompasses terranes and tec-tonic units that are part of the great Central Asian Oro-genic Belt (CAOB; Jahn, 2004) or Altaid orogenic collage(Sengor et al., 1993), or Central Asian Orogenic Supercol-lage (Yakubchuk et al., 2005). More recently, the CAOB

1367-9120/$ - see front matter � 2007 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jseaes.2007.10.012

* Corresponding author. Tel.: +61 8 922 231 55; fax: +61 9 222 36 33.E-mail address: franco.pirajno@doir.wa.gov.au (F. Pirajno).

was described in some detail by Windley et al. (2007).The CAOB extends from the Uralides in the west to thePacific Ocean margin of eastern Asia and is bounded tothe north by the Siberian Craton and to the south by theTarim–North China cratonic blocks (Fig. 1). The CAOBis a complex collage of fragments of ancient microconti-nents and arc terranes, fragments of oceanic volcanicislands (e.g. seamounts), perhaps also volcanic plateaux(e.g. Junggar block, discussed below), oceanic crust(ophiolites), and successions formed at passive continental

Fig. 1. Simplified geological map of northwestern China, showing main tectonic units. After Pirajno et al. (1997) and Bureau of Geology and MineralResources of Xingjiang Uygur Autonomous Province (1993). Extent of flood basalts on the Tarim platform is simplified and taken from Chen et al. (2006).

166 F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183

margins. The amalgamation of these terranes occurred atvarious times in the Palaeozoic and Mesozoic and wasaccompanied by episodes of magmatism, ranging in agefrom Ordovician (ca. 450 Ma) to Triassic–Cretaceous (ca.220–120 Ma) that resulted in the emplacement of large vol-umes of granitoid intrusions (Jahn, 2004) and mafic volca-nic rocks (Wang et al., 2007; Zhu et al., 2005),accompanied by lesser volumes of mafic–ultramafic mate-rial. A-type granitic and peralkaline intrusions in theCAOB are common and are associated with post-colli-sional tectonism. Elsewhere, in NE China and Mongolia,A-type granites are associated with extensive Mesozoicand Tertiary volcanism (Jahn, 2004). Nd–Sr isotope studiesindicate that these A-type and peralkaline granites are juve-nile and of mantle origin (Jahn, 2004).

In NW China the Altay Orogen, Junggar Basin and theTian Shan orogenic belt, as mentioned above, are charac-terised by accreted terranes and that include island arcs,oceanic crust and continental fragments. These terraneswere later affected by rifting processes and post-collisionintraplate rift magmatism with granitic, mafic–ultramaficintrusions and continental flood basalts, between the Car-boniferous–Permian and the Late Cretaceous (Allenet al., 1992; Xia et al., 2003, 2004). There are at least fourmajor Phanerozoic intraplate magmatic events that haveaffected the region, in order of decreasing ages: (1) ca.

410–390 Ma; (2) ca. 330–310 Ma, (3) ca. 300–270 Ma; (4)and ca. 250 Ma, as further elaborated in Section 5.

In this paper, the geology and origin of a number ofmineralised mafic–ultramafic systems are discussed. Thesemafic–ultramafic systems are part of a series of intrusivebodies, commonly found along faults or sutures and thathave been generally labelled ophiolites (that is originatingfrom and representing fragments of oceanic crust, e.g.Hsu, 2003; Zhou et al., 2004; Sengor and Natal’in, 2004),that are present in the Altay and Tian Shan orogens. Typ-ically, these intrusions are funnel-shaped and concentri-cally zoned with lenses of ultramafic rocks enclosed in anenvelope of gabbroic composition. We draw informationfrom published literature as well as our own field, unpub-lished and published data to show that most of these sys-tems are not ophiolites sensu stricto, but formed asintrusive complexes that were emplaced as a result ofwithin-plate magmatic activity. We examine the possibility,in the light of mantle dynamics, that these funnel-shapedand zoned mafic–ultramafic intrusions are a variant ofAlaskan-type complexes and suggest a geodynamic modelinvoking the upward movement of a series of ‘‘plumelets’’rising from the head of a mantle plume(s) that impingedonto the lithosphere in Permian–Triassic times, followingthe collision or amalgamation of the terranes that consti-tute the Altay and Tian Shan orogens.

F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183 167

In a separate section, we describe features of A-typemagmatism in the two orogens. These rocks are becomingincreasingly recognised as forming a substantial fraction ofthe granitic rocks in the orogens. These granites are lesspopular as a research subject, perhaps because they arenot perceived as having the same economic importance interms of metallic mineral deposits as the mafic–ultramaficintrusions. Therefore, we have drawn from the recent liter-ature on A-type granitic rocks, largely because these rockslend themselves to precise U–Pb zircon dating, thereby pro-viding important constraints on the geodynamic history ofthe orogenic belts. We further suggest that the space–timeassociation of mafic–ultramafic intrusions with A-type gra-nitic rocks is due to common mantle-linked thermal events,probably a mantle plume(s). If our model is correct, the A-type magmatism in NW China also may have produced arange of intrusion-related hydrothermal deposits, includingiron–oxide–copper–gold (IOCG) deposits of the OlympicDam type.

2. Regional overview of NW China

Northwest China is occupied by the Xingjiang Uygur,autonomous region, approximately between latitudes 49�and 35�N, and longitudes of 74� and 97�E. Its area of1.6 million km2 accounts for one sixth of the entire Peoples’Republic China. Historically, the Province, located alongthe famous Silk Road, was an important gateway. Xinjianghas borders with eight countries: Mongolia, Russia,Kazakhstan, Kyrgyzstan, Tajikistan, Afghanistan, Paki-stan and India. The physiography of NW China is domi-nated by the three economically important basins, ofTarim, Turpan and Junggar, rich in coal and hydrocarbonresources. These basins are surrounded by spectacularmountain chains. These mountains are the Altay in thenorth, the Tian Shan in the centre and the Kunlun–Karakorum in the south. The Altay mountain range marksthe political borders between NW China, Mongolia, Russiaand Kazakhstan. In Turkic the word Altay means goldmountain (al = gold, tau = mount). The Tian Shan(Tian = heaven, Shan = mountain), extends across the cen-tre, geographically separating the Province into tworegions, southern and northern Xinjiang.

The Altay and Tian Shan form a complex system of oro-genic belts within the CAOB, crisscrossed by strike–slipfaults and sutures. The history of convergence and assem-bly of the microcontinental blocks and subduction–accre-tion complexes in NW China has been investigated byColeman (1989), Nie et al. (1990), Shi et al. (1994), Mossa-kovsky et al. (1993), Carroll et al. (1995), Xiao et al.(2004a,b).

The main tectonic elements of NW China are shown inFig. 1. They are the Tarim craton and the stable massifs orplatform domains of the Ili Block, Junggar and Turpanbasins, surrounded by the network of the Kunlun–Karako-rum, Altay and Tian Shan orogenic belts, bound anddefined by fossil convergent zones, now preserved as major

sutures and fault zones. This network of orogenic belts is ofeconomic importance, because the rocks of these belts hostmost of the metallic mineral deposits in the region (Ruiet al., 2002; Mao et al., 2003). In Xinjiang Province, theorogenic belts of the Altay-East Junggar and West Junggarare in the north, and the complex Tian Shan lies on thenorth margin of the Tarim basin (Fig. 1). On the southernand southwestern sides of the Tarim Block are the Kunlun–Karakorum orogenic belts and the ENE-trending AltynTagh strike slip fault (Fig. 1). The basement geology ofthe Tarim, Turpan and Junggar is poorly known becauseof their thick cover of Mesozoic and Cenozoic sedimentaryrocks (Allen et al., 1992).

The orogens of NW China are intruded by numerousgranitic plutons and batholiths of Palaeozoic age. The dis-tribution of granites in northwest China is schematicallyshown in Fig. 2. Many of these are considered by Coleman(1989) to be A-type anorogenic granites. However, this isdisputed by Allen et al. (1992), on geochemical grounds.They consider the granitic rocks of the Tian Shan orogensas being due to arc magmatism. It is more likely that thegranites of the region include arc-related syn- to post-oro-genic, as well as anorogenic types, as shown in the pub-lished geological maps of Xingjiang (Bureau of Geologyand Mineral Resources of Xingjiang Uygur AutonomousProvince, 1993). Radiometric dating of these granites indi-cates that they were emplaced during the Carboniferous tothe Triassic (ages range approximately from 330 to230 Ma; Jahn, 2004).

The Junggar basin has on its northern and eastern sidesthe Altay–Kelamaili fold belt, and on its western side thenortheast-trending West Junggar fold belt. These belts con-tain highly deformed and dismembered ophiolitic rocks,associated with cherts, acid-intermediate volcanics and tur-bidite facies sedimentary sequences of Permo–Carbonifer-ous age. The Junggar basin has a Mesozoic–Cenozoicsedimentary cover which may exceed 11-km in thickness(Carroll et al., 1990). There is a lack of detailed strati-graphic and sedimentological data for the Junggar base-ment rocks. Carroll et al. (1990), who have carried outstudies of the Junggar rocks, suggested that the basinmay have originated either as a remnant oceanic basin oras a Mid-Carboniferous back-arc basin, which was emerg-ing to the north of the Turpan region. These authors pro-posed a number of possible tectonic scenarios, in all ofwhich the Junggar appears as a remnant oceanic basin ofthe Palaeotethys ocean, between two magmatic arcs, theBogda arc in the south (North Tian Shan) and the Kelama-ili arc (East Junggar)–Altay arcs in the east and north. Inthe model considered more likely, subduction of the Jung-gar ocean during the Carboniferous was directed south-ward, forming the North Tian Shan arc and northwardbeneath the Kelamaili arc. The latter was sutured to theAltay arc, by closure of a Kelamaili ocean, prior to theCarboniferous. Some authors suggested that it was partof a microcontinent and may have originated ether as aremnant oceanic plateau or as a Mid-Carboniferous

Fig. 2. Simplified geological map of a selected region encompassing parts of the Tian Shan and Altay orogens, Junggar, Tuha basins in northwest China,showing distribution of granitic rocks, mafic–ultramafic intrusions and intraplate mafic volcanics. Boxed areas, KA = Kalatongke, HU = Huangshan, POPoyi = Poshi, are discussed in this paper. Based on Ren et al. (1999); distribution of intraplate volcanics is after Xia et al. (2004).

