- Email: [email protected]
Neoproterozoic crustal evolution in Sri Lanka: Insights from petrologic, geochemical and zircon U–Pb and Lu–Hf isotopic data and implications for Gondwana assembly
Accepted Manuscript Title: Neoproterozoic crustal evolution in Sri Lanka: Insights from petrologic, geochemical and zircon U-Pb and Lu-Hf isotopic data and implications for Gondwana assembly Author: M. Santosh T. Tsunogae Sanjeewa P.K. Malaviarachchi Zeming Zhang Huixia Ding Li Tang P.L. Dharmapriya PII: DOI: Reference:
S0301-9268(14)00333-7 http://dx.doi.org/doi:10.1016/j.precamres.2014.09.017 PRECAM 4092
To appear in:
Precambrian Research
Received date: Revised date: Accepted date:
2-8-2014 13-9-2014 15-9-2014
Please cite this article as: Santosh, M., Tsunogae, T., Malaviarachchi, S.P.K., Zhang, Z., Ding, H., Tang, L., Dharmapriya, P.L.,Neoproterozoic crustal evolution in Sri Lanka: Insights from petrologic, geochemical and zircon U-Pb and Lu-Hf isotopic data and implications for Gondwana assembly, Precambrian Research (2014), http://dx.doi.org/10.1016/j.precamres.2014.09.017 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Page |1
Neoproterozoic crustal evolution in Sri Lanka: Insights from petrologic, geochemical and zircon U‐
ip t
Pb and Lu‐Hf isotopic data and implications for
cr
Gondwana assembly
us
an
M. Santosh1,2*, T. Tsunogae3,4, Sanjeewa P.K. Malaviarachchi5,6,
M
Zeming Zhang7, Huixia Ding7, Li Tang1, P.L. Dharmapriya6
1
School of Earth Science and Resources, China University of Geosciences Beijing, No. 29
d
Xueyuan Road, Haidian District, Beijing, 100083, China 2
780-8520, Japan 3
te
Division of Interdisciplinary Science, Faculty of Science, Kochi University, Akebono-cho, Kochi Faculty of Life and Environmental Sciences, University of Tsukuba, Ibaraki 305-8572, Japan
4
Ac ce p
Department of Geology, University of Johannesburg, Auckland Park 2006, South Africa
5
Department of Geology, Faculty of Science, University of Peradeniya, Peradeniya 20400, Sri Lanka 6 Postgraduate Institute of Science, University of Peradeniya, Peradeniya 20400, Sri Lanka 7
State Key Laboratory of Continental Tectonic and Dynamics, Institute of Geology, Academy of
Geological Science, No. 26 Baiwanzhuang Road, Beijing 100037, China
Corresponding author e-mail: [email protected]
Page 1 of 111
Page |2
ip t
Research Highlights
Neoproterozoic arc magmatism prior to Gondwana assembly in Sri
cr
Lanka
us
Crustal growth through both juvenile and reworked components Highland Complex is an accretionary suture
an
Double-sided subduction in Neoproterozoic and collision in
te
Abstract
d
M
Cambrian
Ac ce p
Sri Lanka, the ‘pendant’ of Gondwana, is a collage of distinct crustal blocks that preserve important records of major Neoproterozoic tectonothermal events.
Here we present the
petrology, geochemistry, zircon U-Pb geochronology and Lu-Hf isotopes on a suite of metaigneous rocks including granodiorite, diorite and garnet amphibolite from the Kadugannawa Complex (KC), granodiorite from the Wanni Complex (WC) and mafic granulites, gabbros and garnet-bearing charnockite from the Highland Complex (HC) along a NW – SE transect. The regional metamorphic peak P-T conditions were estimated from garnet-clinopyroxeneplagioclase-quartz assemblage in the metagabbro as 830-860°C and 9.4-9.8 kbar. Slightly lower temperature ranges of 700-780°C were obtained from garnet amphibolite, metagranodiorite and metadiorite, corresponding retrograde conditions. Trace element and rare earth element patterns as well as Rb-Y-Nb and Rb-Yb-Ta discrimination plots show volcanic arc affinity for the granodiorite, diorite and garnet charnockite suggesting that the protoliths of the rocks were formed from felsic to intermediate arc magmas. The mafic granulites and magnesian metagabbro also suggest volcanic arc affinity and indicate subduction-related mafic magmatism and magma underplating. The garnet-bearing metagabbro shows N-MORB signature, whereas the garnet
Page 2 of 111
Page |3 amphibolite displays oceanic island alkali basalt affinity. These rocks therefore represent accretion of the remnants of oceanic lithosphere during the subduction-collisional event. Zircons in a metadiorite and the surrounding metagranodiorite from the KC yield ages of 980 ± 16 Ma to 916 ± 57 Ma marking early Neoproterozoic magmatism followed by metamorphism at 532 ± Zircons in the garnet amphibolite from this complex show extensive metamorphic
recrystallization yielding a weighted mean
206
ip t
18 Ma.
Pb/238U age of 520.7 ± 6.6 Ma. From the WC,
zircons in a metagranodiorite define three groups of weighted mean
206
Pb/238U ages at 805 ± 12
cr
Ma (emplacement of the magmatic protolith), 734.0 ± 4.6 Ma (Cryogenian thermal event) and 546.0 ± 5.7 Ma (latest Neoproterozoic-Cambrian metamorphism). Zircons from the HC record
us
multiple late Neoproterozoic – Cambrian thermal events with weighted mean
206
Pb/238U ages of
576.8 ± 9.3 Ma and 523.0 ± 7.1 Ma (metagabbro) and 579 ± 10 Ma, 540.4 ± 6.0 Ma and 511.1 ± 5.9 Ma (garnet charnockite), 553.0 ± 3.2 Ma (mafic granulite), and 539.1 ± 4.4 Ma (mafic granulite Lu-Hf data reveal dominantly positive εHf(t) values for zircons in the metadiorite and
an
sill).
metagranodiorite from the KC (-1.1 to 7.2) and Hf crustal model ages (TDMC) in the range of 1206 to 1733 Ma suggesting a mixed source from both juvenile and Paleo-Mesoproterozoic
M
components. However, zircons in the garnet amphibolite from this complex show dominantly negative εHf(t) values (mean -16.5) with TDMC in the range of 2356 to 2828 Ma suggesting reworked Neoarchean-Paleoproterozoic crustal source. Zircons in metagranodiorite of the WC
d
also possess negative εHf(t) values (mean -6.2) with TDMC in the range of 1799 to 2498 Ma
te
suggesting reworked Paleoproterozoic crust as the magma source. The HC rocks also preserve distinct imprints of reworking of older crust. The zircon εHf(t) values in mafic granulite show a tight
Ac ce p
cluster from -2.2 to 0.1 with TDMC in the range of 1501-1651 Ma suggesting a mixed source from both juvenile Neoproterozoic and reworked Mesoproterozoic components. Zircons in the metagabbro from this complex show negative εHf(t) values (mean -6.3) and TDMC of 1847-1978 Ma. Zircons in the garnet charnockite also display highly negative εHf(t) values (mean -17.7) and older TDMC) (mean 2614 Ma) suggesting reworked Paleoproterozoic crust. Zircons in the mafic granulite sample show negative εHf(t) values (mean -14.1) and TDMC between 2263 to 2790 Ma indicating that the source material for the magma evolved from the Neoarchean-Paleoproterozoic crust. In summary, the 916 to 980 Ma ages from the KC represent arc magmatism during early Neoproterozoic, followed by the 805 Ma granodioritic magma emplacement in the WC. Repeated thermal events during mid and late Neoproterozoic are also recorded from the Wanni and Highland Complexes, culminating in Cambrian high-grade metamorphism that reached ultrahightemperature conditions. We propose a model of double-sided subduction during the Neoproterozoic, where the Wanni Complex to the west and the Vijayan Complex to the east represent continental arcs, culminating in collision along the HC during late NeoproterozoicCambrian.
Page 3 of 111
Page |4
Key words: Petrology and geochemistry; Zircon U-Pb geochronology and Lu-Hf isotopes; Arc magmatism; Neoproterozoic; Sri Lanka.
ip t
1. Introduction
cr
Crustal evolution models on both Archean and Phanerozoic Earth identify the
us
role of arc magmas derived from both juvenile and reworked older components (e.g., Condie and Kröner, 2013; Condie and Aster, 2013; Santosh et al., 2013a; In convergent margins, continental growth occurs
an
Xiao and Santosh, 2014).
through vertical addition of magmas derived by the melting of downgoing oceanic
M
slabs, as well as lateral accretion of oceanic and trench material, together with other exotic oceanic and crustal fragments.
For instance, the Central Asian
te
d
Orogenic Belt (CAOB) representing the largest Phanerozoic accretionary orogen on the globe is a typical example of continental growth through lateral accretion
Ac ce p
of young arc complexes and old micro-continents together with vertical growth through underplating of mantle-derived magmas (e.g., Xiao and Santosh, 2014, and references therein). Many of the major Precambrian collisional orogenic belts were also constructed through prolonged subduction-accretion processes, similar to the modern analogue of the ongoing subduction and accretion in the Western Pacific (Santosh et al., 2009). Accretionary orogens also characterize the margins of the major supercontinents built through the amalgamation of continental fragments (Cawood and Buchan, 2007).
The geology of Sri Lanka provides important insights into continental growth in
Page 4 of 111
Page |5 the Neoproterozoic Earth. Tectonics of Sri Lanka has also been in focus in relation to the history of assembly of supercontinents, particularly because of its central position within the India – Madagascar – Africa – East Antarctica collage
ip t
of the late Neoproterozoic Gondwana supercontinent (Yoshida et al., 1992). The Sri Lankan Precambrian basement has been subdivided into four major terrains
cr
(e.g. Cooray, 1994): the Wanni Complex (WC) to the west together with the
us
Kadugannawa Complex (KC), the Highland Complex (HC) at the middle, and the Vijayan Complex (VC) to the east (Fig. 1), largely based on Nd- model age
an
mapping and regional structural interpretations (Milisenda et al., 1988, 1994; Kröner et al., 1987, 1991; Kehelpannala, 1991, 2003, 2004). The WC is
M
considered to represent a higher crustal level than that of the HC although there
d
is no clear structural break between the rocks of the two complexes, and the
te
contact between these two has been obliterated by later events (Voll and Kleinschrodt, 1991). The boundary between the VC and the HC is well defined as
Ac ce p
a thrust/shear contact (Hatherton et al., 1975; Vitanage, 1985; Voll and Kleinschrodt, 1991; Kriegsman, 1994). Three tectonic klippens of similar petrological and structural features to the HC are exposed in the south eastern part of Sri Lanka as inliers, namely Kataragama Klippe, Buttala Klippe and Kuda Oya Klippe. Some authors (e.g. Kehelpannala, 1991, 1997) have suggested that the KC forms part of the WC based on geochronological, geochemical and structural evidence. The HC domain is interpreted to be a part of a supracrustal basin developed in a suture zone with the Lützow-Holm Complex in East Antarctica during the final phase of Gondwana assembly (Shiraishi et al. 1994).
Page 5 of 111
Page |6 The WC, KC and VC domains are considered as Grenville-aged terranes of arcrelated settings at the outer margin of the Rodinia supercontinent (Kehelpannala, 2003, 2004; Kröner et al., 2003; Willbold et al., 2004; Kröner et al., 2013).
ip t
Although Sri Lanka is a relatively small continental fragment, mimicking a pendant on the heart of Gondwana, the several distinct tectonic units in this
cr
island and their history of amalgamation offer important clues for evaluating
us
Precambrian crustal evolution history.
an
In this study, we present results from our field investigations and petrologic, geochemical, zircon U-Pb and Lu-Hf isotopic studies on a suite of
Our data reveal Neoproterozoic crustal growth through protracted
d
Complexes.
M
metamorphosed magmatic rocks from the Wanni, Kadugannawa and Highland
te
arc magmatism derived from both juvenile and reworked sources, and lateral accretionary growth by the incorporation of oceanic and continental material.
Ac ce p
The subduction-accretion process culminated in continental collision and highgrade metamorphism during late Neoproterozoic – Cambrian.
2. Geological background
2.1 Overview of geology and petrology
The HC is dominantly composed of granulitic meta-quartzites, marbles, calc-silicates and metapelitic gneisses, in association with charnockites (Cooray,
Page 6 of 111
Page |7 1962, 1984; 1994; Perera, 1984; Mathavan et al. 1999; Mathavan and Fernando, 2001; Dharmapriya et al. 2014). The metasedimentary rocks such as marble and quartzite can be traced for more than 40 km in the central and northeastern part
ip t
of the HC, whereas marble and quartzite are scarce in the southwestern part of this complex (Mathavan et al. 1999). Bands of wollastonite-scapolite, diopside-
cr
and scapolite-bearing calc-granulites and cordierite-bearing gneisses occur in the
us
south and southwestern parts of the HC (Cooray, 1962, 1984; Hapuarachchi, 1968; Perera, 1984; Prame, 1991a; Mathavan et al., 1999; Mathavan and
an
Fernando, 2001).
Metamorphism under granulite facies conditions has been well established
M
in the HC from meta-basaltic and gabbroic to intermediate rocks (Sandiford et al.,
d
1988; Schumacher et al., 1990; Schumacher and Faulhabar, 1994), charnockites
te
(Prame, 1991b) and metamorphosed pelitic rocks (Perera, 1984, 1987; Prame, 1991a, Hiroi et al., 1994, Raase and Schenk, 1994). A P-T zonation across the
Ac ce p
HC has been established where pressures and temperatures decrease from 9-10 kbar and 830 °C in the east and southeast to 5-6 kbar and 700 °C in the northwest (Faulhaber and Raith, 1991; Schumacher and Faulhaber, 1994). Sajeev and Osanai (2005) identified a thermal gradient of 591 - 996 °C across south western to the central region of the HC. A clockwise P-T path is inferred for the pelitic rocks based on the sequence of kyanite and staurolite (occurring as inclusions in garnet) followed by sillimanite and andalusite (Perera, 1987; Hiroi et al., 1994, Raase and Schenk, 1994; Mathavan et al., 1999, Kriegsman and Schumacher, 1999). In contrast, reaction textures involving pyroxenes,
Page 7 of 111
Page |8 plagioclase, garnet and quartz in some meta-igneous rocks (Perera, 1987; Schumacher et al., 1990; Prame, 1991b) and high temperatures (>900 °C) from pyroxene exsolution textures (Schenk et al., 1988) have been used to suggest
ip t
isobaric cooling prior to the uplift. Although isobaric cooling is apparently not recorded in pelitic rocks of the HC apart from the report of Karunaratne et al.
cr
(2002), Dharmapriya et al. (2014) showed a rare evidence from a khondalite as
us
evidence for isobaric cooling from T of ~900 down to 770 °C at P of 7.5 kbar before uplift. Voll et al. (1994) derived peak temperatures of metamorphism at
an
850-900 °C using two-feldspar thermometry. Ultrahigh temperature (UHT) granulites have been reported in few localities in the central HC and rarely from
M
the southwestern part of the HC. These rocks witnessed extreme crustal
d
metamorphism at temperature of 925 °C to 1150 °C and pressures of 9-12.5
te
kbar (e.g. Kriegsman and Schumacher, 1999; Osanai et al., 2006; Sajeev and
Ac ce p
Osanai, 2004a, 2004b; Sajeev et al., 2007).
The WC occurs to the west-northwest of the HC and is dominantly
composed of upper amphibolite to granulite facies orthogneisses and minor meta-sediments.
The meta-igneous rocks show varied protolith-chemistry
ranging from granitic, granodioritic, monzonitic, tonalitic, charnockitic and enderbitic composition (e.g. Pohl and Emmermann, 1991). Garnet sillimanite gneisses, cordierite gneisses, quartzites and calc-silicate rocks occur as minor bands, particularly close to the inferred boundary with the HC. Metacarbonates are absent in the WC although apatite bearing unmetamorphosed carbonatite
Page 8 of 111
Page |9 body is exposed in the northern WC. The western part of the WC is mainly composed of less-deformed granites, including the unmetamorphosed posttectonic K-feldspar rich granite at Tonigala (e.g. Hölzl et al., 1991, Cooray, 1994).
ip t
Migmatization is also widespread within the unit. The peak metamorphic conditions of the WC are estimated to be 700-830 °C and 5-7 kbar (Schenk et al.,
cr
1991; Raase and Schenk, 1994). Classic outcrops of incipient charnockites
us
represent granulite formation within the host amphibolite facies gneisses in the central regions of the WC. In these zones, well-foliated light gray hornblende-
an
gneisses show veins and patches of dark, greenish-brown and coarse-grained incipient charnockite formed along mesoscopic shear zones and foliation planes,
M
representing pervasive fluid influx in the Sri Lankan lower crust (e.g. Hanson et
d
al., 1987; Burton and O’Nions, 1990; Baur et al., 1991; Milisenda et al., 1991;
te
Santosh et al., 1991; Kehelpannala, 1999; Perchuk et al., 2000).
Ac ce p
The KC, often referred as ‘Arenas’ (Vitanage, 1972; Almond, 1991), a
synonym used in Sri Lanka for doubly plunging synforms, is exposed in the northwestern part of Kandy. The complex includes upper amphibolite to granulite facies basement rocks (Kröner et al., 1991; Cooray, 1994; Malaviarachchi and Takasu, 2005, 2011a). Perera (1983) argued a sedimentary origin of the wellbanded hornblende biotite gneisses in the KC, whereas Cooray (1984) regarded their possible protoliths as basaltic lavas or volcanic ash. Relics of primary magmatic layering are preserved in these synforms mainly composed of hornblende gneisses, amphibolites and minor pyroxenites (Kröner et al., 1991).
Page 9 of 111
P a g e | 10 Hornblende- and biotite-bearing orthogneissic rocks with gabbroic, dioritic and trondhjemitic rocks of calc-alkaline affinity (Pohl and Emmermann, 1991) are interlayered within granodioritic to granitic gneisses, charnockites, enderbites and
ip t
minor metasediments (Kröner et al., 2003). Minor layered mafic and ultramafic rocks have also been reported (Stosch, 1991). The KC is considered by some
cr
authors as a migmatized and metamorphosed equivalent of a layered mafic-
us
ultramafic complex, named the ‘Kandy Layered Intrusion’ (KLI: Vol and Kleinschrodt, 1991; Stosch, 1991; Kleinschrodt et al., 1991).
an
The VC exposed in eastern Sri Lanka consists predominantly of metaigneous rocks including upper amphibolite facies calc-alkaline suite of granitoid
M
gneisses, augen-gneisses and minor amphibolites possibly derived from mafic
d
dykes. This granitoid suite was interpreted as a product of subduction-related
te
magmatism (Milisenda, 1991; Pohl and Emmermann, 1991). However, in contrary to previous interpretations, Kröner et al. (2013) reported that the Vijayan
Ac ce p
gneisses show evidence for granulite-facies metamorphism. Minor sedimentary enclaves such as metaquartzite, calc-silicate rocks and marble also occur within the VC (Dahanayake, 1982; Dahanayake and Jayasena, 1983; Kröner et al., 1987,
1991).
The
granitoid
gneisses
show
TTG
(tonalite-trondhjemite-
granodiorite)-dominated signature (Milisenda, 1991; Pohl and Emmermann, 1991) whereas the hornblende gneisses display compositions between diorite and leucogranite with the dominant composition being granodiorite and granite (Kröner et al., 1991; Milisenda, 1991; Kleinschrodt, 1994; Milisenda et al., 1994). Minor metagabbroic rocks and mafic dykes occur interlayered and infolded with
Page 10 of 111
P a g e | 11 the granitoid gneisses. Many exposures of the Vijayan gneisses show evidence of migmatization. Prame (1997) noted that many of the granitic rocks in the VC have A-type characteristics and inferred their formation through melting of felsic
ip t
crust. Igneous charnockite bodies occurring at the SE margin of the VC are interpreted as tectonically infolded or inter-sliced fragments of rocks of the HC
cr
similar to those of the HC-klippens present in the VC, during the thrusting of the
us
HC over the VC (Kröner et. al. 1991).
an
2.2 Summary of previous geochronological studies and tectonic models
d
M
2.2.1 Geochronology
te
The Nd model ages of 3.0-2.2 Ga reported from the HC (Milisenda et al., 1988, 1994) suggest that the rocks in this complex were derived from late
Ac ce p
Archean sources. Hölzl et al. (1991, 1994) and Köhler et al. (1991) presented Rb-Sr whole rock, Rb-Sr biotite and Sm-Nd garnet indicating ages of ~2 Ga. Cordani and Cooray (1990) reported Rb-Sr isochron age of 1100 Ma and De Maesschalck et al. (1990) interpreted a Rb-Sr whole rock isochron age of 1930 Ma as the timing of high-grade metamorphism. Rb-Sr and Sm-Nd isochron ages of 2.6-2.3 Ga and 500-450 Ma were also reported from the HC (Kagami et al. 1990). Garnet-whole rock Sm-Nd ages for widely separated HC paragneisses and metabasites show metamorphic ages of ca. 600 Ma (Hölzl et al., 1991; Milisenda, 1991). Ion microprobe U-Pb study on zircons by Kröner et al. (1987)
Page 11 of 111
P a g e | 12 documented Archean to early Proterozoic ages (3.2-2.4 Ga) from detrital zircons in the HC metasediments. They also reported an intrusion age of 1.1 Ga and new zircon growth at about 550 Ma from meta-igneous rocks. Köhler et al. (1991)
ip t
reported U-Pb zircon and monazite data indicating depositional ages of ~2 Ga and ~550 Ma ages for the high-grade metamorphism. Baur et al. (1991) reported
cr
U-Pb zircon crystallization ages of 1940 Ma and a Pb loss event at 560-550 Ma
us
indicating high-grade metamorphism. Kröner and Williams (1993) reported SHRIMP zircon ages for granitic gneiss from the HC as 1.9 Ga for the
an
crystallization of the igneous protolith and 531 Ma as the age of metamorphism. The lower intercept ages in concordia plots fall between 550-610 Ma for
M
discordant single zircon and zircon fractions in the granulites from the HC, whereas multigrain fractions of metamorphic zircon, monazite and rutile gave Pb/206Pb ages of 541-608 Ma (Baur et al., 1991, Hölzl et
d
207
te
concordant U-Pb and
al., 1991, 1994; Kröner et al., 1994). Burton and O’Nions (1990) analysed U and
Ac ce p
Pb isotopes in ilmenite giving an isochron age of 1.1 Ga which they interpreted as the metamorphic age. Hölzl et al. (1994) used U-Pb method of zircon and monazite from the same rock and reported ages of 555-592 Ma. In a recent study, Malaviarachchi and Takasu (2011b) obtained ca. 728-460 Ma ages for metamorphism using electron microprobe monazite CHIME method. Sajeev et al. (2003) obtained the Sm-Nd isochron whole-rock age of ca. 1.5 Ga from the UHT granulites of the HC and inferred that the granulites could be relics of a pre-Pan African metamorphism. In another study, Sajeev et al. (2007) reported U-Pb zircon metamorphic age of ca. 580 Ma from relatively
Page 12 of 111
P a g e | 13 higher pressure mafic granulites which is in agreement with zircon and monazite ages of ca. 570 Ma, determined from sapphirine granulites of the central HC
HC with the assembly of the Gondwana supercontinent.
ip t
(Sajeev et al. 2010), correlating the ultrahigh-temperature metamorphism in the
From the WC, Nd-modal ages of ca. 700 Ma -1.1 Ga have been reported,
cr
with multiple magmatic events as documented by zircon ages of 1100-530 Ma
us
(Burton and O’Nions, 1990; Kröner et al., 1991, 2003). Massive charnockites from the northern part of the WC have also older protolith emplacement ages as
an
from zircon evaporation (Kröner et al. 1994). Deposition of detrital zircons in metapelites and the emplacement of magmatic zircons of charnockitic and
M
enderbitic rocks yield ion microprobe U-Pb concordia upper intercept age of ~1.1
d
Ga considered as the timing of emplacement of the precursor magmas. The
te
lower intercept ages range between 540-590 Ma correlate with the high-grade metamorphism (e.g. Hölzl et al., 1991). Typical arrested (in-situ) charnockites and
Ac ce p
its host granitic/tonalitic gneiss from the Udadigana quarry in Kurunegala show ages on a common regression line with concordia intercepts at 771 Ma and 563 Ma, considered as the intrusion age and the metamorphic age respectively (Baur et al. 1991). Burton and O’Nions (1990) dated zircons from charnockitized patches and inferred that the in-situ charnockitization in the WC has occurred at 535 Ma. Thus, the arrested or in-situ charnockitization of the WC is considered to post-date the peak granulite facies metamorphism, and principally controlled by fluid influx along shear zones (Kehelpannala, 1999). Nd-model ages of the rocks from the KC range between ca. 900 Ma and
Page 13 of 111
P a g e | 14 1.6 Ga (Milisenda et al., 1988, 1994). Pb-Pb zircon evaporation ages from the KC show ca. 770 Ma -1.1 Ga indicating multiple calc-alkaline magmatic events covering a period of approximately 300 million years (Kröner et al., 2003; Willbold
ip t
et al., 2004). Most of the dioritic to granodioritic rocks of the KC display ca. 900 Ma ages. Xenomorphic zircons of gabbroic gneisses show relatively younger
cr
ages of ~700 Ma whereas idiomorphic zircons from the same sample gave ages
us
of ca. 900 Ma. Perera and Kagami (2001) argued that the regional high grade
isotope systematics on charnockitic rocks.
an
metamorphism of the KC must be older than 1100 Ma based on a study of Nd-Sr
Whole-rock Nd isotope analyses from widely separated localities show
M
mean crustal residence ages of 1.0–1.8 Ga for the VC rocks. Sr isotope data are
d
in line with the Nd isotope results suggesting that the majority of the VC rocks
te
were juvenile and derived from a typical LREE-depleted, mantle-derived precursor and not from anatexis of older crustal lithologies (Milisenda et al.,
Ac ce p
1988, 1994; Kröner et al., 1991). Rb–Sr whole-rock isochron ages of ca. 800 Ma reported by De Maesschalck et al. (1990) were ascribed to metamorphic resetting whereas the magmatic zircons reflect emplacement of the protoliths of the Vijayan gneisses at 1000–1100 Ma (Kröner et al., 2003, 2013). Zircon U-Pb ages of ca. 591- 456 Ma (Hölzl et al., 1991; Kröner et al., 1987, 1991, 2013) are now regarded as the age of metamorphism of the VC.
2.2.2. Structural and tectonic models
Page 14 of 111
P a g e | 15 The HC granulites have been subjected to a major folding and thrusting event (e.g. Berger and Jayasinghe, 1976; Kriegsman, 1994; Kleinschrodt, 1994). The large folds with amplitudes of about 7-10 km and an exposed length parallel
ip t
to their axes of up to 50 km in the central Highlands formed after the peak of metamorphism (800-850 °C, 8-9 kbar) at slightly lower temperature (700-750 °C)
cr
(Kleinschrodt, 1994). These large upright folds were generated during the last
us
deformation event following high-grade metamorphism (e.g. Berger and Jayasinghe, 1976; Kriegsman, 1994). During this event, horizontal shorting and
an
vertical thickening of crust took place with unroofing of the crustal rocks into upper level (Kleinschrodt, 1994; Kriegsman, 1994). Hiroi et al. (2014) reported
M
evidence for fast exhumation of lower-crustal rocks to andalusite-stable upper-
d
crustal conditions by channel flow in a continental collision orogeny, based on
of the HC.
te
felsic/granitic inclusions within high grade ordinary granulites and UHT granulites
Ac ce p
Transposition of structures in the WC is less severe, compared to those in
the HC. In ductile shear zones, gneissic foliations are severely flattened resulting in the obliteration of the previous intrusive and deformation histories. However, in some regions, multiple phases of gabbroic to granitoid intrusions and their deformation are preserved in multiple foliations (Kriegsman, 1991; Kröner et al., 1994). Macroscopic folds of the open to tight type with gently plunging hinges are well developed in the WC similar to those in the HC and their axial traces continue for several kilometers (Tani and Yoshida, 1996). The contact between the WC and the HC is still controversial due to absence of any structural
Page 15 of 111
P a g e | 16 discontinuity in between. The combined HC – WC unit is considered to represent a tilted section of the former lower–middle crust, with the WC representing the
ip t
higher level.
The entire rock suite of the KC is folded into large-scale doubly plunging
cr
five NW-trending synforms (Arenas). These macroscopic plunging folds with
us
upright axial surfaces were generated after the formation of the major compositional layering. The interior domains of the KC preserve prograde
an
amphibolite facies conditions compared to the surrounding HC (Perera, 1983; Sandiford et al., 1988; Voll and Kleinschrodt, 1991). Hence, the KC is considered
M
to be at a higher crustal level than the HC (Sandiford et al., 1988; Kriegsman,
d
1994) whereas Kehelpannala (1997) interpreted that the KC represents a deeper
te
crustal level compared to the WC. Although the boundary between the HC and the KC is well defined as a structural break, that between the WC and the KC is
Ac ce p
not clear e.g. Kröner et al., 2003).
