Xu-xun M, Li-E Go, Zhong-bo Zho,b, Xi-jie Chen, Hi-bing Li,b

a Institute of Geology, Chinese Academy of Geological Sciences, Beijing 100037, China

b Southern Marine Science and Engineering Guangdong Laboratory, Guangzhou 511458, China

c Development and Research Center of China Geological Survey, Beijing 100037, China

Keywords:Leucocratic sill/dike swarm Early Eocene Indo-Asian collision Gangdese Tibet China

ABSTRACT The timing of the initial Indo-Asian collision is a subject of debate for a long time. Besides, the magmatic trace of the collisional process is also unclear. In the present study, the authors report Early Eocene leucocratic sill/dike swarms in the northern edge of the Nymo intrusive complex of the Gangdese belt,southern Tibet. The Nymo intrusive complex was emplaced at ca. 50-47 Ma and surrounded by the metamorphosed Jurassic-aged Bima Formation volcano-sedimentary sequence along its northern side. At outcrops, the leucocratic sills/dikes intruded along or truncated the deformed foliations of the host Bima Formation, which has been subject to high-temperature amphibolite-facies metamorphism at ca. 50-47 Ma. Detailed cathodoluminescence image analyses reveal that the zircon grains of the leucocratic sills/dikes have core-mantle textures. The cores yield the Jurassic ages comparable to the protolith ages of the Bima Formation. In contrast, the mantles of zircon grains yield weighted mean ages of ca. 49-47 Ma,representing the crystallization timing of these leucocratic sills/dikes. The coeval ages for the Nymo intrusive complex, the high-temperature metamorphism, and the leucocratic sills/dikes indicate that a close relationship exists among them. The authors tentatively suggest that these leucocratic sills/dikes were generated from partial melting of the Jurassic-aged Bima Formation volcanic rocks, triggered by the high heat from the magma chamber of the Nymo intrusive complex. This Early Eocene tectono-thermal event of coeval magmatism, metamorphism and partial melting was most likely formed during the Indo-Asian collisional setting.©2021 China Geology Editorial Office.

1. Introduction

As the natural laboratory of the continental dynamics, the Tibetan Plateau is the prototype of the continent-continent collision on Earth (Mo XX et al., 2007; Tapponnier P et al.,2001; Xu ZQ et al., 2013). During the past century,voluminous works focusing on the Indo-Asian collision have been implemented in the Tibetan Plateau. However, the initial timing of the Indo-Asian collision remains an open question(Aitchison JC et al., 2007; Ao SJ et al., 2018; Hu XM et al.,2015; van Hinsbergen DJJ et al., 2019). The reported ages of the initial collision between Indian and Asian plates range from ca. 70 Ma to ca. 40 Ma (Ao SJ et al., 2018; van Hinsbergen DJJ et al., 2019). This hot debate has hindered our understanding on the closure of the Neo-tethyan Ocean and the subsequent continent-continent collisional process. Thus,more work is needed to address the Indo-Asian collision process.

In addition, the ca. 50-40 Ma collision-related metamorphic rocks and S-type granites are mainly reported from the Tethyan Himalaya, immediately south of the Indus-Yarlung Tsangpo suture zone (Carosi R et al., 2013; Ding HX et al., 2016; Gao LE et al., 2016; Liu XC et al., 2016; Zeng LS et al., 2011). The high-pressure metamorphic rocks were considered to be by-products of the Indo-Asian collision.Meanwhile, the Himalayan sequences experienced partial melting due to the crustal thickening triggered by the Indo-Asian collision. The remarkable partial melting is represented by the widespread occurrence of the leucogranites of the gneiss domes within the Tethyan Himalayan belt (Zeng LS et al., 2011). In contrast, main collision-related partial melting and corresponding S-type granites are rarely reported in the Gangdese belt, southern Tibet (Wang Q et al., 2015). Then,why the collision-related partial melting just occurred in the underlying plate of the Indo-Asian collisional zone? And,where are the collision-related partial melting and its byproducts (e.g., leucogranite) in the Gangdese belt?

In the present study, the authors have documented leucocratic sill/dike swarms in the Nymo region, central Gangdese belt, southern Tibet. Detailed geochronological analyses reveal that these sills/dikes were emplaced at ca.49-47 Ma. The authors suspect that these leucocratic sills were formed through partial melting of the Jurassic-aged Bima Formation due to the Indo-Asian collision.

