Geochronology and Tectonic Implications of the Urgamal Eclogite, Western Mongolia

Journal of Mineralogical and Petrological Sciences, Volume 115, page 357–364, 2020

Geochronology and tectonic implications of the Urgamal eclogite, Western Mongolia

† ‡ Kosuke NAEMURA*,**,***, Choindonjamts ERDENEJARGAL , Terbishiinkhen O. JAVKHLAN**, , Takenori KATO§ and Takao HIRAJIMA***

*Nagoya University Museum, Nagoya 464–0814, Japan **School of Geology and Mining Engineering, Mongolian University of Science and Technology, Ulaanbaatar 14191, Mongolia ***Department of Geology and Mineralogy, Graduate School of Sciences, Kyoto University, Kyoto 606–8502, Japan †Institute of Geology, Mongolian Academy of Sciences, Ulaanbaatar 18080 POB, Mongolia ‡Institute of Mineral Resources, Chinese Academy of Geological Sciences, Beijing 100037, China §Institute for Space–Earth Environmental Research, Nagoya University, Nagoya 464–8601, Japan

The Urgamal eclogite in western Mongolia, described here for the ﬁrst time, occurs in the Altai allochthon (also known as the Urgamal subzone) in the Zavkhan terrane of the Central Asian Orogenic belt. The eclogite consists of garnet, omphacite, white mica, quartz, epidote, rutile, and barroisite. We measured zircon U–Th–Pb ages in a barroisite–rich sample of the eclogite using an electron probe microanalyzer, which yielded an age of 1113 +31/ −52 Ma. This suggests that the eclogite protolith is slightly older than the recorded metamorphic and igneous activity in the Zavkhan autochthon (880–780 Ma). To constrain the timing of eclogite facies metamorphism, K– Ar geochronology was applied to white mica separates, yielding an age of 395 ± 7.9 Ma. This Devonian age is much younger than the time at which the Lake terrane and the Zavkhan terrane collided (545–525 Ma). After the collisional event, the Zavkhan terrane was subjected to extensional tectonics, resulting in alkaline volcanism until the beginning of the Silurian (440 Ma). The Urgamal eclogite records a subduction or collision event at 440–395 Ma, which has not been recognized before. This convergent tectonic environment may have been the result of collision between the Zavkhan terrane and the eastern Tarvagatai terrane. During collision, part of the Altai allochthon may have been subducted beneath the Zavkhan autochthon, thereby forming the Urgamal eclogite.

Keywords: Eclogite, Central Asian Orogenic Belt, Zavkhan terrane, K–Ar geochronology, U–Th–total Pb zircon geochronology