Kehelpannala (2003) noted that the homogeneous to compositionally well-
layered gneisses experienced strong ductile deformation during more than one phase of folding. The boundary between the VC and the HC is considered to be a sub-horizontal ductile shear zone/thrust (Hatherton et al., 1975; Vitanage, 1985; Voll and Kleinschrodt, 1991; Kriegsman, 1995), formed when the HC collided with the VC and produced late upright folds in the former and ductile deformation in the latter (Kehelpannala, 1997, 2003, 2004). Shear sense indicators and sheath folds found at the suture between the VC and the HC show
Page 16 of 111
P a g e | 17 that the HC, namely the hanging wall, apparently moved from N to S over the VC (Kehelpannala,
2003)
accompanying
widespread
mylonitization
and
ip t
migmatization along the shear zones (Kleinschrodt, 1994).
cr
2.3 Present study area and description of sampling locations
us
The present study focuses on petrology, geochemistry and geochronology of meta-igneous rocks collected from a traverse running from the WC through the
an
KC into the HC in a direction of increasing elevation (Fig. 1). Fresh samples were chosen from quarries and road-cut outcrops along this traverse. The NW-SE
M
sampling in our study is oblique to the general trend of foliation so that a
d
meaningful stratigraphic section is traversed.
te
In this paper, we present analysis of three samples from the KC, four samples from the HC, and one sample from the WC (Fig. 2). A summary of the
Ac ce p
location and salient features of the samples analyzed for geochemistry and geochronology are given in Table 1. Samples of the KC represent the Dumbara Synform (KDG-1), Huluganga Synform (KDG-2) and Metibokka Synform (KDG4). The HC samples (HC-1, HC-7, HC-8, HC-10 and HC-11) were selected from the areas which represent strong ductile deformation and relatively higher grade metamorphism in the HC. The sample of the WC (WC-2) comes from a classical outcrop of in-situ/incipient charnockitization at Udadigana in Kurunegala. Representative field photographs are shown in Figures 3 and 4.
Page 17 of 111
P a g e | 18 2.3.1 Kadugannawa Complex
ip t
Metagranodiorite (KDG-4) and metadiorite (KDG-1)
The metagranodiorite sample KDG-4 was collected from a quarry at Maussawa
cr
along Harankahawa Road in Galagedara. The quarry exposes evidence for mafic
us
magma intrusion into a felsic magma chamber and subsequent mixing, mingling and fractionation. The presence of mafic-felsic layers in the outcrop may indicate
an
lower crustal melting in a granodiorite-granite magma chamber where the restite portion is represented by the hornblendite-rich mafic layers. The felsic magma
M
chamber was intruded by mafic/intermediate (dioritic) magma as dykes/sills
d
resulting in magma mixing and mingling as evidenced from the mafic magmatic
te
enclaves (MMEs) of various sizes within the granodiorite. The rocks display intense deformation as seen by refolded isoclinal folds and eye-like sheath or
Ac ce p
doubly plunging folds while preserving the magmatic history.
The metadiorite KDG-1 was collected from a road side exposure at Kundasale (about 5 km from Thennekumbura) along the Kandy-Mahiyangana road. The outcrop lies at the lower south-western part of the doubly plunging Dumbara synform of the KC. The rock shows strongly flattened and stretched compositional layers, and bedding parallel with the surrounding TTG gneisses that show compositional layering. Cross cutting dykes of aplitic and granitic composition are also common in the highly deformed parts of the metadiorite.
Page 18 of 111
P a g e | 19
Garnet amphibolite (KDG-2)
ip t
This sample was collected from a newly excavated road cut at the Mahaberiyatenna Airport site. The rock occurs as a several tens of meter thick-
cr
layer within migmatitic hornblende-biotite gneiss in the south western margin of
us
the doubly plunging Huluganga synform of the KC. Coarse-grained calcic amphibole and porphyroblastic red-eye garnets are visible in the outcrop, where garnets
show
breakdown
textures
forming
an
some
plagioclase-bearing
symplectites. Hornblende-rich cumulate layers also occur within the magmatic
d
M
layers associated with the amphibolite.
te
2.3.2. Wanni Complex
Ac ce p
Metagranodiorite (WC-2)
The sample was collected from an abandoned quarry at Udadigana in Kurunegala, tectonically located on the western limb of the Kurunegala Synform of the WC. The quarry exposes excellent evidence for in-situ/incipient charnockitization with greenish patches and veins of orthopyroxene-bearing anhydrous zones overprinting the host gneiss. The sample collected for this study is from the tonalitic to granodioritic host rock of the incipient charnockite. A gradational transition of the host gneiss to charnockite is evidenced at the quarry
Page 19 of 111
P a g e | 20 where some parts are totally converted into a coarse-grained charnockite, suggesting three dimensional networks of fluid- filled veins or tubes.
ip t
2.3.3. Highland Complex
us
cr
Mafic granulite (HC-1, HC-11)
Mafic granulite bands, boudins and enclaves of variable thickness (0.1 to 3m)
an
occur within marbles at the Digana quarry. Our sample HC-1 is from one of these bands. The marble is composed of dolomitic and calc-silicate layers parallel to
M
the regional compositional layering of the HC. Mafic granulite enclaves are
d
located parallel to bedding, and are fragmented with angular edges. They show
te
strong deformation including evidence for flattening and stretching of minerals like pyroxene and hornblende. In some parts of the quarry, the mafic granulite
Ac ce p
shows boudinaged features.
The sample HC 11 is from a mafic sill along a roadside exposure ~2 km from Kivullinda Junction towards Randenigala, along the Kandy-Randenigala road. This location is close to the hinge on the west limb of the Digana Antiform of the HC. Several metadioritic sills ranging in thickness from few tens of cm to several meters occur within alternating layers (5 to 40 cm thickness) of quartzite and psammitic gneiss, parallel to bedding with sharp contacts. Some of the mafic sills contain garnet.
Page 20 of 111
P a g e | 21
Metagabbro (HC-7 and HC-8)
ip t
The samples of metagabbro were collected from road side outcrops along the Kandy-Hantana Road. These rocks occur as discontinuous lenses (few tens of
cr
metres thick) within quartzite layers of approx. 2 km thickness and biotite gneiss
us
(partly charnockitic at places), close to the hinge on the western limb of the Kandy Antiform of the HC. Reddish porphyroblastic garnets are extensively
an
distributed in the rock, partly reflecting a cumulate-textured mafic granulite. Minor quartz-bearing domains (sample HC-8) occur within the metagabbroic host rock
te
Garnet charnockite (HC-10)
d
M
(HC-7).
Ac ce p
The garnet- and orthopyroxene- bearing charnockite sample was collected from a road outcrop near Hakurutale, between 12 and 13 km along the KandyRandenigala Road. The 0.5 – 1 km thick layer is interbanded between a granitic gneiss and a biotite gneiss layer. In some domains, coarse deep red almandinerich garnets occur and some of the garnet grains show plagioclase-bearing symplectites.
Symplectitic
garnets
and
coarse
grained
dark
brownish
hypersthene grains occur along the foliation plane. The rock is generally medium to coarse grained showing brownish to gray-green greasy charnockitic luster.
Page 21 of 111
P a g e | 22 3. Analytical techniques
ip t
3.1 Petrography and mineral chemistry
Polished thin sections were prepared for petrographic study at the State Key
cr
Laboratory of Geological Processes of Mineral Resources of China University of
us
Geosciences Beijing, and at the University of Tsukuba, Japan. Mineral chemical analyses were carried out using an electron microprobe analyzer (JEOL
an
JXA8530F) at the Chemical Analysis Division of the Research Facility Center for Science and Technology, the University of Tsukuba. The analyses were
M
performed under conditions of 15 kV accelerating voltage and 10 nA sample
te
supplied by JEOL.
d
current, and the data were regressed using an oxide-ZAF correction program
Ac ce p
3.2 Whole-rock geochemistry
Representative samples were chosen for whole-rock geochemical analyses from the WC, HC, and KC which are devoid of surface alteration or weathering. The size of the samples were initially reduced in a jaw crusher, and then manually fine-powdered in an agate mortar. Major oxides were analyzed by Thermo JarrellAsh ENVIRO II ICP, and minor and trace elements by Perkin Elmer Sciex ELAN 9000 ICP/MS at Actlabs, Canada, using lithium metaborate/tetraborate fusion technique.
Page 22 of 111
P a g e | 23
3.3 Zircon U-Pb and Lu-Hf isotopes
ip t
Zircon grains were separated by gravimetric and magnetic separation from crushed rock samples, and then purified by hand picking under a binocular
cr
microscope at the Yu’neng Geological and Mineral Separation Survey Centre,
us
Langfang City, Hebei Province, China. The grains were mounted in epoxy resin discs and polished to reveal mid-sections, followed by gold sputter coating.
an
Zircons were imaged under both transmitted and reflected light, and were imaged using cathodoluminescence (CL) to identify internal structures and choose
M
potential target sites for U-Pb analyses. The CL imaging was carried out at the Beijing Geoanalysis Centre. Individual grains were mounted along with the Pb/238U age of 417 Ma (Black et al., 2003), onto
d
206
te
standard TEMORA 1, with
double-sided adhesive tape and enclosed in epoxy resin discs. The discs were
Ac ce p
polished to a mid depth and gold coated for cathodoluminescence (CL) imaging and U-Pb isotope analysis. Zircon morphology, inner structure and texture were examined by using a JSM-6510 Scanning Electron Microscope (SEM) equipped with a backscatter probe and Chroma CL probe. The zircon grains were also examined under transmitted and reflected light images using a petrological microscope.
Zircon U-Pb dating and trace element analysis were simultaneously conducted by LA-ICP-MS at the Stage Key Laboratory of Geological Processes
Page 23 of 111
P a g e | 24 and Mineral Resources, China University of Geosciences. Detailed operating conditions for the laser ablation system and the ICP-MS instrument are as described by Liu et al. (2008, 2010). Laser sampling was conducted using an
ip t
Excimer 193 nm GeoLas 2005 System with a spot size of 32 μm. An Agilent 7500a ICP-MS instrument was used to acquire ion-signal intensities. Nitrogen
cr
was introduced into the central gas flow (ArtHe) of the Ar plasma in the LA-ICP-
us
MS analysis, which increases the sensitivity for most elements by a factor of ~2 compared with the results without adding nitrogen (Hu et al., 2008). To keep time-
an
dependent elemental fractionation at a low level, a laser frequency of 4 Hz and a laser energy of 60 mJ were applied. The uncertainty of analysis is within ~1% for
M
the zircon standard.
d
Zircon 91500 was used as external standard for U-Pb dating, and was
te
analyzed twice every five analyses. Time-dependent drifts of U-Th-Pb isotopic ratios were corrected using a linear interpolation (with time) for every five
Ac ce p
analyses according to the variations of 91500 (i.e. two zircon 91500 + five samples + two zircon 91500). Preferred U-Th-Pb isotopic ratios used for 91500 are from Wiedenbeck et al. (1995). Uncertainty in the preferred values for the external standard 91500 was propagated to the ultimate results for the samples. Concordia diagrams and weighted mean calculations were made using Isoplot/Ex_ver3 (Ludwig, 2003). Trace element compositions of zircons were calibrated against multiple-reference materials (BCR-2G and BIR-1G) combined with internal standardization (Liu et al., 2010). The preferred values of element concentrations for the USGS reference glasses used are from the GeoReM
Page 24 of 111
P a g e | 25 database (http://georem.mpch-mainz.gwdg.de/). In situ Hf isotope compositions of zircon were determined using a Neptune Plus multicollector (MC)-ICP-MS system (Thermo Fisher Scientific, Germany)
ip t
and a Geolas 2005 Excimer ArF laser ablation system (Lambda Physik, Götingen, Germany) at the State Key Laboratory of Geological Processes and
cr
Mineral Resources, China University of Geosciences, Wuhan. The energy
us
density of laser ablation used was 5.3 J cm-2. Helium was used as the carrier gas in the ablation cell and was merged with argon (makeup gas) after the ablation
an
cell. All data for zircon in this study were acquired in the single-spot ablation mode at a spot size of 32 mm. Each measurement consisted of 20 s of
M
acquisition of the background signal followed by 50 s of ablation signal
d
acquisition. The operating conditions for the laser ablation system and the MC-
te
ICP-MS instrument, and the analytical method, are the same as those described
Ac ce p
in detail by Hu et al. (2012).
4. Results
4. 1 Petrology and mineral chemistry
4.1.1 Petrography
A summary of the petrographic features of each of the rock types analyzed in this study is given below. Representative photomicrographs are shown in Fig. 5.
Page 25 of 111
P a g e | 26
Metagranodiorite and metadiorite
ip t
The metagranodiorite and metadiorite in our study are generally coarse-grained with quartz, plagioclase, biotite and calcic amphibole (hornblende) as the
cr
dominant minerals. Sample WC-2, a typical metagranodiorite, is composed of
us
plagioclase (35-45%), K-feldspar (20-30%), quartz (20-25%), biotite (5-10%), and hornblende (2-5%), with accessory Fe-Ti oxides (1-2%), zircon, and apatite (Fig.
an
5a). Plagioclase and K-feldspar are coarse-grained (grain size range of 0.3-1.2 mm and 0.4-1.3 mm, respectively) showing granoblastic texture. Fine, rounded
M
grains of quartz (0.2-0.7 mm) are present along their grain boundaries, although
d
elongated coarse-grained (~4.2 mm) quartz also occurs aligned along the rock
te
foliation. Myrmekite textures can be commonly observed in some of the samples.
Ac ce p
The metagranodiorite (sample KDG-4) comprises plagioclase (30-40%), Kfeldspar (15-25%), hornblende (15-25%), biotite (10-15%), quartz (10-15%), and magnetite (1-2%). Weak foliation of the rock is defined by aligned flakes of hornblende (0.2-1.9 mm) and biotite (0.2-1.8 mm). Plagioclase and K-feldspar are coarse-grained (0.2-1.2 mm), sub-equigranular, and granoblastic (Fig. 5b). Quartz is relatively fine grained (0.2-0.8 mm) compared to that of the sample WC2.
The sample KDG-1 is an orthopyroxene-bearing and K-feldspar-free metadiorite
Page 26 of 111
P a g e | 27 showing textures similar to those of the sample KDG-4. The mineral assemblage is plagioclase (50-60%), hornblende (15-25%), biotite (10-15%), quartz (5-10%), and ilmenite (2-3%), with accessory orthopyroxene, pyrite, apatite, and zircon
ip t
(Fig. 5c). The rock does not have a distinct foliation, and is characterized by granoblastic and semi-equigranular texture of plagioclase (0.6-2.2 mm) and
cr
quartz (0.2-0.8 mm). Both hornblende (0.5-1.5 mm) and biotite (0.2-0.8 mm) are
us
subhedral and randomly oriented along the grain boundaries of plagioclase and quartz. Orthopyroxene is also fine-grained (0.2-0.4 mm) and intergrown with
an
biotite or hornblende.
d
M
Garnet amphibolite
te
Sample KDG-2 is composed of very coarse-grained (~9 mm) poikiloblastic garnet (5-15%) in the matrix of aggregates of hornblende (20-30%) and plagioclase (30-
Ac ce p
40%) (Fig. 5d). Foliation is not obvious in thin section. Accessory minerals are ilmenite-magnetite intergrowth (2-5%), apatite (2-5%) and zircon, and quartz is present only as rare inclusions in garnet. The garnet contains numerous inclusions of ilmenite, apatite, hornblende, and quartz. Hornblende (0.3-2.4 mm) and plagioclase (0.5-1.7 mm) are coarse grained, but is fine-grained compared to the garnet. The garnet is surrounded by medium-grained intergrowth of hornblende + plagioclase, suggesting the progress of the following CFMASH continuous reaction (1); Grt + Qtz + H2O => Hbl + Pl
(1)
Page 27 of 111
P a g e | 28
ip t
Mafic granulite
The sample HC-1 is a medium-grained garnet-free mafic granulite that is
cr
composed of plagioclase (30-40%), clinopyroxene (15-25%), hornblende (15-
us
25%), orthopyroxene (10-20%), ilmenite (2-5%) and magnetite (<1%) (Fig. 5e). Calcite occurs as an accessory mineral, in accordance with the field occurrence
an
of the rock as enclave within marbles. The rock shows a weak foliation defined by aligned aggregates of ferromagnesian minerals. All the dominant minerals are
d
M
medium to fine grained (0.2-0.8 mm).
te
Sample HC-11 is also a garnet-free mafic granulite without any marked foliation and composed of plagioclase (30-40%), clinopyroxene (20-30%), orthopyroxene
Ac ce p
(10-20%), hornblende (10-20%), and accessory ilmenite and zircon (Fig. 5f) in an equigranular matrix (0.2-0.8 mm). Fine-grained aggregates of orthopyroxene and plagioclase occur as grayish spots of about 4 mm in diameter, typically associated with quartz (Fig. 5g), suggesting the progress of the following decomposition reaction of garnet: Grt + Qtz => Opx + Pl
(2)
This reaction involved the near total consumption of garnet possibly associated with post-peak decompression (e.g., Harley, 1989).
Page 28 of 111
P a g e | 29 Metagabbro
The sample HC-7 is a coarse-grained garnet-bearing metagabbro with variable
ip t
quartz-bearing and absent domains. The rock is composed of clinopyroxene (4050%), garnet (30-40%), orthopyroxene (5-10%), and plagioclase (5-10%), with
cr
accessory Fe-Ti oxide (2-5%) and hornblende (1-2%) (Fig. 5h). No obvious
us
foliation can be seen. Garnet and clinopyroxene are coarse grained (~1.7 mm) and semi-equigranular. Garnet is subhedral, and almost free from inclusions
an
except for minor apatite, hornblende, ilmenite, and quartz. Clinopyroxene is also subhedral, and contains thin exsolution lamellae of orthopyroxene. Plagioclase
M
(0.2-0.7 mm) is anhedral and fills the matrix of garnet and pyroxenes. Fine-
d
grained and rounded Fe-Ti oxide (intergrowth of ilmenite and magnetite) grains
te
are scattered throughout the thin section. These textures are similar to metagabbros reported from the Palghat-Cauvery Suture Zone in Southern India
Ac ce p
(e.g., Sajeev et al., 2009; Santosh et al., 2010; Saitoh et al., 2011; Koizumi et al., 2014).
Sample HC-8 is a quartz-rich domain of sample HC-7. It comprises plagioclase (30-40%), clinopyroxene (30-40%), garnet (10-20%), quartz (5-10%), and orthopyroxene (2-5%) with accessory ilmenite, magnetite, rutile, and zircon. Foliation is defined by elongated quartz grains (~4 mm) and garnet aggregates. Garnet is coarse-grained (0.3-3.3 mm), subhedral to anhedral, and often surrounded by fine-grained (~0.2 mm) symplectitic intergrowth of vermicular
Page 29 of 111
P a g e | 30 orthopyroxene and plagioclase or thin corona of orthopyroxene + plagioclase (Fig. 5i). Such intergrowths can be readily observed in quartz-rich portion of the rock, confirming the progress of reaction (2) probably related to rapid
ip t
decompression. Clinopyroxene (0.2-1.1 mm) is subhedral, and often occur as aggregates elongated along the foliation particularly at plagioclase-rich portion of
cr
the rock. Ilmenite occurs as fine-grained (<0.3 mm) rounded to sub-rounded
us
grains, or irregular mineral intergrowing with clinopyroxene.
an
Garnet charnockite
M
Sample HC-10 is characterized by coarse-grained (~2.5 mm) porphyroblastic
d
garnet in hand specimen (Fig. 3), although the modal abundance varies from 0 to
te
5 %. The examined sample corresponds to a garnet-absent portion of the rock, and comprises plagioclase (30-40%), quartz (10-20%), clinopyroxene (10-15%),
Ac ce p
orthopyroxene (5-10%), hornblende (5-10%), with accessory ilmenite, pyrite, calcite, zircon, and retrograde biotite (Fig. 5j). K-feldspar is absent in the rock, which is consistent with its dioritic composition as discussed later. The rock is also characterized by ribbon quartz elongated parallel to the rock foliation defined by thin alternation of pyroxene-rich and plagioclase-rich layers. Plagioclase is coarse-grained (0.7-2.5 mm), granoblastic, and semi-equigranular in plagioclaserich layers, while it is fine-grained (0.2-0.7 mm), subhedral, and locally intergrowing with fine-grained orthopyroxene (~0.2 mm) in pyroxene-rich layers as a product of reaction (2) (Fig. 5j). The lack of garnet in the sample probably
Page 30 of 111
P a g e | 31 implies complete consumption of the mineral during decompression. Mediumgrained (0.2-0.8 mm) and subhedral ortho- and clinopyroxenes, and fine-grained (0.1-0.5 mm) hornblende are dominant minerals in the pyroxene-rich layers,
ip t
whereas their grain size decreases in plagioclase-rich layers. Individual pyroxene and plagioclase grains show no obvious alignment, whereas hornblende is
cr
weakly oriented along the rock foliation. Minor biotite mantles pyroxene and
an
us
hornblende grains as a product of post-peak retrograde hydration.
M
4.1. 2 Mineral chemistry
d
Representative compositions of minerals from electron microprobe analyses are
Ac ce p
below.
te
given in Supplementary Tables 2 to 4, plotted in Figure 6, and briefly discussed
Garnet
Garnet in the metagabbros (HC-7, HC-8) and garnet amphibolite (KDG-2) is essentially almandine-rich with minor pyrope, grossular, and spessartine contents whose abundance varies slightly depending on the assemblage and core-rim positions (Supplementary Table 2, Figs. 6a,b). Garnet in sample HC-7 shows the highest XMg (= Mg/(Mg+Fe)) content of 0.24-0.28. Its almandine content increases slightly from core (Alm57-58 Prp22-23 Grs18-19 Sps1-2) toward rim (Alm59-60
Page 31 of 111
P a g e | 32 Prp19-20 Grs18-19 Sps1-2), which is a typical zoning pattern for garnet grains in mafic granulites. Garnet in sample HC-8 shows nearly consistent core-rim variation, but the core is slightly grossular rich as Alm59-60 Prp14-15 Grs24-25 Sps1-2 (core) and
ip t
Alm64-65 Prp13-14 Grs20-21 Sps1-2 (rim). In contrast, garnet in sample KDG-2 shows slightly Mn-rich rim (Alm58-59 Prp15-16 Grs19-20 Sps5-6) than core (Alm60-61 Prp15-16
us
cr
Grs20-21 Sps3-4), although their XMg values are nearly consistent (0.20-0.23).
an
Pyroxenes
Clinopyroxene in the mafic suites of present study is Mg-rich (XMg = 0.53-0.82),
M
and all are classified as augite (Fig. 6c), although their XMg varies depending on
d
samples. Clinopyroxene in the metagabbros (samples HC-7 and HC-8) shows
te
lower XMg of 0.61-0.64 and 0.53-0.59, than that in the mafic granulite (samples HC-1 and HC-11, XMg = 0.69-0.82). Single grains do not display any
Ac ce p
compositional zoning.
Orthopyroxene also shows a wide compositional range of XMg = 0.39-0.72. Orthopyroxene in mafic granulite (sample HC-1) shows the highest XMg of 0.710.72. Orthopyroxene in sample HC-11 is slightly less magnesian as XMg = 0.600.61, whereas that associated with symplectite in the same sample shows lower XMg of 0.58-0.59. A similar XMg range was also obtained for orthopyroxene in metadiorite (sample KDG1; XMg = 0.57-0.58). Orthopyroxene in metagabbro (sample HC-8) displays the lowest XMg of 0.39-0.42. Al content in the
Page 32 of 111
P a g e | 33 orthopyroxene is generally low (Al <0.1 pfu).
ip t
Plagioclase
Plagioclase in the examined rocks shows significant compositional variations
cr
depending on lithologies (Supplementary Table 4, Fig. 6d). Plagioclase in
us
symplectite-bearing metagabbro and mafic granulite (e.g., samples HC-8 and HC-11, respectively) shows higher anorthite contents of An61-77. Particularly, the
an
coarse-grained matrix plagioclase in sample HC-11 (An68-77) displays the highest anorthite contents of An68-77 as compared to that associated with symplectite
M
(An64-65). Plagioclase in metadiorite (sample KDG-1) shows the highest albite
Ac ce p
Amphibole
te
the samples as Or80-86.
d
content of An24-27. K-feldspar is compositionally nearly homogeneous throughout
Amphibole in metagranodiorite (samples WC-2 and KDG-4), metadiorite (sample KDG-1), mafic granulite (samples HC-1 and HC-11), and garnet amphibolite (sample KDG-2) shows notable compositional variation depending on its occurrence (Supplementary Table 2, Fig. 6e). The mineral is characterized by high Ca and (Na+K) contents (1.79-1.92 pfu and 0.54-0.94 pfu, respectively). Calcic amphibole in metagranodiorite and metadiorite is Si-rich, and classified as edenite – pargasite (sample KDG-1), ferro-edenite (sample KDG-4), and ferro-
Page 33 of 111
P a g e | 34 edenite – ferro-pargasite (sample WC-2). Amphibole in garnet amphibolite is slightly Al-rich and classified as ferro-pargasite. Calcic amphiboles in garnet-free mafic granulite (samples HC-1 and HC-11) are Mg- and Al-rich as pargasite. Ti
ip t
content of amphiboles in metagranodiorite, metadiorite, and amphibolite are low (0.18-0.25 pfu), but that in mafic granulites is slightly enriched (0.27-0.41 pfu).
cr
Calcic amphibole in metagabbro (sample HC-7), which occurs as inclusions in
us
poikiloblastic garnet, is significantly Ti-rich (0.52-0.54 pfu) and classified as
an
kaersutite.
M
Biotite
d
The biotite in metagranodiorite (samples WC-2 and KDG-4) and metadiorite
te
(sample KDG-1) exhibits very high TiO2 content of 3.9-5.6 wt.% (Fig. 6f). XMg of biotite in samples WC-2 and KDG-4 (metagranodiorite) is 0.44-0.51, which is
Ac ce p
lower than that in the sample KDG-1 (metadiorite) (XMg = 0.56-0.61).
4.2 Geothermobarometry
The garnet-clinopyroxene-plagioclase-quartz geothermobarometers were applied to porphyroblastic garnet and clinopyroxene with plagioclase and quartz assemblage in samples HC-7 and HC-8 (metagabbro). The estimated temperature range for garnet-clinopyroxene pairs are 830-860°C for sample HC7 and 840-860°C for sample HC-8 (at 10 kbar) based on the method of Ellis and
Page 34 of 111
P a g e | 35 Green (1979), which is a widely applied thermometer for mafic granulites and relies on experimental calibration of Fe–Mg fractionation between garnet and clinopyroxene at 750°C to 1350°C and 24 to 30 kbar. Calculated temperature
ip t
ranges using the method of Ganguly et al. (1996), which is based on revised solution model of garnet, are slightly lower but nearly consistent, 820-850°C
cr
(sample HC-7) and 790-800°C (sample HC-8). The temperatures were calculated
us
at 10 kbar, a reference pressure inferred from the peak pressure condition of the samples estimated by garnet-clinopyroxene-plagioclase-quartz geobarometer of
M
nearly consistent among all the samples.
an
Moecher et al. (1988) as discussed below. The estimated temperature range is
d
Metamorphic pressure for the metagabbro was calculated using garnet-
te
clinopyroxene-plagioclase-quartz assemblage in the samples HC-7 and HC-8 based on experimental calibration of Perkins and Newton (1981). The estimated
Ac ce p
results are 7.1-7.6 kbar for sample HC-7 and 6.6-6.8 kbar for sample HC-8 at 850ºC. These results are, however, significantly low if compared to the petrographic observation of quartz in the rock because the stability of quartz in mafic granulite possibly indicates high-pressure condition of P > 9 kbar at 850°C (e.g., Green and Ringwood, 1967). We therefore adopted the method of Moecher et al. (1988), which is based on improved garnet-clinopyroxene-plagioclasequartz geobarometer using new thermodynamic and experimental data. Application of the method yielded a pressure range of 9.5-9.8 kbar (sample HC7) and 9.4-9.6 kbar (sample HC-8) at 850°C, suggesting relatively high-pressure
Page 35 of 111
P a g e | 36 conditions.
The Fe-Mg cation exchange between garnet and hornblende has been calibrated
ip t
as a geothermometer by Graham and Powell (1984). We applied this method to garnet amphibolite (sample KDG-2) and obtained a temperature range of 700-
cr
720°C. The estimated temperatures are significantly lower than the results
us
obtained from garnet-clinopyroxene assemblage, and probably record retrograde
an
conditions.
As hornblende coexists with plagioclase in metagranodiorite (samples WC-2 and
M
KDG-4) and metadiorite (sample KDG-1), the mineral pair has been used for
d
geothermometry. Based on hornblende solid-solution models and well-
te
constrained natural and experimental studies, two geothermometers were calibrated by Holland and Blundy (1994) based on the edenite-tremolite reaction,
Ac ce p
which is applicable to quartz-bearing rocks, and edenite-richterite reaction, which is applicable to both quartz-bearing and quartz-free rocks. As quartz is present in the samples, the two thermometers were used for the calculation of temperature of the felsic rocks. The calculated results are 730−770°C for sample KDG-1, 700−740°C for sample KDG-4, and 720−780°C for sample WC-2, probably indicate retrograde re-equilibration.
