The Mongol–Okhotsk Belt in Mongolia — An appraisal of the geodynamic development by the study of sandstone provenance and detrital zircons

Tectonophysics 510 (2011) 132–150

Contents lists available at ScienceDirect

Tectonophysics

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

The Mongol–Okhotsk Belt in Mongolia — An appraisal of the geodynamicdevelopment by the study of sandstone provenance and detrital zircons

Denise Bussien, Nergui Gombojav 1, Wilfried Winkler ⁎, Albrecht von QuadtDepartment of Earth Sciences, ETH Zentrum, Sonneggstrasse 5, ETH Zurich, 8092 Zurich, Switzerland

⁎ Corresponding author. Tel.: +41 44 632 36 97; faxE-mail address: [email protected] (W.

1 Present address: Pöyry Energy AG, Hardturmstrasse

0040-1951/$ – see front matter © 2011 Elsevier B.V. Adoi:10.1016/j.tecto.2011.06.024

a b s t r a c t
a r t i c l e i n f o
Article history:Received 4 October 2010Received in revised form 21 June 2011Accepted 22 June 2011Available online 6 July 2011

Keywords:Mongol–Okhotsk BeltMongoliaGeodynamicsDetrital zirconsU/Pb laser ablation ICP-MS datingHf isotopes and trace elements

The Mongol–Okhotsk Belt formed in a late stage of Jurassic orogeny in the composite Central Asian OrogenicBelt. The present paper investigates the Late Palaeozoic–Mesozoic sandstones associated with the belt inMongolia, aimed at reconstructing the time and mode of ocean formation, subduction and collision. We applyprovenance analysis including (1) heavy mineral and sandstone framework grain analysis, (2) U/Pb laserablation ICP-MS dating of detrital zircons for identifying contemporaneous volcanic arc activity and therecycling of older crustal material, and (3) on dated zircon grains, we analyze trace element contents and Hfisotopic ratios in order to characterize the rock types in which they crystallized and the magma sources,respectively.The investigated samples are derived from (1) the Adaatsag and Doschgol terranes, which represent thesuture zone, (2) the Hangai–Hentei belt to the northwest, and (3) the Ereendavaa terrane and theMiddle Gobivolcanic belt to the southeast of the suture. The latter two are concurrent with the northern and southernmargins of the Mongol–Okhotsk Ocean. Tectono-stratigraphic arguments suggest that the Mongol–Okhotskocean opened during the Silurian within the Early Palaeozoic collage. In the suture zone, Permian syn-sedimentary zircon trace element contents confirm mafic rock sources, and the mantle involvement in themagmatism (epsilon Hf(t) from +13 up to +20). N and S directed, bi-vergent subduction developed asrevealed by contemporaneous zircons: (1) along the northern margin (Hangai–Hentei), from Silurian–EarlyCarboniferous, subduction and accretion prevailed (epsilon Hf(t) from +3 up to +12 in associated zircons),whichwas re-initiated during the Permian, and (2) the contemporaneous Silurian–Devonian southernmargin(Ereendavaa–Middle Gobi) still represented an extensional continental margin showing reworking ofNeoproterozoic–Early Palaeozoic zircons from the basement. It turned into an active continental margin withstarting arc magmatism in the Carboniferous (zircon mean epsilon Hf(t) +5.75).

: +41 44 632 14 22.Winkler).161, 8037 Zürich, Switzerland.

ll rights reserved.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

The Central Asian Orogenic Belt (CAOB) is a spectacular example oflong-lasting continental growth between the Eastern European,Siberian, Tarim and North China continental blocks (e.g., Sengör etal., 1993). It formed by multiple ocean subduction associated withcontinental margin magmatism, accretion and collision of intraocea-nic volcanic arcs, and continental fragments with the large boundingcontinental plates (e.g., Jahn et al., 2000; Khain et al., 2003; Parfenovet al., 1995; Windley et al., 2007). In plate tectonics terms, the CAOBgenerally is considered to be the product of evolution of the Palaeo-Asian Ocean at least since the Neoproterozoic until the Permian. TheMongol–Okhotsk Belt (MOB) represents a young, still enigmaticdivision within the CAOB. It formed by the closure of the Mongol–

Okhotsk ocean (Natal'in, 1993; Parfenov et al., 2001; Zonenshain et al.,1990; Zorin, 1999).

The MOB as a major structural element extends over 3000 kmfrom central Mongolia in northeastern direction to the Gulf ofOkhotsk. The core is represented by a ribbon-like ophiolite-bearingsuture zone and accretionary wedges (Natal'in, 1993). Parfenov et al.(2001) termed it the Aga terrane, which includes the Onon andTukuringra terranes, and the Nilanskiy fragments (Fig. 1). The Ononterrane comprises the intra-oceanic Onon arc of assumed Devonian–Early Carboniferous age (Zorin, 1999). The suture is considered as arelic of the Mongol–Okhotsk ocean (MOO), which was closed as aresult of the collision of the continental Amurian superterrane withthe North Asia (Siberian) craton (Natal'in, 1993; Parfenov et al., 1995,2001) (Fig. 1).

The opening of the MOO is a matter of speculation and LateProterozoic to Late Palaeozoic ages are suggested (e.g., Badarch et al.,2002; Parfenov et al., 2001; Sengör et al., 1993; Zonenshain et al.,1990; Zorin, 1999; Zorin et al., 1993). However, Zorin et al. (1993)advocated that the MOB did not exist yet in the Middle Palaeozoic. On

http://dx.doi.org/10.1016/j.tecto.2011.06.024

mailto:[email protected]

http://dx.doi.org/10.1016/j.tecto.2011.06.024

http://www.sciencedirect.com/science/journal/00401951

Fig. 1. Tectonic sketch map of the Mongol–Okhotsk Belt and the framing units.Modified from Parfenov et al. (2001) and Zorin (1999), and including information on the Mongolian sector from Badarch et al. (2002).

133D.Bussien

etal./

Tectonophysics510

(2011)132

–150

134 D. Bussien et al. / Tectonophysics 510 (2011) 132–150

the other hand it appears there exist no positive biostratigraphic andgeochronometric evidences for rocks older than Silurian–Devonianwithin the suture zone (Natal'in, 1993; Parfenov et al., 2001). A clearargument for the Silurian opening of the MOO comes from datedradiolarites (conodonts and radiolarians) in the Gorkhi Fm. of theHangai–Hentei unit at the northern margin of the MOO, which overlieoceanic hyaloclastites, basalts and dolerites (Kurihara et al., 2008).Accretionary wedges including mainly turbidite formations (Hangai,Hentei–Daurian, Unya–Bom, Ulban terranes), and the North Gobiforearc basin are suggested to have formed during previoussubduction along themargins of theMOO. Calc-alkaline to subalkalinevolcano-plutonic belts and granite intrusions in the units framing thesuture zone to the N and S (e.g., Selenge and East Mongolia units) aresupposed to attest north- and southward-directed subduction of theMongol–Okhotsk oceanic lithosphere from Late Palaeozoic to EarlyJurassic (Fig. 1) (Parfenov et al., 2001; Sal'nikova et al., 2006; Sorokinet al., 2002, 2003, 2005; Zorin, 1999). Younger, Late Jurassic–EarlyCretaceous granites are interpreted to represent post-collisionalintrusions (Sorokin et al., 2004).

There exist two contrasting models for the development of theentire CAOB in literature. Sengör et al. (1993) and Sengör and Natal'in(1996) propose that the CAOB formed in one giant Neoproterozoic toCarboniferous subduction–accretion complex along the so-calledKipchak, Tuva–Mongol and Manchuride magmatic arcs, which finally,until the Permian, should have been contorted between the Siberianand Baltic cratons. With reference to the present subject, the MOB,these authors suggested that the Hangai–Hentei zone formed an EarlyPalaeozoic to Carboniferous accretionary wedge and magmatic arcassemblage due to subduction of Vendian–Early Cambrian ophiolitic

Fig. 2. Tectonostratigraphic map of Mongolia compiled after Badarch et al. (2002) and WiPalaeozoic and a Late Palaeozoic domain, which were amalgamated by the end of Ordovicrelated volcano-sedimentary margin units are superposing the Early Palaeozoic domain in

crust. However, recent palinspastic reconstructions of the CAOB,supported by increasing geochronometric, palaeomagnetic andpalaeontological data sets, instead infer the existence and time-stepped closure of various oceanic basins in the Palaeo-Asian Oceanwith associated subduction zones, volcanic arcs, forearc and backarcbasins (e.g., Badarch et al., 2002; Filippova et al., 2001; Kröner et al.,2005, 2007; Windley et al., 2007; Zorin et al., 2007). According toBadarch et al. (2002) and Windley et al. (2007), a major step ofamalgamation in the history of the CAOB occurred during the Silurian.At that time the Main Mongolian Lineament (Fig. 2) represented thesouthern margin of an Early Palaeozoic collage, whereas the terranesto the south would have been accreted during the Late Palaeozoic. Aswe are going to show, the Late Palaeozoic–Early Mesozoic MOBdeveloped within the Early Palaeozoic collage. This is difficult toreconcile with earlier, long-lasting Kipchak geodynamic model of theCAOB (Sengör et al., 1993; Sengör and Natal'in, 1996).

The size and age of the MOO also are enigmatic matters. In severalpaleogeographical reconstructions, the MOO is drawn as an enormous,often pivotal oceanic domain between the Siberian continent and theBureya–Jiamusi cratonal massif (the Amur superterrane, Fig. 1).According to a compilation of Shi (2006) two main opinions on thepalaeogeography prevail: (a) a large size MOO as part of the Palaeo-Pacific Ocean (Panthalassa), or (2) a pivotal, remnant oceanic basin,which remained after partial closing of the Palaeo-Asian Ocean. In theAdaatsag terrane, tectonically dismembered occurrences of serpenti-nized dunites and hartzburgites associated with Carboniferous layeredgabbro, sheeted mafic dykes and basalts overlain by red cherts andclastic sediments (Kurihara et al., 2008) suggests a highly evolvedoceanic crust existed until Carboniferous (Tomurtogoo et al., 2005).

ndley et al. (2007). The Main Mongolian Lineament divides the country into an Earlyian and during the Permian, respectively. The suture of the Mongol–Okhotsk Belt andcentral and northeastern Mongolia.

image of Fig.�2

Fig. 3. Sample map with terranes and volcanic belts according to Badarch et al. (2002).