2. Geological setting

The Tibetan Plateau, as the roof of the Earth, is comprised of several microcontinents or terranes such as Kunlun,Songpan-Ganze, Qiangtang, Lhasa, and Himalayan terranes from north to south. Among these terranes, the Lhasa terrane was the last one accreted to the southern margin of the Eurasian continent prior to the Indo-Asian collision (Fig. 1).Immediately south of the Lhasa terrane, the Tethyan Himalaya is separated to the south by the Indus-Yarlung Tsangpo ophiolitic mélanges, which represent the remnants of the Neotethyan oceanic lithosphere (Fig. 1).

The Gangdese magmatic belt, extending approximately 2000 km along E-W trending with a width ranging from 30 km to 100 km, is located along the southern margin of the Lhasa terrane (Fig. 1). The Gangdese belt not only records the subduction of the Neotethyan oceanic lithosphere beneath the Lhasa terrane but also documents the Indo-Asian collision.Thus, the arc magmatism of the Gangdese belt spanned a large age range from ca. 245 Ma to ca. 40 Ma (Coulon C et al., 1986; Mo XX et al., 2005b; Ma XX et al., 2017a, 2018,2020a; Meng YK et al., 2018). The 65-40 Ma magmatism was probably formed in the transitional stage from oceancontinent subduction to continent-continent collision (Lee HY et al., 2009; Ma XX et al., 2017b, 2020a, 2020b). After ca.40 Ma, the magmatic rocks in the Gangdese show geochemical characters akin to those of post-collisional setting (Chung SL et al., 2005; Ma XX et al., 2016).

3. Petrography and sampling

The Nymo gabbro-dioritic complex, occupying an area of ca. 60 km in length and 20 km in width, is located in the southern margin of the Gangdese magmatic belt, southern Tibet (Fig. 2). The northern side of the intrusive complex is surrounded by the Jurassic-aged Bima Formation volcanosedimentary sequences, which have been subject to hightemperature amphibolite-facies metamorphism (Fig. 3; Ma XX et al., 2020c).

3.1. Nymo intrusive complex

Fig. 1. Simplified geological map of southern Tibet showing Himalayan-Gangdese orogen (modified from Guillot S et al., 2008). Data in Himalaya: Annapurna (Kohn MJ and Corrie SL, 2011), Everest (Cottle JM et al., 2009), Jomolhari (Regis D et al., 2014), Kaghan (Kaneko Y et al.,2003), Kali Gandaki (Iaccarino S et al., 2015), Lopu Range (Laskowski AK et al., 2016), Mabja dome (Lee J and Whitehouse MJ, 2007),Namche Barwa (Zhang ZM et al., 2015), Sikkim (Rubatto D et al., 2013), Tso Morari (Donaldson DG et al., 2013), and Yardoi dome (Ding HX et al., 2016; Gao LE et al., 2012). Data in Gangdese: Lilong (Zhang ZM et al., 2014), Milin (Zhang ZM et al., 2010), and Nymo (Dong X et al.,2018). STDS-South Tibetan Detachment System; MCT-Main Central Thrust; MBT-Main Boundary Thrust; MFT-Main Frontal Thrust;MP-median pressure; HP-high-pressure; UHP- ultrahigh-pressure; MT-median temperature; HT-high temperature.

This complex is dominated by diorite, quartz diorite,granodiorite, and minor gabbronorite, which intruded the remnant deformed Jurassic granite (Fig. 3; Zhang HF et al.,2007). The dioritic pluton has a fairly homogeneous composition and mineral assemblages including hornblende,plagioclase, biotite, and quartz. The diorite is fresh, without obvious deformation or alteration (Fig. 4b). Both the Jurassic granite and the Nymo intrusive complex have been intruded by ca. 41 Ma dioritic porphyrite dikes (Ma XX et al., 2016).