INTRODUCTION (Schulmann and Paterson, 2011). As the Paleo–Asian Ocean closed, numerous arcs were amalgamated to form The Central Asian Orogenic Belt (CAOB), which is one the CAOB during the latest Permian (Xiao et al., 2003) or of Earth’s largest ancient mountain belts, formed adjacent Jurassic (Van der Voo et al., 2015) (Fig. 1b). to the Siberian craton between 1250 and 220 Ma (Fig. 1a) Mongolia is located in the CAOB, and its oldest (Xiao et al., 2015). The CAOB is composed mainly of tectonic unit consists of Proterozoic cratonic terranes that Precambrian cratons, island arcs, accretionary complexes, are surrounded by Paleozoic arcs and accretionary zones and ophiolites (Xiao et al., 2015). Tectonic activity in (Xiao et al., 2015). In southwestern Mongolia, Protero- the CAOB started with the opening of the Paleo–Asian zoic crust occurs within the Tuva–Mongolia, Zavkhan, Ocean, when oceanic arcs formed in response to plate sub- Baidrag, and Tarvagatai terranes (Fig. 1a). These terranes duction (Xiao et al., 2015). During this time, the CAOB were amalgamated with the Lake terrane during the early produced immense amounts of juvenile crust (Jahn, 2004) Paleozoic, with the Gobi–Altai terrane and Trans–Altai that today makes up ~ 20 vol% of the Eurasian continent terranes during the Ordovician–Carboniferous, and finally with the South Gobi terrane during the Permian (Gibson doi:10.2465/jmps.191126 et al., 2013; Xiao et al., 2015). The autochthon and al- K. Naemura, [email protected] Corresponding author lochthon units in the Zavkhan terrane are separated by a 358 K. Naemura, C. Erdenejargal, T.O. Javkhlan, T. Kato and T. Hirajima key geological boundary, yet the structure and geological basement of the Zavkhan terrane consists of migmatitic history of these units are poorly constrained (Fig. 1a) orthogneiss. The Khavchig complex in the Altai alloch- (e.g., Bold et al., 2016). The autochthonous part of the thon yields a U–Pb zircon age of 840 ± 9 Ma for zircon Zavkhan terrane (hereafter referred to as the Zavkhan au- rims that formed during anatexis (Zhao et al., 2006) and tochthon) is located in the northeast. The allochthonous 1967 ± 13 Ma for the cores of magmatic zircons (Bold et part occupies the southwest part of the Zavkhan terrane, al., 2016). In the Zavkhan autochthon, samples of granite and it borders the Lake terrane to the west. The allochth- and trondhjemite near Zavkhan–Mandal village yield U– onous part is also referred to as the ‘Urgamal subzone’ by Pb zircon ages of 856 ± 2, 786 ± 6, and 862 ± 3 Ma (Ko- Badarch et al. (2002) and the ‘Altai allochthon’ by Bold zakov et al., 2014, 2017). Two distinct magmatic events et al. (2016). In this study, ‘Altai allochthon’ is used here- are recognized in this area. The older igneous rock is gab- after. It is not clear whether the Zavkhan autochthon and brodiorite that cuts the migmatitic gneiss in the northern Altai allochthon share the same Proterozoic basement part of the Budun Formation (Kozakov et al., 2014). The (Bold et al., 2016). These two units are separated by a zircon U–Pb age of the Budun Formation (~ 860 ± 3 Ma) large Cambrian fault, which named as ‘Fault–1’ by Bold is interpreted to represent the start of regional metamor- et al. (2016) where they have suggested the reactivation phism (Kozakov et al., 2014). During or following this of the fault after Middle Devonian. We have revised po- regional metamorphic event, granitic magmatism pro- sition of ‘Fault–1’ in the middle latitude part (~ N48°) to duced the Zavkhan batholith. Zircon U–Pb dating of fit the terrane subdivision by Kozakov et al. (2017). trondhjemites