168 F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183

back-arc basin (Zhao et al., 1996). As pointed out by Jahn(2004), it is an important fact that mafic–ultramafic andophiolitic melanges rock are located along the margins ofthe Basin (Fig. 1). This is especially evident on the nort-western margin, where a zone of ophiolitic rocks constitutethe well-known Talabute ophiolite belt, northwest of thetown of Karamai (Fig. 2).

The Tarim is the largest of the northwest China basins(Figs. 1 and 2). It has an area of about 560,000 km2, andis covered by a sedimentary succession, estimated to besome 15-km thick (Sengor et al., 1996). Oil and gas fieldsare present in the Palaeozoic marine sedimentary succes-sion (Li et al., 1996). Although the basement of the Tarimbasin is largely unknown, rocks of Proterozoic andArchaean age are exposed along the northern margin.Sengor et al. (1996), on structural and geophysical grounds,proposed that it may be underlain by an oceanic plateau.Between the Early and Late Permian, the Tarim blockwas situated on the northern active margin of the Palaeot-

ethys ocean, with the Tian Shan magmatic arcs and mar-ginal basins being part of north-and south-dippingsubduction systems (see below). Palaeomagnetic data indi-cate that the Tarim block, together with the South ChinaBlock, was joined to the Australian margin of Gondwana,forming a composite block, situated at low latitudes (Zhaoet al., 1996). The collision of the Tarim block with theJunggar block (Kazakhstania microcontinent) occurredduring the Late Devonian–Early Carboniferous (Allenet al., 1992; Carroll et al., 1995), so that the accretion ofthe subduction systems between the two blocks, resultedin the formation of the Tian Shan orogenic belts.

The Turpan basin (also called Tu-Ha Basin (Fig. 2), fromTurpan and Hami towns), has the second lowest land sur-face on Earth (150 m below sea level), and is positionedbetween the Junggar basin and the northeast side of theTarim basin, is surrounded on all sides by the Tian Shanfold belts. The evolution of the Turpan basin has beenstudied by Windley et al. (1990) and Allen et al. (1991),

F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183 169

who proposed that the Turpan was formed as a forelandbasin, during the Palaeozoic collisions that formed theTian Shan orogenic belts.

3. Altay orogen

The Altay orogen extends for more than 2500 km fromKazakhstan, across northeastern Xinjiang province to wes-tern Mongolia and was formed through a complex series ofevents that include accretion, subduction and opening andclosing of small basins (Windley et al., 2002; Goldfarbet al., 2003). The Chinese part of the Altay orogen wasdivided by Windley et al. (2002) into five fault-bound terr-anes. These are summarised below: (1) The Altayshan ter-rane consists of Late Devonian–Early Carboniferousmetasedimentary rocks, overlain by shale, siltstone, grey-wackes and limestone (Kumasu Group). These rocks arein fault contact with sedimentary and intermediate-felsicvolcanic rocks. All are intruded by granitic plutons. Theterrane is interpreted to represent the remnants of twoisland arcs; (2) The NW Altayshan terrane contains sedi-mentary and volcanic rocks of Neoproterozoic to EarlyDevonian age forming a sequence of metasediments andschist rocks up to 6000-m thick (Habahe Group), inter-preted as continental-derived turbidites. These rocks areunconformably overlain by shale, limestone, andesite andporphyry volcanic rocks (Baihaba Group), associated withtonalite, granodiorite and hornblende granite plutons. Oneof these granites has a Sm/Nd isochron age of 390 Ma. Theentire succession is interpreted as a continental volcanicarc; (3) The Central Altayshan terrane forms the centralpart of the Chinese Altay orogen and contains high-grademetamorphic rocks and granites ranging in age from Neo-proterozoic to Silurian. In one area (west of the FuyunFault) is a belt of gneiss, migmatites, quartzite and fossilif-erous marble (Habahe Group), overlain by a succession ofa continental turbidites, approximately 8000-m thick. Inthe Keketuohai area are major pegmatite fields, includingone of the largest pegmatites in the world, associated withpost-tectonic granites. Undeformed granites in the Keketu-ohai area have U–Pb SHRIMP and Sm–Nd ages of ca.248 Ma (Zhu et al., 2005); (4) The Qiongkuer-Abagong ter-rane consists of upper Silurian to lower Devonian arcandesitic volcanic and volcanic clastic rocks with lesserbasaltic rocks with a Rb–Sr isochron ages ranging from307 to 285 Ma. These rocks are overlain by the Altay For-mation which consists of a turbiditic sandstone-shale suc-cession, associated with basaltic pillow lavas; (5) TheErqis terrane is wedged in between terrane 4 and the Irtyshfault (see below). This terrane consists of a high-grademetamorphic Precambrian basement (gneiss and schist)with Pb model ages ranging from 1849 and 1791 Ma. Thisbasement is overlain by Devonian and Carboniferous fos-siliferous sedimentary rocks, intruded by post-orogenicgranites.

An important structural element of the Chinese Altayorogen is the Irtysh (also spelt Erqis or Yrtys) transcurrent

fault zone (Figs. 1 and 3), which extends for more than1000 km into Mongolia and Kazakhstan and can reachwidths of up to 50 km. Both sinistral and dextral strike slipmovements occurred along the Irtysh fault zone. Goldfarbet al. (2003) suggested that the terranes of the Altay andJunggar regions were amalgamated into a Cordilleran typeorogen by the early Permian. South of the Irtysh fault is aDevonian–Carboniferous terrane that is interpreted as anisland arc (felsic-intermediate volcanic rocks and calc-alka-line plutons) that is accreted onto the eastern margin of theJunggar Basin (Windley et al., 2002).

An important part of the geodynamic evolution of theAltay orogen took place at the Permian–Triassic transition,when oblique collision occurred between the Tarim blockand the Altay collage and the North China craton (Yakub-chuk et al., 2005). Alkaline magmatism of Yanshanian age(Triassic–Cretaceous) extended far inland from the NorthChina craton and affected the Altay orogen as well (Yak-ubchuk et al., 2005). The zoned mafic–ultramafic plutonsof the Kalatongke belt (Fig. 3 and see below) are tempo-rally and spatially associated with Permian A-type granites.The Kalatongke mafic rocks have been dated using the Rb–Sr method (285 and 298 Ma), and by Sm–Nd isochron(298 Ma with eNd (t) = + 6.0; Han et al., 1997; and refer-ences therein). Han et al. (1997) pointed out that the Ulun-gur A-type granites and the mafic rocks in the region,considered to be ophiolites, have a similar range of eNd

(t) values and therefore these granites and the mafic rocksmay both originate from the same long-lived depletedupper mantle source. These authors go on to state thatthese A-type granites, and by inference the associated maficrocks, are anorogenic, mantle-derived and emplaced in riftsettings. Further discussion of this scenario is givenbelow.

3.1. Kalatongke Ni–Cu deposit

The Kalatongke region in the Altay orogen is character-ised by a 200-km long and 20-km wide zone of mafic–ultra-mafic intrusions, along the Irtysh fault (Fig. 3) (Yan et al.,2003). These mafic–ultramafic rocks intruded sedimentaryand volcanic rocks of Lower Carboniferous age (Yanet al., 2003). In the area of the Kalatongke Ni–Cu depositat least eleven zoned and differentiated mafic–ultramaficcomplexes have been identified. The main Kalatongkeintrusion hosting the sulphide ores is funnel-shaped andcharacterised by a fine-grained biotite–hornblende gabbroshell enclosing biotite–hornblende diorite, biotite–horn-blende norite and olivine norite. The boundaries betweenthese zones are gradational.

Economic magmatic Ni–Cu sulphide deposits occur inthree of the above mentioned eleven zoned mafic intru-sions, constituting the Kalangtoke deposit, one of the larg-est in Xinjiang Province (Jinchuan is the largest Ni–Cudeposit in China; Li et al., 2005). The geology, geochemis-try and isotope systematics of the Kalatongke ores andtheir host rocks have been studied by Yan et al. (2003)

Fig. 3. Simplified geology of the southeastern portion of the Altay orogen; note location of Kalatongke and distribution of mafic–ultramafic intrusions.After Yan et al. (2003).

170 F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183

and Zhang Z-H (unpublished data). The Kalatongke maficrocks have total alkali contents of up to 7%, suggesting analkaline affinity. Geochemical data published by Yan et al.(2003) show that these intrusions contain less MgO andmore silica and alkalies suggesting some degree of differen-

tiation. In addition, these authors also report increases inore-forming elements (S, Cu, Ni, Co) with increases inthe MgO contents of the rocks (i.e. from diorite to norite).Sulphide ores form massive to disseminated zones, withhigh-grade massive ores generally in the centre, grading

F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183 171

outward from strong to weak disseminations. The ore min-erals are typically chalcopyrite, pyrrhotite and pentlanditewith varying amounts of pyrite, violarite, magnetite, Au–Ag mineral and tellurides.