135D. Bussien et al. / Tectonophysics 510 (2011) 132–150

With regard to the timing of opening of the MOO, a primordialpoint to consider is the tectonic relationship of the MOB in generaland its suture with the neighboring units. In accepting that theEarly Palaeozoic domain north of the Main Mongolia Lineamentformed in the Silurian period (Windley et al., 2007), the MOBsuture (Adaatsag and Dochgol terranes) physically separates theNeoproterozoic–Cambrian Ereendavaa and Idermeg terrane base-ment blocks in the southeast from the Precambrian continentalblocks and ophiolitic collage (Late Proterozoic–Early Palaeozoicisland arcs, ophiolites and accretionary complexes) in the north-west (Fig. 2). Consequently, the MOO must have opened later. Thisis corroborated by the following observations: (1) the Mongol–Okhotsk oceanic lithosphere subduction related plutonism andvolcanism (as present in the Selenge and Middle Gobi volcanic–plutonic belts; Fig. 1) penetrates and covers the framing blocks tothe north and south of the suture (Badarch et al., 2002; Parfenovet al., 2001; Zorin, 1999), and (2) the Silurian–Carboniferoussediments in the Hangai–Hentei belt, and in the Ereendevaa terrane(Fig. 2), presumably coeval with the ocean formation andsubduction, unconformably cover older elements of the EarlyPalaeozoic domain (Badarch et al., 2002), i.e. they represented themargins of the evolving MOO.

General consensus exists on the closing history of the MOO. Aprogressive Jurassic to Early Cretaceous closing of the ocean andassociated mountain building from the west (Central Mongolia)towards the Okhotsk Gulf in the east is proposed from stratigraphicarguments (e.g., Natal'in, 1993; Parfenov et al., 2001; Zorin, 1999).These include the general younging trend of deep-water sedimentaryfacies comprised in the suture zone, and of the subduction-relatedmagmatic suites in eastward direction (Fig. 1). However, the modeand timing of ocean formation is still to explore.

The present paper investigates the southwestern extension of theMOB in Mongolia aimed at revealing the geodynamic conditions

leading to the formation and closure of the MOO.We apply a two-foldapproach: i) the tectonic relationship of the belt with respect to theolder, framing units of the CAOB implies that theMOO opened in post-Ordovician time by cutting into and separating older units of the EarlyPalaeozoic CAOB, and ii) the Silurian/Devonian to Jurassic clasticsediments preserved in the suture zone and bordering units provideinsights into the tectonic development of the MOB. For that goal weinvestigate the modal and heavy mineral compositions of sandstones,date detrital zircons by U/Pb ICP-MS laser ablation, and wheneverpossible, trace element content and in-situ Hf isotope analyses areperformed on zircons in the dated magmatic domains. Such data shallreveal the composition and the age of the source rocks of the detritalformations, the type of igneous rocks in which zircon grains hadcrystallized, and the magma source, respectively. In particular, syn-sedimentary ages of detrital zircons will be used to constrain thevolcanic–plutonic activity due to subduction of the MOO lithosphereunder the framing units. Finally, we combine the obtained argumentsand data for proposing a more consistent geodynamic model of theMOB in Mongolia.

2. Geological framework of Mongolia

According to tectono-stratigraphic compilations of Mongolia(Badarch et al., 2002; Kröner et al., 2005; Windley et al., 2007), theMain Mongolian Lineament divides the country in an Early Palaeozoicdomain in the north and a Late Palaeozoic domain in the south (Fig. 2).The eastern part of the northern domain, in addition includes thepresumed suture and deformed margins of the Mongol–Okhotsk Belt.In the current Mongolian tectono-stratigraphic terminology thesuture zone is represented by the Adaatsag and Dochgol terranes ofBadarch et al. (2002), or the Onon zone of Zorin (1999) in thenortheastern prolongation in Russia, respectively. To the north, thesuture is bounded by the Hangai–Hentei unit, and in the south by the

image of Fig.�3

Table 1Sample list including information on localities (UTM), terrane affiliation, chronostratigraphic age and lithology.

SampleNo.

Latitudedeg:min:sec

Longitudedeg:min:sec

Terrane Stratigraphic age Lithology/inferred depositional environment (where possible) Analysis

302MN06 48:51:54 105:27:50 Bayangol;Hangai–Henteioverlap

Carboniferous Fine grained sandstone, strongly weathered DZA,HM,MFA

308MN06 46:47:18 107:03:14 Ereendavaa/Middle Gobi

Permian Greenish-gray, medium grained sandstone DZA,HM


Permian Sandstones, greenish gray HM


Devonian Sandstones, metamorphosed and sheared HM,MFA


Permian Sandstones, yellowish-gray HM


Permian Sandstones, yellowish-gray DZA,MFA


Triassic Brownish-gray, medium grained sandstones HM


Jurassic Sandstones, light yellowish-gray DZA,HM,MFA

323MN06 472714 109:48:23 Ereendavaa/Middle Gobi

Devonian Coarse grained sandstones, greenish gray DZA,HM,MFA


Permian Dark-gray sandstones, slightly sheared HM


Triassic Bluish-gray, coarse grained sandstones DZA,MFA


Devonian Sandstones, light gray, slightly metamorphosed DZA,HM,MFA

329MN06 47:53:18 105:43:01 Haraa; Hangai–Hentei overlap

Devonian Turbiditic sandstones medium-grained, brownish-gray, in alternation with green(minor red) clayey shales

DZA

332MN06 47:17:75 103:36:11 Hangai–Hentei Devonian 1–2 Greenish-gray turbiditic sandstones, alternated with green bioturbated shales DZA,HM,MFA

333MN06 47:16:55 103:35:25 Hangai–Hentei Carboniferous 1–2 Brownish-gray turbiditic sandstones, associated with shales DZA336MN06 46:06:10 102:52:22 Hangai–Hentei Jurassic 1–2 Graded cycles of grain-supported conglomerate to medium-coarse sandstones;

alluvial plain; unconformably overlying folded Devonian sedimentsHM

337MN06 46:25:27 103:08:11 Hangai–Hentei Jurassic 2 Massive, coarse conglomerates with fine-grained matrix associated with irregularcoarse sandstones; alluvial fan

HM,MFA


Carboniferous 1–2 Turbiditic sandstones in spotted (bioturbated) green shales DZA,HM,MFA


Permian 2 Greenish-brown coarse sandstones associated with matrix-supportedconglomerate (rich in vein-quartz pebbles); alluvial fan

HM,MFA

340MN06 47:01:59 107:22:08 Ereendavaa/MiddleGobi

Permian 2 Channelized, and large-scale planar and sigmoidal fine-coarse grained sandstones,greenish gray; sand-dominated fluvial environment

HM


Permian 2 Medium-grained, greenish-brown immature, graded sandstones, ball structures;delta slope

HM

348MN06 47:54:14 106:39:15 Hangai–Hentei Early Carboniferous The well-known 81 gram outcrop, coarse sandstones from ca. 50 m fining-upwardcycle (grain-supported conglomerates to green shales); presumably deltaicchannel-mouth bar.

DZA,HM

349MN06 47:51:45 107:06:34 Hangai–Hentei Carboniferous 1–2 Massive green, medium-coarse grained sandstones facies DZA,MFA

352MN06 47:42:25 106:55:09 Hangai–Hentei Carboniferous Massive immature, non-graded, middle and coarse grained sandstones, quartz-rich;

HM,MFA

353MN06 47:44:08 106:53:31 Hangai–Hentei Carboniferous Thin-bedded (cm) silt-sized beds in fine shales; presumably pro-delta deposits HM401MN07 47:42:16 109:01:21 Ereendavaa/

Middle GobiPermian 2 Siliceous shales and turbiditic sandstones DZA,

HM403MN07 49:21:12 113:46:58 Ereendavaa Triassic Alluvial, coarse conglomerates with subordinate sandstones HM405MN07 50:03:59 114:23:06 Doshgol Permian Brown sandstones HM406MN07 50:03:59 114:23:06 Doshgol Permian Quartz-bearing sandstones DZA,

MFA407MN07 49:18:01 111:20:11 Adaatsag Triassic Graded, mudchips-bearing turbidite beds DZA,

HM408MN07 49:17:30 111:32:52 Adaatsag Permo-Triassic (according

to present detrital zirconages)

Turbiditic sandstones and shales showing low-grade metamorphic overprint DZA,HM

409MN07 48:43:53 111:09:14 Adaatsag Jurassic Coarse sandstones DZA,HM,MFA


Triassic 3 Coarse immature conglomerates and green sandstones; braided river-alluvial fan

DZA,MFA


Table 1 (continued)

SampleNo.

Latitudedeg:min:sec

Longitudedeg:min:sec

Terrane Stratigraphic age Lithology/inferred depositional environment (where possible) Analysis


Triassic 2 Grain-supported conglomerate–sandstone cycles; alluvial fan DZA,HM,MFA

DZA: detrital zircon ages (U/Pb) and geochemistry; HM: heavy minerals; and MFA: modal framework grains.


Ereendavaa terrane and the Middle Gobi volcanic belt. Thesepaleogeographic domains represent the opposing margins of theoriginal MOO. North- and southward subduction along these marginsis documented by Late Palaeozoic magmatic arc remnants andintrusions on both sides of the suture (Badarch et al., 2002; Parfenovet al., 2001; Zorin, 1999) (Fig. 1).