from the Zavkhan batholith yielded an In the middle latitude part of the Zavkhan terrane age of 786 ± 6 Ma (Kozakov et al., 2014). Kozakov et (Fig. 1b), previous studies on U–Pb zircon geochronolo- al. (2014, 2017) postulated that magmatism and metamor- gy has suggested that the regional metamorphism and phism in the Zavkhan terrane occurred mainly between folding ended at 786 ± 6 Ma (Kozakov et al., 2014, 880 and 780 Ma. Accretion–collision between the Lake 2017). However, we discovered an eclogite in the Altai terrane and the Zavkhan terrane took place between 545 allochthon that may indicate a different timing of events. and 525 Ma (Bold et al., 2016). After the collision, the This paper aims to constrain the timing of high–P meta- Zavkhan terrane was subjected to extension until the early morphism of the Urgamal eclogite, to identify a previous- Silurian (~ 440 Ma). Finally, the Zavkhan terrane was in- ly unknown subduction event recorded by the Zavkhan truded by the Hangai batholith between 270 and 240 Ma terrane and its bearing on tectonic evolution of CAOB. (Yarmolyuk et al., 2011). The area targeted in this study has a complex struc- TECTONIC SETTING ture (Fig. 1b), comprising geologically distinct fault– bounded units of the Zavkhan autochthon and the Altai The Lake terrane, which has previously been described as allochthon that were covered by Tsagaan Oloom Forma- an island arc, consists of ophiolite and a volcanic arc tion which deposited from Cryogenian to Ediacaran. In complex that formed during the early Paleozoic (Xiao this area, Urgamal–Shubun and Tsagannur Formations et al., 2015). The oldest event recorded by the Lake ter- are classified as part of the Altai allochthon (Samozvant- rane is the emplacement of plagiogranites in the Dariv– sov et al., 1982). The allochthon includes garnet–biotite– Khantaishir ophiolite (568 ± 4 Ma, Gibsher et al., 2001; muscovite gneiss and muscovite gneiss of sedimentary 573 ± 6 Ma, Kozakov et al., 2012). At the SE margin of origin (Kozakov et al., 2014), intercalated with garnet the Lake terrane, the Chandmani (Ch.) eclogite is closely amphibolite, quartzite, and marbles. The allochthon has related to the Erdene–Uul ophiolite (Fig. 1a; Buriánek et been metamorphosed to the epidote amphibolite and am- al., 2017). The Ch. eclogite experienced major prograde phibolite facies (Kozakov et al., 2017). Kozakov et al. metamorphism from blueschist to amphibole eclogite fa- (2014) estimated that M1 metamorphism of rocks in the cies (22–24 kbar at 520–570 °C) at ~ 600 Ma, followed Altai allochthon took place at 810–785 Ma. These al- by a medium–P/T overprint (10 kbar at ~ 600 °C) during lochthonous units unconformably overlay Budun Forma- the collision between the Baidrag and Lake terranes at tions of the Zavkhan autochthon. ~ 540 Ma (Štípská et al., 2010; Javkhlan et al., 2014, We discovered an eclogite in the Shubun Formation 2019). Granitoid magmatism in the Lake terrane started which is located in the eastern Urgamal–Shubun Forma- at ~ 530 Ma (Rudnev et al., 2009). tion of the Altai allochthon. Shubun Formation made The Zavkhan terrane has been described as a cratonic up of metabasite–siliceous–terrigenous rock association terrane (e.g., Badarch et al., 2002) and a microcontinent (Yarmolyuk et al., 2017). Kozakov et al. (2017) reported (e.g., Lehmann et al., 2010; Bold et al., 2016) that formed 960–930 Ma ages for zircon from granitic and gabbroic mostly at the margin of the Rodinia supercontinent. The gneiss from the Urgamal–Shubun Formation (rocks types Geochronology and tectonic implications of eclogite, Western Mongolia 359