3.2. Amphibolite-facies metamorphic rocks

The high-temperature amphibolite-facies metamorphic rocks are mainly composed of garnet-biotite gneisses and plagioclase-rich amphibolites (Figs. 4c, d). The metamorphic rocks show NNW foliation and are intruded by undeformed dioritic porphyrite dikes (Ma XX et al., 2016, 2020c). The garnet-biotite gneisses (Fig. 4c), sandwiched by the pluton and the plagioclase amphibolites, are comprised mainly of garnet, biotite, plagioclase, and quartz, with minor muscovite,sillimanite, and chlorite in some samples (Ma XX et al.,2020c). All these metamorphic sequences have been intruded by later granite along the northern margin (Figs. 3, 4e), which could have been emplaced at ca. 34 Ma (Ji WQ et al., 2009).

3.3. Leucocratic sill/dike swarm

Leucocratic sill/dike swarms are observed within the metamorphic sequence (Figs. 4f, 5). Petrographically, these sills/dikes or veins (tails of sill or dike, generally with width＜10 cm) or dikes (dip ＞45°) are mainly composed of quartz,biotite, plagioclase, K-feldspar, and minor magnetite (Fig. 6).Sill will be used for convenience in the following discussion.The leucocratic sills of variable attitude, trending, and thickness (centimeters up to ca. 3 m), are dominantly Ndipping along or truncated the deformed foliations of the metamorphic rocks (Figs. 4, 5). In addition, some sills are not deformed, while others have foliations, boudinages, and little folding (such as a ptygmatic appearance). Based on the aforementioned observations, these sills could be distinguished into two suites, one suite being concordant with the foliation and the other series being discordant with the foliation. Therefore, the authors suggest that these sill emplacements may have started during or close after the culmination of the major tectonic-metamorphic event (Ma XX et al., 2020c).

Fig. 2. Simplified map for the central Gangdese belt, southern Tibet. Reported ages are collected from Dong X et al., 2018; Ji WQ et al., 2009;Ma XX et al., 2020a, 2020c; Meng YK et al., 2019; Mo XX et al., 2005a; Wang C et al., 2019; Wen DR et al., 2008; Xu WC et al., 2015; Zhang HF et al., 2007.

Fig. 3. Cross-section of the Nymo intrusive complex showing the leucocratic sill swarm. The ages of the ca. 178 Ma granite, ca. 47 Ma diorite,ca. 42 Ma dikes, and ca. 34 Ma granite are referred to Ji WQ et al., 2009; Ma XX et al., 2016, 2020c; Zhang HF et al., 2007.

4. Analytical method

The zircon U-Pb dating analyses were completed at Institute of Geology, Chinese Academy of Geological Sciences, Beijing 100037. These samples include samples m16-22, m17-24, m17-27, and m17-30. U-Pb abundance data were measured by the latest Neptune Plus multiple collector ICP-MS (MC-ICP-MS) from Thermo Fisher Co. Ltd. The GeoLasPro 193 nm laser-ablation system of U.S. Coherent Co. Ltd was used in the measurement. Helium was used as a carrier gas to enhance transport efficiency of ablated material.The spot size of a laser ablation beam is approximate 32 μm.Laser energy adopted in the measurements was 10 J/cm2in energy density and 8 Hz in frequency. Uranium (U) and lead(Pb) in zircon were ionized in plasma with hyperthermia up to 10000 K. Using a dynamic zoom and extended dispersion, the instrument could collect U-Pb isotopes’ mass numbers of great differences.In-situmeasurement of the U-Pb isotopes can be done within the micro area of zircon. During the measurement, the instrument spent four seconds measuring the background and 23 seconds measuring every spot. The MC-ICP-MS operating conditions were optimized with the measurements of standard zircon 91500, to provide the maximum sensitivity and maintain a minimum oxide production rate of ThO+Th ＜2% as well as the lowest background. The accuracy of the data is verified by using GJ-1 as an auxiliary standard. The MC-ICP-MS measurement was carried out using time-resolved analysis operated in fast peak-hopping and DUAL detector mode using a short integration time. The Harvard standard zircon 91500 and one GJ-1 were measured for every 5-10 sample analyses. The data were calculated by the ICPMSDataCal program (Liu YS et al., 2008) and the Isoplot program in ref. of (Ludwig KR,2003). The analytical results are listed in Table S1 (supporting information).

Fig. 4. Field photos for the Nymo intrusive complex. a-deformed granite; b-undeformed diorite; c-garnet-biotite gneiss; d-plagioclase amphibolite; e-about 34 Ma granite; and f-leucocratic sill swarm within the amphibolitic rocks. Grt-garnet.