4.3 Geochemistry
Page 36 of 111
P a g e | 37 Whole rock geochemical data, including major, minor, trace and rare earth elements, on nine representative samples (two metagranodiorites, one metadiorite, one garnet amphibolite, two mafic granulites, two metagabbros and
ip t
one garnet charnockite) are given in Supplementary Table 5. The salient
cr
geochemical features of the various rock types are summarized below.
us
4.3.1 Major and trace elements
an
Major element contents show that the metagranodiorite has high SiO2 content in a restricted range (71.0-71.3 wt.%) (Fig. 7), whereas the metadiorite displays
M
intermediate SiO2 of 55.8 wt.%. The lowest SiO2 content is displayed by the
d
garnet amphibolite (41.5%). The two mafic granulites have SiO2 content in the
te
range of 49.86-49.94 wt.%, whereas the metagabbro is slightly SiO2 poor as 44.7-48.8 % (Fig. 7). Interestingly, the metagranodiorites in the two different
Ac ce p
crustal units (KDG-4 and WC-2) show almost identical composition. P2O5 contents of the garnet amphibolite shows relatively high values, which is consistent with the abundant occurrence of apatite in the sample (~5%). Garnet amphibolite (KDG-2), metagabbro (HC-7) and metagranodiorites clearly plot away from the general trends displayed by other meta-mafic rocks except for their Na2O-SiO2 relationship. These meta-mafic rocks have systematic positive correlation of Al2O3, K2O and P2O5 and a negative correlation of MgO, Fe2O3, MnO, CaO, TiO2 with increasing silica content. This geochemical variability is clearly reflected by the SiO2 versus Fe2O3T/MgO diagram (Fig. 7j) suggesting
Page 37 of 111
P a g e | 38 three distinct geochemical groups of rocks: magnesian (Fe2O3T/MgO <3), ferroan (Fe2O3T/MgO >3) and felsic (SiO2 >70 %). The three compositional types can also be distinguished in terms of their SiO2 (magnesian: 48.8-54.5 wt.%; ferroan:
ip t
41.5-44.7 wt.%; felsic: >70 wt. %) and TiO2 (magnesian: 0.55-1.2 wt.%; ferroan:
cr
3.0-3.4 wt.%; felsic: 0.31-0.41 wt.%) contents.
us
The major element behaviour shown by the three groups of samples is also reflected in their trace elements (Fig. 8). In the primitive-mantle normalized (Sun
an
and McDonough, 1989) trace element plot, magnesian and felsic samples are characterized by enrichment of large ion lithophile elements (LILE) like Rb, Ba
M
and K, negative Nb, Ta, P and Ti anomalies, and relatively constant high-field
d
strength elements (HFSE: Zr, Hf) (Figs. 8a,c). Nevertheless, the ferroan
te
metagabbro (sample HC-7) is depleted in LILE and does not show Nb and Ta negative anomalies although Zr, Hf, Y and Yb are depleted (Fig. 8b). The garnet
Ac ce p
amphibolite (sample KDG-2) shows a characteristic depletion of LILE whereas the other trace elements are relatively enriched and yield a smooth pattern (Fig. 8b). Another striking feature is marked Sr enrichment in the ferroan samples and no systematic behaviour in the magnesian samples. The felsic samples however, are enriched in LILE with obvious negative anomalies of Ta and Nb, P, Ti and Sr. The felsic samples show approximately one magnitude higher contents of LILE as compared to the other two groups. Rb-Y-Nb and Rb-Yb-Ta diagrams of Pearce et al. (1984) for granitic rocks (Fig. 9) suggest that metagranodiorite and metadiorite are characterized by low Y, Yb, Nb, and Ta, and moderate Rb
Page 38 of 111
P a g e | 39 contents.
In the chondrite normalized (McDonough and Sun, 1995) REE plot (Figs. 8d-f),
ip t
the magnesian samples are characterized by light REE (LREE, e.g. La) enriched patterns with smooth constant heavy REE (HREE, e.g. Yb), except for the HREE
cr
depleted KDG-1 sample. There is no obvious Eu anomaly in the magnesian
us
samples although the garnet charnockite shows a slight positive Eu anomaly (Fig. 8d). The ferroan sample of the KC (KDG-2) has LREE enriched and HREE
an
depleted pattern whereas that of the HC (HC-7) has a LREE depleted pattern with smooth HREE. The garnet amphibolite (sample KDG-2) suggests
M
continuous fractionation from LREE to HREE ((La/Sm)cn = 2.0 and (Gd/Yb)cn =
d
2.4) (Fig. 8e). The felsic samples are highly enriched in LREE and show clear
te
negative Eu anomaly. Further, these samples have relatively flat HREE pattern. Although the samples of the felsic group are from two complexes (WC-1 and
Ac ce p
KDG-4), whole rock chemistry, trace element and REE patterns are remarkably similar.
4.3.2. Petrogenetic implications
The metagranodiorite and metadiorite fall in the field of volcanic-arc granites (VAG) in the Rb-Y-Nb and Rb-Yb-Ta diagrams of Pearce et al. (1984) (Fig. 9). The arc magmatic signature is also supported by the primitive-mantle normalized trace element plot (Figs. 8a,c) characterized by the enrichment of LILE, negative
Page 39 of 111
P a g e | 40 Nb, Ta, P and Ti anomalies, and relatively constant HFSE of the metagranodiorite and metadiorite. LREE enrichment relative to middle REE and relatively constant HREE of the metagranodiorite and metadiorite in the chondrite normalized REE
ip t
plot (Figs. 8d,f) also support volcanic arc affinity of the felsic to intermediate
cr
rocks.
us
In the TiO2-MnO-P2O5 triangular diagram (Fig. 10a) for basaltic rocks, the mafic granulite, magnesian metagabbro, and garnet charnockite (samples HC-1, HC-8,
an
HC-10, and HC-11) show island-arc tholeiite affinity, whereas the ferroan metagabbro (sample HC-7) and garnet amphibolite (sample KDG-2) are
M
classified as MORB and oceanic island alkali basalt, respectively. The arc
d
signature of the magnesian rocks is also supported by enrichment of LILE
te
elements in Fig. 8a, and LREE enrichment relative to middle REE and relatively constant HREE in Fig. 8d. On the other hand, continuous fractionation from
Ac ce p
LREE to HREE, high P2O5 content in TiO2-MnO-P2O5 plot, and high Nb content in Nb-Zr-Y plot support oceanic-island affinity of the garnet amphibolite. The MORB signature of ferroan metagabbro is supported by depleted in LILE and positive Nb and Ta anomalies in Fig. 8b, and slightly LREE-depleted and nearly flat HREE pattern in Fig. 8e.
4.4. Zircon U-Pb geochronology and Lu-Hf isotopes
4.4.1. U-Pb geochronology
Page 40 of 111
P a g e | 41 The zircon U-Pb analytical results are given in Supplementary Table 6, and zircon REE data are given in Supplementary Table 7. Representative CL images of the zircons from different rock types in this study are shown in Figs. 11-13, and
ip t
the age data are plotted in Figs. 14-17. Chondrite-normalized REE patterns are
cr
shown in Figs. 18 and 19.
us
Mafic granulite
Zircon grains in the mafic granulite sample HC-1 are translucent and colorless to
an
pale pink. Most of the grains are subhedral, displaying spherical to weakly elongate features with lengths ranging from 30 to 100 μm and an aspect ratio of
M
1.5:1 to 1:1. In CL images, the zircon grains are dark gray and structureless. A
d
total of 21 spots from 21 zircons were analyzed from this rock. Their Th, U and
te
Th/U values are in the range of 474-933 ppm, 1559-2280 ppm and 0.24-0.37, respectively (Supplementary Table 7). They exhibit fractionated REE patterns
Ac ce p
with LREE depletion and HREE enrichment, and weakly negative Eu anomalies (Fig. 18a). All of the analyzed spots form a coherent group on the concordia, yielding a weighted mean
206
Pb/238U age of 553.0 ± 3.2 Ma (MSWD = 2.4) (Fig
14a,b), which we interpret as the timing of metamorphism.
Metagabbro The zircon grains in metagabbro sample HC-7 are translucent, colorless or light brown. The euhedral to anhedral grains show ellipsoidal or prismatic morphology with a length of 50-200 μm and a length to width ratio of 2:1 to 1:1. In
Page 41 of 111
P a g e | 42 CL images, the grains show slight patchy zoning patterns or structureless domains with homogeneous gray color. A total of 15 spots from 15 zircons were analyzed from this rock. Their Th, U and Th/U contents show ranges of 5-95
ip t
ppm, 32-298 ppm, and 0.14-0.33, respectively (Supplementary Table 7). They exhibit slight fractionated REE patterns with LREE depletion and HREE
cr
enrichment, and variable negative Eu anomalies (Fig. 18b). All the spots are
us
distributed near the concordia line and can be divided into two groups (Fig 14c,d). Seven spot ages range from 555 ± 16 Ma to 589 ± 12 Ma and define a
an
weighted mean 206Pb/238U age of 576.8 ± 9.3 Ma (MSWD = 0.92) with Th/U ratios (0.15-0.33). Another eight spots form a coherent tight group on the concordia and 206
Pb/238U age of 523.0 ± 7.1 Ma (MSWD = 0.54) with
M
yield a weighted mean
Ac ce p
Garnet charnockite
te
distinct thermal events.
d
Th/U ratios (0.14-0.32). These two groups of ages are regarded to represent two
The zircon grains in garnet charnockite sample HC-10 are translucent,
colorless or light brown. The subhedral to anhedral grains show ellipsoidal or irregular morphology with a length of 50-300 μm and a length to width ratio of 3:1 to 1:1. In CL images, some grains possess dark inherited core and bright rim texture. A total of 29 spots from 29 zircons were analyzed from this rock. Their Th, U and Th/U contents show a range of 5-289 ppm, 36-758 ppm, and 0.021.42, respectively (Supplementary Table 7). The inherited cores of zircon exhibit distinct fractionated REE patterns and with negative Eu anomalies (Fig. 18b),
Page 42 of 111
P a g e | 43 whereas the bright rims of zircon shows slightly fractionated and even flat REE patterns with variable negative Eu anomalies (Fig. 18c). The majority of U-Pb data are distributed along the concordia or along the Pb loss line and yield a
ip t
lower intercept age of 533 ± 33 Ma (MSWD = 4.6) which is taken to indicate the timing of metamorphism of this rock. Twenty spots show ages ranging from 509.0
206
Pb/238U ages of 511.1 ±
us
be divided into three groups yielding weighted mean
cr
± 5.8 Ma to 624.0 ± 11.4 Ma and form a cluster near the lower intercept and can
5.9 Ma (MSWD = 0.13), 540.4 ± 6.0 Ma (MSWD = 1.5) and 579 ± 10 Ma (MSWD
an
= 0.13) (Fig 15a,b) with Th/U ratios 0.21-0.52, 0.24-1.42, and 0.02-0.24, respectively. These ages might represent three distinct thermal events of zircon
d te
Mafic granulite sill
M
growth.
The zircons from mafic-granulite sill sample HC-11 are colorless or light
Ac ce p
brown, transparent to translucent. The subhedral to anhedral grains show ellipsoidal, irregular or elongate morphology with a length of 40-300 μm and a length to width ratio of 3:1 to 1:1. In CL images, most zircons are bright with structureless or slight patchy zoning, and a few zircons show dark core and bright rim textures, suggesting their complex origin. A total of 22 spots from 22 zircons were analyzed from this rock. Their Th, U and Th/U contents show a range of 11-205 ppm, 47-720 ppm, and 0.09-0.91, respectively (Supplementary Table 7). Most spots exhibit slightly fractionated or flat REE patterns with variable negative Eu anomalies (Fig. 18d). Sixteen analyses of CL-bright, structureless
Page 43 of 111
P a g e | 44 domains exhibit coherent age and yield a weighted mean
206
Pb/238U age of 539.1
± 4.4 Ma (MSWD = 1.6) (Fig 15 c, d). Their Th/U ratios range between 0.14 and 0.91 with 10 of the 16 spot showing Th/U>0.3. The remaining six analyses of CL-
ip t
dark core or bright patchy domains show older ages between 563 ± 4 Ma and 665 ± 10 Ma. These spots have variable Th/U ratios from 0.09 to 0.35. The
cr
oldest core age of ca. 665 Ma might represent the timing of magmatic
us
emplacement of these rocks, and the weighted mean age of ca. 539 Ma from the
an
bright structureless zircons marks the metamorphic age.
M
Metadiorite
d
The zircons from metadiorite sample KDG-1 are colorless or light brown,
te
transparent to translucent. The euhedral to subhedral grains show near-spherical or prismatic morphology with a length of 40-200 μm and a length to width ratio in
Ac ce p
the range of 3:1 to 1:1. In CL images, most zircons show oscillatory-zoned core and bright thin rim texture. A total of 34 spots from 33 zircons were analyzed from this rock. Their Th, U and Th/U contents show a range of 119-806 ppm, 226-1417 ppm, and 0.13-1.06, respectively (Supplementary Table 7). They exhibit distinct fractionated patterns with variable negative Eu anomalies (Fig. 19a). All of the data when regressed define an upper intercept age of 916 ± 57 Ma and a lower intercept age of 462 ± 180 Ma (MSWD = 2.7) (Fig. 16 a,b). Among these, thirty spots exhibit a wide range of 206Pb/238U ages between 659 ± 5 Ma and 1086 ± 11 Ma suggesting variable Pb loss after the emplacement of the protolith in early
Page 44 of 111
P a g e | 45 Neoproterozoic. The remaining four spots form a coherent group near the lower intercept and yield a weighted mean
206
Pb/238U age of 532 ± 18 Ma (MSWD =
ip t
6.2), representing the metamorphic age.
cr
Garnet amphibolite
us
Zircons in the garnet amphibolite sample KDG-2 are colorless or light brown, transparent to translucent. The subhedral to anhedral grains show near-spherical
an
or irregular morphology with a length of 50-250 μm and a length width ratio of 3:1 to 1:1. Most grains are structureless and light gray in CL images. A total of 21
M
spots from 21 zircons were analyzed from this rock. Their Th, U and Th/U
d
contents show a range of 1-96 ppm, 37-195 ppm, and 0.03-0.60, respectively
te
(Supplementary Table 7). Their REE patterns can be divided into two types, the first type is characterized by LREE depletion and HREE enrichment, whereas the
Ac ce p
second type by flat HREE (Fig. 19b). Two spot ages show 923 Ma (Th/U=0.03) and 848 Ma (Th/U=0.07). Nineteen spots with
206
Pb/238U age ranging from 501 ±
9 Ma to 549 ± 14 Ma form a coherent tight group and yield a weighted mean 206
Pb/238U age of 520.7 ± 6.6 Ma (MSWD = 1.6) with Th/U ratios 0.32-0.60
(Supplementary Table 7, Fig. 16 c,d ). The Cambrian ages might mark the timing of metamorphism of this rock.
Meta-granodiorite
Page 45 of 111
P a g e | 46 Zircons in meta-granodiorite sample KDG-4 are colorless or light brown, and transparent to translucent. The euhedral to subhedral grains show near-spherical or prismatic morphology with a length of 30-200 μm and a length to width ration
ip t
of 3:1 to 1:1. In CL images, most zircons show clear oscillatory zoning and some of the grains display heterogeneous fractured domains. A total of 33 spots from
of
182-2242
ppm,
385-2485
ppm,
and
0.35-1.51,
respectively
us
range
cr
31 zircons were analyzed from this rock. Their Th, U and Th/U contents show a
(Supplementary Table 7), suggesting magmatic crystallization. Moreover, these
anomalies (Fig. 19c). Spot 4.4 yields
206
an
spots exhibit distinct fractionated REE patterns with strong negative Eu Pb/238U spot age of 538 ± 6 Ma, probably
M
corresponding to metamorphism. The remaining 32 spots are distributed along 207
Pb/206Pb age of 980 ± 16 Ma
d
the concordia and yield a weighted mean
this rock.
te
(MSWD = 1.13) (Fig. 16 e,f), which is taken to indicate the emplacement age of
Ac ce p
Zircons in the meta-granodiorite sample WC-2 are colorless or light brown
and translucent. The euhedral to subhedral grains show near-spherical or prismatic morphology with a length of 50-250 μm and a length to width ratio of 3:1 to 1:1. In CL images, most zircons show core-rim texture with weakly light rim and clear oscillatory zoning or show heterogeneous fractured core. A total of 37 spots from 37 zircons were analyzed from this rock. Their Th, U and Th/U contents show a range of 56-776 ppm, 55-1550 ppm, and 0.08-1.05, respectively (Supplementary Table 7). They have variable LREE contents but similar and fractionated HREE patterns with negative Eu anomalies (Fig. 19d). Six spots
Page 46 of 111
P a g e | 47 form a tight group (Group 1) with and a weighted mean
206
206
Pb/238U ages between 789 Ma and 822 Ma
Pb/238U age of 805 ± 12 Ma (MSWD = 4.0) with high
Th/U ratios 0.34-0.96. Six spots define the second group (Group 2) with a narrow
206
206
Pb/238U ages between 723 Ma and 745 Ma and a weighted mean
ip t
range of
Pb/238U age of 734.0 ± 4.6 Ma (MSWD = 1.1) (Th/U ratios 0.19-1.05) (Fig. 17). 206
Pb/238U ages
cr
Eighteen spots constitute another tight group (Group 3) with
us
between 524 Ma and 565 Ma and a weighted mean 206Pb/238U age of 546.0 ± 5.7 Ma (MSWD = 7.8) and relatively low Th/U ratios 0.08-0.48 (15 of the 18 spots
an
show Th/U<0.27, 12 of the 18 spots show Th/U<0.18). The result suggests magmatic emplacement of the protolith at ca. 805 Ma and another thermal event
Ac ce p
te
4.4.2 Zircon Lu-Hf isotopes
d
M
at ca. 734 Ma, followed by metamorphism ca. 546 Ma.
In situ Hf isotope compositions of the dated zircons are given in Table 8, and
plotted in Fig 20. In the eight samples analyzed, the fLu/Hf values of zircons vary from -0.92 to -1.0, which are obviously lower than the fLu/Hf values of mafic crust (0.34, Amelin et al., 2000) and sialic crust (-0.72, Vervoort and Patchett, 1996). Therefore, the two-stage model age is considered to be a better representation than the single stage model age to evaluate the time of source material extraction from the depleted mantle or the residence time of the source material in the crust (Blichert-Toft and Albarede, 1997).
Page 47 of 111
P a g e | 48 Mafic granulite
Seventeen zircons were analyzed from mafic granulite sample HC-1 for Lu176
Hf/177Hf ratios in the range of 0.282365 to
ip t
Hf isotopes. The result show initial
0.282432. The εHf(t) values show a tight cluster from -2.2 to 0.1 with an average
cr
value of -1.1, falling close the line of the Chondrite Uniform Reservoir (CHUR)
us
(Fig. 20). The Hf depleted model ages (TDM) are between 1135 Ma and 1227 Ma (mean 1184 Ma), and Hf crustal model ages (TDMC) show a range of 1501-1651
an
Ma (mean 1582 Ma). The results suggest a mixed source from both juvenile
M
Neoproterozoic and reworked Mesoproterozoic components.
te
d
Metagabbro
Twelve zircons were analyzed from metagabbro sample HC-7 for Lu-Hf
Ac ce p
isotopes. The result shows initial
176
Hf/177Hf ratios ranging from 0.282221 to
0.282280. The εHf(t) values are in the range of -7.5 to -5.5 with an average value of -6.3. The Hf depleted model ages (TDM) are between 1338 Ma and 1418 Ma (mean 1372 Ma), and Hf crustal model ages (TDMC) in the range of 1847-1978 Ma (mean 1902 Ma). The predominantly negative εHf(t) values suggest that the source material involved reworked Paleoproterozoic crust.
Garnet charnockite
Page 48 of 111
P a g e | 49 Twenty-five zircons were analyzed from the garnet charnockite sample HC10 for Lu-Hf isotopes. The result show initial
176
Hf/177Hf ratios ranging from
0.281500 to 0.282144. The εHf(t) values are in the range of -33.3 to -10.5 with an
ip t
average value of -17.7. The Hf depleted model ages (TDM) are between 1522 Ma and 2401 Ma (mean 1809 Ma), and Hf crustal model ages (TDMC) varies 2156-
cr
3580 Ma (mean 2614 Ma averagely). The highly negative εHf(t) values suggest
us
that the source material for the charnockite was reworked Paleoproterozoic crust. Twenty zircons were analyzed from mafic granulite sample HC-11 for Lu-Hf 176
Hf/177Hf ratios in the range of 0.281857 to
an
isotopes. The result show initial
0.282095. The εHf(t) values are in the range of -20.5 to -12.1 with an average
M
value of -14.1. The Hf depleted model ages (TDM) are between 1590 Ma and
d
1922 Ma (mean 1666 Ma), and Hf crustal model ages (TDMC) varies from 2263 to
te
2790 Ma (mean 2386 Ma). The dominantly negative εHf(t) values suggest that the source material for the magma evolved from the Neoarchean-Paleoproterozoic
Ac ce p
crust.
Metadiorite
Twenty-six zircons were analyzed from metadiorite sample KDG-1 for Lu-Hf
isotopes. The result show initial
176
Hf/177Hf ratios ranging from 0.282270 to
0.282505. The εHf(t) values are in the range of -1.1 to 7.2 with an average value of 1.5. The Hf depleted model ages (TDM) are between 1031 Ma and 1360 Ma (mean 1259 Ma), and Hf crustal model ages (TDMC) varies from 1206 to 1733 Ma
Page 49 of 111
P a g e | 50 (mean 1569). The results suggest a mixed source from both juvenile and Proterozoic components.
ip t
Garnet amphibolite
176
Hf/177Hf ratios in the range of
us
for Lu-Hf isotopes. The result show initial
cr
Twenty zircons were analyzed from the garnet amphibolite sample KDG-2
0.281845 to 0.282058. The εHf(t) values are in the range of -21.3 to -13.8 with an
an
average value of -16.5. The Hf depleted model ages (TDM) are between 1641 Ma and 1931 Ma (mean 1744 Ma), and Hf crustal model ages (TDMC) varies from
M
2356 to 2828 Ma (mean 2526 Ma). The predominantly negative εHf(t) values
Ac ce p
Meta-granodiorite
te
Paleoproterozoic crust.
d
suggest that the source material was evolved from the Neoarchean-
Twenty-six zircons were analyzed from meta-granodiorite sample KDG-4 for
Lu-Hf isotopes. The result show initial
176
Hf/177Hf ratios ranging from 0.282230 to
0.282341. The εHf(t) values are in the range of 2.5 to 6.4 with an average value of 3.9. The Hf depleted model ages (TDM) are between 1266 Ma and 1417 Ma (mean 1365 Ma), and Hf crustal model ages (TDMC) varies from 1431 to 1679 Ma (mean 1592 Ma). The results suggest a mixed source from both juvenile and Paleo-Mesoproterozoic components. The dominantly positive εHf(t) values
Page 50 of 111
P a g e | 51 suggest that the source material was evolved from the Mesoproterozoic juvenile material. Thirty-two zircons were analyzed from metagranodiorite sample WC-2 for 176
Hf/177Hf ratios range from 0.281968 to
ip t
Lu-Hf isotopes. The result show initial
0.282283. The εHf(t) values are in the range of -15.0 to -3.9 with an average
cr
value of -6.2. The Hf depleted model ages (TDM) are between 1344 Ma and 1789
us
Ma (mean 1440 Ma), and Hf crustal model ages (TDMC) varies from 1799 to 2498 Ma (mean 1944 Ma). The dominantly negative εHf(t) values suggest that the
an
source material was evolved from reworked Paleoproterozoic crust.
d
M
5. Discussion
te
The samples analyzed in this study are petrographically classified as metagranodiorite (samples KDG-4 and WC-2; Pl + Kfs + Qtz + Bt + Hbl),
Ac ce p
metadiorite (sample KDG-1; Pl + Qtz + Bt + Hbl), mafic granulite (samples HC-1 and HC-11; Cpx + Opx + Pl + Hbl), metagabbro (samples HC-7 and HC-8; Grt + Cpx + Opx + Pl), garnet amphibolite (sample KDG-2; Grt + Hbl + Pl) and garnet charnockite (sample HC-10; Grt + Pl + Qtz + Opx + Cpx). Pressure and temperature conditions of the study area were estimated based on conventional geothermobarometry on mafic granulite, amphibolite, metagranodiorite, and metadiorite. The peak P-T conditions were estimated from garnet-clinopyroxeneplagioclase-quartz assemblages in samples HC-7 and HC-8 as 830-860°C and 9.4-9.8 kbar. Slightly lower temperature ranges of 700-720°C obtained for garnet-
Page 51 of 111
P a g e | 52 hornblende pairs in garnet amphibolite (sample KDG-2) and 720−780°C for hornblende-plagioclase pairs in metagranodiorite and metadiorite probably correspond to retrograde conditions. The peak metamorphic conditions of ~860
ip t
°C and ~9.8 kbar estimated in this study are consistent with previous P-T estimates from the Highland Complex (~850-900 °C and ~8-10 kbar; Schenk et
cr
al., 1988; Schumacher et al., 1990; Kriegsman, 1991; Hiroi et al., 1994, Raase
us
and Schenk, 1994; Schumacher and Faulhaber, 1994; Kriegsman and Schumacher, 1999; Dharmapriya et al., 2014). The UHT metamorphic conditions
an
of >900 °C or even >1100 °C estimated based on the stability of diagnostic mineral assemblages (e.g., sapphirine + quartz) and phase equilibria (e.g.,
M
Kriegsman and Schumacher, 1999; Osanai et al., 2006; Sajeev and Osanai,
te
Highland Complex.
d
2004a, 2004b; Sajeev et al., 2007) is mainly restricted to some localities in the
Ac ce p
The petrographic and geochemical features discussed in previous sections suggest distinct petrogenetic history for the protoliths of the meta-igneous suite from the crustal blocks of Sri Lanka. The REE and trace element patterns (Fig. 8) as well as Rb-Y-Nb and Rb-Yb-Ta diagrams (Fig. 9) reveal volcanic arc affinity for the metagranodiorite, metadiorite and garnet charnockite, suggesting that the protoliths of the rocks were derived from felsic to intermediate arc magmas. The two mafic granulites (HC-1 and HC-11) and magnesian metagabbro (HC-8) also suggest volcanic arc affinity. The garnet-free mafic granulite (sample HC-1) occurring within marble and magnesian metagabbro (sample HC-8) enclosed in
Page 52 of 111
P a g e | 53 quartzite shows evidence for subduction-related mafic magmatism in the crustal blocks of WC, KC and the HC.
ip t
In contrast, the protolith of sample HC-7 (ferroan metagabbro from the Highland Complex) is inferred as N-MORB, suggesting the accretion of remnant accreted
cr
oceanic lithosphere accreted along with the arc components and trench
us
sediments during the subduction-collision event. The garnet amphibolite (KDG-2)
an
displays a signature of oceanic island alkali basalt.
From the Kadugannawa Complex, a metadiorite (KDG-1), garnet amphibolite
M
(KDG-2), and a meta-granodiorite (KDG-4) were analyzed in this study. Zircons in
d
the metadiorite define an upper intercept age of 916 ± 57 Ma and a lower
te
intercept age of 462 ± 180 Ma (MSWD = 2.7). Metamorphic zircons in this rock form a coherent group and yield a weighted mean
206
Pb/238U age of 532 ± 18 Ma
Ac ce p
(MSWD = 6.2). Two zircon grains from the garnet amphibolite show spot ages of 923 Ma and 848 Ma (Th/U=0.07). The majority of zircons in this rock are of metamorphic origin and define a weighted mean
206
Pb/238U age of 520.7 ± 6.6
Ma (MSWD = 1.6). Zircons in the meta-granodiorite sample yield a weighted mean
207
Pb/206Pb age of 980 ± 16 Ma (MSWD = 1.13) marking the timing of
emplacement of the protolith. From the Wanni Complex, zircons in a metagranodiorite (WC-2) define three groups with a weighted mean
206
Pb/238U
age of 805 ± 12 Ma (MSWD = 4.0) (emplacement of the magmatic protolith) 734.0 ± 4.6 Ma (MSWD = 1.1) (Cryogenian thermal event) and 546.0 ± 5.7 Ma
Page 53 of 111
P a g e | 54 (MSWD = 7.8) (latest Neoproterozoic-Cambrian metamorphism). Three samples were analyzed from the Highland Complex. All of the analyzed spots in a mafic granulite (HC-1) form a coherent group on the 206
Pb/238U age of 553.0 ± 3.2 Ma (MSWD =
ip t
concordia, yielding a weighted mean
2.4) marking the timing of metamorphism. Zircons in a metagabbro sample (HC206
Pb/238U ages of 576.8 ± 9.3 Ma
cr
7) form two groups with weighted mean
us
(MSWD = 0.92) and 523.0 ± 7.1 Ma (MSWD = 0.54) suggesting two distinct thermal events. Zircons from a garnet charnockite (HC-10) show Pb loss with a
an
lower intercept age of 533 ± 33 Ma (MSWD = 4.6) marking the timing of metamorphism. Another group of zircons from this rock show spot ages in the
M
range of 509.0 ± 5.8 Ma to 624.0 ± 11.4 Ma with weighted mean
206
Pb/238U ages
d
of 511.1 ± 5.9 Ma (MSWD = 0.13), 540.4 ± 6.0 Ma (MSWD = 1.5) and 579 ± 10
te
Ma (MSWD = 0.13), suggesting different stages of zircon growth during the late Neoproterozoic. Zircons in a mafic granulite sill (HC-11) yield a weighted mean Pb/238U age of 539.1 ± 4.4 Ma (MSWD = 1.6). Another group with CL-dark
Ac ce p
206
core or bright patchy domains shows older ages between 563 ± 4 Ma and 665 ± 10 Ma, and a weighted mean metamorphic age of 539 Ma from the bright structureless zircons.