Figure 1. (a) Simplified geological map of Western Mongolia (modified after Bold et al., 2016). The eclogite localities are indicated as stars. The Zavkhan autochthon is bounded by the Altai allochthon across a regional tectonic boundary (‘Fault–1’ of Bold et al., 2016). We have revised the shape of the tectonic boundary based on geological sketch map of Kozakov et al. (2017) (see text for more information). (b) Simplified geological map of the Urgamal eclogite locality, after Kozakov et al. (2017), which is based on the geological map of Samoz- vantsov et al. (1982). Characters in circles represent first few letters of three villages: U, Urgamal; ZM, Zavkhan–Mandal; Dur, Durvuljin. Star represents a location of the Urgamal eclogite. Some geological terranes are labelled by Formation names. 360 K. Naemura, C. Erdenejargal, T.O. Javkhlan, T. Kato and T. Hirajima

Figure 2. Photomicrographs, back–scattered electron (BSE) image, and a concentration map of potassium of the Urgamal eclogite. (a) Photo- micrograph of a moderately retrogressed eclogite sample 8028. Garnet (Grt) coexists with omphacite (Omp), rutile (Rt), epidote (Ep), barroisite (Brs), white Mica (Wm), and quartz (Qtz), which have been retrogressed into a symplectite of hornblende (Hb), albite (Ab), and biotite (Bt). (b) Photomicrograph of eclogite sample 2602, in which well–preserved Omp coexists with Grt, Qtz, and Wm, with accessory Ep, Brs, and Rt. (c) BSE image of zircon (Zrn) crystals in a thin section of sample 8028. Ilm Ilmenite. (d) K2O concentration map of eclogite sample 2602, showing the distribution of Wm.

were not given in the original article and were provided gressed to black–colored amphibolite via the transforma- by one of our authors, Erdenjargal, C.). Enormous quartz- tion of omphacite to hornblende and plagioclase. Inves- ite blocks (~ 4 km width) are exposed at Mt. Tsakhir, tigated samples are representative of the eclogite in this which lies along the Zavkhan river at the SE margin of area: (1) light–brown amphibolitized eclogite (sample the Shubun Formation. On the east side of Mt. Tsakhir 8028, Fig. 2a), in which omphacite grains were variably (47.6°N, 95.6°E), there are 15 occurrences of layers and replaced by hornblende and plagioclase, and (2) green– lenses in 9 outcrops along the valley. The quartzite shows colored eclogite (sample 2602, Fig. 2b), which occurs as a weak foliation deﬁned by alignment of white mica. At a20cm–wide lens (the length is not traceable beyond 1 the sampled outcrop, quartzite displays a N–S trending m) surrounded by dark–colored amphibolite. and E dipping (~ 60°) foliation that is overprinted by Eclogite facies minerals in sample 8028 consist of crenulations. Thin eclogite layers are generally parallel garnet, omphacite, quartz, epidote, white mica, rutile, and to the foliation in the quartzite, whereas larger eclogite green–blue barroisite (Fig. 2a), which have been variably layers are sometimes discordant to the foliation. Eclogite retrogressed to an amphibolite facies mineral assemblage. sample for this study came from the largest lens (~ 20 × Detailed petrography, mineral compositions, and P–T es- 60 m), whose long axis is nearly parallel to the foliation timates for the eclogite will be reported in a subsequent in the host quartzite. The lens has been variably retro- study. Coarse–grained, typically euhedral zircons, whose Geochronology and tectonic implications of eclogite, Western Mongolia 361