5. Analytical results

5.1. Zircon cathodoluminescence (CL) images

Representative zircon CL images are shown in Fig. 7. In general, these zircon grains have prismatic and euhedral shapes with oscillatory zoning, indicating a magmatic origin.Specifically, most of the zircon grains of samples m16-22,m17-24, and m17-27 have core-mantle textures, with cores are inherited from a basement or country rocks. In contrast,zircon grains from sample m17-30 are dark in color, showing blurry oscillatory zoning. These zircon grains could be magmatic in origin, but with extremely high U contents,which induced the spongy textures indicative of metamictimation (Fig. 7).

5.2. Zircon U-Pb dating

Samples m16-22, m17-24, m17-27, and m17-30 of leucocratic sills have been collected for zircon U-Pb dating analyses. The dating spots are chosen based on the analyses of CL images of these zircon grains (Fig. 7). The results yield weighted mean ages of 49.43 ± 0.99 Ma, 47.56 ± 0.57 Ma,48.08 ± 0.38 Ma, and 48.25 ± 0.26 Ma for these samples,respectively (Fig. 8). Within error, these ages are indistinguishable from each other, implying that these sills are emplaced coevally. Thus, it interprets that these ages could be used to represent the crystallization ages of the leucocratic sills.

Fig. 5. Field photos for the studied leucocratic sills. a, b-leucocratic sill swarm; c-stoped block warped by the sill; d-leucocratic sill intruded along the deformed foliation; e-necking of the sill, and; f-leucocratic sill truncated the deformed foliation.

6. Discussion

6.1. Emplacement timing and petrogenesis

The Nymo leucocratic sills comprise a series of undeformed and/or weakly deformed sills, veins, or dikes within the deformed Jurassic-aged amphibolite-facies Bima Formation. At the outcrop scale, the principal regional foliations in the Bima Formation volcano-sedimentary sequences are vertical or high-angle N-dipping (Figs. 3, 4d,5). Generally, these leucocratic sills intruded along the deformed foliation within the host sequences (Fig. 5c).Sometimes, some sills truncated the deformed foliations, with a stoped block of host plagioclase amphibolite or gneiss being wrapped by the sill (Fig. 5d). These observations suggest that these leucocratic sills were formed coevally with or a little later than the deformed event of the host rocks. The Jurassic Bima Formation has been subject to high-temperature (HT)amphibolite-facies metamorphism, which took place at ca.50-47 Ma (Ma XX et al., 2020c). In addition, the Nymo intrusive complex was emplaced at ca. 50-47 Ma, too. Thus,the Early Eocene HT metamorphism could be triggered by the coeval high heat from the magma chamber of the Nymo intrusive complex (Ma XX et al., 2020c).

Fig. 6. Photomicrographs (cross-polarized light image) to illustrate petrographic features of the leucocratic sills (a, c, and e). Photomicrographs (plane-polarized light image) to illustrate petrographic features of the leucocratic sills (b, d, and f). Qtz-quartz; Bt-biotite; Pl-plagioclase; Mag-magnetite; Kfs-K-feldspar.

Several sill samples have been collected for zircon U-Pb dating analyses. These samples yield indistinguishable crystallization ages falling into the 49-47 Ma range,synchronous with the crystallization timing of the Nymo intrusive complex and the metamorphic timing of the Bima Formation. This strong synchronism among magma emplacement, HT metamorphism, and partial melting reveals that these tectonothermal events have a close interrelationship. Furthermore, as a whole, the Early Eocene Nymo intrusive complex, HT metamorphic rocks, and the leucocratic sills have all been intruded by ca. 41 Ma dioritic porphyrite dikes (Ma XX et al., 2016). This age constraint from dioritic porphyrite dikes further corroborates that the Nymo leucocratic sills were emplaced at Early Eocene.

Fig. 7. Representative cathodoluminescence (CL) images of the zircon grains and the corresponding dating spots. The CL images were taken at Institute of Geology, Chinese Academy of Geological Sciences.

Fig. 8. U-Pb weighted mean ages of zircon grains from the leucocratic sills/dikes (a, c, e, and g) and field photos of sampling location (b, d, f,and h).