Zircons in the metadiorite (KDG-1) and metagranodiorite (KDG-4) from the
Kadugannawa complex show mostly positive εHf(t) values in the range of -1.1 to 7.2 and Hf crustal model ages (TDMC) varies from 1206 to 1733 Ma suggesting a mixed source from both juvenile and Paleo-Mesoproterozoic components. However, zircons in the garnet amphibolite (KDG-2) show dominantly negative
Page 54 of 111
P a g e | 55 εHf(t) values (mean -16.5) with TDMC in the range of 2356 to 2828 Ma suggesting reworked Neoarchean-Paleoproterozoic crustal source. Zircons in the Wanni Complex metagranodiorite (WC-2) also possess negative εHf(t) values (mean -
ip t
6.2) with TDMC in the range of from 1799 to 2498 Ma suggesting reworked Paleoproterozoic crust as the magma source.
cr
The Highland Complex rocks preserve distinct imprints of reworking of
us
older crust. The zircon εHf(t) values in mafic granulite show a tight cluster from 2.2 to 0.1 with TDMC in the range of 1501-1651 Ma suggesting a mixed source
an
from both juvenile Neoproterozoic and reworked Mesoproterozoic components. Zircons in the metagabbro (HC-7) from this complex show negative εHf(t) values
crust as the source.
M
(mean -6.3) and TDMC of 1847-1978 Ma suggesting reworked Paleoproterozoic Zircons in the garnet charnockite (HC-10) also display
te
d
highly negative εHf(t) values (mean -17.7) and older TDMC) (mean 2614 Ma) suggesting reworked Paleoproterozoic crust. Zircons in the mafic granulite sill
Ac ce p
(HC-11) also show negative εHf(t) values (mean -14.1) and TDMC between 2263 to 2790 Ma suggesting that the source material for the magma evolved from the Neoarchean-Paleoproterozoic crust. The εNd(t) and εHf(t) isotopic data reported in previous studies and those
from the present study from the Sri Lankan crustal blocks provide evidence for magma generation from mixed sources involving both juvenile and reworked components. The Nd isotope data reported elsewhere for the KC and the WC show highly heterogeneous εNd(t) values, ranging from +2 to -10 (Willbold et al., 2004), and +4 to -8 (Millisenda et al., 1994). These data suggest variable crustal
Page 55 of 111
P a g e | 56 contamination of mantle-derived melts in a continental magmatic arc setting. However, in clear contrast, the HC rocks show dominantly negative εNd(t) values from -7 to -25 suggesting a different tectonic genetic setting from that of the KC
ip t
and the WC, with substantial contribution from older reworked crustal components.
cr
In summary the 916 to 980 Ma ages from the Kadugannawa Complex
us
represent arc magmatism during early Neoproterozoic. Zircon in the diorites from this complex show positive εHf(t) values suggesting juvenile addition. This was
an
followed by the 805 Ma tonalitic magma emplacement in the Wanni Complex. Multiple thermal events during Neoproterozoic are recorded from the Wanni and
M
Highland Complexes, with subduction-related signature, culminating in Cambrian
te
d
high-grade metamorphism (~550 Ma) that overprinted all the three complexes.
We propose a tectonic model where we envisage that the Wanni and Vijayan
Ac ce p
complexes as continental arcs underlain by Neoarchean – Paleoproterozoic basement (Fig. 21). The similarity in the arc-related geochemical features of the magmatic rocks emplaced broadly and coevally in both Wanni and Vijayan complexes prompt us to suggest a double-sided subduction of an oceanic lithosphere during early to mid Neoproterozoic. The broadly negative εNd(t) and εHf(t) values of these magmatic suites, together with the Neoarchean – Paleoproterozoic two stage Hf crustal model ages are also in accordance with melting of underlying older basement in both of these blocks, as well as the presence of older crust within the WC. The mafic/intermediate suites in the KC
Page 56 of 111
P a g e | 57 complex
(metadiorites)
show
input
of
juvenile
components
within
a
suprasubduction setting. In our model, the WC is envisaged as an accretionary suture with both oceanic and continental components, together with fragments of
ip t
older continental crust, analogous to the tectonic milieu in modern Phanerozoic accretionary belts such as the Central Asian Orogenic Belt (e.g., Xiao and The N-MORB signature of the garnet-bearing metagabbro
cr
Santosh, 2014).
us
(sample HC-7) and the oceanic alkali basalt chemistry of the garnet amphibolite (sample KDG-2) support our model of the accretion of remnants of the oceanic
an
lithosphere along with trench sediments onto the arc (continent). Occurrence of granitic-ring structures, serpentinites, anorthosites, and Cu-magnetite deposits in
M
association with chert bands and apatite in the VC at the boundary with the HC
d
(Pathirana, 1980; Cooray, 1984) indicates active continental margin volcanism
te
and supra-subduction zone magmatism at the junction of the HC and the VC. The submarine canyon with wall heights >1 km in the north east VC and hot
Ac ce p
springs are contiguous with the HC-VC contact zone. These features lend further support our subduction model.
Importantly, our study identifies juvenile magmatic additions during mid
Neoproterozoic (KDG-1, associated with subduction) and latest Neoproterozoic – Cambrian (HC-1; magmatism in post-collisional asthenospheric upwelling). The post-collisional slab-break off and asthenospheric upwelling is also envisaged as the principal contributor of heat that triggered widespread ultrahigh-temperature metamorphism within the HC, similar to the scenario envisaged for the formation
Page 57 of 111
P a g e | 58 of the UHT rocks elsewhere (e.g., Santosh et al., 2013).
As discussed in earlier sections, previous studies had identified the HC, WC and
ip t
VC as distinct tectonic units based largely on Nd model age mapping, combined with regional structural interpretations. The Nd model ages are as follows: HC=2-
cr
3.4 Ga; WC=1-2 Ga; VC=1-1.8Ga; and KC=1.4-1.8Ga. The zircon ages from
us
these three different tectonic units (either magma emplacement ages or depositional ages of sediments) are HC =~ 2.3Ga; WC=750-1.1Ga; VC=650Clearly the HC preserves records of the oldest
an
1.9Ga; and KC=0.8-1.0Ga.
sources of sedimentary units in Sri Lanka, and/or the oldest recycled continental
M
crust in this island. In the present study also, the oldest TDMC (2790 Ma) is Such older
d
obtained from zircons in a garnet charnockite within the HC.
te
components are common in many younger accretionary belts where ancient micro-continents or arcs are accreted and admixed during the final collision In the present case, the subduction-accretion continued until late
Ac ce p
stage.
Neoproterozoic and the final architecture of the HC was built during the Cambrian collisional event.
The ‘flower-style’ extrusion of the HC is clearly
marked by the thrust zone mapped in previous studies at the border of the HC with the VC; the structural break is also corroborated by geochronological studies that show a distinct boundary between the HC and the VC. However, the boundary is unclear on the western margin of the HC at its contact with the WC, where the large arc magma chamber of the KC is found emplaced at the inferred boundary. Away from the KC to the north and south-west also the HC-WC
Page 58 of 111
P a g e | 59 boundary is not clear. However, the doubly plunging synforms defined by the KC are considered to mark the structural boundary in the central region along the HC-WC inferred boundary. Such large-scale sheath folds are considered as the
ip t
hallmarks of major accretionary orogens elsewhere (e.g., Chetty et al., 2012). According our model, KC is not a discrete crustal block, but part of a disrupted
cr
huge arc magma chamber that was exhumed and transposed along the margin
us
of the WC.
an
Some of the existing models suggest that the WC, HC and VC were discrete terranes (Vitanage, 1972, 1985). The WC is proposed to have collided with HC
M
first, and subsequently, both WC and HC together as a unified block collided
d
with, and was thrust over, the VC (Vitanage, 1972; Voll and Kleinschrodt, 1991;
te
Kröner and Jaeckel, 1994; Kriegsman, 1995) during the amalgamation of Gondwana (Kriegsman, 1995; Kehelpannala 1997, 2003, 2004; Kröner et al.
Ac ce p
2003). However, there is no robust geochronologic and tectonic evidence to support this speculation, particularly the two discrete docking events, which should produce two separate metamorphic orogens with different ages along both margins of the HC. In fact, our data, coupled with available information clearly show that the WC and the VC were coeval Neoproterozoic arcs which were welded along the HC during late Neoproterozoic-Cambrian. The Cambrian collision event was accompanied by metamorphic overprint in all the rocks, formation of UHT metamorphic orogen along the HC, and post-collisional mantlederived magmatism within the HC (e.g. sample HC-1 in our study), all suggesting
Page 59 of 111
P a g e | 60 the HC as a collisional suture.
Whether the older continental basement
underlying WC and VC were contiguous prior to Neoproterozoic, or whether these bocks were derived independently from other continental blocks, remains
ip t
to be investigated in future studies.
cr
In previous studies, the magmatic episodes ranging from the Grenvillian
us
(ca. 1.1 Ga) to ca. 882 Ma in the KC (e.g. Kröner et al., 2003; Willbold et al., 2004) were considered to mark the beginning of the breakup of the Rodinia
an
supercontinent. However, the Neoproterozoic magmatic pulses in the Sri Lankan crustal complexes are clearly related to convergent margin setting as shown in
M
our study, as against the extensional setting required for Rodinia breakup. Except
d
for the Late Neoproterozoic-Cambrian mafic dykes (sample HC-1), and some
te
latest Neoproterozoic syn- and post-collisional granitoids in the HC, there are no other rock records of widespread extension-related mafic or bimodal magmatism
Ac ce p
in Sri Lanka during mid Neoproterozoic, and therefore a Rodinia break-up model to explain the magmatic suites is untenable. Accretion of oceanic components and arc magmatism in a convergent margin setting are also clearly indicated from our results. The global assembly of the Rodinia supercontinent with South China at its center is considered to have occurred at ca. 900 Ma (Li et al., 2008). Initial rifting of Rodinia started at ca. 860 with widespread breakup during 830800 Ma. From the Eastern Ghats Belt in India, Korhonen et al. (2011) reported SHRIMP ages of 980 Ma in monazites associated with garnet and cordierite from ultrahigh-temperature Mg–Al granulites, marking the timing of collisional
Page 60 of 111
P a g e | 61 amalgamation of this terrane within the Rodinia assembly. They also reported a range of
207
Pb/235U monazite ages (from ca. 1014 Ma to 959 Ma for one sample
and ca. 1043 Ma to 922 Ma for the second sample) suggesting protracted
ip t
monazite growth during the high-temperature retrograde evolution, which would broadly correspond with the global assembly of the Neoproterozoic Rodinia However, the data from crustal blocks in Sri Lanka indicate
cr
supercontinent.
us
protracted arc magmatism in a convergent margin setting, commencing in early Neoproterozoic, continuing through mid Neoproterozoic and culminating in
suture (HC).
an
regional metamorphism in all the blocks, with UHT imprint along the collisional The arc-related signature of these magmatic suites typical of
M
convergent margin magmas prompt us to suggest that the Neoproterozoic
d
magmatic events culminating in Cambrian collisional metamorphism might be
te
related to the prolonged Pacific-type orogeny associated with the assembly of Gondwana supercontinent, as also proposed for the Southern Granulite Terrane
Ac ce p
in India (Santosh et al., 2009).
Thus, our model deviates from the existing
concepts that link Sri Lanka with Rodinia rifting, and correlates the Neoproterozoic arc-related magmatic suites with convergent margin processes associated with the assembly of Gondwana.
8. Conclusions
Petrological and geochemical analyses and zircon U-Pb and Lu-Hf data from a suite of metamorphosed magmatic rocks from the major tectonic units in Sri
Page 61 of 111
P a g e | 62 Lanka, coupled with the available information from the crustal blocks, lead us to the following conclusions.
ip t
1. Regional metamorphism in the crustal bocks reached up to 860°C and 9.8 kbar, together with the extruded metamorphic orogen in the Highland
cr
Complex showing ultrahigh-temperature signature, which we correlate to
us
heat input from asthenospheric sources in a post-collisional slab-break off
an
setting.
2. Trace element and rare earth element patterns as well as Rb-Y-Nb and
M
Rb-Yb-Ta discrimination plots show volcanic arc affinity and protolith
te
d
crystallization from felsic to intermediate arc magmas.
3. The mafic magmatic units including those incorporated within the
Ac ce p
metasedimentary belt of Highland Complex show N-MORB signature, or oceanic island alkali basalt affinity, suggesting accretion of the remnants of oceanic lithosphere during the subduction-collisional event.
4. Zircon U-Pb ages ranging from 916 to 980 Ma from the Kadugannawa Complex represent arc magmatism during early Neoproterozoic, followed by the 805 Ma granodioritic magma emplacement in the Wanni Complex. Repeated thermal events during mid and late Neoproterozoic are also recorded from the Wanni and Highland Complexes, culminating in
Page 62 of 111
P a g e | 63 Cambrian high-grade metamorphism that reached ultrahigh-temperature conditions.
ip t
5. Zircon Lu-Hf data reveal mixed sources with the involvement of both juvenile and reworked Neoarchean – Paleoproterozoic basement. There
cr
is also clear indication of juvenile addition during both Neoproterozoic and
us
Cambrian tectonothermal events.
6. According to our data, the KC is part of a disrupted huge arc magma
an
chamber that was exhumed and transposed along the margin of the WC
d
simply a part of the WC.
M
which was previously regarded to be either an exotic crustal block or
te
7. A model of double-sided subduction during the Neoproterozoic, where the Wanni Complex to the west and the Vijayan Complex to the east represent
Ac ce p
continental arcs during the Neoproterozoic, culminating in collision along the Highland Complex during late Neoproterozoic-Cambrian is proposed. The occurrence of supra-subduction zone magmatic complexes and volcanic suites formed in a convergent margin setting along the contact of the VC with the HC lend support to our model.
8. The data presented in this study are consistent with a prolonged convergent margin setting from early to late Neoproterozoic, followed by final collisional amalgamation in the Cambrian, with the Highland Complex
Page 63 of 111
P a g e | 64 defining an accretionary belt as well as the collisional suture.
9. Both vertical (arc magmatic) and lateral (accretionary) growth of
ip t
continental crust during the Neoproterozoic is well recorded in the crustal
cr
blocks of Sri Lanka.
us
Acknowledgments
an
We thank Prof. Guochun Zhao, Editor and two referees for constructive and helpful comments. We extend our thanks to Qiongyan Yang and Xueming Teng at
M
the China University of Geosciences Beijing for help with zircon analyses. This study contributes to the Talent Award to M. Santosh from the Chinese
te
d
Government under the 1000 Talent Plan. Partial funding for this project was produced by a Grant-in-Aid for Scientific Research (B) from Japan Society for the
Ac ce p
Promotion of Science (JSPS) to Tsunogae (No. 26302009).
References
Amelin, Y., Lee, D.C., Halliday, A.N., 2000. Early-Middle Archaean crustal evolution deduced from Lu-Hf and U-Pb isotopic studies of single zircon grains. Geochim. Cosmochim. Acta 64, 4205-4225. Almond, D.C., 1991. Arena Gneiss and Kandy Gneiss-a proposed subdivision of the Highland Series around Kandy, and its significance. J. Geol. Soc. Sri Lanka 3, 41-50. Blichert-Toft, J., Albarede, F., 1997. The Lu-Hf isotope geochemistry of chondrites and the evolution of the mantle–crust system. Earth and Planetary Science Letters 148, 243-258.
Page 64 of 111
P a g e | 65
Baur, N, Kröner, A, Todt, W, Liew, T.C., Hofman, A.W., 1991. U-Pb isotopic systematics of zircons from prograde and retrograde transition zones in high grade orthogneisses, Sri Lanka. J. Geol. 99, 527-545.
ip t
Berger, A.R., Jayasinghe, N.R., 1976. Precambrian structure and chronology in the Highland Series of Sri Lanka. Precambrian Res. 3, 559-576.
cr
Burton, K.W., O’Nions, R.K., 1990. The timescale and mechanism of granulite formation at Kurunegala, Sri Lanka. Contrib. Mineral. Petrol. 106, 66-89.
us
Cawood, P.A., Buchan, C., 2007. Linking accretionary orogenesis with supercontinent assembly. Earth-Science Reviews 82, 217-256.
an
Chetty, T.R.K., Yellappa, T., Mohanty, D.P., Nagesh, P., Sivappa, V.V., Santosh, M., Tsunogae, T., 2012. Mega Sheath Fold of the Mahadevi Hills, Cauvery Suture Zone, Southern India: Implication for Accretionary Tectonics. J. Geol. Soc. India 80, 747-758.
M
Condie, K.C., Aster, R.C., 2013. Refinement of the supercontinent cycle with Hf, Refinement of the supercontinent cycle with Hf, Nd and Sr isotopes: Geoscience Frontiers 4, 667-680.
te
d
Condie, K.C., Kröner, A., 2013. The building blocks of continental crust: Evidence for a major change in the tectonic setting of continental growth at the end of the Archean: Gondwana Res. 23, 394-402.
Ac ce p
Cordani, U., Cooray, P.G., 1990. Rb-Sr ages of granulites and gneisses from the Precambrian of Sri Lanka. J. Geol. Soc. Sri Lanka 2, 35-43. Cooray, P.G., 1962. Charnockites and their associated gneisses in the Precambrian of Ceylon. J. Geol. Soc. London 118, 239– 273. Cooray, P.G., 1984. An introduction to the geology of Sri Lanka. National Museum Sri Lanka Publication, 340. Cooray, P.G., 1994. The Precambrian of Sri Lanka: a historic review. Precambrian Res. 66, 3-18. De Maesschalck, A.A, Oen, I.S., Hebeda, E.H., Verschure, R., Arps, E.S., 1990. Rb-Sr whole rock ages of Kataragama and Pottuvil Charnockites and east Vijayan gneiss: indication of a 2 Ga metamorphism in the highlands of south east Sri Lanka. J. Geol. 98, 772-779. Dahanayake, K., Jayasena, H.A.K., 1983. General geology and petrology of some Precambrian crystalline rocks from the Vijayan Complex of Sri Lanka.
Page 65 of 111
P a g e | 66 Precambrian Res. 19, 301–315. Dahanayake, K., 1982. Structural and petrological studies on the Precambrian Vijayan Complex of Sri Lanka. Revista Brasiliano Geociencias 12, 89–93.
cr
ip t
Dharmapriya, P.L., Malaviarachchi, S.P.K., Galli, A., Su, B., Subasinghe, N.D., Dissanayake, C.B., Nimalsiri, T.B., Zhu, B., 2014. P-T evolution of a spinel + quartz bearing khondalite from the Highland Complex, Sri Lanka: Implications for non-UHT metamorphism, J. Asian Earth Sci., Doi: http://dx.doi.org/10.1016/j.jseaes.2014.05.003.
us
Ellis, D.J., Green, D.D., 1979. An experimental study of the effect of Ca upon garnet-clinopyroxene Fe-Mg exchange equilibria. Contrib. Mineral. Petrol 71, 13-22.
an
Faulhaber, S., Raith, M., 1991. Geothermometry and geobarometry of high grade rocks: a case study on garnet pyroxene granulites in southern Sri Lanka. Min. Mag. 55, 33-56.
M
Ganguly, J., Cheng, W., Tirone, M., 1996. Thermodynamics of aluminosilicate garnet solid solution: new experimental data, an optimized model, and thermometric applications. Contrib. Mineral. Petrol. 126, 137-151.
te
d
Graham, C.M., Powell, R., 1984. A garnet-hornblende geothermometer: calibration, testing, and application to Pelona Schist, Southern California. J. Metamorph. Geol. 2, 13-31.
Ac ce p
Green, T.H., Ringwood, A.E., 1967. An experimental investigation of the gabbro to eclogite transformation and its petrological applications. Geochim. Cosmochim. Acta 31, 767-833. Hansen, E.C., Janardan, A.S., Newton, R.C., Prame, W.K.B.N., Ravindrakumar, G.R., 1987. Arrested charnockite formation in southern India and Sri Lanka. Contrib. Mineral. Petrol. 196, 66-89. Hapuarachchi, D.J.A.C., 1968. Cordierite and wollastonite-bearing rocks of south-west of Ceylon. Geol. Mag. 105, 317-324. Harley, S.L., 1989. The origin of granulites: a metamorphic perspective. Geol. Mag. 126, 215-247. Hatherton, T., Pattiarachchi, D.B., Ranasinghe, V.V.C., 1975. Gravity map of Sri Lanka, 1:1,000,000. Geol. Surv. Dept., Sri Lanka, Prof. Paper 3, 39. Hiroi, Y., Ogo, Y., Namba, L., 1994. Evidence for prograde metamorphic evolution of Sri Lankan politic granulites and implications for the development of
Page 66 of 111
P a g e | 67 continental crust. Precambrian Res. 66, 245-263.
ip t
Hiroi, Y., Yanagi, A., Kato, M., Kobayashi, T., Prame, B., Hokada, T., SatishKumar, M., Ishikawa, M., Adachi, T., Osanai, Y., Motoyoshi, Y., Shiraishi, K., 2014. Super cooled melt inclusions in lower-crustal granulites as a consequence of rapid exhumation by channel flow. Gondwana Res. 25, 226234.
cr
Holland, T., Blundy, J., 1994. Non-ideal interactions in calcic amphiboles and their bearing on amphibole-plagioclase thermometry. Contrib. Mineral. Petrol. 116, 433-447.
us
Hölzl, S., Köhler, H., Kröner, A., Jaeckel, P., Liew, T.C., 1991. Geochronology of the Sri Lankan basement. Geological Survey Department, Sri Lanka, Professional Paper 5, 237-257.
an
Hölzl, S., Hofmann, A.W., Todt, W., Köhler, H., 1994. U–Pb Geochronology of the Sri Lankan basement. Precambrian Res. 66, 123–149.
M
Hu, Z.C., Gao, S., Liu,Y.S., Hu, S.H., Chen, H.H., Yuan, H.L., 2008. Signal enhancement in laser ablation ICP-MS by addition of nitrogen in the central channel gas. Journal of Analytical Atomic Spectrometry 23, 1093–1101.
Ac ce p
te
d
Hu, Z.C., Liu,Y.S., Chen, L., Zhou, L., Li, M., Zong, K.Q., Zhu, L., Gao, S., 2011. Contrasting matrix induced elemental fractionation in NIST SRM and rock glasses during laser ablation ICPMS analysis at high spatial resolution. Journal of Analytical Atomic Spectrometry 26, 425–430. Kagami, H., Owada, M., Osanai, Y., Hiroi, Y., 1990. Preliminary geochronological study of Sri Lankan rocks. Interim Report of Japan-Sri Lanka Joint Research, p. 55-70. Karunaratne, D.N., Perera, L.R.K. and Fernando, G.W.A.R., 2002. The missing link between Prograde decompression (?) and Retrograde Isobaric Cooling in the Highland Complex Granulites of Sri Lanka. Proc. 18th Ann. Technical Sessions of the Geol. Soc. of Sri Lanka, p. 3 (Abstract). Kehelpannala, K.V.W., 1991. Structural evolution of high-grade terrains in Sri Lanka with special reference to the areas around Dodanhaslanda and Kandy. Geological Survey Department, Sri Lanka, Professional Paper 5, 69-88. Kehelpannala, K.V.W., 1997. Deformation of a high-grade Gondwana fragment, Sri Lanka. Gondwana Res. 1, 47–68. Kehelpannala, K.V.W., 1999. Shear-zone controlled arrested charnockitization,
Page 67 of 111
P a g e | 68 retrogression and metasomatism of high-grade rocks. Gondwana Res. 2, 573-577. Kehelpannala, K.V.W., 2003. Structural evolution of the middle to lower crust in Sri Lanka—a review. J. Geol. Soc. Sri Lanka 11, 45–86.
ip t
Kehelpannala, K.V.W., 2004. Arc accretion around Sri Lanka during the assembly of Gondwana. Gondwana Res. 7, 1323–1328.
cr
Kleinschrodt, R., Voll, G., Kehelpannala, W., 1991. A layered basic intrusion, deformed and metamorphosed in granulite facies of the Sri Lanka basement. Geol. Rundschau 80, 779-800.
us
Kleinschrodt, R., 1994. Large-scale thrusting in the lower crustal basement of Sri Lanka. Precambrian Res. 66, 39-57.
an
Köhler, H., Dratch, V., Holzl, S., Fehr, T., 1991. Geochronology of Sri Lanka basement. Part II: Rb-Sr and Sm-Nd dating on mineral and whole rock samples. Terra Abstracts 3, 504.
te
d
M
Koizumi, T., Tsunogae, T., Santosh, M., Tsutsumi, Y., Chetty, T.R.K., Saitoh, Y., 2014. Petrology and zircon U-Pb geochronology of metagabbros from a mafic-ultramafic suite at Aniyapuram: Neoarchean to Early Paleoproterozoic convergent margin magmatism and Middle Neoproterozoic high-grade metamorphism in southern India. J. Asian Earth Sci., doi: 10.1016/j.jseaes.2014.04.013.
Ac ce p
Korhonen, F.J., Saw, A.K., Clark, C., Brown, M., Bhattacharya, S., 2011. New constraints on UHT metamorphism in the Eastern Ghats Province through the application of phase equilibria modelling and in situ geochronology. Gondwana Res. 20, 764-781. Kriegsman, L.M., 1991. Structural geology of the Sri Lankan basement – a preliminary review. Geological Survey Department, Sri Lanka, Professional Paper 5, 52-68. Kriegsman, L., 1994. Evidence for a fold nappe in the high-grade basement of central Sri Lanka: terrane assembly in the Pan-African lower crust? Precambrian Res. 66, 59-76. Kriegsman, L., 1995. The Pan-African events in East Antarctica: a review from Sri Lanka and the Mozambique Belt. Precambrian Res. 75, 263–277. Kriegsman, L.M., Schumacher, J.C., 1999. Petrology of sapphirine-bearing and associated granulites from central Sri Lanka. J. Petrol. 40, 1211–1239.
Page 68 of 111
P a g e | 69 Kroner, A and Jaeckel, P., 1994. Zircon ages from rocks of the Wanni Complex, Sri Lanka. J.Geol. Soc. Sri Lanka 5, 41-57.
ip t
Kröner, A., Williams, I.S., Compston, W., Baur, N., Vithanage, P.W., Perera., L.R.K., 1987. Zircon ion microprobe dating of high grade rocks in Sri Lanka. J. Geol. 95, 775-791.
cr
Kröner, A., Williams, I.S., Compston, W., Baur, N., Vithanage, P.W., Perera., L.R.K., 1987. Zircon ion microprobe dating of high grade rocks in Sri Lanka. J. Geol. 95, 775-791.
an
us
Kröner, A., Cooray, P.G., Vitanage, P.W., 1991. Lithotectonic subdivision of the Precambrian basement in Sri Lanka. Geological Survey Department, Sri Lanka, Professional Paper 5, 5-21. Kröner, A., Williams, I.S., 1993. Age of metamorphism in the high-grade rocks of Sri Lanka. J. Geol. 101, 513–521.
te
d
M
Kröner, A., Jaeckel, P., Williams, I.S., 1994. Pb-loss patterns in zircons from a high-grade metamorphic terrain as revealed by different dating methods: UPb and Pb-Pb ages for igneous and metamorphic zircons from northern Sri Lanka. Precambrian Res. 66, 151-181.
Ac ce p
Kröner, A., Kehelpannala, K.V.W., Hegner, A., 2003. Ca. 750–1100 Ma magmatic events and Grenville-age deformation in Sri Lanka: relevance for Rodinia supercontinent formation and dispersal, and Gondwana amalgamation. J. Asian Earth Sci. 22, 279-300. Kröner, A., Rojas-Agramonte, Y., Kehelpannala, K. V. W., Zack, T., Hegner, E., Geng, H. Y., Wong, J., Barth, M., 2013. Age, Nd–Hf isotopes, and geochemistry of the Vijayan Complex of eastern and southern Sri Lanka: A Grenville-age magmatic arc of unknown derivation, Precambrian Res. 234, 288-321. Li, Z.X., Bogdanova, S.V., Collins, A.S., Davidson, A., De Waele, B., Ernst, R.E., Fitzsimons, I.C.W., Fuck, R.A., Gladkochub, D.P., Jacobs, J., Karlstrom, K.E., Lu, S., Natapov, L.M., Pease, V., Pisarevsky, S.A., Thrane, K., Vernikovsky, V., 2008. Assembly, configuration, and break-up history of Rodinia: a synthesis. Precambrian Res. 160, 179–210. Liu, Y.S., Hu, Z.C., Gao, S., Gunther, D., Xu, J., Gao, C.G., Chen, H.H., 2008. In situ analysis of major and trace elements of anhydrous minerals by LA-ICPMS without applying an internal standard. Chemical Geology 257, 34–43.