2.1. The Early Palaeozoic domain north of the Main MongolianLineament

This broad belt contains Archean–Proterozoic continental blocks(Gargan, Sayan, Tuva–Mongolia and Baydrag) with gneisses dated at3.15–1.87 Ga and a granulite peak metamorphism at 1.85–1.84 Ga(Khain et al., 2003; Sal'nikova et al., 2001). A derivation of themicrocontinents from eastern Gondwana is assumed (e.g., Ruzhent-sev and Mossakovskiy, 1996) but a Precambrian rifting-relateddisconnection of the blocks from Siberia is probable (Zonenshain etal., 1990). The continental basement blocks are tectonically wrappedby Late Proterozoic ophiolite-bearing complexes (dated at 880–570 Ma, see Windley et al., 2007 for a summary), including remnantsof seamounts and oceanic plateaus, island arcs, subduction relatedintrusions and accretionary complexes (e.g., Buchan et al., 2002;Khain et al., 2002; Kovach et al., 2005). There is general consensus thatthese units originated from the northward-directed consumption ofthe Palaeo-Asian oceanic domain surrounding the Siberian continentto the south (in present coordinates) and continental blocks within. InMongolia, cross-cutting relationships of intrusions and dykes alongthe southern boundary fault (MainMongolian Lineament) dated at ca.540 Ma are interpreted to match the final assembly of the units withthe Siberian margin during Cambrian to Ordovician (Buchan et al.,2002; Kröner et al., 2005; Windley et al., 2007).

2.2. The Mongol–Okhotsk domain

In the vicinity of the aforementioned Mongol–Okhotsk suture(Fig. 1), the Early Palaeozoic domain (collage) forms the basement ofthe opposing marginal Silurian/Devonian–Jurassic rock formations ofthe MOB. The northern margin Hangai–Hentei basin series consists ofSilurian–Devonian radiolarian cherts and Carboniferous–Permianturbiditic sandstones and continental beds (Badarch et al., 2002;Kelty et al., 2008; Kurihara et al., 2008). Post-depositional tectonicimbrication of ophiolitic blocks with pelagic cherts and turbiditespresumably imply the formation of an accretionary wedge (Kuriharaet al., 2008), which in turn was unconformably overlain by Triassiccontinental facies deposits. Further to the northwest, the Selenge beltwith a Middle Carboniferous to Triassic volcano-plutonic suite andassociated marine to continental sediments are believed to representAndean type volcanic arc formations related to the northwardsubduction of Mongol–Okhotsk oceanic lithosphere under thecomposite Siberian margin (Parfenov et al., 2001; Zorin, 1999; Zorinet al., 2007). Erroneously, Badarch et al. (2002) interpreted the entireHangai–Hantai sediment series as unconformably deposited oversome unknown basement. However, according to Kurihara et al.(2006) and own observations, the main mass of the Hangai–Henteybelt formed in an trench-accretionary wedge environment, which istectonically arranged between the Early Palaeozoic Haraa andBayangol terranes and the MOB suture. But coeval Carboniferous

and Triassic sandstones topping the Haraa and Bayangol terranes weconsider as Hangai–Hentei overlap deposits (Fig. 3, Table 1).

On the southern margin, in the Ereendevaa terrane (Fig. 2),Silurian and Devonian marine sedimentary and volcanic rocks overliePalaeoprotereozoic gneisses and Neoproterozoic schists, metasand-stones and marbles (Badarch et al., 2002). Permian volcanic andmarine sedimentary rocks of the Middle Gobi volcanic–plutonic beltin turn unconformably overlie these turbiditic and pelagic series.Finally, Triassic to Jurassic non-marine sediments (Badarch et al.,2002) discordantly top the terrane. According to Zorin (1999) andParfenov et al. (2001), the Permian Middle Gobi suite represents thesouthern, Andean type margin of the MOO. Further to the southeast,the Mongol–Okhotsk domain is fringed by a Neoproterozoic–Cambrian shelf series (Fig. 2; Windley et al., 2007), correlating withthe Idermeg terrane of Badarch et al. (2002), This unit is assumed torepresent the Early Palaeozoic southern margin of the Siberianamalgamated complex (Kröner et al., 2005; Windley et al., 2007).

Between the above described deformed, active continental marginrealms, the Adaatsag and Dochgol terranes (Badarch et al., 2002) andthe Onon zone (Zorin, 1999) show a highly deformed and metamor-phosed ophiolite-bearing rock association, which represent the sutureof the MOB. The Adaatsag unit contains questionable Ordovicianshales, metasandstones and metavolcanic rocks, Silurian corallimestones, Carboniferous red cherts, siltstones and turbiditic sand-stones. However, these lithostratigraphic age correlations are hard toconceive with the biostratigraphically dated Silurian–Devonianradiolarites, which cover ophiolitic rocks on the northern margin ofthe MOO (e.g., Kurihara et al., 2006, 2008). Associated mélangescontain serpentines, fragments of serpentinized dunite and hartzbur-gite, gabbro, metabasalts, tholeiitic pillow basalts. In the Adaatsagmélange, Tomurtogoo et al. (2005) dated a leucogabbro pegmatitedyke in layered gabbro with 325.4±1.1 Ma (zircon 207Pb/206Pb). Thisage suggests that the igneous crystallization of the plutonic suite, i.e.,the formation of Mongol–Okhotsk oceanic crust may have lasted untilCarboniferous (Namurian, transition C1–C2). The associated youngerMiddle–Late Triassic marine turbiditic clastics are intruded byTriassic–Jurassic granites (Badarch et al., 2002). The Dochgol terranecomprises questionable Devonian to Triassic marine sandstone andshale series with a coarsening trend in the Permian and Triassic. LateJurassic–Early Cretaceous post-accretionary granite and leucograniteplutons intrude the Dochgol terrane. In the eastern continuation, theOnon zone in Russia, a MORB and OIB geochemistry of the ophiolites isassumed (Zorin, 1999).

2.3. The Late Palaeozoic domain south of the Main Mongolian Lineament

This composite domain is assumed to have formed later, bystepped accretion of intraoceanic arcs and microcontinents out of theremaining Palaeo-Asian Ocean along the earlier established Siberianmargin (correlating with the Main Mongolian Lineament). The GobiAltai terrane (Fig. 2) is controversially interpreted either as anaccretionary wedge (Badarch et al., 2002) or a Devonian, carbonaticpassive continental margin system (Windley et al., 2007). The smallTseel terrane is a dated Devonian arc (ca. 397–385 Ma) with ametamorphic root (Kröner et al., 2007). According to Xiao et al.(2003) and Windley et al. (2007), the backbone of the domaincontains Silurian–Carboniferous accretionary wedges and welded

Table 2Synopsis of sedimentary facies, geodynamic characteristics, and the provenance of detrital material as depicted by the tectono-stratigraphic units of the Mongol–Okhotsk Belt inMongolia. The nature and chronostratigraphic range of the basement and sedimentary cover of the Adaatsag and Dochgol terranes is fragmentary.


intraoceanic arcs (Fig. 2; Lamb and Badarch, 2001), which supposedlyformed by northward subduction of the Palaeo-Asian ocean under theMongolian margin coincident with the Main Mongolian Lineament.The South Gobi terrane (Hutag Uul and Tsagaan Uul terranes ofBadarch et al., 2002) with a Grenvillian basement (ca. 950 Ma;Yarmolyuk et al., 2005)may represent amicrocontinent, which earlierwas detached from the North China craton. The Solonker suture to thesouth records the final, Late Permian closure of the Palaeo-AsianOcean. Presumably, this was produced by the collision of the southand north facing accretionary wedges that fringed the remainingPalaeo-Asian Ocean during the Late Palaeozoic (Xiao et al., 2003).

3. Samples and methods

We apply an advanced provenance analysis on Devonian toJurassic sandstones included in the MOB. We assume that theirdetrital content and variable sedimentary facies reveal the openingand closing history of the MOO. Most samples are derived from theHangai–Hentei and the Ereendavaa–Middle Gobi volcanic beltterranes, which frame the suture zone to the north and south,respectively (Table 1). Few Permian–Jurassic samples fromthe Adaatsag–Dochgol suture zone are completing the data set. Thechronostratigraphic attribution of the samples is according to thegeological map of Mongolia 1:1,000,000 (Tomurtogoo et al., 1999). Inusing this map there are uncertainties to account for because of oftenmissing fossil and radiometric age data. However, in most cases thechronostratigraphic correlations could be confirmed by the hereachieved youngest detrital zircon U/Pb ages. In one case the detrital

Fig. 4. Detrital zircon U/Pb laser ablation ICP-MS age distributions (relative-age–probabilityaccording to the geological map of Mongolia 1:1,000,000 (Tomurtogoo et al., 1999). Time s

zircons revealed significantly younger ages than the mapped“Devonian” unit in the Dochgol terrane (see sample 408MN07 inTable 1). With respect to the samples from the Hangai–Hentei unittwo points we want to emphasize: Badarch et al. (2002) interpretedthe Hangai–Hentei sediment series as to be deposited in an overlapbasin. According to our finding, this applies to unconformableoccurrences topping the Bayangol and Haraa terranes (see oursamples 302MN06, 410MN07, 413MN07 and 329MN06 in Fig. 3 andTable 1). The other samples are derived from the main mass of theHangai–Hentei belt, which forms the northern accretionary wedge ofthe MOO as pointed out also by Kurihara et al. (2006).

The lithologic composition of the source terranes is evaluated bytwo independent methods: (1) standard modal framework grainanalysis in sandstones distinguishing monomineralic (quartz andfeldspar) and polymineralic/polycrystalline lithic fragments. For eachsample, 300 grains were statistically counted in thin sections stainedfor feldspars and carbonate. Standard triangular plots combiningsignificant grain types allow the classification of the sandstones andthe interpretation of the plate tectonic position of the source terranes(e.g., Dickinson, 1985; Zuffa, 1980). (2) Detrital heavy minerals(density≥2.9 g/cm3, 0.063–04 mm grain size) were isolated (usingacetic acid rock dissolution and heavy liquid separation) andquantified under the petrographic microscope (e.g., Mange andMaurer, 1992). The heavy minerals in sandstones (typically b1% ofthe bulk rock) correlate with different parent basement lithologies asplutonic, metamorphic and volcanic rocks, in which these mineralsoccur as rock-forming or accessory components. The erosion of oldersedimentary rocks in the source areas is manifested by the relative

curves) measured in sandstones of the Hangai–Hentei overlap basin. Stratigraphic agescale after Ogg et al. (2008).


image of Fig.�4


increase of the ultra-stable minerals zircon, tourmaline, rutile(referred to as the ZTR association). To both methods there are limitsset by metamorphic overprint of the analyzed sandstones thateliminates unstable mineral grains (e.g., pyroxenes, hornblendes, K-feldspars) and lithic fragments. Hence, in the present study we carriedout statistical counts on sandstones not showing evident tectoniccleavage in thin sections.