Table 1. Result of K–Ar geochronology for white mica in eclogite sample 2602

size is up to 100 µm in the long axis, occur sporadically and 0.0001167, respectively (Steiger and Jäger, 1977). in the matrix (Fig. 2c). The cores of large zircon crystals The K–Ar age result for white mica in eclogite is 395 ± appear to be homogeneous, but the surrounding rims con- 7.9 Ma (at a 2σ confidence level) (Table 1). Because ar- tain minute holes and inclusions of thorium oxides and gon retention temperatures in white mica is generally as- apatite (Fig. 2c). These features indicate that the zircon sumed to be ~ 350–430 °C (e.g., Kelley, 2002), the meas- crystals have been damaged by their own radioactivity, ured K–Ar age value of high–grade metamorphic rocks, and have subsequently reacted with fluids to precipitate including eclogite, have been interpreted as cooling ages apatite and thorium oxide via metamictization (e.g., when the radiogenic Ar diffusion in white mica was Suzuki and Kato, 2008). closed. However, Itaya (2020) proposed the closure tem- The major constituent minerals in sample 2602 perature of argon diffusion in white mica is much higher, are garnet, omphacite, white mica, quartz, epidote and approximately 600 °C. Adopting the latter value, it is rutile (Fig. 2b). The foliation in the eclogite is defined critical important whether metamorphic peak temperature by the alignment of white mica (Fig. 2d). There are small of the Urgamal eclogite exceed more than 600 °C or not. amounts of calcic amphibole in the matrix. The absence of lawsonite and glaucophane suggests that the peak eclogite metamorphic temperature exceeded AGE DETERMINATIONS FOR THE ~ 550–600 °C. Thus, the K–Ar age likely reflect the tim- URGAMAL ECLOGITE ing of eclogite facies metamorphism. If excess argon still remains in muscovite due to insufficient recrystallization, The main aim of this study was to use K–Ar geochronol- the obtained K–Ar age is served as an upper limit of the ogy to determine the metamorphic age of white mica in eclogite facies metamorphism. the Urgamal eclogite. We also determined U–Th total Pb ages for relict zircons to constrain their igneous proto- U–Th–total Pb age for zircon liths. The results are presented below. A high degree of metamictization makes it difficult to ap- K–Ar geochronology of white mica ply conventional laser ablation ICP–MS technique, due to the large beam diameter (generally >10 um). Chemical U– Sample 2602 was sliced and crushed using a stamp mill. Th–total Pb Isochron method of dating (CHIME) with an The fragments were sieved (75–150 µm), and after re- electron microprobe (EMP) makes it possible to deter- moving magnetic minerals, the white mica fraction was mine an age for a microvolume of the zircon domain separated. See Yagi (2006) for details of the mineral sep- (Suzuki and Adachi, 1991; Suzuki and Kato, 2008). Al- aration procedure. We chose the most abundant, purest though CHIME zircon ages generally have large errors fractions for K and Ar analyses. Analyses and calculation (up to 100 Ma at 2σ precision for low Pb concentrations), of ages and errors followed Nagao et al. (1984) and Itaya the method is precise enough to distinguish whether the et al. (1991). Potassium was analyzed by flame photome- relict zircons formed after 880 Ma as part of the Zavkhan try using a 2000 µg g−1 Cs buffer with an analytical error autochthon, or formed as part of the Altai allochthon of 2% at a 2σ confidence level. Argon isotopes were an- whose basement granitic gneiss sometimes records old alyzed on a 15 cm–radius sector type mass spectrometer history prior to 1000 Ma. with a single collector system which called as HIRU Zircon grains in conventional thin sections of sample (Itaya et al., 1991), using the isotopic dilution method 8028 were analyzed using an EMP (JCXA–733; JEOL, with a 38Ar spike. Multiple runs of the standard (JG–1 bi- Tokyo) with five wavelength–dispersive spectrometers otite, 91 Ma) gave an error of 1% at a 2σ confidence level (radius of the Rowland circle = 140 mm) at the Institute (Itaya et al., 1991). The decay constants of 40Kto40Ar, for Space–Earth Environmental Research, Nagoya Uni- 40Ca, and the ratio of 40K to total K used in the age cal- versity, Nagoya, Japan. The instrument operating condi- culations are 0.581 × 10−10 year−1, 4.962 × 10−10 year−1, tions were 15 kV accelerating voltage, 500 nA probe cur- 362 K. Naemura, C. Erdenejargal, T.O. Javkhlan, T. Kato and T. Hirajima rent, and 3 µm probe diameter. Matrix correction was per- formed using the method of Bence and Albee (1968) with the α–factor table from Kato (2005) modified for U Mβ line emission. The procedure for calculating CHIME ages from concentrations of Pb, Th, and U is described by Suzuki and Adachi (1991) and Suzuki and Kato (2008). We selected the two least–metamictized grains of zircon and identified 21 points that do not contain inclusions or holes when viewed under an optical microscope and in secondary electron images. However, among the 21 analyses, 12 yielded a significant amount (0.1–0.3 wt%) of CaO (Fig. 3a). Large calcium cations (ionic radius = 1 Å) cannot be partitioned into the zircon structure, in which both Zr and Si have small ionic radii (0.59 and 0.26 Å, respectively). Suzuki and Kato (2008) attributed similar Ca–rich zones to secondary xenotime (YPO4)–rich do- mains in zircon grains. Xenotime easily degrades into a metamict state, so Suzuki and Kato (2008) concluded that zircon analyses with higher CaO content should be avoid- ed due to the potential loss of Pb. We used the remaining nine zircon analyses for age calculations, each of which contain <0.1 wt% CaO and can be plotted on the Cocherie plot (Fig. 3) (Cocherie and Albarede, 2001). The protolith age is constrained to 1113 −52/+31 Ma (1σ). The age suggests that the protolith of the Urgamal eclogite formed at the end of the Mesoproterozoic.