Fig. 9. Tentative cartoon showing the Early Eocene tectonic regime of the Indo-Asian collision and the relationship between the Nymo intrusive complex and the leucocratic sill/dike swarm.

The magma source of the Nymo leucocratic sills could be the “so-called” basement rocks of the Gangdese belt, or the Indian crust indented beneath the Lhasa terrane, or the Jurassic Bima Formation volcano-sedimentary rocks.However, the Gangdese belt is dominated by the juvenile crust (Ji WQ et al., 2009), no obvious basement rocks have been revealed until now, except the scattered Precambrian basement rocks in the eastern Himalayan Syntaxis region(Dong X et al., 2020). Previous geochemical data of ca. 50 Ma igneous rocks in the Gangdese belt exhibit typical arc-affinity features (Zhou LM et al., 2018), indicating continuing subduction of the Neotethyan oceanic lithosphere instead of the Indian continent beneath the Lhasa terrane at ca. 50 Ma(Aitchison JC et al., 2007; Ao SJ et al., 2018; van Hinsbergen DJJ et al., 2019). Thus, the ancient Indian continent material shouldn ’t have been involved in the magma source of the Nymo leucocratic sills. In our opinion, the magma source of the Nymo leucocratic sills was most likely derived from partial melting of the Jurassic Bima Formation volcanosedimentary rocks. This contention is corroborated by (1) the lack of muscovite for the leucocratic sills (Fig. 6), ruling out the partial melting of pure sedimentary sequences (Aikman AB et al., 2008); (2) the abundance of Early-Middle Jurassicaged inherited zircon grains in the leucocratic sills (Fig. 7),and these inherited zircon grains have coeval ages with the protoliths of the Bima Formation in the Gangdese belt (Kang ZQ et al., 2014; Ma XX et al., 2017a; Wang C et al., 2019);and (3) thein-situmelting of the high-temperature amphibolite and gneiss (530-750 MPa and 700-800°C) (Figs.8b, d; Ma XX et al., 2020c).

As discussed above, these leucocratic sills were probably produced by partial melting of the Jurassic-aged Bima Formation volcanic rocks due to voluminous heating from the magma chamber of the Nymo intrusive complex (Ma XX et al., 2020c) and/or the crustal thickening of the Gangdese belt at ca. 50 Ma (Ma XX et al., 2021). The Early Eocene crustal thickening in the Gangdese belt was probably triggered by the magma addition due to Early Eocene magmatic flare-up (Ma XX et al., 2021; Zhu DC et al., 2017). Of course, the magma heat from the magma chamber of the Nymo intrusive complex could have played a pivotal role in the partial melting of the Bima Formation (Fig. 9; Ma XX et al., 2020c).

In summary, the formation of the Nymo leucocratic sills probably resulted from partial melting of the Jurassic Bima Formation volcanic rocks.

6.2. Tectonic implications

As mentioned in the introduction section, the Indo-Asian collision processes are hotly debated among the geological community. The major debate centers on the timing of the initial Indo-Asian collision. Some workers strongly intend that the initial Indo-Asian collision took place at ca. 70-65 Ma(Mo XX et al., 2003; Yin A and Harrison TM, 2000).However, the 50-40 Ma igneous rocks within the Gangdese belt still exhibit clear arc-affinity characters (Wang YF et al.,2019; Zhou LM et al., 2018), seemingly arguing against the ca. 70-65 Ma initial collisional model. In other words, these igneous rocks were probably still formed in a subductionrelated arc setting, rather than a collisional or post-collisional setting.