For constraining the stratigraphic age of the eroded rocks, U/Pb laserablation ICP-MS dating of detrital zircons is applied. Due to the greatstability of zircons, multiple reworking of these ultra-stable grains insedimentary and magmatic environments is known (e.g., Chew et al.,2007; Martin-Gombojav and Winkler, 2008a). However, we assumethat syn-sedimentary detrital zircon grains portray the plutonic–volcanic activity in subduction-related arcs in the framing margins oftheMOObasin (Martin-Gombojav andWinkler, 2008b). In addition, thesyn-sedimentary U/Pb grain age spectra allow verifying or re-interpreting the chronostratigraphic age correlation of the sampledformations, thus contributing to an improved stratigraphic framework.

Detrital zircons were extracted from ~2 kg of sand- or siltstoneusing standard mechanical mineral separation techniques. Prior toisotopic analyses, all zircon grains were inspected by cathodolumi-nescence imaging to prove their homogeneous composition andrecognize the magmatic growth by its oscillatory patterns. Onlymagmatic domains were analyzed (Th/UN0.2; Rubatto and Gebauer,2000) and inherited cores were avoided. Laser ablation ICP-MSanalyses were performed at the Institute of Isotope Geochemistry andMineral Resources (IGMR), ETH Zurich. We utilized an Elan 6100instrument coupled to an in-house-built 193 nm Excimer laser tomeasure Pb/U and Pb isotopic ratios in detrital zircons. Ablation wasperformed in He gas (1.1 l/min), utilized a pulse rate of 10 Hz with0.5 mJ/pulse, and a focused laser beamwith a diameter of 40 μm. Rawcounts were corrected for an electron multiplier dead time of 20 ns.Data reduction was performed using the LAMTRACE code (S. Jackson,Macquarie University, Australia). The accuracy and reproducibility ofU–Pb zircon analyses were monitored by periodic measurements ofthe BR266 external standard, with 206Pb/238U and 207Pb/206Pb agesbetween 559.0±0.3 Ma and 562.2 Ma (Stern, 2001). The standardmean U, Th, and Hf concentrations were 909 ppm, 201 ppm, and8220 ppm, respectively.

Trace element contents on the dated zirconwere analyzed in orderto determine the igneous rock type, in which grains had crystallized,following the Belousova et al. (2002) method. In this study, eight rocktypes were determined according to the classification of the authorsbased on limits of the element contents (CART Treemethod). Analyseswere carried out at the IGMR, ETHZ, using the Elan 6100DRC ICP-MScoupled to a GeoLas ArF Excimer laser (wavelength of 193 nm). As forU/Pb analyses, ablated material was transported by a helium flow(1.0 l/min), spot size was 40 μm, and a laser repetition rate of 10 Hzwith 0.5 mJ/pulse was applied. Each run was composed of 4 externalstandard (NIST 610) measurements framing 16 unknown. SiO2 wasused as an internal standard and data were processed with the SILLsoftware (written by A. Murray, ETH Zürich, Switzerland andUniversity of Leeds, UK). 1% of error on the obtained values had noincidence on the rock types determined by the classification.

Finally, a Nu plasma MC-ICP-MS (Nu instrument Ltd.) attached toan Excimer laser (GeoLas ArF, wavelength of 193 nm)was used for thein-situ Hf isotopic analyses in dated zircon grains at the IGMR, ETHZ.Helium was used as carrier gas (0.8–1.1 l/min), spot size of a 60 μmdiameter was chosen, and the laser repetition rate was 4 Hz with anenergy range of 10–20 J/cm2. Mud Tank zirconwas used as a referencestandard. Baseline was measured within 30 s and ablated zirconwithin 60 s. Isobaric interference of 176Yb and 176Lu on 176Hf werecorrected by measuring 171Yb (176Yb/171Yb=0.897145) and 175Lu(176Lu/175Lu=0.026549), respectively. Age correction to calculateinitial 176Hf/177Hf ratio were obtained using a 176Lu decay constantof 1.867×10−11 year−1, the 176Lu/177Hf ratio measured and with

chondritic values of 176Hf/177Hf=0.0332 and 176Lu/177Hf=0.282772(Blichert-Toft and Albarède, 1997).

4. Results

4.1. The northern margin: Hangai–Hentei belt

The Silurian–Devonian formations consist of red radiolarian chertbeds and deep-sea turbidite deposits (Table 2) (e.g., Badarch et al.,2002; Kelty et al., 2008; Kurihara et al., 2006, 2008). The Carbonif-erous series shows a shallowing-upward trend from turbidites toproximal delta deposits including channel mouth bars. Triassiccontinental river and alluvial fan deposits overlie the foldedDevonian–Carboniferous series by angular unconformity. In spite ofmapped small occurrences of Permian deposits (Tomurtogoo et al.,1999), no Permian sedimentary formation could be sampled in theHangai–Hentei unit. They all showed to be of primary volcanic origin.The conspicuous observation of reworked Silurian–Devonian redchert and sandstone pebbles in Carboniferous conglomerates atteststhat parts of the basin were inverted during syn-sedimentaryaccretionary tectonics. The general shallowing upward trend accom-panied by unconformity development and reworking suggestsdeposition in a trench and forearc basin of Great Valley-sequencetype (Ingersoll, 1978).

The Devonian–Triassic sandstones show narrowly peaked distri-butions of Silurian–Devonian detrital zircons (Fig. 4) with a minorCarboniferous age component. In both Triassic samples, Early Permianzircons are present. Ordovician and minor Cambrian zircon ages aredetermined in the Devonian sandstones, and in a Carboniferoussample (302MN06; Fig. 4), Riphean to Ordovician ones (10 out of 56grains) are abundant. The derivation of this sample from theCarboniferous overlap formation on the Bayangol terrane indicateslocal reworking of Neoproterozoic–Early Palaeozoic sediments orbasement. Zircon trace element contents allow predicting majorigneous rock types in which they had grown: dolerite (n=53),larvikite (n=24), granitoid with 65–70% SiO2 (n=11), and basalt(n=8) (Fig. 5A). Granitoid-type with 70–75% SiO2 (n=3), nephelinsyenite- (n=2), and syenite- (n=1) are less abundant. Mafic rocks(including the larvikite type, cf. Belousova et al., 2002) are highlydominant and are present over a larger period than felsic ones. Syn-sedimentary Silurian–Devonian zircon grains show that mafic andfelsic intrusions were contemporaneous (Fig. 5A). Younger magma-tism (Carboniferous–Permian) was dominated by mafic products. Allbut one Hf isotopic ratio present positive epsilon values (Fig. 6A). Fewdata of Riphean and Sinian ages were obtained and are positive. Onevalue is close to the depleted mantle line (Griffin et al., 2002), butno trend can be interpreted from these few values. Cambrianzircons show values close to bulk earth composition (CHUR values,Blichert-Toft and Albarède, 1997). Ordovician values, as well as Siluro-Devonian ones, are clustered in two groups (mean values +5.4and +11.2 epsilon Hf), pointing to a mantellic source with variablecontinental crust influence, respectively. Carboniferous data have alsovalues close to the CHUR (Late Carboniferous) and a major clusterwith a mean value of +10 epsilon Hf. Lower values are recorded inPermian zircons. In general, the Hangai–Hentei zircon grains sinceSilurian–Devonian present a mantellic source with influence ofcontinental crust in variable proportions. This indicates the derivationof the syn-sedimentary zircons from a continental margin arccomplex.

The heavy mineral associations are dominated by zircon, tourma-line and rutile (ZTR) with 34–97% of the total (Figs. 7 and 8A), whichimplies variable supply from reworked older sediment formations andshallow continental granitic crust blocks, respectively. In severalDevonian, Carboniferous and Triassic samples intermediate to basicvolcanic source rocks are indicated by the presence (16–62%) ofamphiboles and pyroxenes (hypersthene and diopsidic augite in the

Fig. 5. Diagrams of the rock types versus the 206Pb/238U age (in Ma) for the Hangai–Hentei, Adaatsag–Dochgol and Ereendavaa–Middle Gobi zones. The types of rock were obtainedbased on the REE zircon grain contents according to the Belousova et al. (2002) classification (“Long” CART Tree classification). Time scale after Ogg et al. (2008).


Triassic sample 413MN06). Metamorphic rock heavy mineral grains(mainly epidote group andminor garnet) show awide range from 3 to60% pointing to variable volumes of regional medium-grade meta-morphic rocks (presumably schists) in the source terranes. However,the importance of regionally metamorphosed rocks in the sourceareas cannot be inferred from that alone, because part of epidote mayhave been produced by hydrothermal overprint of intermediate tobasic volcanic rocks in the source terranes. According to the modalframework grain quantification, all Devonian to Triassic sandstonesshow an arkosic to litharenitic composition (Fig. 8C) with adominance of volcanic rock fragments (60–100% of total clasts;Fig. 8B). The sandstones mostly plot in the dissected and transitionalvolcanic arc provenance field (Fig. 8C), which is in agreement with theheavy mineral results.