DISCUSSION

New U–Th–Pb ages for zircon grains in the Urgamal eclogite suggest that the protolith formed before ~ 1000 Ma (at the 95% confidence interval). This age is slightly older than U–Pb zircon ages from the Urgamal–Shubun Forma- tion (960–930 Ma from granitic and gabbroic gneisses, Kozakov et al., 2017). The Urgamal–Shubun ages, includ- Figure 3. Results of Th–U–total Pb geochronology of zircon from ing our zircon age, are generally older than the Zavkhan the Urgamal eclogite. (a) CaO versus Th–U–total Pb age dia- autochthon in the NE Zavkhan terrane, which formed be- gram for zircon analyses of sample 8028. Note that CaO–poor tween 880 and 780 Ma (Kozakov et al., 2014). U–Pb age (<0.1 wt%) spots indicate older ages than CaO–rich spots. (b) data obtained by Bold et al. (2016) (their Table 1) also Th/Pb versus U/Pb diagram for zircon crystals from the eclogite – suggest that the oldest magmatic ages of the granitic sample 8028. Error ellipsoids represent age data for Ca poor spots in (a). The best–fit regression line (heavy line) for the el- gneiss in the Altai allochthon (1967 ± 13 Ma) are much lipsoids has the slope of 1113 Ma. The 1σ error curves are older than those of the Zavkhan autochthon (839 ± 11 shown as parabolas. The age is calculated at the weighted mean Ma). This age difference may indicate that the Altai al- point (X bar, Y bar) of the best–fit line, and the error is calcu- lochthon was formed separately to the Zavkhan autoch- lated on the hyperbolas, close to the weighted mean point, where the distance between the hyperbolas is smallest. thon, and that the two were later juxtaposed through collision. The newly discovered eclogite in the Urgamal–Shu- ages from the Altai allochthon granites, using the meta- bun Formation can be taken as evidence for subduction. morphic rims of zircon grains. In their study, granite An obtained white mica K–Ar age suggests that the Urga- gneiss from the Urd Sharga Gorge records an metamor- mal eclogite formed during the early Devonian (~ 395 ± 8 phic age of 529 ± 22 Ma, and an unnamed granitic gneiss Ma). To put this age result into context, previous geochro- from the Dund Sharga Gorge also records a metamorphic nological data are described. Bold et al. (2016) reported age of 507.07 ± 0.66 Ma. This metamorphism was prob- Geochronology and tectonic implications of eclogite, Western Mongolia 363 ably the result of collision between the Zavkhan terrane tween 440 and 395 Ma. We speculate that part of the and the Lake terrane, which occurred from 545 to 525 Ma. Altai allochthon was subducted beneath the Zavkhan au- One of the aim of our study was to investigate whether tochthon, resulting in the formation of eclogite. This con- eclogite facies metamorphism took place in response vergence could have been driven by collision of the Zav- to this collision. The Ch. eclogite in the southeastern part khan and the Tarvagatai terranes. To test this hypothesis, of the Lake terrane records 547.9 ± 2.6 Ma (1σ) age (Štíp- it is necessary to carry out detailed field mapping of met- ská et al., 2010) and likely record this collision event. In amorphic rocks in the Zavkhan terrane. contrast, the K–Ar age of the Urgamal eclogite is much younger than the collision between the Lake and Zavkhan ACKNOWLEDGMENTS terranes. Thus, it is highly unlikely that the Urgamal eclogite formed as a result of collision between these terranes. We thank the Educational Research Activity of Nagoya After the collision with the Lake terrane, the Zav- University Museum during 2014–2019, which allowed khan terrane was subjected to extensional tectonics. This the first author to study Mongolian geology. Prof. T. Oji resulted in the formation of graben that were filled with and Prof. E. Yoshida are acknowledged for their manage- bimodal igneous rocks and sediments of the late Ordovi- ment of this project. We also thank Ms. Yo. Majigsuren cian–Silurian Teel Formation (Bold et al., 2016). Syn–rift for her kind assistance