Sulphur isotope analyses carried out by Yan et al. (2003)on 96 samples from an ore-bearing Kalatongke intrusiongive a range of d34S from �3.5 to +3.0&, with an averageof 0.2&. These values are consistent with mantle-derivedsulphur. Lead isotopic compositions of sulphide mineralsreported by the same authors indicate that the lead wasderived from the subcontinental mantle. Zhang et al.(2008) showed that the Kalatongke intrusions have initial87Sr/86Sr ratios ranging from 0.70375 to 0.7054 and eNd(t)from +6.3 to +8.2, implying that the magmas originatedfrom depleted asthenospheric mantle, but with strongcrustal contamination as revealed by Re–Os isotopesystematics. In addition, precise Re–Os dating on sulphideores yielded isochron ages ranging from �282 to 284 Ma(Zhang et al., 2008). These authors pointed out thatthese ages are, within errors, close to those of A-type gran-ites in the Altay orogen and also of related Cu and Audeposits.

4. Tian Shan

The Tian Shan orogen is �2500-km long and up to 500-km wide, extending from NW China (Xinjiang Province)through to Kyrgyzstan, Uzbekistan and Kazakhstan,where the orogen is represented by a collage of island arcsof Cambrian–Ordovician age that host important porphyryCu deposits (Jenchureva, 1997). The NW China sector ofthe Tian Shan orogen is shown in Figs. 1 and 2.

The Tian Shan orogenic belt can be divided into North,Central and South Tian Shan, separated by major E-W-trending thrusts (Fig. 1). The North Tian Shan, best repre-sented in the Bogda Shan range east of Xinjiang’s capitalcity, Urumqi, consists of Carboniferous calc-alkaline vol-canic and sedimentary rocks, intruded by mafic and inter-mediate-felsic plutons. The Bogda Shan, together withthe Kanggur terrane (named the Jueluotage belt or terraneby local geologists), are interpreted as volcanic arcs, whichcould have formed either by north-directed or by a south-directed subduction of oceanic crust and separated by anocean from other island arcs, such as the Aqishan–Yamansu arc. The Central Tian Shan is a wedge-shapedzone comprising, from north to south, a forearc melange,a Carboniferous volcanic arc and a Silurian–Devonian vol-canic arc. The Central Tian Shan also contains basementinliers of Late Proterozoic age, which may have been partof an earlier microcontinent (i.e. Ili block or microconti-nent) (Zhang et al., 1984). The Ili block (also called IliBasin) is considered to have been rifted from the Tarimplate in the Ordovician, later to be amalgamated andsutured between the Tarim and Junggar–Kazakhstanblocks in the Late Palaeozoic (Wang et al., 2007). The Ilirift is of metallogenic importance because it hosts epither-mal precious metal mineralisation (e.g. Axi and Jinxi-Yel-

mand Au deposits, Xiao et al., 2005). The South TianShan contains fragments of oceanic crust material in faultcontact with sandstone, shale, chert and limestone ofMid-Silurian to Mid-Carboniferous age, which were possi-bly deposited on a passive margin on the north side of theTarim block (Carroll et al., 1995). The South Tian Shanbelt consist of Carboniferous felsic to intermediate volcanicand volcaniclastic rocks attributed to the Kanggur andAqishan–Yamansu volcanic arcs, and Silurian volcanogen-ic sedimentary rocks, deposited in a back-arc setting. Theserocks are separated from the North Tian Shan by a majorE-W-trending suture, the Kanggur Fault. The Central TianShan collided with a north-facing passive margin on thenorth side of the Tarim during Late Devonian–Early Car-boniferous. In this way, the South and Central Tian Shanwere amalgamated with the Tarim Block. Between the LateCarboniferous and Early Permian, the intervening oceanbetween the North Tian Shan arc and the amalgamatedTarim–South–Central Tian Shan tectonic unit, closed andthe North Tian Shan arc was accreted to the Central TianShan orogen. These collision events resulted in the forma-tion of foreland basins on the Junggar and Tarim blocksand of intramontane basins (e.g. Turpan) (Windley et al.,1990).

As mentioned above the Tian Shan orogen consists of anumber of accreted terranes that were amalgamated anduplifted during the collision between the Tarim Blockand the Junggar Block (or Kazakhstan microcontinent),between the Late Devonian and Early Permian (Carrollet al., 1995; Chen, 1997). In detail, the geodynamic evolu-tion of the Tian Shan orogen is still not well known,although numerous regional studies (Coleman, 1989;Windley et al., 1990; Allen et al., 1992; Carroll et al.,1995; Jahn, 2004) have helped to unravel aspects of itscomplex geological history. The Tian Shan was upliftedduring the last 10 million years (uplift is ongoing), inresponse to horizontal stresses linked with the India–Asiacollision (Abdrakhmatov et al., 1996).

The arc settings of the Central Tian Shan, Bogda Shan,North Tian Shan and the Ili rift have been disputed.These are viewed by several authors as continental rifts(Xia et al., 2004 and references therein). Indeed, it hasbeen suggested that the Tian Shan orogenic belt evolvedinto a series of post-orogenic rift structures, in whichvoluminous continental rift-type volcanism took place(see below). Furthermore, associated with these post-oro-genic rifts are numerous mafic–ultramafic intrusions,arranged along the length of the rifts structures andincluding the Huangshan group, the Baishiquan intru-sions in the eastern Tian Shan, and the Poyi–Poshi intru-sions near the NE margin of the Tarim basin (Fig. 2). Inthis paper we consider further the Huangshan and Poyi–Poshi mafic–ultramafic intrusive complexes. Chai et al.(this issue) provide details of the geology and mineralisa-tion of the Baishiquan intrusions, for which eNd(t = 284)ranging from 1.67 to 7.61 and Sri ratios ranging from0.704005 to 0.717559 are reported.

172 F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183

4.1. Huangshan Ni–Cu deposits

Gu et al. (1995) and Zhou et al. (2004) have describedthe geology and mineralisation of the mafic–ultramafic sys-tems in the Huangshan-Jingerquan district in the NorthTian Shan domain. A summary of their observation fol-lows. The Huangshan-Jingerquan district, about 140 kmsoutheast of the town of Hami, contains more than 25mafic–ultramafic complexes, some of which have subeco-nomic and economic magmatic Ni–Cu–(PGE) mineralisa-tion. The Xianshan, Huangshanxi (xi means west) andHuangshandong (dong means east) are part of a series ofmafic–ultramafic complexes along the Aqishan-Yamansusuture zone. These complexes are spatially associated with

Fig. 4. Geology of the Huangshan region (A) and detail of Huangshan East

the east–northeast-trending Kanggur–Gandun faults andthe main rock types forming them are: peridotite, olivinepyroxenite, websterite, gabbronorite, troctolite, gabbroand diorite. It is significant that all these rock types containhornblende and Zhou et al. (2004) regard this as an indica-tion that water must have been added to the source mag-mas prior to their emplacement. These mafic–ultramaficrocks intrude flysch-type sedimentary rocks of the Mid-Carboniferous Gangdun Group and basaltic lavas (spi-lite-keratophyre) and pyroclastic rocks of the WutongwoziGroup (Fig. 4).

The funnel-shaped Huangshanxi intrusion is about 3.8-km long and 800-m wide and is composed of an early ultra-mafic unit, intruded by a layered ultramafic–mafic unit and

Ni–Cu-bearing mafic–ultramafic intrusions. After Zhang et al. (2008).

F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183 173

a late mafic unit. The rock type of the first unit is perido-tite, the second unit comprises, from base to top, wehrlite(ca. 150 m), olivine websterite (ca. 570 m), plagioclase web-sterite (ca. 350 m), gabbro and noritic gabbro (ca. 200 m)and finally diorite (ca. 150 m). Zhou et al. (2004) suggesteda crystallisation sequence as follows: cpx ± ol fi cpx-opx ± ol fi cpx-opx ± pl fi cpx-opx-pl fi pl-qtz-bt. Thelast unit forms the lower margin of the intrusion and is afine-grained gabbronorite. At least 40 ore zones have beenidentified in the Huangshanxi intrusion with resources ofabout 80 Mt grading 0.54% Ni and 0.30% Cu. The ore min-erals are pyrrhotite, pentlandite and lesser chalcopyrite.The Huangshandong intrusion (Fig. 4) is lozenge-shapedin cross-section and lens-shaped on the surface, 3.5-kmlong and 1.2-km wide, and like the Huangshanxi intrusionis composed by a sequence of intrusive phases. From thebase to top, rock types of the first phase are olivine gabbro,hornblende gabbro and diorite. The second phase consistsof gabbronorite dykes that intrude phase one rocks; thethird phase comprises hornblende lherzolite, troctoliteand gabbro. Twenty ore zones are present in the Huang-shandong intrusion with reserves of approximately135 Mt, grading 0.30% Ni and 0.16% Cu. The ore mineralsare pyrrhotite, pentlandite, chalcopyrite and pyrite.

SHRIMP U–Pb dating of zircons from the more evolvedrocks of the Huangshanxi intrusion by Zhou et al. (2004)yielded a mean 206Pb/238U age of 269 ± 2 Ma. In addition,the same investigators carried out Rb–Sr and Sm–Nd anal-yses and obtained 87Sr/86Sr ratio of 0.71023 ± 4 and143Nd/144Nd ratio of 0.511845 and positive eNd(t) valuesranging from +6.7 to +9.3. Zhang et al. (2008) carriedout Re–Os dating of sulphide ore from Huangshandongand obtained an age of 284 ± 14 Ma. A Rb–Sr isochronage of ca. 260 Ma from a muscovite granite that intrudesthe mafic–ultramafic rocks was reported by Gu et al.(1995).

Zhou et al. (2004) suggested that the Huangshan intru-sions are typical of continental settings and, on the basisof geochemical data, that the parental magmas werehigh-Mg tholeiitic basaltic magmas. The Huangshan Ni–Cu mineralisation is of magmatic origin as shown by cumu-lus–intercumulus textures and its stratiform-strataboundstyle. The timing of the Huangshan magmas is post-oro-genic and within-plate, as are many of the A-type graniticrocks in the region (see below). Zhou et al. (2004) proposeda mantle plume model to explain the emplacement of themafic–ultramafic intrusions and associated A-type granites.We return to this topic in the last section of this paper.