Page 69 of 111
P a g e | 70
Liu, Y.S., Gao, S., Hu, Z., Gao, C., Zong, K., Wang, D., 2010. Continental and oceanic crust recycling-induced melt-peridotite interactions in the Trans-North China Orogen: U-Pb dating, Hf isotopes and trace elements in zircons of mantle xenoliths. Journal of Petrology 51, 537–571.
ip t
Ludwig, K.R., 2003. ISOPLOT 3.00: A Geochronological Toolkit for Microsoft Excel. Berkeley Geochronology Center, Special Publications 1a.
cr
Malaviarachchi, S., Takasu, A. 2005. Petrographic study on the high-grade metamorphic rocks from the Highland and Kadugannawa Complexes, central Sri Lanka. Geoscience Reports of Shimane University, Japan, 24, 31-46.
an
us
Malaviarachchi, S.P.K., Takasu, A., 2011a. Petrology of metamorphic rocks from the Highland and Kadugannawa Complexes, Sri Lanka. J. Geol. Soc. Sri Lanka 14 103-122. Malaviarachchi, S.P.K., Takasu, A., 2011b. Electron Microprobe dating of Monazites from Sri Lanka. J. Geol. Soc. Sri Lanka 14, 81-90.
d
M
Mathavan, V, W. K. B. N., Prame, W.K.B.N., Cooray, P.G., 1999. Geology of the high grade Proterozoic terrains of Sri Lanka and the assembly of Gondwana: an update on recent developments, Gondwana Res. 2, 237-250.
te
Mathavan, V., Fernando G.W.A.R., 2001. Reactions and textures in grossularwollastonite-scapolite calc-silicate granulites from Maligawila, Sri Lanka. Lithos, 59, 217-232.
Ac ce p
McDonough, W.F., Sun, S.-S., 1995. The composition of the Earth. Chemical Geology 120, 223-253. Milisenda, C.C., 1991. Compositional characteristics of the Vijayan Complex. Geological Survey Department, Sri Lanka, Professional Paper 5, 135-140. Milisenda, C.C., Pohl, J.R., Hofmann, A.W., 1991. Charnockite formation at Kurunegala, Sri Lanka. Geological Survey Department, Sri Lanka, Professional Paper 5, 141-149. Milisenda, C.C, Leiw, T.C., Hofmann, A.W., Kröner, A., 1988. Isotopic mapping of age provinces in Precambrian high grade terrains: Sri Lanka. J. Geol. 96, 608-615. Milisenda, C.C., Leiw, T.C., Hofmann, A.W., Köhler, H., 1994. Nd isotopic mapping of the Sri Lanka basement: update, and additional constraints from Sr isotopes. Precambrian Res. 66, 95-110.
Page 70 of 111
P a g e | 71 Moecher, D.P., Essene, E.J., Anovitz, L.M., 1988. Calculation and application of clinopyroxene-garnet-plagioclase-quartz geobarometers. Contrib. Mineral. Petrol. 100, 92–106.
ip t
Osanai, Y., Sajeev, K., Owada, M., Kehelpannala, K.V.W., Prame, W.K.B. Nakano, N., Jayatileke, S., 2006. Metamorphic evolution of ultrahightemperature and high-pressure granulites from Highland Complex, Sri Lanka. J. Asian Earth Sci. 28, 20-37.
cr
Pathirana, H.D.N.C., 1980. Geology of Sri Lanka in relation to plate tectonics. J. National Science Council Sri Lanka. 8, 75-85.
us
Pearce, J.A., Harris, N.B.W., Tindle, A.G., 1984. Trace element discrimination diagrams for the tectonic interpretation of granitic rocks. J. Petrol. 25, 956– 983.
an
Perchuk, L.L., Safonov, O.G., Gerya, T.V., Fu, B., Harlov, D.E., 2000. Mobility of components in metasomatic transformation and partial melting of gneisses: an example from Sri Lanka. Contrib. Mineral. Petrol. 140, 212-232.
M
Perkins, D., Newton, R.C., 1981. Charnockite geobarometers based on coexistence garnet-pyroxene-plagioclase-quartz. Nature 292, 144–146.
te
d
Perera, L.R.K., 1983. The origin of pink granites of Sri Lanka – another view. Precambrian Res. 20, 17-37.
Ac ce p
Perera, L.R.K., 1984. Co-existing cordierite-almandine-a key to the metamorphic history of Sri Lanka. Precambrian Res. 25, 349-364. Perera, L.R.K., 1987, Petrogenesis of granulite-facies metamorphic rocks in Sri Lanka. University of Peradeniya (Mphil. thesis) (unpubl.). Perera, L.R.K., Kagami, H., 2011. Centimetre- and metre-scale Nd and Sr isotopic homogenization in Kadugannawa Complex, Sri Lanka. J. Geol. Soc. Sri Lanka 14, 129-141. Pohl, J.G., Emmermann, R., 1991. Chemical composition of the Sri Lanka Precambrian Basement. Geological Survey Department, Sri Lanka, Professional Paper 5, 94-123. Prame, W.K.B.N, 1991a. Metamorphism and nature of the granulite facies crust in south western Sri Lanka: Characterization by pelitic/ psammopelitic rocks and associated granulites. Geological Survey Department, Sri Lanka, Professional Paper 5, 188-200. Prame, W.K.B.N., 1991b. Petrology of the Kataragama Complex, Sri Lanka:
Page 71 of 111
P a g e | 72 evidence for high pressure granulite facies metamorphism and subsequent isobaric cooling. Geological Survey Department, Sri Lanka, Professional Paper 5, 200-224.
ip t
Prame, W.K.B.N., 1997. Geochemistry and genesis of granitoid rocks from southern Sri Lanka. Min. Petrol. 60, 245–265.
cr
Raase, P., Schenk, V., 1994. Petrology of granulite facies metapelites of the Highland Complex, Sri Lanka: implication for the metamorphic zonation and the P-T path. Precambrian Res. 66, 265-294.
an
us
Saitoh, Y., Tsunogae, T., Santosh, M., Chetty, T.R.K., Horie, K., 2011. Neoarchean high-pressure metamorphism from the northern margin of the Palghat-Cauvery Suture Zone, southern India: petrology and zircon SHRIMP geochronology. J. Asian Earth Sci. 42, 268-285.
M
Sajeev, K., Osanai, Y., 2004a. Ultrahigh-temperature metamorphism (1150° C,12 kbar) and multistage evolution of Mg-, Al-rich granulites from the Central Highland Complex Sri Lanka. J. Petrol. 45, 1821–1844.
te
d
Sajeev, K., Osanai, Y., 2004b. 'Osumilite' and 'spinel+quartz' from Highland Complex, Sri Lanka: a case of cooling and decompression after ultrahightemperature metamorphism. J. Min. Petrol. Sci. 99, 320–327.
Ac ce p
Sajeev, K., Osanai, Y., 2005. Thermal gradients in the Sri Lankan granulite terrane: a garnet–biotite thermometric approach, J. Metamorph. Geol. 23, 383-397. Sajeev, K., Osanai, Y., Suzuki, S., Kagami, H., 2003. Geochronological evidence for multistage metamorphic events in ultra-high temperature granulites from central Highland Complex, Sri Lanka. Polar Geosci. 16, 137-148. Sajeev, K., Osanai, Y., Connolly, J.A.D., Suzuki, S., Ishioka, J., Kagami, H., Rino, S., 2007. Extreme Crustal Metamorphism during a Neoproterozoic Event in Sri Lanka: A study of dry mafic granulites. J. Geol. 115, 563–582. Sajeev, K., Windley, B.F., Connolly, J.A.D., Kon, Y., 2009. Retrogressed eclogite (20 kbar, 1020ºC) from the Neoproterozoic Palghat-Cauvery suture zone, southern India. Precambrian Res. 171, 23-36. Sajeev, K., Williams, I.S., Osanai, Y., 2010. Sensitive high-resolution ion microprobe U–Pb dating of prograde and retrograde ultrahigh-temperature metamorphism as exemplified by Sri Lankan granulites. Geology 38, 971– 974.
Page 72 of 111
P a g e | 73
Sandiford, N., Powell, R., Martin, S.F., Perera, L.R.K., 1988. Thermal and baric evolution of garnet granulites from Sri Lanka. J. Metamorph. Geol. 6, 351364.
cr
ip t
Santosh, M., Yoshida, M., Nanda-Kumar, V., 1991. Fluid characteristics across a gneiss-charnockite reaction front in Sri Lanka: Implication for granulite formation in Gondwana deep crust. J. Mineral. Petrol. Econnomic Geol. 86, 27-44.
us
Santosh, M., Tsunogae, T., Shimizu, H., Dubessy, J., 2010. Fluid characteristics of retrogressed eclogites and mafic granulites from the Cambrian Gondwana suture zone in southern India. Contrib. Mineral. Petrol. 159, 349-369.
an
Santosh, M., Maruyama, S., Sato, K., 2009. Anatomy of a Cambrian suture in Gondwana: Pacific-type orogeny in southern India? Gondwana Res. 16, 321341.
d
M
Santosh, M., Yang, Q.Y., Shaji, E., Tsunogae, T., Ram Mohan, M., Satyanarayanan, M., 2013a. An exotic Mesoarchean microcontinent: The Coorg Block, southern India. Gondwana Res., DOI: 10.1016/j.gr.2013.10.005
Ac ce p
te
Santosh, M., Liu, S.J., Tsunogae, T., Li, J.H., 2013b. Paleoproterozoic ultrahightemperature granulites in the North China Craton: Implications for tectonic models on extreme crustal metamorphism. Precambrian Res. 222-223, 77106. Schenk, V., Raase, P., Schumacher, R., 1988. Very high temperatures and isobaric cooling before tectonic uplift in the Highland Series of Sri Lanka. Terra Cognita 8, 265. Schenk, V., Raase, P., Schumacher, R., 1991. Metamorphic zonation and P–T history of the Highland Complex in Sri Lanka. Geological Survey Department, Sri Lanka, Professional Paper 5, pp. 150–163. Schumacher, R., Shenk, V., Raase, P., Vithanage, P.W., 1990. Granulite facies metamorphism of metabasic and intermediate rocks in the Highland Series of Sri Lanka. In: Ashworth, J.R., Brown, M. (Eds.), High temperature metamorphism and Crustal Anatexis. The Mineralogical Society series 2, 235271. Schumacher ,R., Faulhaber, S., 1994. Summary and discussion of P-T estimates from garnet-pyroxene-plagioclase-quartz bearing granulite facies rocks from Sri Lanka. Precambrian Res. 66, 295-308.
Page 73 of 111
P a g e | 74
Shiraishi, K., Ellis, D.J. ,Hiroi, Y., Fanning, C.M., Motoyoshi, Y., Nakai, Y., 1994. Cambrian orogenic belt in east Antarctica and Sri Lanka: implications for Gondwana assembly. J. Geol. 102, 47-65.
ip t
Stosch, H.G., 1991. Geochemistry of the mafic intrusion in the synforms of the Kandy area. Geological Survey Department, Sri Lanka, Professional Paper 5, 125-134.
us
cr
Sun, S.-S., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In Saunders, A.D., Norry M.J. (Eds.), Magmatism in the Ocean Basins. Geol. Soc. London, Spec. Publ. 42, 313-345
an
Tani, Y., Yoshida, M., 1996. The structural evolution of the Arena Gneisses and its bearing on Proterozoic tectonics of Sri Lanka. J. Southeast Asian Earth Sci. 14, 309-329.
M
Vervoort, J.D., Patchett, P.J., Gehrels, G.E., Nutman, A. P., 1996. Constraints on early Earth differentiation from hafnium and neodymium isotopes. Nature 379, 624-627.
te
d
Voll, G., Evangelakakisb, C., Kroll, H., 1994. Revised two-feldspar geothermometry applied to Sri Lankan feldspars. Precambrian Res. 66, 351367.
Ac ce p
Voll, G., Kleinschrodt, R., 1991. Sri Lanka: Structural, magmatic and metamorphic development of a Gondwana fragment. Geological Survey Department, Sri Lanka, Professional Paper 5, 22-52. Vitanage, P.W., 1972. Post-Precambrian uplift and regional neotectonic movements in Ceylon. 24th Int. Geol. Congr., Montreal, Sect. 3, 642-654. Vitanage, P.W., 1985. Tectonics and mineralization in Sri Lanka. Bull. Geol. Soc. Finland 57, 157-168. Wiedenbeck, M., Alle, P., Corfu, F., Griffin, W.L., Meier, M., Oberli, F., Quadt, A.V., Roddick, J.C., Spiegel, W., 1995. Three natural zircon standards for U-Th-Pb, Lu-Hf, trace element and REE analyses. Geostandards and Geoanalytical Research 19, 1-23. Willbold, M., Hegner, E., Kleinschrodt, R., Stosch, H.-G., Kehelpannala, K.V.W., Dulski, P., 2004. Geochemical evidence for a Neoproterozoic magmatic continental margin in Sri Lanka-relevence for the Rodinia-Gondwana
Page 74 of 111
P a g e | 75 supercontinental cycle. Precambrian Res. 130, 185-198. Xiao, W.J., Santosh, M., 2014. The western Central Asian Orogenic Belt: A window to accretionary orogenesis and continental growth. Gondwana Res. 25, 1429-1444.
an
us
cr
ip t
Yoshida, M., Funaki, M., Vitanage, P.W., 1992. Proterozoic to Mesozoic East Gondwana: The juxtaposition of India, Sri Lanka, and Antarctica. Tectonics 11, 381-391.
M
Figure captions
Fig. 1 Generalized geological and tectonic framework of Sri Lanka showing the
d
major crustal blocks and their boundaries (after Cooray, 1994). The study area is
te
shown in the Fig. 2 is marked by box. Sample localities are shown by open
Ac ce p
circles. (Sanjeewa)
Fig. 2 Detailed geological map and sample locations of the study areas (Figure modified after Kröner et al. 2003).
Fig. 3 Representative field photographs of locations KDG-4, KDG-1, KDG-2 from the Kadugannawa Complex and WC-2 from the Wanni Complex.
(a)
Metagranodiorite;
(d)
(b)
metadiorite;
(c)
garnet-bearing
amphibolite;
metagranodiorite with patches of incipient charnockite. See text for discussion.
Page 75 of 111
P a g e | 76 Fig. 4 Representative field photographs of locations HC-1, HC-11, HC-7 and HC10 from the Highland Complex. (a) Blocks and boudins of mafic granulite within
garnet-bearing charnockite. See text for discussion.
ip t
marble; (b) mafic sills in metasediments; (c) garnet-bearing metagabbro; (d)
cr
Fig. 5. Photomicrographs showing textures of representative samples. (a) A
us
typical granoblastic texture of hornblende + biotite + quartz + plagioclase + Kfeldspar in metagranodiorite (sample WC-2). (b) Metagranodiorite from the
an
Kadugannawa Complex (sample KDG-4). (c) Orthopyroxene-bearing and Kfeldspar-absent metadiorite (sample KDG-1). (d) Porphyroblastic garnet in the
M
matrix of fine-grained hornblende and plagioclase in garnet amphibolite (sample
d
KDG-2). (e) Medium-grained foliated pyroxene + hornblende + plagioclase
te
assemblage in mafic granulite (sample HC-1). (f) Equigranular texture of
Ac ce p
pyroxene + hornblende + plagioclase in mafic granulite (sample HC-11).
Fig. 5 (continued). (g) Orthopyroxene + plagioclase symplectite present as grayish spots in sample HC-11 as an evidence of decomposition of garnet during exhumation. (h) Coarse-grained massive metagabbro with subidioblastic garnet, clinopyroxene, and minor orthopyroxene (sample HC-7). Plagioclase and ilmenite fill the matrix of the minerals. (i) Medium-grained garnet and quartz which are separated by orthopyroxene + plagioclase corona in metagabbro (sample HC-8). (j) Granoblastic plagioclase + clinopyroxene + orthopyroxene + hornblende with symplectitic aggregates of orthopyroxene + plagioclase garnet charnockite
Page 76 of 111
P a g e | 77 (sample HC-10).
Fig. 6. Compositional diagrams showing chemistry of representative minerals. (a) (b)
Compositional
diagrams
showing
Ca/(Fe+Mg+Ca+Mn)
and
ip t
and
Mn/(Fe+Mg+Ca+Mn) versus XMg of garnet. (c) Wollastonite-enstatite-ferrosilite
cr
diagram showing compositions of clinopyroxene and clinopyroxene. (d) Anorthite-
us
albite-orthoclase diagram showing compositions of plagioclase. (e) Si (pfu) versus XMg diagram showing compositions of calcic amphibole. (f) XMg versus
an
TiO2 (wt.%) diagram showing biotite chemistry. GR: metagranodiorite, GD:
M
metadiorite, GA: garnet amphibolite, MG: mafic granulite, MGB: metagabbro.
d
Fig. 7. Major element variation diagrams showing concentrations of major
te
elements in the examined samples.
Ac ce p
Fig. 8. (a) to (c) Primitive mantle-normalized multi-element variation diagrams for magnesian (a), ferroan (b), and felsic (c) groups. Normalizing values are from Sun and McDonough (1989). (d) to (f) Chondrite-normalized REE spider plots for magnesian (d), ferroan (e), and felsic (f) groups. Normalizing values are from McDonough and Sun (1995).
Fig. 9. Discrimination diagrams for granites based on Rb-Y-Nb and Rb-Yb-Ta variations after Pearce et al. (1984). (a) Nb-Y diagram. (b) Ta-Yb diagram. (c) RbY+Nb diagram. (d) Rb-Yb+Ta diagram. VAG: volcanic-arc granites, syn-COLG:
Page 77 of 111
P a g e | 78 syn-collisional granites, WPG: within plate granites, ORG: ocean-ridge granites
Fig. 10. Triangular diagrams showing compositions of mafic granulites and
ip t
amphibolite. (a) TiO2-MnOx10-P2O5x10 diagram. OIT: oceanic-island tholeiite or seamount tholeiite, OIA: oceanic-island alkali basalt or seamount alkali basalt,
cr
CAB: island-arc calc-alkaline basalt, IAT: island-arc tholeiite, Bon: boninite. (b)
us
2Nb-Zr/4-Y diagram. AI: within-plate alkali basalt, AII: within-plate alkali basalt and within-plate alkali tholeiite, B: E-type MORB, C: within-plate tholeiite and
an
volcanic-arc basalt, D: N-type MORB and volcanic-arc basalt. (c) Y/15-La/10Nb/8 diagram. 1A: calc-alkali basalt, 1C: volcanic arc tholeiite, 1B: an area of
M
overlap between 1A and 1C. 2A: continental basalt, 2B: back-arc basin basalt,
d
3A: alkali basalt from intercontinental rift, 3B: enriched E-MORB, 3C: weakly
te
enriched E-MORB, 3D: N-MORB.
Ac ce p
Fig. 11 Cathodoluminescence (CL) images of representative zircons in samples HC -1, HC- 7, and HC- 10. Analytical points (white circles for U-Pb and red circles for Lu-Hf), ages in Ma and εHf (t) values are also shown.
Fig. 12 Cathodoluminescence (CL) images of representative zircons in samples HC-11, KDG-1, and KDG-2. Analytical points (white circles for U-Pb and red circles for Lu-Hf), ages in Ma and εHf (t) values are also shown.
Fig. 13 Cathodoluminescence (CL) images of representative zircons in samples
Page 78 of 111
P a g e | 79 KDG-4 and WC-2. Analytical points (white circles for U-Pb and red circles for LuHf), ages in Ma and εHf (t) values are also shown.
ip t
Fig. 14 Zircon U-Pb concordia plots, age data histograms and probability curves
cr
for samples HC-1, HC-7.
us
Fig. 15 Zircon U-Pb concordia plots, age data histograms and probability curves
an
for sample HC-10, HC-1.
Fig. 16 Zircon U-Pb concordia plots, age data histograms and probability curves
d
M
for samples KDG-1, KDG-2, KDG-4.
te
Fig. 17 Zircon U-Pb concordia plots, age data histograms, probability curves and bar charts for sample WC-2.
Ac ce p
Fig. 18 REE patterns of zircons in samples HC-1, HC- 7, HC-10, HC-11. Fig. 19 REE patterns of zircons in samples KDG-1, KDG-2, KDG-4, WC-2. Fig. 20 εHf (t) versus
207
Pb/206Pb age diagram of zircons from the magmatic
suites of Sri Lanka.
Fig. 21 Plate tectonic cartoons illustrating the model of Neoproterozoic doublesided subduction (a) culminating in Cambrian collision (b) proposed in this study. See text for details.
Page 79 of 111
P a g e | 80 Table captions
Table 1 Location, rock type and mineralogy of samples from the crustal blocks in
ip t
Sri Lanka analysed in this study.
Table 8 LA-MC-ICPMS Lu-Hf isotope data on zircons from the magmatic suites
Ac ce p
te
d
M
an
us
cr
of Sri Lanka.
Page 80 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Graphical Abstract (for review)
Page 81 of 111
Table-1
HC-11 WC-2
cr
metagabbro
Cpx Grt Opx Pl Ilm Mag Hbl Zrn
o
o
metagabbro
Pl Cpx Grt Qtz Opx Ilm Mag Rt Zrn
o
o
garnet charnockite
Pl Qtz Cpx Opx Hbl Ilm Py Cal Zrn Bt
o
o
mafic granulite
Pl Cpx Opx Hbl Ilm Zrn
o
o
metagranodiorite
Pl Kfs Qtz Bt Hbl Ilm Zrn Ap
N 7 15' 59.9"; E 80 37' 58.4"
Hantana
N 7 15' 00.1"; E 80 37' 59.0"
Hantana
N 7 12' 22.3"; E 80 47' 54.9"
Hakurutale 2km SE from Kivullinda JCT Udadigana
Pl Cpx Hbl Opx Ilm Mag Zrn
o
us
HC-10
mafic granulite
o
N 7 14' 21.0"; E 80 43' 48.1" N 7 30' 09.3"; E 80 20' 43.3"
an
HC-8
Pl Kfs Hbl Bt Qtz Mag Zrn
o
M
HC-7
metagranodiorite
o
N 7 18' 07.4"; E 80 44' 12.1"
Diagana
Hbl Pl Grt Ilm Mag Ap Zrn Qtz
ed
HC-1
garnet amphibolite
pt
KDG-4
o o Mahaberiyatenna Airport site N 7 16' 36.8"; E 80 45' 34.3" N 7o 22' 47.2"; E 80o 31' 51.8" Maussawa
Ac ce
KDG-2
ip t
Table 1 Location, rock type and mineralogy of samples from the crustal blocks in Sri Lanka analysed in this study. U-Pb geochronology and Lu-Hf isotope analysis Sample No. Locality Co-ordinates Rock type Assemblage o o N 7 16' 44.1"; E 80 40' 49.1" KDG-1 Kundasale metadiorite Pl Hbl Bt Qtz Ilm Opx Py Ap Zrn
Page 82 of 111
Yb/177Hf
176
Lu/177Hf
176
Hf/177Hf
WC-2-01
608.0
0.034894
0.001226
0.282199
WC-2-02
608.0
0.021266
0.000788
0.282249
WC-2-03
608.0
0.020237
0.000744
0.282257
WC-2-04
608.0
0.018627
0.000675
0.282168
WC-2-05
608.0
0.027675
0.000978
WC-2-06
608.0
0.016021
WC-2-07
608.0
WC-2-08 WC-2-09
1s
176
Hf/177Hfi
εHf(0)
εHf(t)
TDM (Ma)
TDMC (Ma)
fLu/Hf
0.282185
-20.3
-7.4
1495
2018
-0.96
0.000011
0.282240
-18.5
-5.4
1409
1895
-0.98
0.000011
0.282249
-18.2
-5.1
1395
1875
-0.98
0.000010
0.282161
-21.3
-8.2
1516
2071
-0.98
0.282292
0.000009
0.282281
-17.0
-4.0
1355
1803
-0.97
0.000569
0.282262
0.000012
0.282256
-18.0
-4.9
1382
1860
-0.98
0.015693
0.000541
0.281985
0.000012
0.281979
-27.8
-14.7
1762
2475
-0.98
608.0
0.027203
0.000977
0.282197
0.000012
0.282186
-20.3
-7.3
1487
2014
-0.97
608.0
0.028383
0.001019
0.282255
0.000010
0.282243
-18.3
-5.3
1409
1888
-0.97
WC-2-10
608.0
0.028674
0.000987
0.282172
0.000011
0.282161
-21.2
-8.2
1523
2070
-0.97
WC-2-12
608.0
0.018053
0.000683
0.282233
0.000010
0.282225
-19.1
-6.0
1427
1928
-0.98
WC-2-13
608.0
0.035602
0.001250
0.282257
0.000011
0.282243
-18.2
-5.3
1414
1888
-0.96
WC-2-14
608.0
0.014895
0.000568
0.282273
0.000008
0.282267
-17.6
-4.5
1367
1834
-0.98
WC-2-15
608.0
0.045200
0.001550
0.282143
0.000014
0.282125
-22.2
-9.5
1587
2149
-0.95
WC-2-16
608.0
0.019193
0.000718
0.282256
0.000012
0.282248
-18.2
-5.1
1396
1876
-0.98
WC-2-17
608.0
0.029196
0.001037
0.282291
0.000011
0.282279
-17.0
-4.0
1359
1807
-0.97
WC-2-18
608.0
0.020072
0.000761
0.282205
0.000010
0.282196
-20.1
-7.0
1469
1993
-0.98
608.0
0.033156
0.001157
0.282253
0.000011
0.282240
-18.4
-5.4
1417
1895
-0.97
608.0
0.019062
0.000709
0.282288
0.000010
0.282280
-17.1
-4.0
1351
1805
-0.98
608.0
0.015912
0.000597
0.282251
0.000009
0.282244
-18.4
-5.3
1398
1885
-0.98
608.0
0.030750
0.001084
0.282255
0.000010
0.282243
-18.3
-5.3
1410
1887
-0.97
608.0
0.013215
0.000543
0.282282
0.000009
0.282276
-17.3
-4.2
1354
1814
-0.98
WC-2-24
608.0
0.018073
0.000667
0.282273
0.000011
0.282265
-17.6
-4.5
1371
1838
-0.98
WC-2-25
608.0
0.013856
0.000551
0.282289
0.000009
0.282283
-17.1
-3.9
1344
1799
-0.98
WC-2-26
608.0
0.021998
0.000825
0.282281
0.000012
0.282272
-17.4
-4.3
1365
1824
-0.98
WC-2-20 WC-2-21 WC-2-22 WC-2-23
d
ep te
Ac c
WC-2-19
an
0.000009
M
Age (Ma)
us
176
No.