during fieldwork, Dr. Uyanga Bold igneous activity resulted in the emplacement of the Num- for her suggestion on the tectonic implications of the Ur- rug granite (442.10 ± 0.19 Ma), which lies in the Khasagt gamal eclogite, and Prof. Y. Takahashi for his introduction Khairkhan range of the Zavkhan autochthon (Bold et al. to Mongolian geology. This study was carried out by the 2016). Rift–related igneous activity has also been reported joint research program of the Institute for Space–Earth in the adjacent Lake zone (Yarmolyuk et al., 2011; Soe- Environmental Research (ISEE), Nagoya University. We jono et al., 2017). Rifting is thought to have lasted until at also thank Dr. K. Yagi and Dr. T. Fujiwara of the Hiruzen least the early Silurian (~ 440 Ma), so it is also unlikely Institute for Geology and Chronology for their kind sup- that the Urgamal eclogite formed in relation to this event. port during K–Ar dating. Special thanks are also due to It is difficult to explain the formation of Devonian Ms. N. Takagi and Mr. M. Takaya for making thin sec- eclogite within the Neoproterozoic cratonic terrane. Dur- tions. We are also grateful to Dr. N. Nakano for his con- ing the Devonian, subduction–related arc volcanism took structive review comments and Prof. M. Satish–Kumar for place in the Zavkhan area, but the driving force behind the his positive review comments and patient editorial works. volcanism remains unknown (e.g., Bold et al., 2016). Our This study was supported in part by JSPS KAKENHI working hypothesis is that a previously unrecognized sub- Grant Number JP19H01999 to T.H. duction/collision event produced the eclogite and magmatism during the Devonian. A possible driving force for REFERENCES the convergence was the collision between the Zavkhan and the Tarvagatai terranes. During convergence, part of Badarch, D., Cunningham, D.W. and Windley, B. (2002) A new the Altai allochthon may have subducted beneath the terrane subdivision for Mongolia: Implications for the Phaner- ozoic crustal growth of Central Asia. Journal of Asian Earth Zavkhan autochthon. At present, there is no convincing Sciences, 21, 87–110. explanation for the eclogite in the Zavkhan terrane. Thus, Bence, A.E. and Albee, A.L. (1968) Empirical correctio factors for the Zavkhan terrane’s tectono–metamorphic history, as it the electron microanalysis of silicates and oxides. Journal of is currently understood, may require major revisions. Geology, 76, 382–403. Bold, U., Crowly, J.L., Smith, E.F., Sambuu, O. and Macdonald, F.A. (2016) Neoproterozoic to early Paleozoic tectonic evolu- CONCLUSION tion of the Zavkhan terrane of Mongolia: Implications for continental growth in the Central Asian orogenic belt. Litho- Kozakov et al. (2014, 2017) suggested that the basement sphere, 8, 729–750. rocks of the Zavkhan terrane were mainly formed during Buríanek, B., Schulmann, K., Hrdličková, K., Hanžl, P., et al. (2017) Geochemical and geochronological constraints on dis- the Neoproterozoic, that magmatic and metamorphic ac- tinct Early–Neoproterozoic and Cambrian accretionary events tivity ended prior to 786 ± 6 Ma, and that the Zavkhan along southern margin of the Baydrag Continent in western terrane was metamorphosed to the epidote–amphibolite to Mongolia. Gondowana Research, 47, 200–227. amphibolite facies. Contrary to these views, we discov- Cocherie, A. and Albarede, F. (2001) An improved U–Th–Pb age ered eclogite in the Urgamal–Shubun Formation of the calculation for electron microprobe dating of monazite. Geo- chimica et Cosmochimica Acta, 65, 4509–4522. Altai allochthon. The Zavkhan terrane was in an exten- Gibsher, A.S., Khain, E.V., Kotov, A.B., Salnikova, E.V., et al. sional environment until 440 Ma, so eclogite–forming (2001) Late Devonian age of the Khantaishir Ophiolite Com- high–pressure metamorphism must have taken place be- plex in Western Mongolia. Russian Geology and Geophysics, 364 K. Naemura, C. Erdenejargal, T.O. Javkhlan, T. Kato and T. Hirajima