4.2. Poyi–Poshi and Luodong mafic–ultramafic intrusions

Poyi–Poshi is a funnel-shaped and zoned mafic–ultra-mafic intrusion, about 30-km long and 10-km wide, situ-ated about 300 km southwest of the town of Hami in theTurpan Basin. It intrudes Proterozoic basement rocksand is associated with possibly coeval Permian–Carbonifer-ous granitic and dioritic plutons (Fig. 5). Rock types

include gabbro, olivine gabbro, dunite and peridotite.The ultramafic zones are enclosed in gabbroic rocks.North-trending dioritic and mafic dykes cut the intrusion(Fig. 5). The intrusion is surrounded by skarn developedin Proterozoic carbonate country rocks. Three Ni–Cu pros-pects have been identified by the No. 6 Geological Team ofthe Xinjiang Bureau of Geology and Mineral Resourcesand these are, from west to east, Luodong, Poshi and Poiyi(Fig. 5). At Poshi, stratiform sulphide disseminations arepresent in the ultramafic zone, at the boundary betweenperidotite and pyroxenite, with grades of up to 0.7% Ni(No. 6 Geological Team leader, personal communication,2005). Ore minerals are pyrrhotite, chalcopyrite and minorpentlandite. The Luodong intrusion is about 2 km · 1 kmand consists of gabbro enclosing lenses of dunite andpyroxenite. Recent drilling carried out at Luodong didnot intersect economic mineralisation.

A SHRIMP zircon age of 278 ± 2 Ma is reported for thePoyi intrusion (Mao J-W, unpublished data), and it reason-able to assume that Poshi and Luodong are of the sameage. These intrusions are spatially associated with Perm-ian–Carboniferous granitic plutons.

5. Large igneous provinces (LIPs) in NW China

Three major thermal events affected NW China,between the Carboniferous and the Permian–Triassic peri-ods: Tian Shan, Tarim and Emeishan. These events, brieflydiscussed below, were marked by the eruption of basalticlavas, mafic–ultramafic intrusions and A-type graniticmagmatism.

5.1. Tian Shan

Xia et al. (2003, 2004) reported on the widespread occur-rence of Carboniferous volcanic rocks (ca. 345–325 Ma) inthe Tian Shan orogenic belt (Fig. 2). As mentioned above,the Ili block and Bogda Shan probably formed as Carbon-iferous continental rift systems associated with widespreadtholeiitic volcanism (Xia et al., 2004). In the west, is the IliCarboniferous–Permian rift, and in the east are the Bogdarift (on the north side of the Turpan basin) and the Juelot-age rift (on the south side of the Turpan Basin). This entireregion of Carboniferous rift-related volcanic rocks consti-tutes a LIP of approximately 210,000 km2, named the TianShan LIP. Xia et al. (2004) pointed out that this is only aminimum estimate of the area, because these volcanic rocksalso extend into Kazakhstan, Kyrgyzstan and Mongolia.In the Tian Shan the volcanic succession attains thicknessesvarying from several hundred metres to over 13 km in theIli block. The volcanic rocks of the Tian Shan LIP belongto the tholeiitic, alkaline and calc-alkaline series. The rea-son for these variations is probably due to heterogeneitiesin the basement lithologies and the subcontinental litho-spheric mantle (Xia et al., 2003). The geochronology ofthe Tian Shan LIP is poorly constrained and the only datesavailable, at the time of writing, are Ar–Ar ages of 345–

Fig. 5. Geology of the Poyi–Poshi area and mafic–ultramafic intrusions (A); image of the same area downloaded from Google Earth. Note north-southtrending dyke swarms that are spatially, and possibly co-genetically, associated with the mafic–ultramafic intrusions.

174 F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183

325 Ma and 327–306 Ma and a SHRIMP U–Pb zircon ageof 321–319 (Li et al., 2004). Trace element geochemistryand Sr–Nd isotopic data (87Sr/86Sr(i) = 0.703–0.705 and

eNd(t) = + 4 to +7) reported by Xia et al. (2003, 2004)indicate that the Tian Shan rift-related volcanic rocks orig-inated from asthenospheric OIB-type mantle source, with

F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183 175

contributions from crustal and continental lithosphericmantle. Furthermore, the geochemical data of Xia et al.(2003, 2004) also indicate that high-Ti alkaline lavas arepredominant in the western Tian Shan (Ili rift) and thatlow-Ti lavas predominantly erupted in the eastern TianShan rifts. Xia et al. (2003) concluded that this spatial var-iation may reflect differences in lithospheric thickness andthe thermal structure of the asthenospheric mantle, so thatthe eastern Tian Shan lavas originated from a hot mantleand thinner lithosphere, perhaps in the axis of a mantleplume. By contrast, the western Tian Shan alkaline lavasmay have formed by lower degrees of melting from aperipheral zone of the mantle plume, where temperaturesare lower and the lithosphere was thicker.

5.2. Tarim

In the Tarim Basin, tholeiitic volcanic rocks, maficdykes, ultramafic rocks and syenites form a LIP, estimatedto be greater than 200,000 km2 in the western and south-western parts of the Basin (see Fig. 1; Yang et al., 2006;Chen et al., 2006; Chen et al., 1997). K–Ar dating yieldsa range of ages from 277 ± 4 to 288 ± 10 Ma, whereasAr–Ar dating shows a plateau age of 278.5 ± 1.4 Ma(Yang, personal communication, 2006). One of the intru-sive rocks is a quartz-syenite with an A-type geochemicalaffinity (Li et al., 2006). The Huangshan mafic–ultramaficintrusions dated at ca. 270 Ma and the Kalatongke intru-sions at ca. 284 Ma may be part of this thermal event.The Permian volcanic rocks in the Mongolian orogeniczone, north of Beijing, may also belong to this event sincethey yield Rb–Sr isochrons of ca. 270 Ma (Zhu et al., 2001).Although not yet well defined, this magmatic province maybe of a huge extension, extending from the Tarim block tothe northern margin of the North China Craton, a distancein excess of 3000 km. This is therefore a huge LIP. We alsospeculate that this large scale magmatism may have beenlinked with a mantle plume and the evolution of the Palaeo–Tethys ocean (Yang et al., 2006).

The ca. 260–250 Ma Siberian–Emeishan thermal eventaffected large parts of Central and Western Asia (Dobret-sov, 2005). The Emeishan Large Igneous Province (ELIP)covers an area of at least 250,000 km2 in SW China (Yun-nan, Sichuan and Guizhou Provinces), and in NW Vietnam(Chung et al., 1998), but according to Xiao et al. (2003,2004a,b); this is a conservative estimate. The ELIP consistsof a succession of predominantly tholeiites, locally associ-ated with picritic and rhyolitic lava flows (Zhang et al.,2006). In addition to lava flows, mafic–ultramafic layeredcomplexes, dikes and sills, syenite and other alkaline intru-sions, are part of the ELIP (Xiao et al., 2004a,b). The geo-chronology of the ELIP is fairly well established, withprecise radiometric ages on mafic and ultramafic intrusionsranging from 259 to 262 Ma and those on lavas from 246 to254 Ma; the peak magmatism was probably between 251and 253 Ma (Dobretsov, 2005 and references therein).The ELIP was more or less synchronous with the Siberian

Traps and together these may identify a Permian superp-lume event that affected much of the Asian continent inthe Permian (see Dobretsov, 2005).

6. A-type magmatism

Loiselle and Wones (1979) defined A-type granitic rocksas those that occur within-plate rift settings and are geo-chemically characterised by high alkalies, high abundancesof Nb, Zr, REE but with a negative Eu anomaly and lowCaO and MgO contents. Commonly, A-type granitoidshave high Y/Nb ratios and high halogen contents, suggest-ing that volatiles have an important role in their petrogen-esis (Eby, 1990). Eby (1992) reviewed the main features ofA-type magmatic rocks and further subdivided them intoA1 and A2 groups, based on Rb/Sr and Y/Nb ratios. Theseare lower for A1 types and higher for A2 types. Moreover,Eby (1992) and Wu et al. (2002) proposed that A1 grani-toids are geochemically similar to ocean-island-basalt(OIB) and were emplaced during intraplate rifting andrelated to plume events. In contrast, A2 granitoids havemixed geochemical signatures of continental crust andisland arc and are considered to form in a post-orogenicsetting (Eby, 1992; Wu et al., 2002). Nevertheless it is alsopointed out by Eby (1992) that discriminant ratios, such asY/Nb, can change if the magma interacts with crustalmaterials.

6.1. Altay orogen

In the Altay Orogen along the Ulungur River, approxi-mately parallel with the Kalatongke belt of mafic–ultramaficrocks, is a northwest-trending belt of approximately 200 km,which contains A-type peralkaline granites. Han et al.(1997) interpreted the Ulungur granites as having beenemplaced in an extensional setting. The Ulungur peralka-line granites have Y/Nb ratios from 1.39 to 3.33 and eNd(t)values ranging from +5.1 to + 6.7 (Han et al., 1997. In theSawuer region in the western Junggar, A-type alkali feld-spar granites have SHRIMP ages ranging from ca. 290 to297 Ma (Zhou et al., 2006). These A-type granites correlatewith similar A-type granites in the East Junggar with zirconU–Pb ages of ca. 290 Ma. Zhou et al. (2006) suggested thatthese rocks formed in a post-collisional extensional settingand that they are associated with Permian volcanic rocks(ca. 270–286 Ma; cf. Tarim LIP referred to above). Post-collisional intrusions along the Talabute Fault (also knownas Darabut Fault), in the Western Junggar, are A- and I-type granitic rocks of the Miaogou and Karamai suites,which include gabbro, diorite, tonalite, granodiorite andmonzonite. These rocks have SHRIMP U–Pb ages rangingfrom ca. 297 to 305 Ma and eNd(t) values from +8.4 to+6.6 (Chen and Arakawa, 2005). Chen and Jahn (2004)and Chen and Arakawa (2005) suggested that these rockswere derived from a parental magma of mantle origin,and were formed by partial melting of juvenile lower crust(Miaogou suite) and from the differentiation of a basaltic

176 F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183

underplate (Karamai suite). The magmas thus producedare associated with a post-collisional extensional regime.This partial melting of mixed sources, from mafic–ultra-mafic underplates and lower crust may be a common fea-ture in NW China and perhaps in much of the CAOB.