In a recent detrital zircon U/Pb laser ablation ICP-MS dating workon mainly Silurian–Carboniferous sandstones in the Hentei zone,Kelty et al. (2008) observed dominant age distributions peaking in theEarly Carboniferous (354–340 Ma, Early Mississippian). These resultscall attention to stronger Carboniferous volcanic activity in the sourceareas than our data would imply, where Devonian ages prevail.However, we would not consider their samples 8, 9, and 10 (Kelty etal., 2008) as sandstones related to the dynamics of the MOB. TheseCambrian to Ordovician sample points (according to the map ofTomurtogoo et al., 1999) reveal an untypically large and old age rangeof detrital zircons between Archean and Ordovician. Their measuredage distribution and their tectonic position in the Haraa terrane(Badarch et al., 2002) clearly suggest that the analyzed sediments areolder and have to be attributed to the Early Palaeozoic developmentnorth of the Main Mongolian Lineament.

In summary, and by including attributable results of Kelty et al.(2008), the zircon age spectra and the clastic grains in Silurian/Devonian to Triassic sediments of the Hangai–Hentei accretionary

wedge, developing along the northern margin of the MOO, havemain detrital sources in coeval Devonian–Early Carboniferous andPermian plutonic–volcanic continental arc rock series. The heavyminerals and the abundance of volcanic lithic clasts corroborate thepresence of transitional and dissected volcanic arcs. Potentialsource was the Selenge volcano-plutonic belt (Figs. 1 and 2),however, a pre-Carboniferous start of the volcanic–plutonic activityis implied by our results. The reason may be that the integrated ageinformation from the sandstones is more complete than local datingof in-situ plutonic–volcanic rocks. The prevailing ZTR heavy mineralassociation and metamorphic heavy minerals give evidence forreworking of older sediments and metamorphic schistous basementrocks. Rare Cambrian–Ordovician zircon grains (see also Kelty et al.,2008) testify minor contribution from the Early Palaeozoic domainbasement. The lack of substantial reworking of Proterozoic zircongrains suggests that Precambrian granitic blocks (e.g., Sayan,Gargan, Tuva–Mongolia; Fig. 2) rarely were exposed at the surfaceat that time, and the basin was also isolated from other cratonalsources like e.g., the North Asia Craton (Siberia).

4.2. The southern margin: Ereendavaa and Middle Gobi terranes

The basement consists of Palaeoproterozoic gneisses, amphibo-lites, schists and marbles, overlain by Neoproterozoic black schists,metasandstones, limestones, and minor conglomerates and volcanics,which were intruded by Devonian granites (Badarch et al., 2002).Silurian clastics, Devonian volcanics and turbiditic volcaniclasticseries (with minor limestones) unconformably overlie the Neopro-terozoic. Zonenshain and Jamjandamba (1975) and Nagibina andBadamgarav (1975) observed unconformable contacts at the base ofboth, the Silurian and Devonian formations. These are in turnunconformably overlain by Carboniferous and Permian volcanic and

image of Fig.�5

Fig. 6.Diagrams of epsilon Hf(t) versus 206Pb/238U age (inMa) for the Hangai–Hentei, Adaatsag–Dochgol and Ereendavaa–Middle Gobi zones. Corrections in function of age are basedon chondritic values (CHUR) from Blichert-Toft and Albarède (1997) which become the reference value. The depleted mantle evolution trend is from Griffin et al. (2002) values(dashed line). Time scale after Ogg et al. (2008).


image of Fig.�6

Fig. 7. Heavy mineral frequency percents counted in sandstones from the northern margin (Hangai–Hentei overlap basin) and southern margin (Ereendavaa terrane–Middle Gobibelt) of theMongol–Okhotsk ocean. Sandstones of the Adaatsag and Dochgol terranes are assumed to belong to the oceanic suture. Stratigraphic ages according to the geological mapof Mongolia 1,000,000 (Tomurtogoo et al., 1999).


marine sedimentary rocks of the Middle Gobi volcano-plutonic belt(Fig. 2, Table 2). In the Ereendavaa terrane, Carboniferous deep-seaturbidites are overlain by a shallowing-upward Permian series.Discordant Triassic to Jurassic continental alluvial/fluvial sediments

are interpreted as post-collisional formations. Late Triassic–EarlyJurassic granite plutons again intrude the terrane.

The Devonian–Jurassic sandstones depict a chronostratigraphi-cally broader detrital zircon age distribution than the northernmargin

image of Fig.�7

Fig. 8. Concluding diagrams of detrital provenance indicators according to tectonostratigraphic derivation and age. A: heavy mineral plot comparing ratios of metamorphic (MET;garnet, epidote group), volcanic (VOL; common hornblende, pyroxene group), and stable minerals (ZTR; zircon, monazite, tourmaline, rutile, brookite, anatase, sphene). Apatite isnot included here, see also Fig. 8B and C: standardmodal framework grain analysis in sandstones (after Dickinson, 1985; Zuffa, 1980) depicting proportions of monocrystalline grainsand polycrystalline lithoclasts. Abbreviations: Qm—monomineralic quartz, F—feldspars, Lt—total aphanitic lithoclasts, Lvh—volcanic hypabyssal lithoclasts, Ls—sedimentarylithoclasts, Lm—metamorphic lithoclasts. In figure C the plate tectonic position of the source rocks is inferred (Dickinson, 1985). Stratigraphic ages according to the geological map ofMongolia 1:1,000,000 (Tomurtogoo et al., 1999).


of the MOO (Hangai–Hentei). Both Devonian samples are character-ized by reworked Cambrian–Ordovician zircons associated withminor Proterozoic input (Fig. 9). The younger Devonian sample(323MN06) reveals a beginning of a Silurian–Devonian age compo-nent. The Carboniferous–Permian samples show a younging trend ofcoeval Carboniferous to Early Permian detrital zircons. The Triassicsandstone is dominated by a narrow Permian zircon age population,and in the Jurassic sample, Late Triassic–Early Jurassic zircons prevailwith a Late Jurassic component (sample 322MN06 in Fig. 9). In theCarboniferous and Permian samples, Carboniferous zircons areubiquitous. All Carboniferous to Jurassic samples have a backgroundof reworked Cambrian–Ordovician and Late Proterozoic zircon grains

Fig. 9. Detrital zircon U/Pb laser ablation ICP-MS age distributions (relative-age–probabilStratigraphic ages according to the geological map of Mongolia 1:1,000,000 (Tomurtogoo e

in common. As for the northern margin, mafic zircon-producingrocks prevailed with dolerite- (n=36), larvikite- (n=22), andbasalt-type (n=9) over felsic rocks (syenite, n=3; granitoid with65–70% SiO2, n=16; granitoidwith 70–75% SiO2, n=5) (Fig. 5C). Butin the southern margin, the presence of both groups (bimodalmagmatism) is recorded continuously from Carboniferous untilJurassic, as it holds also for reworked Cambrian–Ordovician zircons(Fig. 5C). Hf isotopic ratios of reworked zircons are widespread fromnegative values in Riphean to largely positive ones in Cambrian andOrdovician (Fig. 6C). An evolution from negative towards CHURvalues tentatively could be drawn in Riphean zircons but very littledata are available. Slightly negative values are recorded in

ity curves) measured in sandstones of the Ereendavaa terrane and Middle Gobi belt.t al., 1999). Time scale after Ogg et al. (2008).

image of Fig.�8


image of Fig.�9


Cambrian–Ordovician zircons (mean −2.6 epsilon Hf) with fewpositive ones (up to +3). Close to depleted mantle values arenoticeable in Ordovician–Devonian (mean +12.1 epsilon Hf) andCarboniferous–Permian zircons (mean +11.6 epsilon Hf). In bothcases, a decreasing trend towards CHUR values can be observed,e.g., increasing continental crust involvement. The wide-spreadCarboniferous–Triassic values suggest the establishment of anactive continental margin setting due to southward subduction ofMOO lithosphere.

With regard to the heavy mineral contents, no stratigraphic orpaleogeographic trend is obvious. However, the majority of Ereenda-vaa and Middle Gobi sediments contain a strong metamorphic grainassociation of garnet with epidote (together 40–80% of the total;Fig. 7) indicating the general presence of high greenshist-grade peliticrocks in the source areas. These metamorphic grains represent onlyminor constituents (b5%) when the reworking-related ZTR associa-tion strongly predominates. An almost unique zircon–tourmalineassociation characterizes the Devonian sample, which is typical forrecycling of older sediments and shallow crustal granitic rocks ingeneral. The Triassic sample (321MN06 in Fig. 7) that is different tothat used for detrital zircon dating, shows an exclusive volcanicassemblage of hornblende and pyroxene, whereas in other samplesthe volcanic grains usually are subordinate (≤8%).

Themodal framework grain analysis reveals a strong dominance ofvolcanic-hypabyssal rock fragments over the others (Fig. 8B, C) exceptfor the two Devonian sandstones, in which metamorphic lithic clastsprevail. In terms of plate tectonic setting of the source terranes, aderivation from transitional to dissected arc and basement upliftenvironments is indicated. However, the method has a major flaw(Dickinson, 1985), because the age of the reworked volcaniclithoclasts cannot be differentiated. This problem can be resolved byaccounting for the age populations of the associated detrital zircons.Whereas the Carboniferous through Jurassic age sandstones of theEreendavaa–Middle Gobi units possess detrital zircons of similarstratigraphic age, the Devonian ones do not, and in contrast show aderivation from metamorphic source rocks (Fig. 8B). These sand-stones also show a dominant amount of reworked Cambrian–Ordovician zircons with a minor Proterozoic age grain contribution(Fig. 9). These age spectra obviously are inherited from reworked,older Neoproterozoic–Early Palaeozoic plutonic–magmatic arc com-plexes, which are included in the Early Palaeozoic domain to the southof the Mongol–Okhotsk suture (Badarch et al., 2002; Windley et al.,2007). Data of Kelty et al. (2008) corroborate this assumption byhaving measured such detrital zircon age distributions and older onesin the northern basement block (Haraa terrane of Badarch et al., 2002)adjacent to the Hangai–Hentei belt.