42, 1178–1185. Schulmann, K. and Paterson, S. (2011) Asian Continental Growth. Gibson, T.M., Myrow, P.M., MacDonald, F.A., Minjin, C. and Nature Geoscience, 4, 827–829. Gehrels, G.E. (2013) Depositional history, tectonics, and de- Soejono, I., Buriánek, D., Janoušek, V., Svojtka, M., et al. (2017) trital zircon geochronology of Ordovician and Devonian strata A reworked Lake Zone margin: Chronological and geochem- in southwestern Mongolia. GSA Bulletin, 125, 877–893. ical constraints from the Ordovician arc–related basement of Itaya, T. (2020) K–Ar phengite geochronology of HP–UHP meta- the Hovd Zone (western Mongolia). Lithos, 294–295, 112– morphic rocks–An in–depth review–. Journal of Mineralogical 132. and Petrological Sciences, 115, 44–58. Steiger, R.H. and Jäger, E. (1977) Subcommission on geochronol- Itaya, T., Nagao, K., Inoue, K., Honjou, Y., et al. (1991) Argon ogy: Convention on the use of decay constants in geo– and isotope analysis by a newly developed mass spectrometric cosmochronology. Earth and Planetary Science Letters, 36, system for K–Ar dating. Mineralogical Journal, 15, 203–221. 359–362. Jahn, B.M. (2004) The Central Asian Orogenic Belt and growth of Štípská, P., Schulmann, K., Lehmann, J., Corsini, M., et al. (2010) the continental crust in the Phanerozoic. Geological Society, Early Cambrian eclogites in SW Mongolia: evidence that the London, Special Publications, 226, 73–100. Palaeo–Asian Ocean suture extends further east than expected. Javkhlan, T.O., Takasu, A., Kabir, M.F. and Batulzii, D. (2014) Journal of Metamorphic Geology, 28, 915–933. K–Ar ages of amphibole–rich metamorphosed veins in eclo- Suzuki, K. and Adachi, M. (1991) The chemical Th–U–total Pb gite bodies from the Lake Zone, SW Mongolia. Earth Sci- isochron ages of zircon and monazite from the gray granite ences (Chikyu Kagaku), 68, 89–96. of the Hida Terrane, Japan. Journal of Earth Sciences, Nagoya Javkhlan, T.O., Takasu, A., Kabir, M.F. and Batulzii, D. (2019) University, 38, 11–37. Multiple Metamorphic Events Recorded within Eclogites of Suzuki, K. and Kato, T. (2008) CHIME dating of monazite, xen- the Chandman District, SW Mongolia. Minerals, 9, 465, doi: otime, zircon and polycrase: Protocol, pitfalls and chemical 10.3390/min9080495. criterion of possibly discordant age data. Gondowana Re- Kato, T. (2005) New accurate Bence–Albee α–factors for oxides search, 14, 569–586. and silicatescalculated from the PAP correction procedure. Van der Voo, R., van Hinsbergen, D.J., Domeier, M., Spakman, W. Geostandards and Geoanalytical Research, 29, 83–94. and Torsvik, T.H. (2015) Latest Jurassic–earliest Cretaceous Kelley, S. (2002) K–Ar and Ar–Ar Dating. In Noble Gas Geochem- closure of the Mongol–Okhotsk Ocean: A paleomagnetic and istry and Cosmochemistry (Porcelli, D., Ballentine, C.J. and seismological–tomographic analysis. In Late Jurassic margin Wieler, R. Eds.). Reviews in Mineralogy and Geochemistry, of Laurasia–A record of faulting accommodating plate rotation 47, Mineralogical Society of America and the Geochemical (Anderson, T.H., et al., Eds.). Geological Society of America Society, Washington, D.C., 785–818. Special Papers, 513, 589–606. Kozakov, I.K., I., Sal’nikova, E., Yarmolyuk, V., Kozlovsky, A., Xiao, W., Windley, B.F., Hao, J. and Zhai, M. (2003) Accretion et al. (2012) Convergent boundaries and related igneous and leading to collision and the Permian Solonker suture, Inner Mon- metamorphic complexes in Caledonides of Central Asia. Geo- golia, China: termination of the central Asian orogenic belt. Tec- tectonics, 46, 16–36. tonics, 22, 1069, http://dx.doi.org/10.1029/2002TC10001484. Kozakov, I.K., Kovach, V.P., Bibikova, E.V., Kirnozova, T.I., et al. Xiao, W., Windley, B.F., Sun, S., Li, J., et al. (2015) A tale of (2014) Late Riphean episode in the formation of crystalline Amalgamation of three Permo–Triassic Collage Systems in rock complexes in the Dzabkhan microcontinent: Geological, Central Asia: Oroclines, Sutures, and Terminal Accretion. An- geochronologic, and Nd isotopic–geochemical data. Petrolo- nual Review of Earth and Planetary Sciences, 43, 477–507. gy, 22, 480–506. Yagi, K. (2006) Manual of mineral separation for K–Ar age dating. Kozakov, I.K., Kuznetsov, A.B., Erdenjargal, Ch., Salnikova, E.B., Engineering geology of Japan, 19–25. et al. (2017) Neoproterozoic complexes of the shelf cover of Yarmolyuk, V.V., Kovach, V.P., Kovalenko, V.I., Salnikova, E.B., the Dzabkhan terrane basement in the Central Asian Orogenic et al. (2011) Composition, Sources, and Mechanism of Con- Belt. Stratigraphy and Geological Correlation, 25, 479–491. tinental Crust Growth in the Lake Zone of the Central Asian Lehmann, J., Schulmann, K., Lexa, O., Corsini, M., et al. (2010). Caledonides: I. Geological and Geochronological Data. Pet- Structural constraints on the Evolution of the Central Asian rology, 19, 55–78. Orogenic Belt in SW Mongolia. American Journal of Science, Yarmolyuk, V.V., Kozlovsky, A.M. and Lebedev, V.I. (2017) Neo- 310, 575–628. proterozoic Magmatic Complexes of the Songino Block Nagao, K., Nishido, H., Itaya, T. and Ogata, K. (1984) An age (Mongolia): a Problem of Formation and Correlation of Pre- determination by K–Ar method. The bulletin of the Hiruzen cambrian Terranes in the Central–Asian Orogenic Belt. Petrol- Research Institute, Okayama University of Science, 9, 19–38. ogy, 25, 365–395. Rudnev, S.N., Izokh, A.E., Kovach, V.P., Shelepaev, R.A. and Ter- Zhao, Y., Song, B. and Zhang, S.H. (2006) The Central Mongolian ent’eva, L.B. (2009) Age, Composition, Sources, and Geody- microcontinent: Its Yangtze aﬃnity and tectonic implications. namic Environments of the Origin of Granitoids in the North- Proceedings Symposium on continental growth and orogeny ern Part of the Ozernaya Zone, Western Mongolia: Growth in Asia, Taipei, Taiwan, 135–136. Mechanisms of the Paleozoic Continental Crust. Petrology, 17, 439–475. Samozvantsov V.A., Tsukernik, A.B., Polyakov V.I., Baatarhuyag Manuscript received November 26, 2019 A., Hyuppenen T.L., et al. (1982) Geological Map of the Manuscript accepted April 20, 2020 Mongolian People’s Republic, Scale 1:200000. M–46–XXXV, Published online July 20, 2020 Tekhnoeksport, Moscow (in Russia). Manuscript handled by M. Satish–Kumar