cr
ip t
Table-8
Page 83 of 111
ip t 0.001009
0.282174
WC-2-28
608.0
0.025933
0.000933
0.281979
WC-2-29
608.0
0.017038
0.000656
0.282261
WC-2-31
608.0
0.016344
0.000586
0.282266
WC-2-32
608.0
0.024456
0.000898
0.282258
WC-2-33
608.0
0.018576
0.000708
0.282240
HC-11-01
539.1
0.000998
0.000031
HC-11-02
539.1
0.000446
0.000013
HC-11-04
539.1
0.000477
0.000014
HC-11-05
539.1
0.000516
HC-11-06
539.1
HC-11-07
0.000011
cr
0.027214
0.282163
us
608.0
-21.1
-8.2
1521
2067
-0.97
0.281968
-28.0
-15.0
1789
2498
-0.97
0.000010
0.282254
-18.1
-4.9
1387
1864
-0.98
0.000010
0.282259
-17.9
-4.8
1378
1852
-0.98
0.000015
0.282247
-18.2
-5.2
1400
1878
-0.97
0.000012
0.282232
-18.8
-5.7
1417
1912
-0.98
0.282058
0.000012
0.282057
-25.3
-13.4
1641
2346
-1.00
0.282028
0.000010
0.282028
-26.3
-14.5
1681
2412
-1.00
0.282059
0.000010
0.282059
-25.2
-13.4
1639
2343
-1.00
0.000015
0.282043
0.000010
0.282043
-25.8
-13.9
1660
2377
-1.00
0.001572
0.000054
0.282058
0.000013
0.282057
-25.3
-13.4
1642
2346
-1.00
539.1
0.000503
0.000015
0.282061
0.000010
0.282060
-25.2
-13.3
1636
2339
-1.00
HC-11-08
539.1
0.000441
0.000013
0.282057
0.000010
0.282057
-25.3
-13.4
1641
2346
-1.00
HC-11-09
539.1
0.000124
0.000003
0.282004
0.000010
0.282004
-27.2
-15.3
1712
2464
-1.00
HC-11-10
539.1
0.000987
0.000031
0.282073
0.000012
0.282072
-24.7
-12.9
1621
2313
-1.00
HC-11-11
539.1
0.000487
0.000017
0.282081
0.000010
0.282081
-24.4
-12.6
1608
2293
-1.00
HC-11-12
539.1
0.000689
0.000020
0.282071
0.000010
0.282071
-24.8
-12.9
1622
2315
-1.00
HC-11-14
539.1
0.000309
0.000009
0.282057
0.000009
0.282057
-25.3
-13.4
1641
2346
-1.00
539.1
0.000939
0.000029
0.282067
0.000010
0.282067
-24.9
-13.1
1628
2324
-1.00
539.1
0.000741
0.000021
0.282079
0.000010
0.282079
-24.5
-12.7
1611
2298
-1.00
539.1
0.000060
0.000002
0.281999
0.000010
0.281999
-27.3
-15.5
1719
2475
-1.00
539.1
0.000902
0.000028
0.282095
0.000011
0.282095
-23.9
-12.1
1590
2263
-1.00
539.1
0.000130
0.000004
0.282024
0.000011
0.282024
-26.5
-14.6
1686
2421
-1.00
HC-11-20
539.1
0.001070
0.000035
0.281947
0.000011
0.281947
-29.2
-17.3
1791
2591
-1.00
HC-11-21
539.1
0.000737
0.000022
0.282072
0.000010
0.282072
-24.8
-12.9
1621
2314
-1.00
HC-11-22
539.1
0.009487
0.000325
0.281860
0.000010
0.281857
-32.2
-20.5
1922
2790
-0.99
HC-11-16 HC-11-17 HC-11-18 HC-11-19
d
ep te
Ac c
HC-11-15
an
0.000012
M
WC-2-27
Page 84 of 111
ip t 0.000496
0.282002
HC-10-03
533.0
0.000208
0.000006
0.282106
HC-10-04
533.0
0.004298
0.000152
0.282137
HC-10-05
533.0
0.004251
0.000164
0.282015
HC-10-06
533.0
0.000561
0.000015
0.282145
HC-10-07
533.0
0.004830
0.000186
HC-10-08
533.0
0.006191
HC-10-09
533.0
0.004883
HC-10-11
533.0
0.000161
0.000005
HC-10-12
533.0
0.012997
HC-10-13
533.0
HC-10-14
0.000019
cr
0.013494
0.281997
us
533.0
-27.2
-15.7
1737
2483
-0.99
0.282106
-23.6
-11.8
1575
2242
-1.00
0.000013
0.282136
-22.4
-10.8
1537
2175
-1.00
0.000010
0.282013
-26.8
-15.1
1705
2447
-1.00
0.000011
0.282144
-22.2
-10.5
1522
2156
-1.00
0.281993
0.000011
0.281991
-27.6
-15.9
1736
2497
-0.99
0.000241
0.281993
0.000011
0.281990
-27.6
-15.9
1738
2498
-0.99
0.000178
0.281933
0.000019
0.281931
-29.7
-18.0
1816
2629
-0.99
0.282017
0.000011
0.282017
-26.7
-15.0
1694
2438
-1.00
0.000450
0.281765
0.000022
0.281760
-35.6
-24.1
2059
3007
-0.99
0.002461
0.000099
0.282142
0.000018
0.282141
-22.3
-10.6
1529
2163
-1.00
533.0
0.008187
0.000293
0.282114
0.000018
0.282111
-23.3
-11.6
1575
2229
-0.99
HC-10-15
533.0
0.000393
0.000011
0.282120
0.000010
0.282119
-23.1
-11.4
1556
2212
-1.00
HC-10-16
533.0
0.006434
0.000236
0.281929
0.000015
0.281927
-29.8
-18.2
1824
2639
-0.99
HC-10-17
533.0
0.004681
0.000178
0.281927
0.000011
0.281925
-29.9
-18.3
1825
2644
-0.99
HC-10-18
533.0
0.034608
0.001208
0.281676
0.000013
0.281664
-38.8
-27.5
2222
3218
-0.96
HC-10-19
533.0
0.005998
0.000237
0.281502
0.000016
0.281500
-44.9
-33.3
2401
3580
-0.99
HC-10-20
533.0
0.007711
0.000309
0.281694
0.000013
0.281691
-38.1
-26.5
2147
3160
-0.99
533.0
0.030751
0.001051
0.281673
0.000011
0.281663
-38.9
-27.5
2217
3221
-0.97
533.0
0.008653
0.000329
0.281826
0.000012
0.281823
-33.5
-21.9
1969
2870
-0.99
533.0
0.018805
0.000679
0.281850
0.000011
0.281844
-32.6
-21.1
1953
2823
-0.98
533.0
0.002831
0.000107
0.281977
0.000010
0.281976
-28.1
-16.5
1754
2531
-1.00
HC-10-22 HC-10-23 HC-10-24 HC-10-25
d
ep te
Ac c
HC-10-21
an
0.000011
M
HC-10-01
533.0
0.004084
0.000161
0.281972
0.000010
0.281970
-28.3
-16.6
1763
2543
-1.00
HC-10-26
533.0
0.003430
0.000131
0.282035
0.000010
0.282034
-26.0
-14.4
1675
2401
-1.00
HC-10-27
533.0
0.003593
0.000141
0.282029
0.000011
0.282028
-26.3
-14.6
1684
2415
-1.00
HC-7-01
543.0
0.001018
0.000031
0.282260
0.000011
0.282260
-18.1
-6.2
1366
1892
-1.00
Page 85 of 111
ip t 0.000022
0.282279
HC-7-05
543.0
0.001113
0.000033
0.282251
HC-7-06
543.0
0.000665
0.000018
0.282254
HC-7-07
543.0
0.000549
0.000016
0.282262
HC-7-08
543.0
0.001852
0.000056
0.282280
HC-7-09
543.0
0.000577
0.000017
0.282255
HC-7-11
543.0
0.000873
0.000024
HC-7-13
543.0
0.000688
0.000020
HC-7-14
543.0
0.000992
0.000030
HC-7-15
543.0
0.001226
HC-7-16
543.0
HC-1-01
0.000010
cr
0.000762
0.282279
us
543.0
-17.4
-5.5
1340
1850
-1.00
0.282251
-18.4
-6.5
1378
1912
-1.00
0.000010
0.282254
-18.3
-6.4
1374
1905
-1.00
0.000011
0.282262
-18.0
-6.1
1363
1888
-1.00
0.000013
0.282279
-17.4
-5.5
1340
1849
-1.00
0.000011
0.282254
-18.3
-6.4
1373
1904
-1.00
0.282280
0.000010
0.282280
-17.4
-5.5
1338
1847
-1.00
0.282255
0.000010
0.282255
-18.3
-6.4
1372
1903
-1.00
0.282233
0.000012
0.282232
-19.1
-7.1
1403
1953
-1.00
0.000040
0.282236
0.000012
0.282236
-19.0
-7.0
1399
1946
-1.00
0.000874
0.000026
0.282221
0.000012
0.282221
-19.5
-7.5
1418
1978
-1.00
553.0
0.005983
0.000268
0.282368
0.000014
0.282365
-14.3
-2.2
1227
1651
-0.99
HC-1-02
553.0
0.006039
0.000256
0.282374
0.000010
0.282371
-14.1
-2.0
1218
1637
-0.99
HC-1-03
553.0
0.006708
0.000282
0.282373
0.000016
0.282370
-14.1
-2.0
1220
1639
-0.99
HC-1-04
553.0
0.007236
0.000298
0.282375
0.000014
0.282372
-14.1
-2.0
1218
1636
-0.99
HC-1-05
553.0
0.005477
0.000231
0.282380
0.000015
0.282378
-13.8
-1.8
1208
1621
-0.99
HC-1-06
553.0
0.005387
0.000228
0.282395
0.000014
0.282392
-13.3
-1.3
1188
1589
-0.99
HC-1-07
553.0
0.003809
0.000138
0.282417
0.000013
0.282416
-12.5
-0.4
1155
1537
-1.00
553.0
0.005521
0.000238
0.282390
0.000015
0.282388
-13.5
-1.4
1195
1600
-0.99
553.0
0.005921
0.000250
0.282393
0.000011
0.282390
-13.4
-1.3
1192
1594
-0.99
553.0
0.004044
0.000142
0.282416
0.000015
0.282414
-12.6
-0.5
1157
1540
-1.00
553.0
0.006902
0.000294
0.282427
0.000011
0.282424
-12.2
-0.1
1146
1519
-0.99
HC-1-09 HC-1-10 HC-1-11 HC-1-12
d
ep te
Ac c
HC-1-08
an
0.000012
M
HC-7-04
553.0
0.005637
0.000229
0.282404
0.000011
0.282401
-13.0
-0.9
1176
1569
-0.99
HC-1-13
553.0
0.007241
0.000308
0.282435
0.000012
0.282432
-11.9
0.1
1135
1501
-0.99
HC-1-14
553.0
0.006267
0.000270
0.282387
0.000012
0.282384
-13.6
-1.5
1200
1607
-0.99
HC-1-15
553.0
0.006233
0.000268
0.282407
0.000012
0.282404
-12.9
-0.8
1172
1562
-0.99
Page 86 of 111
ip t 0.000243
0.282429
HC-1-18
553.0
0.005871
0.000241
0.282399
KDG-4-01
980.0
0.045121
0.001538
0.282369
KDG-4-02
980.0
0.030845
0.001091
0.282290
KDG-4-03
980.0
0.012870
0.000482
0.282338
KDG-4-04
980.0
0.030585
0.001150
KDG-4-05
980.0
0.042430
KDG-4-06
980.0
0.030300
KDG-4-07
980.0
0.024680
0.000915
KDG-4-08
980.0
0.035258
KDG-4-09
980.0
KDG-4-10
0.000015
cr
0.005770
0.282426
us
553.0
-12.1
-0.1
1142
1513
-0.99
0.282396
-13.2
-1.1
1183
1580
-0.99
0.000014
0.282341
-14.3
6.4
1266
1431
-0.95
0.000013
0.282270
-17.1
3.9
1363
1590
-0.97
0.000012
0.282329
-15.4
6.0
1275
1458
-0.99
0.282272
0.000011
0.282251
-17.7
3.2
1389
1631
-0.97
0.001499
0.282335
0.000014
0.282307
-15.5
5.2
1314
1506
-0.95
0.001114
0.282278
0.000011
0.282257
-17.5
3.5
1380
1618
-0.97
0.282246
0.000011
0.282230
-18.6
2.5
1417
1679
-0.97
0.001267
0.282295
0.000012
0.282271
-16.9
4.0
1362
1586
-0.96
0.027611
0.001038
0.282276
0.000013
0.282257
-17.5
3.5
1379
1618
-0.97
980.0
0.033971
0.001216
0.282278
0.000018
0.282255
-17.5
3.4
1384
1622
-0.96
KDG-4-11
980.0
0.051090
0.001764
0.282350
0.000017
0.282317
-14.9
5.6
1301
1483
-0.95
KDG-4-12
980.0
0.025094
0.000901
0.282274
0.000012
0.282257
-17.6
3.5
1378
1618
-0.97
KDG-4-13
980.0
0.035424
0.001379
0.282270
0.000012
0.282245
-17.7
3.0
1401
1645
-0.96
KDG-4-14
980.0
0.010098
0.000429
0.282289
0.000011
0.282281
-17.1
4.3
1340
1564
-0.99
KDG-4-15
980.0
0.029547
0.001075
0.282261
0.000012
0.282242
-18.1
2.9
1402
1653
-0.97
KDG-4-16
980.0
0.030448
0.001091
0.282273
0.000012
0.282253
-17.7
3.3
1386
1628
-0.97
980.0
0.021464
0.000856
0.282307
0.000010
0.282291
-16.5
4.7
1330
1542
-0.97
980.0
0.044247
0.001600
0.282279
0.000014
0.282250
-17.4
3.2
1396
1635
-0.95
980.0
0.043887
0.001672
0.282303
0.000012
0.282272
-16.6
4.0
1365
1584
-0.95
980.0
0.033846
0.001309
0.282305
0.000013
0.282281
-16.5
4.3
1349
1565
-0.96
KDG-4-18 KDG-4-19 KDG-4-20 KDG-4-21
d
ep te
Ac c
KDG-4-17
an
0.000012
M
HC-1-16
980.0
0.028177
0.001111
0.282271
0.000011
0.282250
-17.7
3.2
1390
1633
-0.97
KDG-4-22
980.0
0.033561
0.001188
0.282263
0.000012
0.282241
-18.0
2.9
1404
1653
-0.96
KDG-4-23
980.0
0.024356
0.000897
0.282253
0.000013
0.282236
-18.4
2.7
1407
1665
-0.97
KDG-4-24
980.0
0.075508
0.002571
0.282345
0.000013
0.282297
-15.1
4.9
1338
1529
-0.92
Page 87 of 111
ip t 0.001550
0.282287
KDG-4-26
980.0
0.021552
0.000854
0.282265
KDG-2-01
520.7
0.002531
0.000086
0.281977
KDG-2-02
520.7
0.000244
0.000006
0.282023
KDG-2-03
520.7
0.001201
0.000042
0.282020
KDG-2-04
520.7
0.000201
0.000005
KDG-2-05
520.7
0.003280
KDG-2-06
520.7
0.001328
KDG-2-08
520.7
0.001784
0.000070
KDG-2-09
520.7
0.004295
KDG-2-10
520.7
KDG-2-12
0.000017
cr
0.041104
0.282259
us
980.0
-17.1
3.5
1383
1615
-0.95
0.282249
-17.9
3.2
1389
1636
-0.97
0.000016
0.281976
-28.1
-16.7
1753
2538
-1.00
0.000012
0.282022
-26.5
-15.1
1687
2435
-1.00
0.000014
0.282019
-26.6
-15.2
1693
2442
-1.00
0.281996
0.000011
0.281996
-27.4
-16.0
1723
2493
-1.00
0.000136
0.281896
0.000011
0.281895
-31.0
-19.6
1864
2718
-1.00
0.000048
0.281933
0.000011
0.281933
-29.7
-18.3
1810
2634
-1.00
0.281892
0.000013
0.281892
-31.1
-19.7
1866
2725
-1.00
0.000162
0.281996
0.000015
0.281995
-27.4
-16.1
1730
2497
-1.00
0.000281
0.000007
0.281994
0.000012
0.281994
-27.5
-16.1
1726
2498
-1.00
520.7
0.001944
0.000065
0.281997
0.000013
0.281996
-27.4
-16.0
1725
2493
-1.00
KDG-2-13
520.7
0.000934
0.000030
0.282022
0.000011
0.282022
-26.5
-15.1
1689
2436
-1.00
KDG-2-14
520.7
0.001584
0.000048
0.282058
0.000016
0.282058
-25.2
-13.8
1641
2356
-1.00
KDG-2-15
520.7
0.001525
0.000047
0.282006
0.000009
0.282005
-27.1
-15.7
1712
2473
-1.00
KDG-2-16
520.7
0.002683
0.000108
0.281846
0.000015
0.281845
-32.7
-21.3
1931
2828
-1.00
KDG-2-17
520.7
0.000766
0.000021
0.282039
0.000013
0.282039
-25.9
-14.5
1666
2399
-1.00
KDG-2-18
520.7
0.000173
0.000004
0.282012
0.000011
0.282012
-26.9
-15.4
1701
2458
-1.00
520.7
0.003531
0.000133
0.281983
0.000013
0.281981
-27.9
-16.5
1747
2526
-1.00
520.7
0.000428
0.000011
0.281989
0.000013
0.281989
-27.7
-16.3
1733
2510
-1.00
755.0
0.003763
0.000131
0.282389
0.000012
0.282387
-13.5
3.0
1193
1471
-1.00
755.0
0.002483
0.000095
0.282474
0.000010
0.282473
-10.5
6.1
1075
1279
-1.00
755.0
0.003083
0.000120
0.282408
0.000011
0.282406
-12.9
3.7
1167
1429
-1.00
KDG-1-05
755.0
0.015456
0.000596
0.282344
0.000011
0.282336
-15.1
1.2
1270
1586
-0.98
KDG-1-06
755.0
0.028044
0.001131
0.282370
0.000014
0.282353
-14.2
1.9
1252
1547
-0.97
KDG-1-07
755.0
0.003021
0.000115
0.282507
0.000011
0.282505
-9.4
7.2
1031
1206
-1.00
KDG-2-20 KDG-1-01 KDG-1-03 KDG-1-04
d
ep te
Ac c
KDG-2-19
an
0.000011
M
KDG-4-25
Page 88 of 111
ip t 0.000677
0.282322
KDG-1-09
755.0
0.024695
0.000976
0.282313
KDG-1-10
755.0
0.013257
0.000540
0.282341
KDG-1-11
755.0
0.025974
0.001031
0.282348
KDG-1-12
755.0
0.020893
0.000864
0.282342
KDG-1-13
755.0
0.010112
0.000385
0.282353
KDG-1-14
755.0
0.014504
0.000592
KDG-1-15
755.0
0.014082
0.000570
KDG-1-16
755.0
0.021951
0.000844
KDG-1-17
755.0
0.017256
KDG-1-18
755.0
KDG-1-19
0.000014
cr
0.017363
0.282313
us
755.0
-15.9
0.4
1303
1638
-0.98
0.282300
-16.2
-0.1
1325
1667
-0.97
0.000011
0.282334
-15.2
1.2
1272
1591
-0.98
0.000013
0.282333
-15.0
1.1
1279
1592
-0.97
0.000012
0.282330
-15.2
1.0
1281
1599
-0.97
0.000013
0.282348
-14.8
1.6
1251
1560
-0.99
0.282337
0.000013
0.282329
-15.4
1.0
1279
1602
-0.98
0.282278
0.000013
0.282270
-17.5
-1.1
1360
1733
-0.98
0.282340
0.000012
0.282328
-15.3
0.9
1284
1605
-0.97
0.000737
0.282341
0.000012
0.282330
-15.3
1.0
1279
1599
-0.98
0.018840
0.000774
0.282336
0.000012
0.282325
-15.4
0.9
1287
1610
-0.98
755.0
0.007692
0.000315
0.282320
0.000014
0.282315
-16.0
0.5
1294
1633
-0.99
KDG-1-20
755.0
0.022287
0.000900
0.282337
0.000015
0.282324
-15.4
0.8
1290
1613
-0.97
KDG-1-21
755.0
0.015511
0.000630
0.282291
0.000012
0.282282
-17.0
-0.7
1344
1705
-0.98
KDG-1-22
755.0
0.025541
0.001046
0.282350
0.000014
0.282335
-14.9
1.2
1276
1587
-0.97
KDG-1-23
755.0
0.019186
0.000755
0.282370
0.000014
0.282360
-14.2
2.1
1238
1533
-0.98
KDG-1-24
755.0
0.017021
0.000682
0.282355
0.000012
0.282345
-14.7
1.6
1257
1565
-0.98
KDG-1-25
755.0
0.013953
0.000561
0.282291
0.000013
0.282283
-17.0
-0.6
1342
1704
-0.98
755.0
0.005558
0.000218
0.282343
0.000012
0.282340
-15.2
1.4
1259
1578
-0.99
755.0
0.021774
0.000862
0.282363
0.000012
0.282351
-14.5
1.7
1253
1553
-0.97
KDG-1-27
d
ep te
Ac c
KDG-1-26
an
0.000014
M
KDG-1-08
Page 89 of 111
Ac ce p
te
d
M
an
us
cr
ip t
Figure-1
Page 90 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-2
Page 91 of 111
Ac ce p
te
d
M
an
us
cr
ip t
Figure-3
Page 92 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-4
Page 93 of 111
Ac ce p
te
d
M
an
us
cr
ip t
Figure-5 part I
Page 94 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-5 part 2
Page 95 of 111
Figure-6
(b)
0.26
rim
KDG-1(GD) KDG-2 (GA) KDG-4 (GR) HC-1 (MG) HC-7 (MGB) HC-8 (MGB) HC-11 (MG) WC-2 (GR)
core
0.24
0.22
0.06
0.05
0.04
ip t
(a)
mantle
core
0.03 rim rim
0.02 rim
0.18 0.15
0.2
core
0.25
core
0.01 0.15
0.3
(d)
M
50
augite
Wo
endiopside
d
pigeonite
0.25
an
diopside hedenbergite 50
0.2
core
0.3
Mg/(Fe+Mg)
Mg/(Fe+Mg)
(c)
rim
rim
us
core
cr
0.2
mantle
50
50
Or
orthopyroxene 50
(e)
Ac ce pt e
En
Fs
(f)
0.65
50
Ab
An
6
CaB > 1.5, (Na+K)A > 0.5
5.5
0.6
0.55
0.5
0.45
0.4
5
pargasite
edenite
4.5
kaersutite
ferro-edenite
6.8
6.6
4
ferro-pargasite 6.4
Si (pfu)
6.2
6
3.5
0.4
0.45
0.5
0.55
0.6
0.65
Mg/(Fe+Mg)
Page 96 of 111
Figure-7
Al2O3
16
2
14
1
40
25
45
50
55
60
65
70
(c)
75
40
15
45
50
55
60
65
70
(d)
0.3
Fe2O3
20
0
MnO
0.2
10
10
45
50
55
60
65
70
75
0
40
14
(e)
MgO
8
45
50
8
40
45
50
55
60
65
75
2
40
45
5
(g)
4
70
55
60
(h)
Na2O
4
Ac ce pt e
3
3
50
70
75
65
70
75
70
75
70
75
K2O
d
0
4
M
6
2
65
CaO
10
4
60
(f)
12
6
55
an
40
us
0.1 5 0
75
cr
12
TiO2
(b)
3
ip t
(a)
18
2
2
1
1
40
2
45
(i)
50
55
Metagranodiorite
1.5
Metagabbro
HC-7 (ferroan)
KDG-4 (felsic)
HC-8 (magnesian)
70
75
P2O5
0
40
45
(j)
4
Mafic granulite
50
55
Grt amphibolite
Fe2O3/MgO
Ferroan
Felsic
Grt charnockite
2
HC-10 (magnesian)
Magnesian
1
40
45
50
65
HC-11 (magnesian)
KDG-2 (ferroan)
0
60
3
HC-1 (magnesian)
KDG-1 (magnesian)
0.5
65
WC-2 (felsic)
Metadiorite
1
60
55
60
SiO2 (wt.%)
65
70
75
40
45
50
55
60
65
SiO2 (wt.%)
Page 97 of 111
Figure-8
10 2
10 3 (d) Magnesian
10 2
1
10
10
Sr
K
Rb Ba Th Ta Nb Ce
P
Zr
Hf Sm Ti
Y
Yb
2
1
Ce
Pr
Nd Sm Eu Gd Tb
(e) Ferroan
10 2
Metagabbro
1
Grt amphibolite
Rb Ba Th Ta Nb Ce
P
Zr
Hf Sm Ti
Y
1
M
KDG-2 K
Er Tm Yb
Lu
10
HC-7
Sr
Dy Ho
an
10
10
La
10 3 (b) Ferroan
0.1
HC-8
us
0.1
Grt charnockite
HC-1 HC-10 HC-11 Metadiorite KDG-1 Metagabbro
ip t
10
Mafic granulite
cr
(a) Magnesian
Yb
3
La
Ce
Pr
Nd Sm Eu Gd Tb
Dy Ho
Er Tm Yb
Lu
Pr
Nd Sm Eu Gd Tb
Dy Ho
Er Tm Yb
Lu
10 3
(c) Felsic
(f) Felsic
Metagranodiorite
10
1
Sr
K
Ac ce pt e
10
d
WC-2 KDG-4
2
Rb Ba Th Ta Nb Ce
P
Zr
Hf Sm Ti
Y
Yb
10 2
10
1
La
Ce
Page 98 of 111
Figure-9
100
1000
Metagranodiorite WC-2 KDG-4
WPG
VAG + syn-COLG
syn-COLG 1
10
ORG
1
10
Y (ppm)
100
1000
0.1
0.1
1
1000
1000
syn-COLG WPG
10
10
100
an 1
1
10
100
Yb + Ta (ppm)
d
Y + Nb (ppm)
1000
ORG
Ac ce pt e
1
VAG
M
ORG
100
us
VAG
10
10
WPG
100
100
Yb (ppm)
cr
syn-COLG
1
ORG
VAG
ip t
1
WPG
Metadiorite KDG-1
10
100
Page 99 of 111
TiO2
ip t
Figure-10
(a)
cr
2Nb
MORB
M an
AI OIT
3D 1C
1B
2B
AII
3C
3B
B
IAT OIA
1A
C
Bon CAB
2A
3A
Y
La/10
Nb/8
ce pt
P2O5x10 Zr/4
ed
D
Ac
MnOx10
Y/15
(c)
us
(b)
Mafic granulite HC-1 (magnesian) HC-11 (magnesian) Metagabbro HC-7 (ferroan) HC-8 (magnesian) Grt amphibolite KDG-2 (ferroan) Grt charnockite HC-10 (magnesian)
Page 100 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-11
Page 101 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-12
Page 102 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-13
Page 103 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-14
Page 104 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-15
Page 105 of 111
Ac ce p
te
d
M
an
us
cr
ip t
Figure-16
Page 106 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-17
Page 107 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-18
Page 108 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-19
Page 109 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-20
Page 110 of 111
Ac ce p
te
d
M
an
us
cr
ip t
Figure-21 revised
Page 111 of 111
S0301-9268(14)00333-7 http://dx.doi.org/doi:10.1016/j.precamres.2014.09.017 PRECAM 4092
To appear in:
Precambrian Research
Received date: Revised date: Accepted date:
2-8-2014 13-9-2014 15-9-2014
Please cite this article as: Santosh, M., Tsunogae, T., Malaviarachchi, S.P.K., Zhang, Z., Ding, H., Tang, L., Dharmapriya, P.L.,Neoproterozoic crustal evolution in Sri Lanka: Insights from petrologic, geochemical and zircon U-Pb and Lu-Hf isotopic data and implications for Gondwana assembly, Precambrian Research (2014), http://dx.doi.org/10.1016/j.precamres.2014.09.017 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Page |1
Neoproterozoic crustal evolution in Sri Lanka: Insights from petrologic, geochemical and zircon U‐
ip t
Pb and Lu‐Hf isotopic data and implications for
cr
Gondwana assembly
us
an
M. Santosh1,2*, T. Tsunogae3,4, Sanjeewa P.K. Malaviarachchi5,6,
M
Zeming Zhang7, Huixia Ding7, Li Tang1, P.L. Dharmapriya6
1
School of Earth Science and Resources, China University of Geosciences Beijing, No. 29
d
Xueyuan Road, Haidian District, Beijing, 100083, China 2
780-8520, Japan 3
te
Division of Interdisciplinary Science, Faculty of Science, Kochi University, Akebono-cho, Kochi Faculty of Life and Environmental Sciences, University of Tsukuba, Ibaraki 305-8572, Japan
4
Ac ce p
Department of Geology, University of Johannesburg, Auckland Park 2006, South Africa
5
Department of Geology, Faculty of Science, University of Peradeniya, Peradeniya 20400, Sri Lanka 6 Postgraduate Institute of Science, University of Peradeniya, Peradeniya 20400, Sri Lanka 7
State Key Laboratory of Continental Tectonic and Dynamics, Institute of Geology, Academy of
Geological Science, No. 26 Baiwanzhuang Road, Beijing 100037, China
Corresponding author e-mail: [email protected]
Page 1 of 111
Page |2
ip t
Research Highlights
Neoproterozoic arc magmatism prior to Gondwana assembly in Sri
cr
Lanka
us
Crustal growth through both juvenile and reworked components Highland Complex is an accretionary suture
an
Double-sided subduction in Neoproterozoic and collision in
te
Abstract
d
M
Cambrian
Ac ce p
Sri Lanka, the ‘pendant’ of Gondwana, is a collage of distinct crustal blocks that preserve important records of major Neoproterozoic tectonothermal events.