In contrast to coeval sediments of the Hangai–Hentei basinrecording syn-sedimentary volcanic arc activity, the Devoniansediments of the Ereendavaa–Middle Gobi units not yet showevidence for contemporaneous volcanic input on the southern marginof theMOO at that time. Hence, there extension and basement erosionpresumably prevailed during the Devonian. Contemporaneous zirconpopulations in almost all post-Devonian samples document later,Carboniferous and Permian volcanic arc activity in the supplyingMiddle Gobi volcanic belt, which could be linked to an activecontinental margin. Triassic arc activity is not evident but duringthe Jurassic subduction-related plutonic–volcanic activity continued.The ubiquitous reworking of Cambrian–Ordovician zircons suggeststhat in the source terranes the Carboniferous–Jurassic arc complexeswere associated with older ones in the basement forming EarlyPalaeozoic domain.

4.3. The suture zone: Adaatsag and Dochgol terranes

In this zone the stratigraphic range and density of our samples isstill incomplete due to very poor outcrop conditions. We were not

able to sample rocks older than Permian. A sample mapped asDevonian in the Adaatsag terrane (Tomurtogoo et al., 1999),according to the measured zircon population, correlates with aPermo-Triassic age (sample 408MN07 in Fig. 10). However, thesepreliminary results complete well the data from the framing units.

Both terranes are classified as accretionary wedges by Badarch et al.(2002), and are considered byZorin (1999) and Parfenov et al. (2001) asthe western extremity of the Mongolia–Okhotsk suture zone. Theirlithostratigraphic content is highly deformed and in part similar.However, the lithologic attributions may be confusing, becauseassociated mélanges might contain rock elements from both theMongol–Okhotsk oceanic domain and its framing continental margins.

According to Tomurtogoo (1997) Parfenov et al. (2001) Badarch et al.(2002), the Adaatsag terrane contains Ordovician shales, metasandstonesandmetavolcanic rocks, Silurian coral limestones,mélanges, fragments ofgabbro, metabasalts, tholeiitic pillow basalts and ultramafics. In theeastern continuation (Russia) aMORBandOIB chemistry of theophiolitesis documented (Gusev and Peskov, 1996). The Middle–Late Triassicmarine clastics are intruded by Triassic–Jurassic granites. In theOnon andJargalant river sections, Zonenshain and Jamjandamba (1975), Nagibinaand Badamgarav (1975) describe a quite continuous Triassic–EarlyCretaceous marine and continental clastic series containing severaltrachybasaltic and andesitic intercalations. The Dochgol terrane isdescribed to contain Devonian to Triassic marine sandstone, shale andconglomerate series (Badarch et al., 2002). Mafic lithologies were notobservedby the authors. Late Jurassic–Early Cretaceouspost-accretionarygranite and leucogranite plutons intrude the Dochgol terrane.

In the Permian and Triassic sandstone samples (including themapped “Devonian” sample 408MN07 in Fig. 10), Permian zircongrains dominate, and an irregular mixing with Cambrian–Devonianzircons is observed. The Jurassic sandstone depicts a narrow LateTriassic–Liassic age distribution with a peak in the Rhaetian(≈202 Ma, Fig. 10) as typical for a local volcanic source. Accordingto the trace element contents, only twomeasured zircons are not frommafic rocks (see carbonatite; Fig. 5B). The majority of the zirconspoints to the derivation from larvikite (n=27), dolerite (n=12) andbasalt (n=8) that could suit into an extensional geodynamic context.The epsilon Hf values, although wide-spread, are concordant with amantellic source. Reworked Cambrian–Ordovician zirconswith valuesup to +20.1 suggest involvement of a primitive mantle, whereasother contemporaneous data indicate a CHUR affinity. Devonian toPermian grains record scattered Hf ratios (from +14.9 to +3.2),which point to a mantellic source with increasing influence of thecontinental crust through time. The Rhaetian (latest Triassic) zircons(409MN07) cluster around a mean value of +3.7 much correlatablewith a continental arc subduction system.

The heavy mineral spectra are rich (average ca. 50%) and evendominated by the ZTR association (up to 80–90%). Metamorphicgrains garnet and epidote are fairly frequent (20–30%), whereasvolcanic amphiboles and pyroxenes show a subordinate occurrence.However, modal framework grain analysis classifies the source rocksin the transitional and undisssected continental arc environment(Fig. 8). According to the sandstone composition, a derivation of theclastic material from an intra-oceanic arc (like Onon) is excluded.

5. Synthesis and geodynamic interpretation

The crosscutting manner of the Mongol–Okhotsk suture, separat-ing blocks of the Early Palaeozoic domain north of the MainMongolian Lineament, and the stratigraphic range of the involvedsedimentary rocks from latest Silurian upward suggest that the MOObasin opened after the Ordovician amalgamation of the Siberianmargin. The preceding closing of parts of the Palaeo-Asian Ocean andcollision of the margins is known to have been followed by northwardsubduction (in present coordinates) of the remaining Palaeoasianlithosphere under the Early Palaeozoic Siberian margin until the

Fig. 10. Detrital zircon U/Pb laser ablation ICP-MS age distributions (relative-age–probability curves) measured in sandstones of the Adaatsag and Dochgol terranes. Stratigraphicages according to the geological map of Mongolia 1:1,000,000 (Tomurtogoo et al., 1999). Time scale after Ogg et al. (2008).


Permian (e.g., Hendrix et al., 2001; Lamb and Badarch, 2001; Windleyet al., 2007). Consequently, we hypothesize that the MOO may haveopened by back-arc spreading during the Late Silurian by cutting intothe Early Palaeozoic domain in Mongolia (Fig. 11 a), and on a largerscale into the Transbaikalian collage in Russia, respectively. Latersubduction of MOO lithosphere toward the north and south iscorroborated by the observation that in both margins continentalvolcanic arcs penetrate and overlie deformed rocks of the EarlyPalaeozoic domain. In the north they were represented by the MiddleCarboniferous to Triassic Selenge arc and the Angar–Vitim granitoidsin Transbaikalia (340–280 Ma; e.g., Mazukabzov et al., 2010). Thesouthern continental arc was located in the Middle Carboniferous–Triassic Middle Gobi volcanic–plutonic belt (Badarch et al., 2002)including Carboniferous to Jurassic granitoids in the Kherlen Depres-sion (Murao et al., 1998). An open question remains concerning thepronounced input of contemporaneous Silurian–Devonian zircons

into the Hangai–Hentei (see also Kelty et al., 2008) from the northerncontinental margin arc. A candidate may be located in the NorthHentei metallogenic belt which contains the Devonian Zaanar graniteand Late Devonian–Early Carboniferous subvolcanic rhyolites (Dril etal., 2010).

Our results from the oceanic margin units (Table 2) suggest thatthe MOO was fringed to the north and south by active continentalmargins, but they depict slightly different periods of subduction-related volcanism. Along the northern margin (in the Hangai–Hentei margin), during the Devonian and Early Carboniferousnorthward subduction of MOO lithosphere and continental arcmagmatism prevailed, which was re-initiated during the Permian(Fig. 11B and C). Minor reworking of Early Palaeozoic domainvolcanic–plutonic basement (Cambrian–Ordovician and Late Prote-rozoic) stemming from the earlier consumption of the Palaeo-AsianOcean is documented. However, voluminous supply from Archean–

image of Fig.�10

Fig. 11. Tentative paleotectonic model for the Mongol–Okhotsk Belt in Mongolia from Silurian through Permian. Orientation according to actual coordinates. The Mongol–Okhotskocean is suggested to have opened by back-arc spreading related to the subduction of Palaeo-Asian Ocean lithosphere under the earlier amalgamated Early Palaeozoic domain inMongolia (A). Soon after spreading the northern margin inverted to subduction accretion and volcanic arc formation (Hangai–Hentei) whereas extension persisted along thesouthern margin (Ereendavaa) (B). Presumably, at the samemargin, inversion to subduction and related volcanic arc activity started in the Carboniferous (C). The indicated positionof the Onon arc in Mongolia is speculative.


Proterozoic continental blocks within (as the Gargan, Sayan, Tuva–Mongolia and Baydrag), or from the Siberian Craton is not obvious(see also Kelty et al., 2008). These blocks presumably were stillburied at that time and/or the Hangai–Hentei basin was morpho-logically sheltered from Siberian cratonal sources by the continen-tal arc.

The contemporaneous Silurian–Devonian southern margin (theEreendavaa terrane andMiddle Gobi belt) presumably represented anextensional continental margin, on which clastics from the erosion ofEarly Palaeozoic domain were deposited (Fig. 11B). It turned into anactive continental margin later in the Carboniferous as indicated bystarting input of magmatic arc zircon grains. Continued southwardsubduction (Fig. 11C) is manifested by the occurrence of Permian andTriassic zircons in coeval sediments. The most probable source ofzircons is the Middle Gobi plutonic belt including the Kherlendepression were Carboniferous, Triassic and Jurassic granitic rocksare known to intrude Devonian sediments (Murao et al., 1998).

On both margins, Triassic and Jurassic continental sediment seriesunconformably overlie tectonically deformed forearc series andaccretionary wedges (Badarch et al., 2002). This is corroborated byreworking of Silurian–Devonian lithified sediments as red radiolarianchert pebbles in the associated conglomerates. Within the northern(Hangai–Hentei) margin, however, this kind of reworking is alreadyobserved in Carboniferous turbidite beds.

Permian–Triassicmarine (turbiditic) sediments from theMOO basin(Fig. 11) included in the suture zone (Adaatsag and Dochgol terranes)depict mixed input from contemporaneous but mostly Permianvolcanic–plutonic sources together with older reworked continentalbasement. Therefore, a direct relationship with the assumed intra-

oceanic Onon arc is excluded. The Cambrian–Ordovician ones are to beattributed to the reworking of the Early Palaeozoic domain basement. Ina Jurassic, post-collisional continental sandstone, a dominating localsubduction-related volcanic source of Triassic–Early Jurassic (Rhaetian)age is individualized. The closure of theMOO should have occurred afterthat stage.