6.2. Tian Shan

The ca. 240 Ma Weiya post-orogenic pluton in the east-ern Tian Shan consists of gabbro, quartz syenite, biotitemonzogranite, diorite and fine-grained granite (Zhanget al., 2005). The mineralogical, geochemical and petrolog-ical signatures of the Weiya pluton have led Zhang et al.(2005) to consider two distinct sources for the gabbroand syenite rocks, but both related to extension of the lith-osphere. The gabbro is thought to have derived from a spi-nel-lherzolite source that had been metasomatised by fluidsfrom continental subduction and subsequently partiallymelted (Zhang et al., 2005). The syenite would have beenderived from a granulite source at the base of continentalcrust (Zhang et al., 2005). The granitoid rocks of the ca.323 Ma Tuwu porphyry Cu–Mo prospect, located alongthe major Kangguertag Fault zone in the Eastern TianShan consist of granite porphyry and diorite intrudingbasaltic and andesitic rocks. These rocks have Y/Nb ratiosranging from 1.55 to 4.22, and possibly belong to the A2type of Eby (1992).

Fig. 6. Schematic geology of the Eurasian continent showing Permo–Triassicwithin boxed area. After and slightly modified from Nikishin et al. (2002).

6.3. Discussion

Crustal and lithospheric thinning induce thermal distur-bances, which cause a sequence of partial melting eventsfirst in the mantle, then in the lower crust (Barker et al.,1975). Large mafic magma chambers can rise from mantleplumes and underplate and intrude the lower crust. Seismictransects in the Baltic Shield and Sea provided evidencesupporting such a model of mafic underplating of the lowercrust and the intrusion of mafic sheets (BABEL WorkingGroup, 1993). In this region, high-temperature, dry andfluorine-rich magmas (Rapakivi granites) in associationwith intrusions of mantle-derived mafic magmas are pres-ent (Puura and Floden, 1999).

Extensive post-orogenic magmatism in the entire regionof NW China is characterised by A-type granitic intrusionsand mafic–ultramafic complexes. This, locally alkaline andperalkaline, magmatism extends for thousands of kilome-tres from Xinjiang, through southern Mongolia and InnerMongolia to North China, where granitic rocks wereemplaced in two stages, first in the Permian–Triassic(300–250 Ma) and then during the Late Triassic–Creta-ceous (210–120 Ma) (Hong et al., 1996; Hong et al.,2004; Jahn, 2004). The intrusion of these A-type magmasoccurred in the very large Permian–Triassic rift systemoccupying central Asia (Fig. 6; Nikishin et al., 2002; Vla-dimirov et al., 1997). The formation and emplacement of

continental rifts and areas of flood volcanism. NW China approximately

F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183 177

large quantities of magmas in this rift system are probablyassociated with large scale extension, accompanied bybasaltic underplating, linked to the impingement of mantleplumes (Jahn, 2004). The Siberian–Emeishan plume orsuperplume event is a good candidate (Dobretsov, 2005).As pointed out by Jahn (2004), the magmatic and tectonicmanifestation of mantle plumes or a superplume need notbe confined to a narrow time interval. Quite the contrary,there is ample evidence that these processes can last for sev-eral tens of millions of years. A mantle plume event mayinclude several pulses of magmatism, encompassing floodvolcanism, and a variety of intrusive eventss, although asingle pulse (e.g. flood basalts) may only last a few millionyears (Ernst and Buchan, 2002; Ernst et al., 2005). Further-more, recognition of a mantle plume event can be compli-cated by contamination of the magmatic products duringtheir ascent through a heterogeneous lithosphere and crust

Fig. 7. (A) eNd(t) values (n = 143) plotted against time of a range of graniticages (n = 119) for magmatic and hydrothermal ore deposits in NW China. Seeand references cited therein), Han et al. (1997), Zhang et al. (2005), Chen and

(Ernst et al., 2005). The Siberian–Emeishan mantle plumesaffected large parts of Asia and are characterised by mag-matic events that include, picrites and alkali basalts, tholei-itic flood basalts and mafic–ultramafic and A-type graniticintrusions (Dobretsov, 2005). This Siberian-Emeishan vol-cano-plutonic belt covers large areas of Asia, from theSiberian Traps to the Tianshan, Mongolia, NW, NE andSW China Jahn et al., 2000; Hong et al., 2004; Nikishinet al., 2002), around the Palaeotethys ocean and spanningages from 290 to 250 Ma.

In NW China, most post-orogenic felsic and mafic–ultramafic intrusions have positive eNd(t) values, suggest-ing the participation of juvenile mantle material, as shownin Fig. 7A. From this figure it can also be seen that fourmajor thermal events occurred in the region between 500and 200 Ma, namely at 410–390, 330–310, 300–290 and270–250 Ma. All mafic–ultramafic rocks, except one, have

and mafic–ultramafic rocks in NW China; (B) histogram of mineralisationtext for details. Data for both diagrams were taken from Hong et al. (2003Jahn (2004) and Zhou et al. (2004).

178 F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183

strong positive eNd(t) values. The histogram of mineralisa-tion ages (Fig. 7B) shows that the majority of both hydro-thermal and orthomagmatic deposits formed in the periodfrom 320–250 Ma. Jahn (2004) pointed out that the bestway to explain positive eNd(t) values for granitic rocks isto melt a mixture of new crust, with Nd and Sr ratios sim-ilar to those of the mantle protolith, together with oldercrust (see also Zhou et al., 2004). Similarly, Bowring andHoush (1995) indicated that the positive eNd(t) signaturesreflect derivation from juvenile material in a depleted man-tle, whereas negative values reflect derivation from evolvedcrustal sources. As mentioned in the introduction, mostmodels of continental crustal growth in central Asia sug-gest accretionary and collision processes involving micro-continental fragments, volcanic arcs and oceanic plateauxand crust, resulting in the formation of sutures, some ofwhich contain mafic–ultramafic rocks (ophiolites). We sug-gest that the arrival of a mantle plume induced extensionand rifting, followed by the emplacement of felsic andmafic–ultramafic magmas, a scenario schematically illus-trated in Fig. 8. The dominant ages of 290–250 Ma inNW China, and the recognition of continental mafic volca-nism in rift structures in the Tian Shan and elsewhere inAsia (Fig. 6), all point to the involvement of the Sibe-

Fig. 8. Tectono-magmatic model illustrating the geodynamic evolutionfrom (A) amalgamated terranes in NW China (Altay and Tien Shanorogens) to (B) mantle plume causing underplating of the lower crust withproduction of A-type granitoids and continental flood basalts, during post-orogenic extension. Note that the suture zones in A, act as zones of weaknessduring extensional stresses resulting in rift basins that are intruded by felsicand mafic–ultramafic plutons and filled with flood-type volcanic rocks.

rian-Emeishan mantle plumes, one of which, or possiblythe lateral extent of the Emeishan plume, impinged ontothe orogens of the NW China region.

7. Zoned mafic–ultramafic intrusions and A-type granitic

magmatism: Potential for new discoveries

The orogenic belts of the CAOB have been called‘‘Turkic-type’’ or accretionary type orogens (Sengor andNatal’in, 1996), which are built by the juxtaposition oramalgamation of island arcs-trench systems, oceanic pla-teaux, oceanic crust, microcontinental fragments, andwedges of flysch lithologies. These accretionary complexesare subsequently disrupted by strike–slip faulting and aresubjected to post-orogenic extension, largely due to gravita-tional collapse of the crust that thickened as a result of ter-rane collisions. Within this environment, post-orogenic andanorogenic magmatism, which typically occurs in tensionalzones of plate interiors, affected the region on a large scalemostly in Permian–Triassic times. This magmatism consistsof A-type felsic-mafic intrusions, is associated with floodvolcanism, dyke swarms and mafic–ultramafic intrusions.

The tectono-thermal events at about 290–250 Ma overlarge areas of Asia, including NW China, are suggestiveof a common cause. This is likely to have been lithosphericextension associated with voluminous magmatism and rif-ting. The changes in geochemical features of the magmaticproducts from one area to another probably reflect regio-nal variations in the thickness and nature of the lithosphereand mantle sources. In some cases the lithosphere wouldhave been contaminated by subduction-related compo-nents, giving mixed geochemical signatures. We suggestthat during the Carboniferous–Permian and early Triassicperiods, mantle plumes may have been active in southernScandinavia-North Sea, western Siberia (Siberian Traps),SW China (Emeishan) and the Tian Shan–Altay region(Fig. 8). An alternative view is presented by Chai et al. (thisissue), who envisage lithospheric delamination andasthenospheric upwellings to account for the post-colli-sional anorogenic magmatism in NW China.

The Altay and Tian Shan magmatic Ni–Cu deposits aregenetically associated with and hosted in funnel-shaped,differentiated layered and/or zoned mafic–ultramafic intru-sions. These intrusions occur in intracontinental rifts andare spatially and temporally associated with flood basaltsand A-type granitic rocks. In addition, the volcanic rocksare locally associated with radiolarian chert and reef lime-stone, suggesting that some of this volcanism was oceanicand may have formed seamounts. The Altay and TianShan mafic–ultramafic intrusions have features in commonwith Alaskan-type complexes. Concentrically zoned andfunnel-shaped Alaskan-type mafic–ultramafic intrusionswere first recognised, as the name implies, in southeasternAlaska (Taylor, 1967). A review of Alaskan-type com-plexes is provided by Johan (2002). Ishiwatari and Ichiy-ama (2004) reported on Alaskan-type complexes innortheast China. Typically, Alaskan complexes consist of

Fig. 9. Plane-polarised-light photomicrographs of anorthositic gabbrofrom Huangshandong, showing (A) biotite (bt) infilling microfractures;(B) pale-green chloritic alteration (chl). (C) is ore-bearing dunite,pervasively altered to a chlorite–talc-carbonate assemblage (chl and tc);note the right margin of the sulphide bleb (pyrrhotite; po), deformed andfragmented (plane-polarised light).