Here we present the
petrology, geochemistry, zircon U-Pb geochronology and Lu-Hf isotopes on a suite of metaigneous rocks including granodiorite, diorite and garnet amphibolite from the Kadugannawa Complex (KC), granodiorite from the Wanni Complex (WC) and mafic granulites, gabbros and garnet-bearing charnockite from the Highland Complex (HC) along a NW – SE transect. The regional metamorphic peak P-T conditions were estimated from garnet-clinopyroxeneplagioclase-quartz assemblage in the metagabbro as 830-860°C and 9.4-9.8 kbar. Slightly lower temperature ranges of 700-780°C were obtained from garnet amphibolite, metagranodiorite and metadiorite, corresponding retrograde conditions. Trace element and rare earth element patterns as well as Rb-Y-Nb and Rb-Yb-Ta discrimination plots show volcanic arc affinity for the granodiorite, diorite and garnet charnockite suggesting that the protoliths of the rocks were formed from felsic to intermediate arc magmas. The mafic granulites and magnesian metagabbro also suggest volcanic arc affinity and indicate subduction-related mafic magmatism and magma underplating. The garnet-bearing metagabbro shows N-MORB signature, whereas the garnet
Page 2 of 111
Page |3 amphibolite displays oceanic island alkali basalt affinity. These rocks therefore represent accretion of the remnants of oceanic lithosphere during the subduction-collisional event. Zircons in a metadiorite and the surrounding metagranodiorite from the KC yield ages of 980 ± 16 Ma to 916 ± 57 Ma marking early Neoproterozoic magmatism followed by metamorphism at 532 ± Zircons in the garnet amphibolite from this complex show extensive metamorphic
recrystallization yielding a weighted mean
206
ip t
18 Ma.
Pb/238U age of 520.7 ± 6.6 Ma. From the WC,
zircons in a metagranodiorite define three groups of weighted mean
206
Pb/238U ages at 805 ± 12
cr
Ma (emplacement of the magmatic protolith), 734.0 ± 4.6 Ma (Cryogenian thermal event) and 546.0 ± 5.7 Ma (latest Neoproterozoic-Cambrian metamorphism). Zircons from the HC record
us
multiple late Neoproterozoic – Cambrian thermal events with weighted mean
206
Pb/238U ages of
576.8 ± 9.3 Ma and 523.0 ± 7.1 Ma (metagabbro) and 579 ± 10 Ma, 540.4 ± 6.0 Ma and 511.1 ± 5.9 Ma (garnet charnockite), 553.0 ± 3.2 Ma (mafic granulite), and 539.1 ± 4.4 Ma (mafic granulite Lu-Hf data reveal dominantly positive εHf(t) values for zircons in the metadiorite and
an
sill).
metagranodiorite from the KC (-1.1 to 7.2) and Hf crustal model ages (TDMC) in the range of 1206 to 1733 Ma suggesting a mixed source from both juvenile and Paleo-Mesoproterozoic
M
components. However, zircons in the garnet amphibolite from this complex show dominantly negative εHf(t) values (mean -16.5) with TDMC in the range of 2356 to 2828 Ma suggesting reworked Neoarchean-Paleoproterozoic crustal source. Zircons in metagranodiorite of the WC
d
also possess negative εHf(t) values (mean -6.2) with TDMC in the range of 1799 to 2498 Ma
te
suggesting reworked Paleoproterozoic crust as the magma source. The HC rocks also preserve distinct imprints of reworking of older crust. The zircon εHf(t) values in mafic granulite show a tight
Ac ce p
cluster from -2.2 to 0.1 with TDMC in the range of 1501-1651 Ma suggesting a mixed source from both juvenile Neoproterozoic and reworked Mesoproterozoic components. Zircons in the metagabbro from this complex show negative εHf(t) values (mean -6.3) and TDMC of 1847-1978 Ma. Zircons in the garnet charnockite also display highly negative εHf(t) values (mean -17.7) and older TDMC) (mean 2614 Ma) suggesting reworked Paleoproterozoic crust. Zircons in the mafic granulite sample show negative εHf(t) values (mean -14.1) and TDMC between 2263 to 2790 Ma indicating that the source material for the magma evolved from the Neoarchean-Paleoproterozoic crust. In summary, the 916 to 980 Ma ages from the KC represent arc magmatism during early Neoproterozoic, followed by the 805 Ma granodioritic magma emplacement in the WC. Repeated thermal events during mid and late Neoproterozoic are also recorded from the Wanni and Highland Complexes, culminating in Cambrian high-grade metamorphism that reached ultrahightemperature conditions. We propose a model of double-sided subduction during the Neoproterozoic, where the Wanni Complex to the west and the Vijayan Complex to the east represent continental arcs, culminating in collision along the HC during late NeoproterozoicCambrian.
Page 3 of 111
Page |4
Key words: Petrology and geochemistry; Zircon U-Pb geochronology and Lu-Hf isotopes; Arc magmatism; Neoproterozoic; Sri Lanka.
ip t
1. Introduction
cr
Crustal evolution models on both Archean and Phanerozoic Earth identify the
us
role of arc magmas derived from both juvenile and reworked older components (e.g., Condie and Kröner, 2013; Condie and Aster, 2013; Santosh et al., 2013a; In convergent margins, continental growth occurs
an
Xiao and Santosh, 2014).
through vertical addition of magmas derived by the melting of downgoing oceanic
M
slabs, as well as lateral accretion of oceanic and trench material, together with other exotic oceanic and crustal fragments.
For instance, the Central Asian
te
d
Orogenic Belt (CAOB) representing the largest Phanerozoic accretionary orogen on the globe is a typical example of continental growth through lateral accretion
Ac ce p
of young arc complexes and old micro-continents together with vertical growth through underplating of mantle-derived magmas (e.g., Xiao and Santosh, 2014, and references therein). Many of the major Precambrian collisional orogenic belts were also constructed through prolonged subduction-accretion processes, similar to the modern analogue of the ongoing subduction and accretion in the Western Pacific (Santosh et al., 2009). Accretionary orogens also characterize the margins of the major supercontinents built through the amalgamation of continental fragments (Cawood and Buchan, 2007).
The geology of Sri Lanka provides important insights into continental growth in
Page 4 of 111
Page |5 the Neoproterozoic Earth. Tectonics of Sri Lanka has also been in focus in relation to the history of assembly of supercontinents, particularly because of its central position within the India – Madagascar – Africa – East Antarctica collage
ip t
of the late Neoproterozoic Gondwana supercontinent (Yoshida et al., 1992). The Sri Lankan Precambrian basement has been subdivided into four major terrains
cr
(e.g. Cooray, 1994): the Wanni Complex (WC) to the west together with the
us
Kadugannawa Complex (KC), the Highland Complex (HC) at the middle, and the Vijayan Complex (VC) to the east (Fig. 1), largely based on Nd- model age
an
mapping and regional structural interpretations (Milisenda et al., 1988, 1994; Kröner et al., 1987, 1991; Kehelpannala, 1991, 2003, 2004). The WC is
M
considered to represent a higher crustal level than that of the HC although there
d
is no clear structural break between the rocks of the two complexes, and the
te
contact between these two has been obliterated by later events (Voll and Kleinschrodt, 1991). The boundary between the VC and the HC is well defined as
Ac ce p
a thrust/shear contact (Hatherton et al., 1975; Vitanage, 1985; Voll and Kleinschrodt, 1991; Kriegsman, 1994). Three tectonic klippens of similar petrological and structural features to the HC are exposed in the south eastern part of Sri Lanka as inliers, namely Kataragama Klippe, Buttala Klippe and Kuda Oya Klippe. Some authors (e.g. Kehelpannala, 1991, 1997) have suggested that the KC forms part of the WC based on geochronological, geochemical and structural evidence. The HC domain is interpreted to be a part of a supracrustal basin developed in a suture zone with the Lützow-Holm Complex in East Antarctica during the final phase of Gondwana assembly (Shiraishi et al. 1994).
Page 5 of 111
Page |6 The WC, KC and VC domains are considered as Grenville-aged terranes of arcrelated settings at the outer margin of the Rodinia supercontinent (Kehelpannala, 2003, 2004; Kröner et al., 2003; Willbold et al., 2004; Kröner et al., 2013).
ip t
Although Sri Lanka is a relatively small continental fragment, mimicking a pendant on the heart of Gondwana, the several distinct tectonic units in this
cr
island and their history of amalgamation offer important clues for evaluating
us
Precambrian crustal evolution history.
an
In this study, we present results from our field investigations and petrologic, geochemical, zircon U-Pb and Lu-Hf isotopic studies on a suite of
Our data reveal Neoproterozoic crustal growth through protracted
d
Complexes.
M
metamorphosed magmatic rocks from the Wanni, Kadugannawa and Highland
te
arc magmatism derived from both juvenile and reworked sources, and lateral accretionary growth by the incorporation of oceanic and continental material.
Ac ce p
The subduction-accretion process culminated in continental collision and highgrade metamorphism during late Neoproterozoic – Cambrian.
2. Geological background
2.1 Overview of geology and petrology
The HC is dominantly composed of granulitic meta-quartzites, marbles, calc-silicates and metapelitic gneisses, in association with charnockites (Cooray,
Page 6 of 111
Page |7 1962, 1984; 1994; Perera, 1984; Mathavan et al. 1999; Mathavan and Fernando, 2001; Dharmapriya et al. 2014). The metasedimentary rocks such as marble and quartzite can be traced for more than 40 km in the central and northeastern part
ip t
of the HC, whereas marble and quartzite are scarce in the southwestern part of this complex (Mathavan et al. 1999). Bands of wollastonite-scapolite, diopside-
cr
and scapolite-bearing calc-granulites and cordierite-bearing gneisses occur in the
us
south and southwestern parts of the HC (Cooray, 1962, 1984; Hapuarachchi, 1968; Perera, 1984; Prame, 1991a; Mathavan et al., 1999; Mathavan and
an
Fernando, 2001).
Metamorphism under granulite facies conditions has been well established
M
in the HC from meta-basaltic and gabbroic to intermediate rocks (Sandiford et al.,
d
1988; Schumacher et al., 1990; Schumacher and Faulhabar, 1994), charnockites
te
(Prame, 1991b) and metamorphosed pelitic rocks (Perera, 1984, 1987; Prame, 1991a, Hiroi et al., 1994, Raase and Schenk, 1994). A P-T zonation across the
Ac ce p
HC has been established where pressures and temperatures decrease from 9-10 kbar and 830 °C in the east and southeast to 5-6 kbar and 700 °C in the northwest (Faulhaber and Raith, 1991; Schumacher and Faulhaber, 1994). Sajeev and Osanai (2005) identified a thermal gradient of 591 - 996 °C across south western to the central region of the HC. A clockwise P-T path is inferred for the pelitic rocks based on the sequence of kyanite and staurolite (occurring as inclusions in garnet) followed by sillimanite and andalusite (Perera, 1987; Hiroi et al., 1994, Raase and Schenk, 1994; Mathavan et al., 1999, Kriegsman and Schumacher, 1999). In contrast, reaction textures involving pyroxenes,
Page 7 of 111
Page |8 plagioclase, garnet and quartz in some meta-igneous rocks (Perera, 1987; Schumacher et al., 1990; Prame, 1991b) and high temperatures (>900 °C) from pyroxene exsolution textures (Schenk et al., 1988) have been used to suggest
ip t
isobaric cooling prior to the uplift. Although isobaric cooling is apparently not recorded in pelitic rocks of the HC apart from the report of Karunaratne et al.
cr
(2002), Dharmapriya et al. (2014) showed a rare evidence from a khondalite as
us
evidence for isobaric cooling from T of ~900 down to 770 °C at P of 7.5 kbar before uplift. Voll et al. (1994) derived peak temperatures of metamorphism at
an
850-900 °C using two-feldspar thermometry. Ultrahigh temperature (UHT) granulites have been reported in few localities in the central HC and rarely from
M
the southwestern part of the HC. These rocks witnessed extreme crustal
d
metamorphism at temperature of 925 °C to 1150 °C and pressures of 9-12.5
te
kbar (e.g. Kriegsman and Schumacher, 1999; Osanai et al., 2006; Sajeev and
Ac ce p
Osanai, 2004a, 2004b; Sajeev et al., 2007).
The WC occurs to the west-northwest of the HC and is dominantly
composed of upper amphibolite to granulite facies orthogneisses and minor meta-sediments.
The meta-igneous rocks show varied protolith-chemistry
ranging from granitic, granodioritic, monzonitic, tonalitic, charnockitic and enderbitic composition (e.g. Pohl and Emmermann, 1991). Garnet sillimanite gneisses, cordierite gneisses, quartzites and calc-silicate rocks occur as minor bands, particularly close to the inferred boundary with the HC. Metacarbonates are absent in the WC although apatite bearing unmetamorphosed carbonatite
Page 8 of 111
Page |9 body is exposed in the northern WC. The western part of the WC is mainly composed of less-deformed granites, including the unmetamorphosed posttectonic K-feldspar rich granite at Tonigala (e.g. Hölzl et al., 1991, Cooray, 1994).
ip t
Migmatization is also widespread within the unit. The peak metamorphic conditions of the WC are estimated to be 700-830 °C and 5-7 kbar (Schenk et al.,
cr
1991; Raase and Schenk, 1994). Classic outcrops of incipient charnockites
us
represent granulite formation within the host amphibolite facies gneisses in the central regions of the WC. In these zones, well-foliated light gray hornblende-
an
gneisses show veins and patches of dark, greenish-brown and coarse-grained incipient charnockite formed along mesoscopic shear zones and foliation planes,
M
representing pervasive fluid influx in the Sri Lankan lower crust (e.g. Hanson et
d
al., 1987; Burton and O’Nions, 1990; Baur et al., 1991; Milisenda et al., 1991;
te
Santosh et al., 1991; Kehelpannala, 1999; Perchuk et al., 2000).
Ac ce p
The KC, often referred as ‘Arenas’ (Vitanage, 1972; Almond, 1991), a
synonym used in Sri Lanka for doubly plunging synforms, is exposed in the northwestern part of Kandy. The complex includes upper amphibolite to granulite facies basement rocks (Kröner et al., 1991; Cooray, 1994; Malaviarachchi and Takasu, 2005, 2011a). Perera (1983) argued a sedimentary origin of the wellbanded hornblende biotite gneisses in the KC, whereas Cooray (1984) regarded their possible protoliths as basaltic lavas or volcanic ash. Relics of primary magmatic layering are preserved in these synforms mainly composed of hornblende gneisses, amphibolites and minor pyroxenites (Kröner et al., 1991).
Page 9 of 111
P a g e | 10 Hornblende- and biotite-bearing orthogneissic rocks with gabbroic, dioritic and trondhjemitic rocks of calc-alkaline affinity (Pohl and Emmermann, 1991) are interlayered within granodioritic to granitic gneisses, charnockites, enderbites and
ip t
minor metasediments (Kröner et al., 2003). Minor layered mafic and ultramafic rocks have also been reported (Stosch, 1991). The KC is considered by some
cr
authors as a migmatized and metamorphosed equivalent of a layered mafic-
us
ultramafic complex, named the ‘Kandy Layered Intrusion’ (KLI: Vol and Kleinschrodt, 1991; Stosch, 1991; Kleinschrodt et al., 1991).
an
The VC exposed in eastern Sri Lanka consists predominantly of metaigneous rocks including upper amphibolite facies calc-alkaline suite of granitoid
M
gneisses, augen-gneisses and minor amphibolites possibly derived from mafic
d
dykes. This granitoid suite was interpreted as a product of subduction-related
te
magmatism (Milisenda, 1991; Pohl and Emmermann, 1991). However, in contrary to previous interpretations, Kröner et al. (2013) reported that the Vijayan
Ac ce p
gneisses show evidence for granulite-facies metamorphism. Minor sedimentary enclaves such as metaquartzite, calc-silicate rocks and marble also occur within the VC (Dahanayake, 1982; Dahanayake and Jayasena, 1983; Kröner et al., 1987,
1991).
The
granitoid
gneisses
show
TTG
(tonalite-trondhjemite-
granodiorite)-dominated signature (Milisenda, 1991; Pohl and Emmermann, 1991) whereas the hornblende gneisses display compositions between diorite and leucogranite with the dominant composition being granodiorite and granite (Kröner et al., 1991; Milisenda, 1991; Kleinschrodt, 1994; Milisenda et al., 1994). Minor metagabbroic rocks and mafic dykes occur interlayered and infolded with
Page 10 of 111
P a g e | 11 the granitoid gneisses. Many exposures of the Vijayan gneisses show evidence of migmatization. Prame (1997) noted that many of the granitic rocks in the VC have A-type characteristics and inferred their formation through melting of felsic
ip t
crust. Igneous charnockite bodies occurring at the SE margin of the VC are interpreted as tectonically infolded or inter-sliced fragments of rocks of the HC
cr
similar to those of the HC-klippens present in the VC, during the thrusting of the
us
HC over the VC (Kröner et. al. 1991).
an
2.2 Summary of previous geochronological studies and tectonic models
d
M
2.2.1 Geochronology
te
The Nd model ages of 3.0-2.2 Ga reported from the HC (Milisenda et al., 1988, 1994) suggest that the rocks in this complex were derived from late
Ac ce p
Archean sources. Hölzl et al. (1991, 1994) and Köhler et al. (1991) presented Rb-Sr whole rock, Rb-Sr biotite and Sm-Nd garnet indicating ages of ~2 Ga. Cordani and Cooray (1990) reported Rb-Sr isochron age of 1100 Ma and De Maesschalck et al. (1990) interpreted a Rb-Sr whole rock isochron age of 1930 Ma as the timing of high-grade metamorphism. Rb-Sr and Sm-Nd isochron ages of 2.6-2.3 Ga and 500-450 Ma were also reported from the HC (Kagami et al. 1990). Garnet-whole rock Sm-Nd ages for widely separated HC paragneisses and metabasites show metamorphic ages of ca. 600 Ma (Hölzl et al., 1991; Milisenda, 1991). Ion microprobe U-Pb study on zircons by Kröner et al. (1987)
Page 11 of 111
P a g e | 12 documented Archean to early Proterozoic ages (3.2-2.4 Ga) from detrital zircons in the HC metasediments. They also reported an intrusion age of 1.1 Ga and new zircon growth at about 550 Ma from meta-igneous rocks. Köhler et al. (1991)
ip t
reported U-Pb zircon and monazite data indicating depositional ages of ~2 Ga and ~550 Ma ages for the high-grade metamorphism. Baur et al. (1991) reported
cr
U-Pb zircon crystallization ages of 1940 Ma and a Pb loss event at 560-550 Ma
us
indicating high-grade metamorphism. Kröner and Williams (1993) reported SHRIMP zircon ages for granitic gneiss from the HC as 1.9 Ga for the
an
crystallization of the igneous protolith and 531 Ma as the age of metamorphism. The lower intercept ages in concordia plots fall between 550-610 Ma for
M
discordant single zircon and zircon fractions in the granulites from the HC, whereas multigrain fractions of metamorphic zircon, monazite and rutile gave Pb/206Pb ages of 541-608 Ma (Baur et al., 1991, Hölzl et
d
207
te
concordant U-Pb and
al., 1991, 1994; Kröner et al., 1994). Burton and O’Nions (1990) analysed U and
Ac ce p
Pb isotopes in ilmenite giving an isochron age of 1.1 Ga which they interpreted as the metamorphic age. Hölzl et al. (1994) used U-Pb method of zircon and monazite from the same rock and reported ages of 555-592 Ma. In a recent study, Malaviarachchi and Takasu (2011b) obtained ca. 728-460 Ma ages for metamorphism using electron microprobe monazite CHIME method. Sajeev et al. (2003) obtained the Sm-Nd isochron whole-rock age of ca. 1.5 Ga from the UHT granulites of the HC and inferred that the granulites could be relics of a pre-Pan African metamorphism. In another study, Sajeev et al. (2007) reported U-Pb zircon metamorphic age of ca. 580 Ma from relatively
Page 12 of 111
P a g e | 13 higher pressure mafic granulites which is in agreement with zircon and monazite ages of ca. 570 Ma, determined from sapphirine granulites of the central HC
HC with the assembly of the Gondwana supercontinent.
ip t
(Sajeev et al. 2010), correlating the ultrahigh-temperature metamorphism in the
From the WC, Nd-modal ages of ca. 700 Ma -1.1 Ga have been reported,
cr
with multiple magmatic events as documented by zircon ages of 1100-530 Ma
us
(Burton and O’Nions, 1990; Kröner et al., 1991, 2003). Massive charnockites from the northern part of the WC have also older protolith emplacement ages as
an
from zircon evaporation (Kröner et al. 1994). Deposition of detrital zircons in metapelites and the emplacement of magmatic zircons of charnockitic and
M
enderbitic rocks yield ion microprobe U-Pb concordia upper intercept age of ~1.1
d
Ga considered as the timing of emplacement of the precursor magmas. The
te
lower intercept ages range between 540-590 Ma correlate with the high-grade metamorphism (e.g. Hölzl et al., 1991). Typical arrested (in-situ) charnockites and
Ac ce p
its host granitic/tonalitic gneiss from the Udadigana quarry in Kurunegala show ages on a common regression line with concordia intercepts at 771 Ma and 563 Ma, considered as the intrusion age and the metamorphic age respectively (Baur et al. 1991). Burton and O’Nions (1990) dated zircons from charnockitized patches and inferred that the in-situ charnockitization in the WC has occurred at 535 Ma. Thus, the arrested or in-situ charnockitization of the WC is considered to post-date the peak granulite facies metamorphism, and principally controlled by fluid influx along shear zones (Kehelpannala, 1999). Nd-model ages of the rocks from the KC range between ca. 900 Ma and
Page 13 of 111
P a g e | 14 1.6 Ga (Milisenda et al., 1988, 1994). Pb-Pb zircon evaporation ages from the KC show ca. 770 Ma -1.1 Ga indicating multiple calc-alkaline magmatic events covering a period of approximately 300 million years (Kröner et al., 2003; Willbold
ip t
et al., 2004). Most of the dioritic to granodioritic rocks of the KC display ca. 900 Ma ages. Xenomorphic zircons of gabbroic gneisses show relatively younger
cr
ages of ~700 Ma whereas idiomorphic zircons from the same sample gave ages
us
of ca. 900 Ma. Perera and Kagami (2001) argued that the regional high grade
isotope systematics on charnockitic rocks.
an
metamorphism of the KC must be older than 1100 Ma based on a study of Nd-Sr
Whole-rock Nd isotope analyses from widely separated localities show
M
mean crustal residence ages of 1.0–1.8 Ga for the VC rocks. Sr isotope data are
d
in line with the Nd isotope results suggesting that the majority of the VC rocks
te
were juvenile and derived from a typical LREE-depleted, mantle-derived precursor and not from anatexis of older crustal lithologies (Milisenda et al.,
Ac ce p
1988, 1994; Kröner et al., 1991). Rb–Sr whole-rock isochron ages of ca. 800 Ma reported by De Maesschalck et al. (1990) were ascribed to metamorphic resetting whereas the magmatic zircons reflect emplacement of the protoliths of the Vijayan gneisses at 1000–1100 Ma (Kröner et al., 2003, 2013). Zircon U-Pb ages of ca. 591- 456 Ma (Hölzl et al., 1991; Kröner et al., 1987, 1991, 2013) are now regarded as the age of metamorphism of the VC.
2.2.2. Structural and tectonic models
Page 14 of 111
P a g e | 15 The HC granulites have been subjected to a major folding and thrusting event (e.g. Berger and Jayasinghe, 1976; Kriegsman, 1994; Kleinschrodt, 1994). The large folds with amplitudes of about 7-10 km and an exposed length parallel
ip t
to their axes of up to 50 km in the central Highlands formed after the peak of metamorphism (800-850 °C, 8-9 kbar) at slightly lower temperature (700-750 °C)
cr
(Kleinschrodt, 1994). These large upright folds were generated during the last
us
deformation event following high-grade metamorphism (e.g. Berger and Jayasinghe, 1976; Kriegsman, 1994). During this event, horizontal shorting and
an
vertical thickening of crust took place with unroofing of the crustal rocks into upper level (Kleinschrodt, 1994; Kriegsman, 1994). Hiroi et al. (2014) reported
M
evidence for fast exhumation of lower-crustal rocks to andalusite-stable upper-
d
crustal conditions by channel flow in a continental collision orogeny, based on
of the HC.
te
felsic/granitic inclusions within high grade ordinary granulites and UHT granulites
Ac ce p
Transposition of structures in the WC is less severe, compared to those in
the HC. In ductile shear zones, gneissic foliations are severely flattened resulting in the obliteration of the previous intrusive and deformation histories. However, in some regions, multiple phases of gabbroic to granitoid intrusions and their deformation are preserved in multiple foliations (Kriegsman, 1991; Kröner et al., 1994). Macroscopic folds of the open to tight type with gently plunging hinges are well developed in the WC similar to those in the HC and their axial traces continue for several kilometers (Tani and Yoshida, 1996). The contact between the WC and the HC is still controversial due to absence of any structural
Page 15 of 111
P a g e | 16 discontinuity in between. The combined HC – WC unit is considered to represent a tilted section of the former lower–middle crust, with the WC representing the
ip t
higher level.
The entire rock suite of the KC is folded into large-scale doubly plunging
cr
five NW-trending synforms (Arenas). These macroscopic plunging folds with
us
upright axial surfaces were generated after the formation of the major compositional layering. The interior domains of the KC preserve prograde
an
amphibolite facies conditions compared to the surrounding HC (Perera, 1983; Sandiford et al., 1988; Voll and Kleinschrodt, 1991). Hence, the KC is considered
M
to be at a higher crustal level than the HC (Sandiford et al., 1988; Kriegsman,
d
1994) whereas Kehelpannala (1997) interpreted that the KC represents a deeper
te
crustal level compared to the WC. Although the boundary between the HC and the KC is well defined as a structural break, that between the WC and the KC is
Ac ce p
not clear e.g. Kröner et al., 2003).
Kehelpannala (2003) noted that the homogeneous to compositionally well-
layered gneisses experienced strong ductile deformation during more than one phase of folding. The boundary between the VC and the HC is considered to be a sub-horizontal ductile shear zone/thrust (Hatherton et al., 1975; Vitanage, 1985; Voll and Kleinschrodt, 1991; Kriegsman, 1995), formed when the HC collided with the VC and produced late upright folds in the former and ductile deformation in the latter (Kehelpannala, 1997, 2003, 2004). Shear sense indicators and sheath folds found at the suture between the VC and the HC show
Page 16 of 111
P a g e | 17 that the HC, namely the hanging wall, apparently moved from N to S over the VC (Kehelpannala,
2003)
accompanying
widespread
mylonitization
and
ip t
migmatization along the shear zones (Kleinschrodt, 1994).
cr
2.3 Present study area and description of sampling locations
us
The present study focuses on petrology, geochemistry and geochronology of meta-igneous rocks collected from a traverse running from the WC through the
an
KC into the HC in a direction of increasing elevation (Fig. 1). Fresh samples were chosen from quarries and road-cut outcrops along this traverse. The NW-SE
M
sampling in our study is oblique to the general trend of foliation so that a
d
meaningful stratigraphic section is traversed.
te
In this paper, we present analysis of three samples from the KC, four samples from the HC, and one sample from the WC (Fig. 2). A summary of the
Ac ce p
location and salient features of the samples analyzed for geochemistry and geochronology are given in Table 1. Samples of the KC represent the Dumbara Synform (KDG-1), Huluganga Synform (KDG-2) and Metibokka Synform (KDG4). The HC samples (HC-1, HC-7, HC-8, HC-10 and HC-11) were selected from the areas which represent strong ductile deformation and relatively higher grade metamorphism in the HC. The sample of the WC (WC-2) comes from a classical outcrop of in-situ/incipient charnockitization at Udadigana in Kurunegala. Representative field photographs are shown in Figures 3 and 4.
Page 17 of 111
P a g e | 18 2.3.1 Kadugannawa Complex
ip t
Metagranodiorite (KDG-4) and metadiorite (KDG-1)
The metagranodiorite sample KDG-4 was collected from a quarry at Maussawa
cr
along Harankahawa Road in Galagedara. The quarry exposes evidence for mafic
us
magma intrusion into a felsic magma chamber and subsequent mixing, mingling and fractionation. The presence of mafic-felsic layers in the outcrop may indicate
an
lower crustal melting in a granodiorite-granite magma chamber where the restite portion is represented by the hornblendite-rich mafic layers. The felsic magma
M
chamber was intruded by mafic/intermediate (dioritic) magma as dykes/sills
d
resulting in magma mixing and mingling as evidenced from the mafic magmatic
te
enclaves (MMEs) of various sizes within the granodiorite. The rocks display intense deformation as seen by refolded isoclinal folds and eye-like sheath or
Ac ce p
doubly plunging folds while preserving the magmatic history.
The metadiorite KDG-1 was collected from a road side exposure at Kundasale (about 5 km from Thennekumbura) along the Kandy-Mahiyangana road. The outcrop lies at the lower south-western part of the doubly plunging Dumbara synform of the KC. The rock shows strongly flattened and stretched compositional layers, and bedding parallel with the surrounding TTG gneisses that show compositional layering. Cross cutting dykes of aplitic and granitic composition are also common in the highly deformed parts of the metadiorite.
Page 18 of 111
P a g e | 19
Garnet amphibolite (KDG-2)
ip t
This sample was collected from a newly excavated road cut at the Mahaberiyatenna Airport site. The rock occurs as a several tens of meter thick-
cr
layer within migmatitic hornblende-biotite gneiss in the south western margin of
us
the doubly plunging Huluganga synform of the KC. Coarse-grained calcic amphibole and porphyroblastic red-eye garnets are visible in the outcrop, where garnets
show
breakdown
textures
forming
an
some
plagioclase-bearing
symplectites. Hornblende-rich cumulate layers also occur within the magmatic
d
M
layers associated with the amphibolite.
te
2.3.2. Wanni Complex
Ac ce p
Metagranodiorite (WC-2)
The sample was collected from an abandoned quarry at Udadigana in Kurunegala, tectonically located on the western limb of the Kurunegala Synform of the WC. The quarry exposes excellent evidence for in-situ/incipient charnockitization with greenish patches and veins of orthopyroxene-bearing anhydrous zones overprinting the host gneiss. The sample collected for this study is from the tonalitic to granodioritic host rock of the incipient charnockite. A gradational transition of the host gneiss to charnockite is evidenced at the quarry
Page 19 of 111
P a g e | 20 where some parts are totally converted into a coarse-grained charnockite, suggesting three dimensional networks of fluid- filled veins or tubes.
ip t
2.3.3. Highland Complex
us
cr
Mafic granulite (HC-1, HC-11)
Mafic granulite bands, boudins and enclaves of variable thickness (0.1 to 3m)
an
occur within marbles at the Digana quarry. Our sample HC-1 is from one of these bands. The marble is composed of dolomitic and calc-silicate layers parallel to
M
the regional compositional layering of the HC. Mafic granulite enclaves are
d
located parallel to bedding, and are fragmented with angular edges. They show
te
strong deformation including evidence for flattening and stretching of minerals like pyroxene and hornblende. In some parts of the quarry, the mafic granulite
Ac ce p
shows boudinaged features.