6. Conclusions

We have gathered structural, sedimentary provenance andmagmatism characterization arguments for sketching an improvedgeodynamic model for the development of the MOB in Mongolia(Table 2, Fig. 11):

1. Structural and chronostratigraphic evidences suggest that theMOOwas not a remainder of the Neoproterozoic–Early PalaeozoicPalaeo-Asian Ocean but is a younger, Late Paleozoic feature. Exceptof a Carboniferous age in the ophiolites of the Adaatsag terrane(Tomurtogoo et al., 2005), radiometric data are missing. Indirectevidence for the opening of the MOO comes from the observationthat Silurian distal Hangai–Hentei accretionary wedge depositsonlap the ophiolitic basement (Kurihara et al., 2008).

2. The MOO opened in post-Ordovician time within the EarlyPalaeozoic collage of the CAOB, probably through back-arcspreading due to protracted northward subduction of Palaeo-Asian oceanic lithosphere under the earlier acquired EarlyPalaeozoic margin in Mongolia, which fringed the North AsiaCraton (Siberia) to the south. We suggest the MOO to have

image of Fig.�11


represented a shorter-lived side branch of the Palaeo-Pacific Oceanas recently sketched by Parfenov et al. (2010).

3. Soon after basin spreading, during Silurian–Devonian the northernmargin of the MOO as represented by Early Palaeozoic terranes(e.g., Haraa, Bayangol, Tsetserleg) turned to subduction accretionand the development of continental volcanic arcs (correlatablewith the Zaamar granite and Selenge volcanic–plutonic belts).Coeval Devonian sediments deposited on the southern margin(Ereendavaa–Middle Gobi) do not yet prove the existence ofcontinental volcanic arc source rocks, and a protracted extensionalmargin regime can be inferred for this period.

4. The establishment of a volcanic–plutonic arc along the southernmargin of the MOO is evidenced for the Carboniferous andPermian. The continental margin arcs present in the Middle Gobivolcanic belt and Kherlen depression, formed through southwardsubduction of the MOO lithosphere.

5. Intensive reworking of Permian volcanic arc products in Permianand Triassic marine sandstones of the MOO domain (Adaatsag–Dochgol terranes) and in coeval continental sandstones of theEreendavaa–Middle Gobi belt document a major Permian subduc-tion pulse. In contrast, along the northern margin (Hangai–Hentei)Permian volcaniclastic input had minor importance.

6. Discordant, Jurassic continental facies sandstones of the southernmargin (Ereendavaa–Middle Gobi) contain mainly Triassic volca-no-plutonic zircons and few Jurassic ones. Therefore, the collisionof themargins and suturing should have occurred after the Triassic.This is corroborated by the persistence of Triassic marine turbiditedeposition in the Adaatsag–Dochgol terrane.

On a large scale we find the MOB to have finally formed by thepost-Triassic collision of two opposing active continental margins asbasically supposed by Zorin (e.g. 1999). However, an equivalent of thesupposed intra-oceanic Onon arc in adjacent Siberia could not befound. In the present model, continental blocks as acquired during theEarly Palaeozoic orogeny and newly formed accretionary complexescollided after the bi-vergent subduction of the intervening oceanicdomain. A similar process of accretionary wedge collision wassuggested by Xiao et al. (2003) for the latest Permian closure of thePalaeo-Asian ocean leading to the Sulinheer suture in southernmostMongolia. The MOB, therefore, represented the last step of consoli-dation of the CAOB on Mongolian territory.

Acknowledgments

This work was fully supported by the Swiss Science Foundationgrant no. 2-77927-06. We appreciate the constructive comments onan earlier version of the manuscript by Tatiana Donskaya and ThomasKelty. Journal reviewers Wenjiao Xiao and Thomas Kelty furthercontributed to improve the manuscript. Anaad Chimedtseren andMunktsengel Baatar from the Mongolian University of Science andTechnology are cordially thanked for their efficient support duringfield work.

References

Badarch, G., Cunningham, W.D., Windley, B.F., 2002. A new terrane subdivision forMongolia: implications for the Phanerozoic crustal growth of Central Asia. Journalof Asian Earth Sciences 21, 87–110.

Belousova, E.A., Griffin, W.L., O'Reilly, S.Y., Fisher, N.I., 2002. Igneous zircon: traceelement composition as an indicator of source rock type. Contributions toMineralogy and Petrology 143, 602–622.

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

Buchan, C., Pfänder, J., Kröner, A., 2002. Timing of accretion and collisional deformationin the Central Asian Orogenic Belt: implications of granite geochronology in theBayankhongor ophiolite zone. Chemical Geology 192, 23–45.

Chew, D.M., Schaltegger, U., Kosler, J., Whitehouse, M.J., Gutjahr, M., Spikings, R.A.,Miskovic, A., 2007. U–Pb geochronologic evidence for the evolution of the

Gondwanan margin of the north-central Andes. Geological Society of AmericaBulletin 119, 697–711.

Dickinson, W.R., 1985. Interpreting provenance relations from detrital modes ofsandstones. In: Zuffa, G.G. (Ed.), Provenance of Arenites: NATO AdvancedStudy Institutes Series C; Mathematical and Physical Sciences, 148, pp.333–361.

Dril, S., Khanchuk, A.I., Obolenskiy, A.A., et al., 2010. Late Carboniferous through EarlyJurassic metallogenesis and tectonics of Northeast Asia. In: Nockleberg, W.J. (Ed.),Metallogenesis and Tectonics of Northeast Asia. U.S. Geological Survey ProfessionalPaper 1765.

Filippova, I.B., Bush, V.A., Didenko, A.N., 2001. Middle Paleozoic subduction belts: theleading factor in the formation of the Central Asian fold-and-thrust belt. RussianJournal of Earth Sciences 3, 405–426.

Griffin, W.L., Wang, X., Jackson, S.E., Pearson, N.J., O'Reilly, S.Y., Xu, X., Zhou, X., 2002.Zircon chemistry and magma mixing, SE China: in-situ analysis of Hf isotopes,Tonglu and Pingtan igneous complexes. Lithos 61, 237–269.

Gusev, G.S., Peskov, A.I., 1996. Geochemistry and conditions of ophiolite formations ofeastern Transbaikalia. Geochemistry 8, 723–737.

Hendrix, M.S., Beck, M.A., Badarch, G., Graham, St.A., 2001. Triassic synorogenicsedimentation in southern Mongolia: early effects of intracontinental deformation.Geological Society of America Memoir 194, 389–412.

Ingersoll, R.V., 1978. Submarine fan facies of the Upper Cretaceous Great Valleysequence, northern and central California. Sedimentary Geology 21, 205–230.

Jahn, B., Wu, F., Chen, B., 2000. Granitoids of the Central Asian Orogenic Belt andcontinental growth in the Phanerozoic. Transactions Royal Society Edinburgh 91,181–193.

Kelty, T.K., Yin, A., Dash, B., Gehrels, G.E., Ribeiro, A.E., 2008. Detrital-zircongeochronology of Paleozoic sedimentary rocks in the Hangay–Hentey basin,north-central Mongolia: implications for the tectonic evolution of the Mongol–Okhotsk Ocean in central Asia. Tectonophysics 451, 290–311.

Khain, E.V., Bibikova, E.V., Kröner, A., Zhuravlev, D.Z., Sklyarov, E.V., Fedotova, A.A.,Kravchenko-Bereznoy, I.R., 2002. The most ancient ophiolites of the Central Asianfold belt: U–Pb, and Pb–Pb zircon ages for the Dunzhugur complex, Eastern Sayan,Siberia, and geodynamic implications. Earth and Planetary Science Letters 199,311–325.

Khain, E.V., Bibikova, E.V., Salnikova, E.B., Kröner, A., Gibsher, A.S., Didenko, A.N.,Degtyarev, K.E., Fedotova, A.A., 2003. The Palaeo-Asian ocean in the Neoproterozoicand early Palaeozoic: new geochronologic data and palaeotectonic reconstructions.Precambrian Research 122, 329–358.

Kovach, V.P., Jian, P., Zarmollzuk, V.V., 2005. Magmatism and geodynamics of earlystages of the Paleoasian ocean formation— geochronological and geochemical dataon ophiolites of the Bayankhongor zone. Doklady Earth Sciences 404, 1072–1077.

Kröner, A., Windley, B.F., Badarch, G., Tomurtogoo, O., Hegner, E., Liu, D.Y., Wingate,M.T.D., 2005. Accretionary growth in the Central Asian Orogenic Belt of Mongoliaduring the Neoproterozoic and Palaeozoic and comparison with the Arabian–Nubian Shield and the present Southwest Pacific. Geophysical Research Abstracts 7SRef-ID: 1607-7962/gra/EGU05-A-06650.

Kröner, A., Windley, B.F., Badarch, G., Tomurtogoo, O., Hegner, E., Jahn, B.M.,Gruschka, S., Khain, E.V., Demoux, A., Wingate, M.T.D., 2007. Accretionarygrowth and crust-formation in the Central Asian Orogenic Belt and compar-ison with the Arabian–Nubian shield. Geological Society of America Memoir200, 181–209.

Kurihara, T., Tsukada, K., Otoh, S., Kashiwagi, K., Minjin, C., Sersmaa, G., Dorjsuren,B., Bujinlkham, B., 2006. Middle Paleozoic radiolarians from the GorkhiFormation, central Mongolia. In: Tomurhuu, D., Natal'in, B., Ya, A., Khishigsu-ren, S., Erdenesaikhan, G. (Eds.), Structural and Tectonic Correlation Across theCentral Asian Orogenic Collage: Implications for Continental Growth andIntracontinental Deformation. Abstract and Guidebook Volume, WorkshopIGCP-480. Mongolian University of Science and Technology Press, Ulaanbaatar.67 pp.

Kurihara, T., Tsukada, K., Otoh, S., et al., 2008. Upper Silurian and Devonian pelagicdeep-water radiolarian chert from the Khangai–Khentei belt of CentralMongolia: evidence for Middle Paleozoic subduction accretion activity in theCentral Asian Orogenic Belt. Journal of Asian Earth Sciences 34, 209–225.

Lamb, M.A., Badarch, G., 2001. Paleozoic sedimentary basins and volcanic arc systems ofsouthern Mongolia: new geochemical and petrographic constraints. GeologicalSociety of America Memoir 194, 117–149.