F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183 179

an ultramafic core surrounded by an envelope of earliergabbroic rocks, as observed in the case of the NW Chinamafic–ultramafic intrusions. Alaskan-type complexes areusually of small size ranging from 12 to 40 km2. The areaof the NW China complexes range from 0.1 km2 (Kala-tongke) to about 3 km2 (Huangshandong). The generalzoning of Alaskan-type complexes passes from a dunitecore, through successive zones of olivine-clinopyroxenite,biotite and hornblende-bearing clinopyroxenite, thenhornblendite and finally to an outer zone of monzonite–gabbro (Johan, 2002). A similar pattern, albeit withvariations, characterises the NW China intrusions. In bothAlaskan-type and NW China the ultramafic rocks haverhythmic layering and mineralogical evidence to show thattheir liquidus temperatures decrease outwards through thesuccessive zones, as in a fractional crystallisation sequence,with hornblende being a late product in the outer zones(Johan, 2002; Zhou et al., 2004). A generally acceptedmodel for the origin of Alaskan-type complexes is that ofintrusions of a tholeiitic magma, which gives rise to gabbroor gabbronorite, followed at a later stage by ultramaficmagmas that are emplaced at localised centres as multipleintrusions in order of increasing liquidus temperature, frompyroxenite, then wehrlite to dunite. The mineralisationstyles include massive sulphide, sulphide-matrix breccias,interstitial sulphide networks, and disseminated sulphide.Sulphide veins and dissemination locally penetrate footwallrocks. The environment of emplacement of the multipleore-bearing intrusions of ultramafic magma (probablymantle-derived) is the upper crust in tensional environ-ments associated with rifting. Contamination of themagma was an important factor for sulphur saturationand formation of a sulphide phase (Naldrett, 1997). A dis-cussion of the origin of the magmatic sulphides hosted bythese intrusions is beyond the scope of this contribution,but models that specifically address this topic are presentedand discussed by Chai et al. (this issue).

Another feature of the NW China mafic–ultramaficintrusions is their common and similar hydrothermal alter-ation. This alteration is locally pervasive to fracture-con-trolled and includes talc-carbonate, biotite–chlorite,sericite–muscovite–chlorite, actinolite–tremolite assem-blages (Fig. 9). The reasons for this alteration are not clearand it would be useful to attempt dating of the alterationminerals. One possibility is that these intrusions are a pow-erful heat source that can generate hydrothermal convec-tion cells within and around the intrusive body, once it isemplaced. The convective fluids then cause varying degreesof hydrothermal alteration in the rocks through which theycirculate. It may be that the strong hydrothermal alterationobserved in the ore-bearing Huangshan mafic–ultramaficrocks (Fig. 9) may be due to post-emplacement hydrother-mal activity.

Anorogenic A-type felsic to intermediate magmatism ofalkaline affinity is commonly linked with a variety of oredeposits, of which the most economically important arethe iron–oxide–copper–gold (IOCG) ore systems. In addi-

tion, because A-type magmas are enriched in incompatibleelements (e.g. Ti, P, Y, Nb, K, Th, U, F, Ba, REE) they canproduce peraluminous and peralkaline granites, which con-tain greisen or late-magmatic sub-solidus Sn, W, Zn, Cu,U, Nb mineralisation (Pirajno, 1992 and references citedtherein). The IOCG style mineral systems (Hitzmanet al., 1992; Oreskes and Hitzman, 1993; Williams et al.,2005; Pollard, 2006) include giant ore deposits, such asOlympic Dam and Ernest Henry in Australia, Carajas inBrazil, Candelaria in Chile, Palabora in South Africa (see

180 F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183

Williams et al., 2005, for a review of these deposits). Themain features of the IOCG deposits are the enrichment inFe, Cu, Au, P, F, REE, U, and the widespread alkali(Na–Ca and K) metasomatism in both host rocks and atdistrict scale. IOCG hydrothermal systems form at crustaldepths ranging from near surface to >10 km) and are pro-duced by volatile-rich, alkaline magmas (Hitzman et al.,1992). Typically IOCG deposits are located in major trans-pressional to transtensional fault systems, and appear to betemporally linked with Ni–Cu mineralised mafic–ultra-mafic intrusions (Pollard, 2006). Moreover, mantle-derivedmelts may have provided the necessary heat to generate theparent alkaline-sub-alkaline intrusions. Mineralisationstyles range from breccia pipes, through stratabound,structurally controlled vein systems to pluton-hosted, Theyappear to be linked to planetary-scale rifting events and theassembly and breakup of supercontinents, such as Rodinia(Unrug, 1997; Sears et al., 2005).

8. A conceptual geodynamic model and conclusions

A mantle plume model, proposed by Sutherland (1998)for the Cenozoic magmatism in Eastern Australia, envis-

Fig. 10. Conceptual model of plume-derived magmatism in NW China. Thisplume head with chains of diapirs or upwellings rising towards lower crustachambers from which mafic and ultramafic rocks differentiate and are injected

aged magmatic diapirs (or fingers) rising from a largerplume and penetrating upward in zones of tensional stress.The mafic–ultramafic intrusions would be generated by anelongate plume head from which a series of irregular plum-elets (diapirs or fingers) project upwards. These would par-tially melt with a distinct OIB signature or interact en routewith various other rocks (thereby introducing different geo-chemical signatures), from deep lithosphere to metasoma-tised SCML. With some modifications this model can beapplied to NW China, in which Permian–Triassic magma-tism is related to mantle melts that could have originatedfrom mantle diapirs or fingers that rose from an underlyingplume head, as shown in Fig. 10, and forming Alaskan-type zoned intrusions. Alternatively, there could have beentwo events, one producing the gabbro envelopes fromwhich lava fields were erupted and one in which the laterultramafic intrusions were emplaced. The lava fields wouldhave originated from plume fingers arising from the tails orwings of a mantle plume, whereas the ultramafic magmascame from the mantle plume head. Both models call fora regime of extensional tectonics in response to theimpingement of mantle plumes beneath the Asian conti-nental mass.

model is based on Johnston (1989) and Sutherland (1998). (A) elongatel regions; (B) plumelets rise from these diapirs, each developing magma

in the upper crust. See text for details.

F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183 181

Acknowledgements

Franco Pirajno publishes with the permission of theExecutive Director of the Geological Survey of WesternAustralia. Murray Jones and Michael Prause are thankedfor drafting the figures. Mr Dong Lian-Hui, chief geologistof the Xinjiang Bureau of Geology and Mineral Resources,Wu Hua and Cheng Song-Lin, Directors, of the No. 6 Geo-logical Team of the Xinjiang Bureau of Geology and Min-eral Resources are thanked for their logistical support ofour field visits. Drs D. Hennig and C. Halls are thankedfor their constructive criticism of the paper.

References

Abdrakhmatov, K.Y., Aldazhanov, S.A., Hager, B.H., Hamburger, M.W.,Herring, T.A., Kalabaev, K.B., Makarov, V.I., Molnar, P., Panasyuk,S.V., Prilepin, M.T., Reilinger, R.E., Sadybakasov, I.S., Souter, B.J.,Trapeznikov, Yu., Tsurkov, V.Ye., Zubovich, A.V., 1996. Relativelyrecent construction of the Tian Shan inferred from GPS measurementsof present-day crustal deformation rates. Nature 384, 450–453.

Allen, M.B., Windley, B.F., Zhang, C., Zhao, Z., Wang, C., 1991. Basinevolution within and adjacent to the Tien Shan range, NW China.Journal of the Geological Society, London 148, 369–378.

Allen, M.B., Windley, B.F., Chi, Z., 1992. Palaeozoic collisional tectonicsand magmatism of the Chinese Tien Shan, central Asia. Tectonophys-ics 220, 89–115.

BABEL Working Group, 1993. Deep seismic reflection/refraction inter-pretation of crustal structure along BABEL profiles A and B in thesouthern Baltic sea. Geophysical Journal International 112, 325—343.

Barker, F., Wones, D.R., Sharp, w.N., Desborough, G.A., 1975. ThePikes Peak batholith, Colorado Front Range, and a model for theorigin of the gabbro-anorthosite-syenit-potassic granite suite. Precam-brian Research 2, 97–160.

Bowring, S.A., Housh, T., 1995. The Earth’s early evolution. Science 269,1535–1540.

Bureau of Geology and Mineral Resources of Xingjiang Uygur Auton-omous Region, 1993. Regional Geology of Xingjiang Uygur Auton-omous Region, Geological Memoirs, Series 1, No. 32, Ministry ofGeology and Mineral Resources.

Carroll, A.R., Liang, Y., Graham, S.A., Xiao, X., Hendrix, M.S., Chu, J.,McKnight, C.L., 1990. Junggar Basin, northwest China: trapped LatePaleozoic ocean. Tectonophysics 181, 1–14.

Carroll, A.R., Graham, S.A., Hendrix, M.S., Ying, D., Zhou, D., 1995.Late Paleozoic tectonic amalgamation of northwestern China: sedi-mentary record of the northern Tarim, northwestern Turpan andsouthern Junggar basin. Geological Society of America Bulletin 107,571–594.

Chen, Y.-J., 1997. Mineralization during collisional orogenesis and itscontrol of the distribution of gold and other deposits in the JunggarOcean, Xinjiang, China. Acta Geological Sinica 71, 69–79.