In the present study, the authors tentatively propose that the Nymo leucocratic sills were formed during the early stage of the Indo-Asian collision. In the meantime, the subduction of the Neotethyan oceanic lithosphere was still underway.Thus, the igneous rocks of the Gangdese still show arcaffinity characters (Zhou LM et al., 2018). At this transitional stage, from oceanic subduction to continental collision, the upwelling of the asthenosphere mantle will favor the formation of a big magma chamber, which will trigger the HT metamorphism and partial melting of country rocks adjacent to the chamber. The formation processes of the Nymo leucocratic sill swarm could be referenced to the tentative tectonic model in the present study (Fig. 9). This scenario is corroborated by the sharp decrease of convergence rate between Indian and Eurasian plates at ca. 50 Ma (Meng J et al., 2012; van Hinsbergen DJJ et al., 2011). Furthermore, the ca. 50-40 Ma collision-related ultrahigh and high-pressure metamorphic rocks are widely distributed in the Himalayan terranes, which is the underlying plate of the Indo-Asian collisional zone. For example, the ca. 48-45 Ma mediumpressure schists in the Yardoi gneiss dome and ca. 40 Ma high-pressure metamorphism in the Tethyan Himalayan rocks,which were interpreted to have resulted from subduction of the Tethyan Himalayan continent beneath the Eurasian continent (Ding HX et al., 2016; Gao LE et al., 2012;Laskowski AK et al., 2016). The collision-related ultrahighpressure metamorphism (ca. 53.3 Ma) occurred in the eastern Ladakh region, NW Himalaya (de Sigoyer J et al., 2004;Leech ML et al., 2005). All these observations further suggest that the studied leucocratic sill swarms in the Nymo region were formed during the early stage of the Indo-Asian collision.

In addition, voluminous mafic microgranular enclaves are enclosed within the Quxu batholith, and both enclaves and host granitoid plutons have the same crystallization ages of ca. 50 Ma (Mo XX et al., 2005a; Ma XX et al., 2017b). The occurrence of these mafic microgranular enclaves could be associated with the upwelling of the asthenosphere mantle,triggered by slab breakoff of the Neotethyan oceanic lithosphere beneath the Lhasa terrane (Mo XX et al., 2005a; Ji WQ et al., 2016). Additional evidence for this scenario is provided by the ca. 45 Ma oceanic island basalt-type gabbro in the Tethyan Himalaya, southern Tibet (Ji WQ et al., 2016);ca. 43 Ma highly fractionated granite in the Gangdese belt(Wang Q et al., 2015); and ca. 41 Ma dioritic porphyrite dikes in the Gangdese belt (Ma XX et al., 2016). In addition, the ca.41 Ma dioritic porphyrite dikes contain lots of inherited zircon grains with Precambrian ages, indicating that the ancient material of Indian continental crust has been involved in the magma source of these dioritic dikes in the Gangdese belt,southern Tibet.

7. Conclusions

(i) The Nymo leucocratic sill/dike swarms were emplaced at ca. 49-47 Ma, coevally with the crystallization timing of the Nymo intrusive complex (ca. 50-47 Ma) and the metamorphic timing of the HT amphibolite (ca. 50-47 Ma).

(ii) The Nymo leucocratic sill/dike swarms could be generated by partial melting of the Jurassic Bima Formation sequence, triggered by the crustal thickening or the heating from the magma chamber of the Nymo intrusive complex.

(iii) The Early Eocene Nymo leucocratic sill/dike swarms could be employed as markers for the Indo-Asian collision.

CRediT authorship contribution statement

Xu-xuan Ma, Li-E Gao and Hai-bing Li conceived of the presented idea. Zhong-bao Zhao and Xi-jie Chen carried out the experiment and associated analysis. All authors discussed the results and contributed to the final manuscript.

Declaration of competing interest

The authors declare no conflicts of interest.

Acknowledgment

Fruitful discussions with Ling-sen Zeng shed light on the petrogenesis and tectonic implications of the leucocratic sills,which are much appreciated. The authors are much indebted to two anonymous reviewers for their constructive comments,which helped the improvement of the presentation. Thanks are due to chief editor Zi-guo Hao and guest editor Cong Zhang for their efficient handling of this manuscript. This study was co-supported by the National Key Research and Development Project “Key scientific issues of transformative technology”(2019YFA0708601), the Key Special Project for Introduced Talents Team of the Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou)(GML2019ZD0201), the second Tibetan Plateau Scientific Expedition and Research Program (STEP) Grant(2019QZKK0802), the Research Grants of Chinese Academy of Geological Sciences (J2024), the Geological Survey of China (DD20190057 and DD20190060), the National Natural Science Foundation of China (41502198) and the open fund of State Key Laboratory for Mineral Deposits Research at Nanjing University (2020-LAMD-K04).

Supplementary dataset

Supplementary data (Table S1) to this article can be found online at doi: 10.31035/cg2021019.