Mange, M.A., Maurer, H.F.W., 1992. Heavy Minerals in Colour. Chapman & Hall. 147 pp.Martin-Gombojav, N., Winkler, W., 2008a. Recycling of Proterozoic crust in the Andean

Amazon foreland of Ecuador: implications for orogenic development of theNorthern Andes. Terra Nova 20, 22–31.

Martin-Gombojav, N., Winkler, W., 2008b. Sediment Provenance analysis in theMongol–Okhotsk belt (NE Mongolia): preliminary results and geodynamicimplications. Structural and Tectonic Correlation Across the Central AsianOrogenic Collage: Implications for Continental Growth and IntracontinentalDeformation (Abstracts & Excursion Guidebook). 3 rd International Workshopand Field Excursion for IGCP Project 480, Beijing 6–15 August 2007, 24–26.

Mazukabzov, A.M., Donskaya, T.V., Gladkochub, D.P., Paderin, I.P., 2010. The LatePaleozoic geodynamics of the West Transbaikalian segment of the CentralAsian fold belt. Russian Geology and Geophysics 51, 482–491.

Murao, S., Dorjgotov, D., Tseden, T., 1998. K–Ar dating of granitoids and hydrothermalmicas from the northern part of Kherlen Depression, Mongolia. Bulletin GeologicalSurvey of Japan 49, 249–255.

Nagibina, M.S., Badamgarav, J., 1975. Stratigraphy of Late Mesozoic formations,North-East Mongolia. In: Zaitsev, N.S., Luvsandanzan, B., Martinson, G.G.,Menner, V.V., Pavlova, T.G., Peive, A.V., Timofeev, P.P., Tomurtogoo, O., Yanshin,


A.L. (Eds.), Stratigraphy of Mesozoic Formations in Mongolia. Nauka, Moscow,pp. 184–198.

Natal'in, B., 1993. History and modes of Mesozoic accretion in Southeastern Russia. TheIsland Arc 2, 15–34.

Ogg, J., Ogg, G., Gradstein, F., 2008. The Concise Geological Time Scale. CambridgeUniversity Press. 177 pp.

Parfenov, L.M., Bulgatov, A.N., Gordienko, I.V., 1995. Terranes and accretionary historyof the Transbaikal Orogenic Belts. International Geology Review 37, 736–751.

Parfenov, L.M., Popeko, L.I., Tomurtogoo, O., 2001. Problems of tectonics of the Mongol–Okhotsk orogenic belt. Geology of the Pacific Ocean 16, 797–830.

Parfenov, L.M., Berzin, N.A., Badarch, G., et al., 2010. Tectonic and metallogenic modelfor Northeast Asia. In: Nockleberg, W.J. (Ed.), Metallogenesis and Tectonics ofNortheast Asia. U.S. Geological Survey Professional Paper 1765, 9-1–9-55.

Rubatto, D., Gebauer, D., 2000. Use of cathodoluminescence for U–Pb zircon dating byion microprobe: some examples from the Western Alps. In: Pagel, M., Blanc, Ph.,Barbin, V., Ohnenstetter, D. (Eds.), Cathodoluminescence in Geosciences. SpringerVerlag, pp. 373–400.

Ruzhentsev, S.V., Mossakovskiy, A.A., 1996. Geodynamics and tectonic evolution of theCentral Paleozoic structures as a result of the interaction between the Pacific andIndo-Atlantic segment of the Earth. Geotectonics 29, 294–311.

Sal'nikova, E.B., Kozakov, I.K., Kotov, A.B., Kröner, A., Todt,W., Nutman, A., Yakovlev, S.Z.,Kovach, V.P., 2001. Age of Palaeozoic granites and metamorphism in the Tuvino-Mongolian massif of the Central Asian mobile belt: loss of a Precambrianmicrocontinent. Precambrian Research 110, 143–164.

Sal'nikova, E.B., Larin, A.M., Kotov, A.B., Sorokin, A.P., Sorokin, A.A., Velikoslavinsky, S.D.,Yakovleva, S.Z., Fedoseenko, A.M., Plotkina, Yu.V., 2006. The Toksko–Algominigneous complex of the Dzhugdzhur–Stanovoi folded region: age and geodynamicsetting. Doklady Earth Sciences 409A (6), 888–892.

Sengör, A.M.C., Natal'in, B.A., 1996. Paleotectonics of Asia: fragments of synthesis. In:Yin, A., Harrison, T.M. (Eds.), The Tectonic Evolution of Asia. Cambridge UniversityPress, pp. 486–640.

Sengör, A.M.C., Natal'in, B.A., Burtman, V.S., 1993. Evolution of the Altaid tectoniccollage and Palaeozoic crustal growth in Eurasia. Nature 364, 299–306.

Shi, G.R., 2006. The marine Permian of East and Northeast Asia: an overview ofbiostratigraphy, palaeobiology and palaeogeographical implications. Journal ofAsian Earth Sciences 26, 175–206.

Sorokin, A.A., Kudryashov, N.M., Sorokin, A.P., 2002. Fragments of Paleozoic activemargins at the southern periphery of the Mongol–Okhotsk foldbelt: evidence forthe northeastern Argun terrane, Amur river region. Doklady Earth Sciences 387A,1038–1042.

Sorokin, A.A., Kudryashov, N.M., Sorokin, A.P., Rublev, A.G., Levchenko, O.A., Kotov, A.B.,Kudryashov, N.M., Sal'nikova, E.B., Kovach, V.P., 2003. Geochronology, geochem-istry, and geodynamic setting of Paleozoic granitoids in the eastern segment ofMongol–Okhotsk Belt. Doklady Earth Sciences 393, 1136–1140.

Sorokin, A.A., Yarmolyuk, V.V., Kotov, A.B., Sorokin, A.P., Kudryashov, N.M., Jinyi, L.,2004. Geochronology of Triassic–Jurassic granitoids in the southern framing of theMongol–Okhotsk foldbelt and the problem of Early Mesozoic granite formation inCentral and Eastern Asia. Doklady Earth Sciences 399, 1091–1094.

Sorokin, A.A., Kotov, A.B., Kudryashov, N.M., Kovach, V.P., 2005. Late Paleozoic UrushaMagmatic Complex in the southern framing of the Mongol–Okhotsk Belt (Amurregion): age and geodynamic setting. Petrology 13, 596–610.

Stern, R.A., 2001. A new isotopic and trace element standard for the ion microprobe:preliminary thermal ionization mass spectrometry (TIMS) U–Pb and electron-microprobe data. Radiogenic Age and Isotopic studies: Report 14, Geological Surveyof Canada, Current Research 2001-F1, 11.

Tomurtogoo, O., 1997. A new tectonic scheme of the Paleozoides in Mongolia. In:Zhaqin, X., Yufeng, R., Xiaoping, Q. (Eds.), Proceedings of the 30th InternationalGeological Congress, VSP, pp. 75–82.

Geological map of Mongolia 1:1,000,000. In: Tomurtogoo, et al. (Ed.), MongolianAcademy of Sciences, Institute of Geology and Mineral Resources, MineralResources Authority of Mongolia.

Tomurtogoo, O., Windley, B.F., Kröner, A., Badarch, G., Liu, D.Y., 2005. Zircon age andoccurrence of the Adaatsag ophiolite and Muron shear zone, central Mongolia:constraints on the evolution of the Mongol–Okhotsk ocean, suture and orogen.Journal of the Geological Society of London 162, 125–134.

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

Xiao, W., Windley, B.F., Hao, J., Zhai, M., 2003. Accretion leading to collision and thePermian suture, Inner Mongolia, China: termination of the Central Asian OrogenicBelt. Tectonics 22 (6), 1069. doi:10.1029/2002TC001484.

Yarmolyuk, V.V., Kovalenko, V.I., Sal'nikova, E.B., Kozakov, L.K., Kotov, A.B., Kovach,V.P., Vladykin, N.V., Yakovleva, S.Z., 2005. U–Pb age of syn- and postmeta-morphic granitoids of South Mongolia: evidence for the presence ofgrenvillides in the central Asian foldbelt. Doklady Earth Sciences 404,986–990.

Zonenshain, L.P., Jamjandamba, L., 1975. Stratigraphy of Early Mesozoic formations,North-East Mongolia. In: Zaitsev, N.S., Luvsandan'zan, B., Martinson, G.G., Menner,V.V., Pavlova, T.G., Peive, A.V., Timofeev, P.P., Tomurtogoo, O., Yanshin, A.L. (Eds.),Stratigraphy of Mesozoic Formations in Mongolia. Nauka, Moscow, pp. 198–226 (inRussian).

Zonenshain, L.P., Kuzmin, M.I., Natapov, L.M., 1990. Geology of the USSR: a plate-tectonic synthesis. Geodynamics Series 21, 97–120.

Zorin, Yu.A., 1999. Geodynamics of the western part of the Mongol–Okhotsk collisionalbelt, Trans-Baikal region (Russia) and Mongolia. Tectonophysics 306, 33–56.

Zorin, Yu.A., Belichenko, V.G., Turutanov, E.Kh., Kozhevnikov, V.M., Ruzhentsev, S.V.,Dergunov, A.B., Filippova, I.B., Tomurtogoo, O., Arvisbaatar, N., Bayasgalan, Ts,Biambaa, Ch., Khosbayar, P., 1993. The South Siberia–Central Mongolia transect.Tectonophysics 225, 361–378.

Zorin, Yu.A., Sklyarov, E.V., Belichenko, V.G., Mazukabzov, A.M., 2007. Evolution ofisland arcs and geodynamics of the eastern Central Asian Foldbelt in the Neogea.Doklady Earth Sciences 412, 39–42.

Zuffa, G.G., 1980. Hybrid arenites: their composition and classification. Journal ofSedimentary Petrology 50, 21–29.

http://dx.doi.org/10.1029/2002TC001484

Documents

The Mongol–Okhotsk Belt in Mongolia — An appraisal of the geodynamic development by the study of sandstone provenance and detrital zircons