Chen, B., Jahn, B.-J., 2004. Genesis of post-collisional granitoids andbasement nature of the Junggar Terrane, NW China: Nd–Sr isotopeand trace element evidence. Journal of Asian Earth Sciences 23, 691–703.

Chen, B., Arakawa, Y., 2005. Elemental and Nd–Sr isotopic geochemistryof granitoids from the West Junggar foldbelt (NW China), withimplications for Phanerozoic continental growth. Geochimica etCosmochimica Acta 69 (5), 1307–1320.

Chen, H-L., Yang, S-F., Dong, C-W., Jia, C-Z., Wei, G-Q., Wang, Z-G.,1997. Confirmation of Permian basite zone in Tarim Basin and itstectonic significance. Geochimica 26 (6), 77–87, in Chinese withEnglish abstract.

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. Tarim platform, early to middle

permian basalt magmatism and interaction with sedimenatary rocks(title translated from Chinese). Geology in China 33 (3), 544–551, inChinese.

Chung, S.L., Jahn, B.M., Wu, G.Y., Lo, C.H., Bolin, C., 1998. TheEmeishan flood basalts in SW China: a mantle plume initiation modeland its connection with continental breakup and mass extinction at thePermian–Triassic boundary. American Geophysical Union, Geody-namic Series 27, 47–58.

Coleman, R.G., 1989. Continental growth of north-west China. Tectonics8, 621–635.

Dobretsov, N.L., 2005. 250 Ma large igneous provinces of Asia: siberianand Emeishan traps (plateau basalts) and associated granitoids.Russian Geology and Geophysics 46 (9), 870–890.

Eby, G.N., 1990. The A-type granitoids: a review of their occurrence andchemical characteristics and speculations on their petrogenesis. Lithos26, 115–134.

Eby, G.N., 1992. Chemical subdivision of the A-type granitoids: petro-genetic and tectonic implications. Geology 20, 641–644.

Ernst, R.E., Buchan, K.L., 2002. Maximum size and distribution in timeand space of mantle plumes: evidence from large igneous provinces.Journal of Geodynamics 34, 309–342.

Ernst, R.E., Buchan, K.L., Campbell, I.H., 2005. Frontiers in largeigneous province research. Lithos 79, 271–297.

Goldfarb, R.J., Mao, J-W., Hart, C., Wang, D.-H., Anderson, E., Wang,Z-L., 2003. Tectonic and metallogenic evolution of the Altay Shan,northern Xinjinag Uygur Autonomous Region, northwestern China.In: Mao, J.-W., Goldfarb, R., Seltmann, R., Wang, D.-H., Xiao, W.-J.,Hart, C. (Eds.), Tectonic evolution and metallogeny of the ChineseAltay and Tianshan. International Symposium IGCP-473 ProjectGuidebook Series, 10, 17–30.

Gu, L.-X., Zhu, J.-L., Guo, J.-C., Liao, J.-J., Yan, Z.-F., Yang, H., 1995.Geology and genesis of the mafic–ultramafic complexes in theHuangshan-Jing’erquan (HJ) belt, East Xinjiang. Chinese Journal ofGeochemistry 14, 97–116.

Han, B.-F., Wang, S.-H., Jahn, B.-M., Hong, D.-W., Kagami, H., Sun,Y.-L., 1997. Depleted-mantle source for the Ulungur River A-typegranites from North Xinjiang, China: geochemistry and Nd–Srisotopic evidence, and implications for Phanerozoic crustal growth.Chemical Geology 138, 135–159.

Hitzman, M.W., Oreskes, N., Einaudi, M.T., 1992. Geological character-istics and tectonic setting of Proterozoic iron oxides (Cu–U–Au–REE)deposits. Precambrian Res. 58, 241–287.

Hong, D., Wang, S.-G., Han, B.-F., Jin, M.-Y., 1996. Post-orogenicalkaline granites from China and comparisons with anorogenicalkaline granites elsewhere. Journal of Southeast Asian Earth Sciences13, 13–27.

Hong, D., Zhang, J.-S., Wang, T., Wang, S.-G., Xie, X.-L., 2004.Continental crust growth and the supercontinental cycle: evidencefrom the Central Asian orogenic Belt. Journal of Asian Earth Sciences23, 799–813.

Hsu, K.J., 2003. Role of ophiolites in archipelago model of orogenesis.Geological Society of America Special Paper 373, 159–171.

Ishiwatari, A., Ichiyama, Y., 2004. Alaskan type plutons and ultramaficlavas in Far East Russia, Northeast China and Japan. InternationalGeology Review 46, 316–331.

Jahn, B.-M., 2004. The Central Asian Orogenic Belt and Growth of theContinental Crust in the Phanerozoic. Geological Society, London,Special Publication, 226, 73–100.

Jahn, B.-M., Wu, F.-Y., Chen, B., 2000. Massive granitoid generation inCentral Asia: Nd isotope evidence and implication for continentalgrowth in the Phanerozoic. Episodes 23, 82–92.

Jenchureva, R.J., 1997. Tectonic setting of porphyry type mineralizationand hydrothermal alteration in Paleozoic island arcs and activecontinental margins, Kyrgyz Range (Tien Shan), Kyrgystan. Minera-lium Deposita 32, 434–440.

Johan, Z., 2002. Alaskan-type complexes and their platinum-groupelement mineralization. In: Cabri L.J. (Ed.), The geology, geochem-istry, mineralogy and mineral beneficiation of platinum-group ele-

182 F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183

ments. Canadian Institute of Mining, Metallurgy and PetroleumSpecial 54, 669–719.

Johnston, R.W., 1989. Intraplate volcanism in Eastern Australia and NewZealand. Cambridge University Press, Cambridge.

Li, X.H., Su, L., Chung, S.L., Li, Z.X., Liu, Y., Song, B., Liu, D.Y., 2005.Formation of the Jinchuan ultramafic intrusion and the world’s thirdlargest Ni–Cu sulphide deposit: associated with the �825 Ma southChina mantle plume? Geochemistry Geophysics Geosystems 6, 1–16.

Li, D.-S., Liang, D.-G., Jia, C.-Z., Wang, G., Wu, Q.-Z., He, D.-F., 1996.Hydrocarbon accumulations in the Tarim basin China. AAPG Bulletin80, 15887–16603.

Li, X.-M., Xia, L.-Q., Xia, Z.-C., Xu, X.-Y., Ma, Z.-P., Wang, L.-S., 2004.Zircon U–Pb geochronology of volcanic rocks of the Qi’eshan Groupin the east Tianshan mountains. Geological Bulletin of China 23,1215–1220, in Chinese with English abstract.

Li, Z.-L., Yang, S.-F., Chen, H.-L., Yu, X., 2006. Discovery of a Permianquartz-syenite dyke and bimodal dyke from the Tarim Basin and itstectonic implications. International Conference on Continental Vol-canism, IAVCEI 2006, Guangzhou, China, Abstracts & Program, 64.

Loiselle, M.C., Wones, D.R., 1979. Characteristic and origin of anoro-genic granites. Geological Society of America Abstracts with programs11, 468.

Mao, J.-W., Goldfarb, R., Seltmann, R., Wang, D.-H., Xiao, W.-J., Hart,C. (Eds.), 2003. Tectonic evolution and metallogeny of the ChineseAltay and Tianshan, Int Symp IGCP-473, Project Guidebook Series,10.

Mossakovsky, A.A., Ruhentsev, S.V., Samygin, S.G., Kheraskova, T.N.,1993. Central Asian Fold Belt: geodynamic evolution and formationhistory. Geotectonics 27, 445–474.

Naldrett, A.J., 1997. Key factors in the genesis of Noril’sk, Sudbury,Jinchuan, Voisey’s Bay and other world class Ni–Cu–PGE deposits:implications for exploration. Australian Journal of Earth Sciences 44,283–316.

Nie, S., Rowley, D.B., Ziegler, A.M., 1990. Constraints on the locations inPalaeotethys during the Late Palaeozic. Geological Society Memoir 12,397–409.

Nikishin, A.M., Ziegler, P.A., Abbott, D., Brunet, M.-F., Cloetingh, S.,2002. Permo–Triassic intraplate magmatism and rifting in Eurasia:implications for mantle plumes and mantle dynamics. Tectonophysics351, 3–39.

Oreskes, N., Hitzman, M.W., 1993. A model for the origin of OlympicDam-type deposits. Geol. Assoc. Can. Spec. Pap. 40, 615–633.

Pirajno, F., 1992. Hydrothermal mineral deposits. Springer-Verlag, Berlin,pp. 709.

Pirajno, F., Luo, Z.-Q., Liu, S.-F., Dong, L.-H., 1997. Gold deposits in theEastern Tian Shan, northwestern China. International GeologyReview 39, 891–904.

Pollard, P.J., 2006. An intrusion related origin for Cu–Au mineralizationin iron oxide-copper-gold (IOCG) provinces. Mineralium Deposita 41,179–187.

Puura, V., Floden, T., 1999. Rapakivi-granite-anorthosite magmatism – away of thinning and stabilisation of the Sveofennian crust, Baltic SeaBasin. Tectonophysics 305, 75–92.

Ren, J.-S., Wang, Z.-X., Chen, B.-W., Jiang, C.-F., Niu, B.-G., 1999.Tectonic map of China. Geological Publishing House, Beijing.

Rui, Z.-Y., Goldfarb, R.J., Qiu, Y.-M., Zhou, T.-H., Chen, R.-Y.,Pirajno, F., Yun, G., 2002. Paleozoic-early Mesozoic gold deposits ofthe Xinjiang Autonomous Region, northwestern China. MineraliumDeposita 37, 393–418.

Sears, J.W., St. George, G.M., Winne, J.C., 2005. Continental rift systemsand anorogenic magmatism. Lithos 80, 147–154.

Sengor, A.M.C., Natal’in, B.A., 1996. Turkic type orogeny and its role inthe making of the continental crust. Annual Review Earth andPlanetary Sciences 24, 263–337.

Sengor, A.M.C., Natal’in, B.A., 2004. Phanerozoic analogues of Archaeanoceanic basement fragments: altayd ophiolites and ophirags. In:Kusky, T.M. (Ed.), Precambrian ophiolites and related rocks, Devel-opments in Precambrian Geology 13, 675–726.

Sengor, A.M.C., Graham, S.A., Biddle, K.T., 1996. Is the Tarim Basinunderlain by a Neoproterozoic oceanic plateau?. Geological Society ofAmerica, 1996 Annual Meeting, Abstracts with Programs, A-67.

Sengor, A.M.C., Natal’in, B.A., Burtman, V.S., 1993. Evolution of theAltayd tectonic collage and Paleozoic crustal growth in Eurasia.Nature 364, 299–307.

Shi, Y., Lu, H., Jia, D., Cai, D., Wu, S., Chen, C., Howell, D.G., Valin,Z.C., 1994. Paleozoic plate-tectonic evolution of the Tarim andwestern Tianshan regions, western China. International GeologyReview 36, 1058–1066.

Sutherland, F.L., 1998. Origin of north Queensland Cenozoic volcanism:relationships to long lava flow basaltic fields, Australia. Journal ofGeophysical Research 103, 27347–27358.

Taylor, H.P., 1967. The zoned ultramafic complexes of southeasternAlaska. In: Wyllie, P.J. (Ed.), Ultramafic and related rocks. J. Willey &Sons Inc, New York, pp. 97–121.

Unrug, R., 1997. Rodinia to Gondwana: the geodynamic map of Gondwanasupercontinent assembly. Geol. Soc. Am. GSA Today 7, 1–6.

Vladimirov, A.G., Ponomareav, A.P., Shokalski, S.P., Khalilov, V.A.,Kostitsyn, Yu.A., Pnomarchuk, V.A., Rudnev, S.A., Vystavnoi, S.A.,Kruk, N.N., Titov, A.V., 1997. Late Paleozoic–Early Mesozoicgranitoid magmatism in Altai. Russian Geology and Geophysics 38,755–770.

Wang, B., Shu, L.S., Cluzel, D., Faure, M., Charvet, J., 2007. Geochem-ical constraints on Carboniferous volcanic rocks of the Yili Block(Xinjiang, NW China): implication for the tectonic evolution ofWestern Tianshan. Journal of Asian Earth Sciences 29, 148–159.

Williams, P.J., Barton, M.D., Johnson, D.A., Fontbote, L., De Haller, A.,Mark, G., Oliver, N.H.S., Marschik, R., 2005. Iron oxide copper-golddeposits: geology, space–time distribution, and possible modes oforigin. Economic Geology 100th Anniversary Issue, 371–406.

Windley, B.F., Allen, M.B., Zhang, C., Zhao, Z., Wang, C., 1990.Paleozoic accretion and Cenozoic redeformation of the Chinese TianShan range, Central Asia. Geology 18, 128–131.

Windley, B.F., Kroener, A., Guo, J., Qu, G., Li, Y., 2002. Neoproterozoicto Paleozoic geology of the Altay orogen, NW China: new zircon agedata and tectonic evolution. Journal of Geology 110 (6), 719–737.

Windley, B.F., Alexeiev, D., Xiao, W.-J., Kroner, A., Badarch, G., 2007.Tectonic models for accretion of the Central Asian Orogenic Belt.Journal of the Geological Society, London 64, 31–47.

Wu, F.-Y., Sun, D.-Y., Li, H.-M., Jahn, B.-M., Wilde, S., 2002. A-typegranites in northeastern China: age and geochemical constraints ontheir petrogenesis. Chemical Geology 187, 143–173.

Xia, L.-Q., Xu, X.-Y., Xia, Z.-C., Li, X.-M., Ma, Z.-P., Wang, L.-S., 2003.Carboniferous post-collisional rift volcanism of the Tianshan moun-tains, northwestern China. Acta Geological Sinica 77, 338–360.

Xia, L.-Q., Xu, X.-Y., Xia, Z.-C., Li, X.-M., Ma, Z.-P., Wang, L.-S., 2004.Petrogenesis of Carboniferous rift-related volcanic rocks in theTianshan, northwestern China. Geological Society of America Bulletin116, 419–433.

Xiao, W.-J., Zhang, L.-C., Qin, K.-Z., Sun, S., Li, J.-L., 2004a. Paleozoicaccretionary and collision tectonics of the eastern Tianshan (China):implications for the continental growth of central Asia. AmericanJournal of Science 304, 370–395.

Xiao, W.-J., Windley, B.F., Badarch, G., Sun, S., Qin, K., Wang, Z.,2004b. Palaeozoic accretionary and convergent tectonics of thesouthern Altayds: implications for the growth of Central Asia. Journalof the Geological Society, London 161, 339–342.

Xiao, L., Xu, Y.G., Chung, S.L., He, B., Mei, H., 2003. Chemostrati-graphic correlation of upper Permian lavas from Yunnan Province,China: extent of the Emeishan Large Igneous Province. InternationalGeology Review 45, 754–766.

Xiao, L., Xu, Y.G., Xu, J.F., He, B., Pirajno, F., 2004a. Chemostratig-raphy of flood basalts in the Garze-Litang region and Zongza Block:implications for western extension of the Emeishan Large IgneousProvince, SW China. Acta Geologica Sinica 78, 61–67.

Xiao, L., Xu, Y.G., Mei, H.J., Zheng, Y.F., He, B., Pirajno, F., 2004b.Distinct mantle sources of low-Ti and high Ti basalts from the western

F. Pirajno et al. / Journal of Asian Earth Sciences 32 (2008) 165–183 183

Emeishan large igneous province, SW China: implications for plume-lithosphere interaction. Earth andPlanetary Science Letters 228, 525–546.

Xiao, L., Hayward, N., Begg, G., Fu, M.-L., Wang, F.-Z., Pirajno, F.,2005. The Jinxi-Yelmand high-sulfidation epithermal gold deposit,Western Tianshan, Xinjiang province, P.R. China. Ore GeologyReviews 26, 17–37.

Yakubchuk, A.S., Shatov, V.V., Kirwin, D., Edwards, A., Tomurtogoo,O., Badarch, G., Buryak, V.A., 2005. Gold and base metal metallog-eny of the Central Asian Orogenic Supercollage. Economic Geology100th Anniversary Volume, 1035–1068.

Yan, S.-H., Zhang, Z.-C., Wang, D.-H., Chen, B.-L., He, L.-X., Zhou, G.,2003. Kalatongke magmatic copper-nickel sulphide deposit. In: Mao,J.-W., Goldfarb, R., Seltmann, R., Wang, D.-H., Xiao, W.-J., Hart, C.(Eds.), Tectonic evolution and metallogeny of the Chinese Altay andTianshan. International Symposium IGCP-473 Project GuidebookSeries 10, 131–151.

Yang, S.-F., Chen, H.-L., Li, Z.-L., Dong, C.-W., Jia, C.-Z., Wei, G.-Q.,2006. Early to middle Permian magmatism in the Tarim Basin.International Conference on Continental Volcanism, IAVCEI 2006,Guangzhou, China, Abstracts & Program, 54.

Zhang, Z.-M., Liou, J.-G., Coleman, R.C., 1984. An outline of the platetectonics of China. Geological Society of America Bulletin 95, 295–312.

Zhang, Z.-C., Mahoney, J.J., Mao, J.-W., Wang, F.-S., 2006. Geochem-istry of picritic and associated basalt flows of the Western EmeishanFlood Basalt province, China. Journal of Petrology. doi:10.1093/petrology/eg1034.

Zhang, Z.-H., Gu, L.-X., Wu, C.-Z., Li, W.-Q., Xi, A.-H., Wang, S., 2005.Zircon SHRIMP dating for the Weiya pluton, eastern Tianshan: itsgeological implications. Acta Geologica Sinica 79 (4), 481–490.

Zhang, Z.-H., Mao, J.-W., Du, A., Pirajno, F., Yang, J.-M., Wang, Z.-L.,2008. Re-Os dating of the two largest copper-nickel sulphide depositsin northern Xinjiang, and its geological significance. Journal of AsianEarth Sciences 32 (2–4), 204–217.

Zhao, X., Coe, R.S., Gilder, S.A., Frost, G.M., 1996. Palaeomag-netic constraints on the palaeogeography of China: implicationsfor Gondwanaland. Australian Journal of Earth Sciences 43,643–672.

Zhou, D.-W., Su, L., Jiang, P., Wang, R., Liu, X.-M., Lu, G.-X., Wang,J., 2004. Zircon U–Pb SHRIMP ages of high pressure granulite inYushugou ophiolitic terrane in southern Tianshan and their tectonicimplications. Chinese Science Bulletin 49 (13), 1415–1419.

Zhou, T.-F., Yuan, F., Tan, L.-G., Fan, Y., Yue, S.-C., 2006.Geodynamic significance of the A-type granites in the Sawuer regionin west Junggar, Xinjiang: rock geochemistry and SHRIMP zircon ageevidence. Science in China Ser. D 49, 113–123.

Zhu, Y.-F., Sun, S.-H., Gu, L.-B., Ogasawara, Y., Jiang, N., Honma, H.,2001. Permian volcanism in the Mongolian orogenic zone, northeastChina: geochemistry, magma source and petrogenesis. GeologicalMagazine 138, 101–115.

Zhu, Y.-F., Zhang, L.-F., Gu, L.-B., Guo, X., Zhou, J., 2005. The zirconSHRIMP chronology and trace element geochemistry of the Carbon-iferous volcanic rocks in western Tianshan mountains. Chinese ScienceBulletin 50 (19), 2210–2212.