The sample HC 11 is from a mafic sill along a roadside exposure ~2 km from Kivullinda Junction towards Randenigala, along the Kandy-Randenigala road. This location is close to the hinge on the west limb of the Digana Antiform of the HC. Several metadioritic sills ranging in thickness from few tens of cm to several meters occur within alternating layers (5 to 40 cm thickness) of quartzite and psammitic gneiss, parallel to bedding with sharp contacts. Some of the mafic sills contain garnet.
Page 20 of 111
P a g e | 21
Metagabbro (HC-7 and HC-8)
ip t
The samples of metagabbro were collected from road side outcrops along the Kandy-Hantana Road. These rocks occur as discontinuous lenses (few tens of
cr
metres thick) within quartzite layers of approx. 2 km thickness and biotite gneiss
us
(partly charnockitic at places), close to the hinge on the western limb of the Kandy Antiform of the HC. Reddish porphyroblastic garnets are extensively
an
distributed in the rock, partly reflecting a cumulate-textured mafic granulite. Minor quartz-bearing domains (sample HC-8) occur within the metagabbroic host rock
te
Garnet charnockite (HC-10)
d
M
(HC-7).
Ac ce p
The garnet- and orthopyroxene- bearing charnockite sample was collected from a road outcrop near Hakurutale, between 12 and 13 km along the KandyRandenigala Road. The 0.5 – 1 km thick layer is interbanded between a granitic gneiss and a biotite gneiss layer. In some domains, coarse deep red almandinerich garnets occur and some of the garnet grains show plagioclase-bearing symplectites.
Symplectitic
garnets
and
coarse
grained
dark
brownish
hypersthene grains occur along the foliation plane. The rock is generally medium to coarse grained showing brownish to gray-green greasy charnockitic luster.
Page 21 of 111
P a g e | 22 3. Analytical techniques
ip t
3.1 Petrography and mineral chemistry
Polished thin sections were prepared for petrographic study at the State Key
cr
Laboratory of Geological Processes of Mineral Resources of China University of
us
Geosciences Beijing, and at the University of Tsukuba, Japan. Mineral chemical analyses were carried out using an electron microprobe analyzer (JEOL
an
JXA8530F) at the Chemical Analysis Division of the Research Facility Center for Science and Technology, the University of Tsukuba. The analyses were
M
performed under conditions of 15 kV accelerating voltage and 10 nA sample
te
supplied by JEOL.
d
current, and the data were regressed using an oxide-ZAF correction program
Ac ce p
3.2 Whole-rock geochemistry
Representative samples were chosen for whole-rock geochemical analyses from the WC, HC, and KC which are devoid of surface alteration or weathering. The size of the samples were initially reduced in a jaw crusher, and then manually fine-powdered in an agate mortar. Major oxides were analyzed by Thermo JarrellAsh ENVIRO II ICP, and minor and trace elements by Perkin Elmer Sciex ELAN 9000 ICP/MS at Actlabs, Canada, using lithium metaborate/tetraborate fusion technique.
Page 22 of 111
P a g e | 23
3.3 Zircon U-Pb and Lu-Hf isotopes
ip t
Zircon grains were separated by gravimetric and magnetic separation from crushed rock samples, and then purified by hand picking under a binocular
cr
microscope at the Yu’neng Geological and Mineral Separation Survey Centre,
us
Langfang City, Hebei Province, China. The grains were mounted in epoxy resin discs and polished to reveal mid-sections, followed by gold sputter coating.
an
Zircons were imaged under both transmitted and reflected light, and were imaged using cathodoluminescence (CL) to identify internal structures and choose
M
potential target sites for U-Pb analyses. The CL imaging was carried out at the Beijing Geoanalysis Centre. Individual grains were mounted along with the Pb/238U age of 417 Ma (Black et al., 2003), onto
d
206
te
standard TEMORA 1, with
double-sided adhesive tape and enclosed in epoxy resin discs. The discs were
Ac ce p
polished to a mid depth and gold coated for cathodoluminescence (CL) imaging and U-Pb isotope analysis. Zircon morphology, inner structure and texture were examined by using a JSM-6510 Scanning Electron Microscope (SEM) equipped with a backscatter probe and Chroma CL probe. The zircon grains were also examined under transmitted and reflected light images using a petrological microscope.
Zircon U-Pb dating and trace element analysis were simultaneously conducted by LA-ICP-MS at the Stage Key Laboratory of Geological Processes
Page 23 of 111
P a g e | 24 and Mineral Resources, China University of Geosciences. Detailed operating conditions for the laser ablation system and the ICP-MS instrument are as described by Liu et al. (2008, 2010). Laser sampling was conducted using an
ip t
Excimer 193 nm GeoLas 2005 System with a spot size of 32 μm. An Agilent 7500a ICP-MS instrument was used to acquire ion-signal intensities. Nitrogen
cr
was introduced into the central gas flow (ArtHe) of the Ar plasma in the LA-ICP-
us
MS analysis, which increases the sensitivity for most elements by a factor of ~2 compared with the results without adding nitrogen (Hu et al., 2008). To keep time-
an
dependent elemental fractionation at a low level, a laser frequency of 4 Hz and a laser energy of 60 mJ were applied. The uncertainty of analysis is within ~1% for
M
the zircon standard.
d
Zircon 91500 was used as external standard for U-Pb dating, and was
te
analyzed twice every five analyses. Time-dependent drifts of U-Th-Pb isotopic ratios were corrected using a linear interpolation (with time) for every five
Ac ce p
analyses according to the variations of 91500 (i.e. two zircon 91500 + five samples + two zircon 91500). Preferred U-Th-Pb isotopic ratios used for 91500 are from Wiedenbeck et al. (1995). Uncertainty in the preferred values for the external standard 91500 was propagated to the ultimate results for the samples. Concordia diagrams and weighted mean calculations were made using Isoplot/Ex_ver3 (Ludwig, 2003). Trace element compositions of zircons were calibrated against multiple-reference materials (BCR-2G and BIR-1G) combined with internal standardization (Liu et al., 2010). The preferred values of element concentrations for the USGS reference glasses used are from the GeoReM
Page 24 of 111
P a g e | 25 database (http://georem.mpch-mainz.gwdg.de/). In situ Hf isotope compositions of zircon were determined using a Neptune Plus multicollector (MC)-ICP-MS system (Thermo Fisher Scientific, Germany)
ip t
and a Geolas 2005 Excimer ArF laser ablation system (Lambda Physik, Götingen, Germany) at the State Key Laboratory of Geological Processes and
cr
Mineral Resources, China University of Geosciences, Wuhan. The energy
us
density of laser ablation used was 5.3 J cm-2. Helium was used as the carrier gas in the ablation cell and was merged with argon (makeup gas) after the ablation
an
cell. All data for zircon in this study were acquired in the single-spot ablation mode at a spot size of 32 mm. Each measurement consisted of 20 s of
M
acquisition of the background signal followed by 50 s of ablation signal
d
acquisition. The operating conditions for the laser ablation system and the MC-
te
ICP-MS instrument, and the analytical method, are the same as those described
Ac ce p
in detail by Hu et al. (2012).
4. Results
4. 1 Petrology and mineral chemistry
4.1.1 Petrography
A summary of the petrographic features of each of the rock types analyzed in this study is given below. Representative photomicrographs are shown in Fig. 5.
Page 25 of 111
P a g e | 26
Metagranodiorite and metadiorite
ip t
The metagranodiorite and metadiorite in our study are generally coarse-grained with quartz, plagioclase, biotite and calcic amphibole (hornblende) as the
cr
dominant minerals. Sample WC-2, a typical metagranodiorite, is composed of
us
plagioclase (35-45%), K-feldspar (20-30%), quartz (20-25%), biotite (5-10%), and hornblende (2-5%), with accessory Fe-Ti oxides (1-2%), zircon, and apatite (Fig.
an
5a). Plagioclase and K-feldspar are coarse-grained (grain size range of 0.3-1.2 mm and 0.4-1.3 mm, respectively) showing granoblastic texture. Fine, rounded
M
grains of quartz (0.2-0.7 mm) are present along their grain boundaries, although
d
elongated coarse-grained (~4.2 mm) quartz also occurs aligned along the rock
te
foliation. Myrmekite textures can be commonly observed in some of the samples.
Ac ce p
The metagranodiorite (sample KDG-4) comprises plagioclase (30-40%), Kfeldspar (15-25%), hornblende (15-25%), biotite (10-15%), quartz (10-15%), and magnetite (1-2%). Weak foliation of the rock is defined by aligned flakes of hornblende (0.2-1.9 mm) and biotite (0.2-1.8 mm). Plagioclase and K-feldspar are coarse-grained (0.2-1.2 mm), sub-equigranular, and granoblastic (Fig. 5b). Quartz is relatively fine grained (0.2-0.8 mm) compared to that of the sample WC2.
The sample KDG-1 is an orthopyroxene-bearing and K-feldspar-free metadiorite
Page 26 of 111
P a g e | 27 showing textures similar to those of the sample KDG-4. The mineral assemblage is plagioclase (50-60%), hornblende (15-25%), biotite (10-15%), quartz (5-10%), and ilmenite (2-3%), with accessory orthopyroxene, pyrite, apatite, and zircon
ip t
(Fig. 5c). The rock does not have a distinct foliation, and is characterized by granoblastic and semi-equigranular texture of plagioclase (0.6-2.2 mm) and
cr
quartz (0.2-0.8 mm). Both hornblende (0.5-1.5 mm) and biotite (0.2-0.8 mm) are
us
subhedral and randomly oriented along the grain boundaries of plagioclase and quartz. Orthopyroxene is also fine-grained (0.2-0.4 mm) and intergrown with
an
biotite or hornblende.
d
M
Garnet amphibolite
te
Sample KDG-2 is composed of very coarse-grained (~9 mm) poikiloblastic garnet (5-15%) in the matrix of aggregates of hornblende (20-30%) and plagioclase (30-
Ac ce p
40%) (Fig. 5d). Foliation is not obvious in thin section. Accessory minerals are ilmenite-magnetite intergrowth (2-5%), apatite (2-5%) and zircon, and quartz is present only as rare inclusions in garnet. The garnet contains numerous inclusions of ilmenite, apatite, hornblende, and quartz. Hornblende (0.3-2.4 mm) and plagioclase (0.5-1.7 mm) are coarse grained, but is fine-grained compared to the garnet. The garnet is surrounded by medium-grained intergrowth of hornblende + plagioclase, suggesting the progress of the following CFMASH continuous reaction (1); Grt + Qtz + H2O => Hbl + Pl
(1)
Page 27 of 111
P a g e | 28
ip t
Mafic granulite
The sample HC-1 is a medium-grained garnet-free mafic granulite that is
cr
composed of plagioclase (30-40%), clinopyroxene (15-25%), hornblende (15-
us
25%), orthopyroxene (10-20%), ilmenite (2-5%) and magnetite (<1%) (Fig. 5e). Calcite occurs as an accessory mineral, in accordance with the field occurrence
an
of the rock as enclave within marbles. The rock shows a weak foliation defined by aligned aggregates of ferromagnesian minerals. All the dominant minerals are
d
M
medium to fine grained (0.2-0.8 mm).
te
Sample HC-11 is also a garnet-free mafic granulite without any marked foliation and composed of plagioclase (30-40%), clinopyroxene (20-30%), orthopyroxene
Ac ce p
(10-20%), hornblende (10-20%), and accessory ilmenite and zircon (Fig. 5f) in an equigranular matrix (0.2-0.8 mm). Fine-grained aggregates of orthopyroxene and plagioclase occur as grayish spots of about 4 mm in diameter, typically associated with quartz (Fig. 5g), suggesting the progress of the following decomposition reaction of garnet: Grt + Qtz => Opx + Pl
(2)
This reaction involved the near total consumption of garnet possibly associated with post-peak decompression (e.g., Harley, 1989).
Page 28 of 111
P a g e | 29 Metagabbro
The sample HC-7 is a coarse-grained garnet-bearing metagabbro with variable
ip t
quartz-bearing and absent domains. The rock is composed of clinopyroxene (4050%), garnet (30-40%), orthopyroxene (5-10%), and plagioclase (5-10%), with
cr
accessory Fe-Ti oxide (2-5%) and hornblende (1-2%) (Fig. 5h). No obvious
us
foliation can be seen. Garnet and clinopyroxene are coarse grained (~1.7 mm) and semi-equigranular. Garnet is subhedral, and almost free from inclusions
an
except for minor apatite, hornblende, ilmenite, and quartz. Clinopyroxene is also subhedral, and contains thin exsolution lamellae of orthopyroxene. Plagioclase
M
(0.2-0.7 mm) is anhedral and fills the matrix of garnet and pyroxenes. Fine-
d
grained and rounded Fe-Ti oxide (intergrowth of ilmenite and magnetite) grains
te
are scattered throughout the thin section. These textures are similar to metagabbros reported from the Palghat-Cauvery Suture Zone in Southern India
Ac ce p
(e.g., Sajeev et al., 2009; Santosh et al., 2010; Saitoh et al., 2011; Koizumi et al., 2014).
Sample HC-8 is a quartz-rich domain of sample HC-7. It comprises plagioclase (30-40%), clinopyroxene (30-40%), garnet (10-20%), quartz (5-10%), and orthopyroxene (2-5%) with accessory ilmenite, magnetite, rutile, and zircon. Foliation is defined by elongated quartz grains (~4 mm) and garnet aggregates. Garnet is coarse-grained (0.3-3.3 mm), subhedral to anhedral, and often surrounded by fine-grained (~0.2 mm) symplectitic intergrowth of vermicular
Page 29 of 111
P a g e | 30 orthopyroxene and plagioclase or thin corona of orthopyroxene + plagioclase (Fig. 5i). Such intergrowths can be readily observed in quartz-rich portion of the rock, confirming the progress of reaction (2) probably related to rapid
ip t
decompression. Clinopyroxene (0.2-1.1 mm) is subhedral, and often occur as aggregates elongated along the foliation particularly at plagioclase-rich portion of
cr
the rock. Ilmenite occurs as fine-grained (<0.3 mm) rounded to sub-rounded
us
grains, or irregular mineral intergrowing with clinopyroxene.
an
Garnet charnockite
M
Sample HC-10 is characterized by coarse-grained (~2.5 mm) porphyroblastic
d
garnet in hand specimen (Fig. 3), although the modal abundance varies from 0 to
te
5 %. The examined sample corresponds to a garnet-absent portion of the rock, and comprises plagioclase (30-40%), quartz (10-20%), clinopyroxene (10-15%),
Ac ce p
orthopyroxene (5-10%), hornblende (5-10%), with accessory ilmenite, pyrite, calcite, zircon, and retrograde biotite (Fig. 5j). K-feldspar is absent in the rock, which is consistent with its dioritic composition as discussed later. The rock is also characterized by ribbon quartz elongated parallel to the rock foliation defined by thin alternation of pyroxene-rich and plagioclase-rich layers. Plagioclase is coarse-grained (0.7-2.5 mm), granoblastic, and semi-equigranular in plagioclaserich layers, while it is fine-grained (0.2-0.7 mm), subhedral, and locally intergrowing with fine-grained orthopyroxene (~0.2 mm) in pyroxene-rich layers as a product of reaction (2) (Fig. 5j). The lack of garnet in the sample probably
Page 30 of 111
P a g e | 31 implies complete consumption of the mineral during decompression. Mediumgrained (0.2-0.8 mm) and subhedral ortho- and clinopyroxenes, and fine-grained (0.1-0.5 mm) hornblende are dominant minerals in the pyroxene-rich layers,
ip t
whereas their grain size decreases in plagioclase-rich layers. Individual pyroxene and plagioclase grains show no obvious alignment, whereas hornblende is
cr
weakly oriented along the rock foliation. Minor biotite mantles pyroxene and
an
us
hornblende grains as a product of post-peak retrograde hydration.
M
4.1. 2 Mineral chemistry
d
Representative compositions of minerals from electron microprobe analyses are
Ac ce p
below.
te
given in Supplementary Tables 2 to 4, plotted in Figure 6, and briefly discussed
Garnet
Garnet in the metagabbros (HC-7, HC-8) and garnet amphibolite (KDG-2) is essentially almandine-rich with minor pyrope, grossular, and spessartine contents whose abundance varies slightly depending on the assemblage and core-rim positions (Supplementary Table 2, Figs. 6a,b). Garnet in sample HC-7 shows the highest XMg (= Mg/(Mg+Fe)) content of 0.24-0.28. Its almandine content increases slightly from core (Alm57-58 Prp22-23 Grs18-19 Sps1-2) toward rim (Alm59-60
Page 31 of 111
P a g e | 32 Prp19-20 Grs18-19 Sps1-2), which is a typical zoning pattern for garnet grains in mafic granulites. Garnet in sample HC-8 shows nearly consistent core-rim variation, but the core is slightly grossular rich as Alm59-60 Prp14-15 Grs24-25 Sps1-2 (core) and
ip t
Alm64-65 Prp13-14 Grs20-21 Sps1-2 (rim). In contrast, garnet in sample KDG-2 shows slightly Mn-rich rim (Alm58-59 Prp15-16 Grs19-20 Sps5-6) than core (Alm60-61 Prp15-16
us
cr
Grs20-21 Sps3-4), although their XMg values are nearly consistent (0.20-0.23).
an
Pyroxenes
Clinopyroxene in the mafic suites of present study is Mg-rich (XMg = 0.53-0.82),
M
and all are classified as augite (Fig. 6c), although their XMg varies depending on
d
samples. Clinopyroxene in the metagabbros (samples HC-7 and HC-8) shows
te
lower XMg of 0.61-0.64 and 0.53-0.59, than that in the mafic granulite (samples HC-1 and HC-11, XMg = 0.69-0.82). Single grains do not display any
Ac ce p
compositional zoning.
Orthopyroxene also shows a wide compositional range of XMg = 0.39-0.72. Orthopyroxene in mafic granulite (sample HC-1) shows the highest XMg of 0.710.72. Orthopyroxene in sample HC-11 is slightly less magnesian as XMg = 0.600.61, whereas that associated with symplectite in the same sample shows lower XMg of 0.58-0.59. A similar XMg range was also obtained for orthopyroxene in metadiorite (sample KDG1; XMg = 0.57-0.58). Orthopyroxene in metagabbro (sample HC-8) displays the lowest XMg of 0.39-0.42. Al content in the
Page 32 of 111
P a g e | 33 orthopyroxene is generally low (Al <0.1 pfu).
ip t
Plagioclase
Plagioclase in the examined rocks shows significant compositional variations
cr
depending on lithologies (Supplementary Table 4, Fig. 6d). Plagioclase in
us
symplectite-bearing metagabbro and mafic granulite (e.g., samples HC-8 and HC-11, respectively) shows higher anorthite contents of An61-77. Particularly, the
an
coarse-grained matrix plagioclase in sample HC-11 (An68-77) displays the highest anorthite contents of An68-77 as compared to that associated with symplectite
M
(An64-65). Plagioclase in metadiorite (sample KDG-1) shows the highest albite
Ac ce p
Amphibole
te
the samples as Or80-86.
d
content of An24-27. K-feldspar is compositionally nearly homogeneous throughout
Amphibole in metagranodiorite (samples WC-2 and KDG-4), metadiorite (sample KDG-1), mafic granulite (samples HC-1 and HC-11), and garnet amphibolite (sample KDG-2) shows notable compositional variation depending on its occurrence (Supplementary Table 2, Fig. 6e). The mineral is characterized by high Ca and (Na+K) contents (1.79-1.92 pfu and 0.54-0.94 pfu, respectively). Calcic amphibole in metagranodiorite and metadiorite is Si-rich, and classified as edenite – pargasite (sample KDG-1), ferro-edenite (sample KDG-4), and ferro-
Page 33 of 111
P a g e | 34 edenite – ferro-pargasite (sample WC-2). Amphibole in garnet amphibolite is slightly Al-rich and classified as ferro-pargasite. Calcic amphiboles in garnet-free mafic granulite (samples HC-1 and HC-11) are Mg- and Al-rich as pargasite. Ti
ip t
content of amphiboles in metagranodiorite, metadiorite, and amphibolite are low (0.18-0.25 pfu), but that in mafic granulites is slightly enriched (0.27-0.41 pfu).
cr
Calcic amphibole in metagabbro (sample HC-7), which occurs as inclusions in
us
poikiloblastic garnet, is significantly Ti-rich (0.52-0.54 pfu) and classified as
an
kaersutite.
M
Biotite
d
The biotite in metagranodiorite (samples WC-2 and KDG-4) and metadiorite
te
(sample KDG-1) exhibits very high TiO2 content of 3.9-5.6 wt.% (Fig. 6f). XMg of biotite in samples WC-2 and KDG-4 (metagranodiorite) is 0.44-0.51, which is
Ac ce p
lower than that in the sample KDG-1 (metadiorite) (XMg = 0.56-0.61).
4.2 Geothermobarometry
The garnet-clinopyroxene-plagioclase-quartz geothermobarometers were applied to porphyroblastic garnet and clinopyroxene with plagioclase and quartz assemblage in samples HC-7 and HC-8 (metagabbro). The estimated temperature range for garnet-clinopyroxene pairs are 830-860°C for sample HC7 and 840-860°C for sample HC-8 (at 10 kbar) based on the method of Ellis and
Page 34 of 111
P a g e | 35 Green (1979), which is a widely applied thermometer for mafic granulites and relies on experimental calibration of Fe–Mg fractionation between garnet and clinopyroxene at 750°C to 1350°C and 24 to 30 kbar. Calculated temperature
ip t
ranges using the method of Ganguly et al. (1996), which is based on revised solution model of garnet, are slightly lower but nearly consistent, 820-850°C
cr
(sample HC-7) and 790-800°C (sample HC-8). The temperatures were calculated
us
at 10 kbar, a reference pressure inferred from the peak pressure condition of the samples estimated by garnet-clinopyroxene-plagioclase-quartz geobarometer of
M
nearly consistent among all the samples.
an
Moecher et al. (1988) as discussed below. The estimated temperature range is
d
Metamorphic pressure for the metagabbro was calculated using garnet-
te
clinopyroxene-plagioclase-quartz assemblage in the samples HC-7 and HC-8 based on experimental calibration of Perkins and Newton (1981). The estimated
Ac ce p
results are 7.1-7.6 kbar for sample HC-7 and 6.6-6.8 kbar for sample HC-8 at 850ºC. These results are, however, significantly low if compared to the petrographic observation of quartz in the rock because the stability of quartz in mafic granulite possibly indicates high-pressure condition of P > 9 kbar at 850°C (e.g., Green and Ringwood, 1967). We therefore adopted the method of Moecher et al. (1988), which is based on improved garnet-clinopyroxene-plagioclasequartz geobarometer using new thermodynamic and experimental data. Application of the method yielded a pressure range of 9.5-9.8 kbar (sample HC7) and 9.4-9.6 kbar (sample HC-8) at 850°C, suggesting relatively high-pressure
Page 35 of 111
P a g e | 36 conditions.
The Fe-Mg cation exchange between garnet and hornblende has been calibrated
ip t
as a geothermometer by Graham and Powell (1984). We applied this method to garnet amphibolite (sample KDG-2) and obtained a temperature range of 700-
cr
720°C. The estimated temperatures are significantly lower than the results
us
obtained from garnet-clinopyroxene assemblage, and probably record retrograde
an
conditions.
As hornblende coexists with plagioclase in metagranodiorite (samples WC-2 and
M
KDG-4) and metadiorite (sample KDG-1), the mineral pair has been used for
d
geothermometry. Based on hornblende solid-solution models and well-
te
constrained natural and experimental studies, two geothermometers were calibrated by Holland and Blundy (1994) based on the edenite-tremolite reaction,
Ac ce p
which is applicable to quartz-bearing rocks, and edenite-richterite reaction, which is applicable to both quartz-bearing and quartz-free rocks. As quartz is present in the samples, the two thermometers were used for the calculation of temperature of the felsic rocks. The calculated results are 730−770°C for sample KDG-1, 700−740°C for sample KDG-4, and 720−780°C for sample WC-2, probably indicate retrograde re-equilibration.
4.3 Geochemistry
Page 36 of 111
P a g e | 37 Whole rock geochemical data, including major, minor, trace and rare earth elements, on nine representative samples (two metagranodiorites, one metadiorite, one garnet amphibolite, two mafic granulites, two metagabbros and
ip t
one garnet charnockite) are given in Supplementary Table 5. The salient
cr
geochemical features of the various rock types are summarized below.
us
4.3.1 Major and trace elements
an
Major element contents show that the metagranodiorite has high SiO2 content in a restricted range (71.0-71.3 wt.%) (Fig. 7), whereas the metadiorite displays
M
intermediate SiO2 of 55.8 wt.%. The lowest SiO2 content is displayed by the
d
garnet amphibolite (41.5%). The two mafic granulites have SiO2 content in the
te
range of 49.86-49.94 wt.%, whereas the metagabbro is slightly SiO2 poor as 44.7-48.8 % (Fig. 7). Interestingly, the metagranodiorites in the two different
Ac ce p
crustal units (KDG-4 and WC-2) show almost identical composition. P2O5 contents of the garnet amphibolite shows relatively high values, which is consistent with the abundant occurrence of apatite in the sample (~5%). Garnet amphibolite (KDG-2), metagabbro (HC-7) and metagranodiorites clearly plot away from the general trends displayed by other meta-mafic rocks except for their Na2O-SiO2 relationship. These meta-mafic rocks have systematic positive correlation of Al2O3, K2O and P2O5 and a negative correlation of MgO, Fe2O3, MnO, CaO, TiO2 with increasing silica content. This geochemical variability is clearly reflected by the SiO2 versus Fe2O3T/MgO diagram (Fig. 7j) suggesting
Page 37 of 111
P a g e | 38 three distinct geochemical groups of rocks: magnesian (Fe2O3T/MgO <3), ferroan (Fe2O3T/MgO >3) and felsic (SiO2 >70 %). The three compositional types can also be distinguished in terms of their SiO2 (magnesian: 48.8-54.5 wt.%; ferroan:
ip t
41.5-44.7 wt.%; felsic: >70 wt. %) and TiO2 (magnesian: 0.55-1.2 wt.%; ferroan:
cr
3.0-3.4 wt.%; felsic: 0.31-0.41 wt.%) contents.
us
The major element behaviour shown by the three groups of samples is also reflected in their trace elements (Fig. 8). In the primitive-mantle normalized (Sun
an
and McDonough, 1989) trace element plot, magnesian and felsic samples are characterized by enrichment of large ion lithophile elements (LILE) like Rb, Ba
M
and K, negative Nb, Ta, P and Ti anomalies, and relatively constant high-field
d
strength elements (HFSE: Zr, Hf) (Figs. 8a,c). Nevertheless, the ferroan
te
metagabbro (sample HC-7) is depleted in LILE and does not show Nb and Ta negative anomalies although Zr, Hf, Y and Yb are depleted (Fig. 8b). The garnet
Ac ce p
amphibolite (sample KDG-2) shows a characteristic depletion of LILE whereas the other trace elements are relatively enriched and yield a smooth pattern (Fig. 8b). Another striking feature is marked Sr enrichment in the ferroan samples and no systematic behaviour in the magnesian samples. The felsic samples however, are enriched in LILE with obvious negative anomalies of Ta and Nb, P, Ti and Sr. The felsic samples show approximately one magnitude higher contents of LILE as compared to the other two groups. Rb-Y-Nb and Rb-Yb-Ta diagrams of Pearce et al. (1984) for granitic rocks (Fig. 9) suggest that metagranodiorite and metadiorite are characterized by low Y, Yb, Nb, and Ta, and moderate Rb
Page 38 of 111
P a g e | 39 contents.
In the chondrite normalized (McDonough and Sun, 1995) REE plot (Figs. 8d-f),
ip t
the magnesian samples are characterized by light REE (LREE, e.g. La) enriched patterns with smooth constant heavy REE (HREE, e.g. Yb), except for the HREE
cr
depleted KDG-1 sample. There is no obvious Eu anomaly in the magnesian
us
samples although the garnet charnockite shows a slight positive Eu anomaly (Fig. 8d). The ferroan sample of the KC (KDG-2) has LREE enriched and HREE
an
depleted pattern whereas that of the HC (HC-7) has a LREE depleted pattern with smooth HREE. The garnet amphibolite (sample KDG-2) suggests
M
continuous fractionation from LREE to HREE ((La/Sm)cn = 2.0 and (Gd/Yb)cn =
d
2.4) (Fig. 8e). The felsic samples are highly enriched in LREE and show clear
te
negative Eu anomaly. Further, these samples have relatively flat HREE pattern. Although the samples of the felsic group are from two complexes (WC-1 and
Ac ce p
KDG-4), whole rock chemistry, trace element and REE patterns are remarkably similar.
4.3.2. Petrogenetic implications
The metagranodiorite and metadiorite fall in the field of volcanic-arc granites (VAG) in the Rb-Y-Nb and Rb-Yb-Ta diagrams of Pearce et al. (1984) (Fig. 9). The arc magmatic signature is also supported by the primitive-mantle normalized trace element plot (Figs. 8a,c) characterized by the enrichment of LILE, negative
Page 39 of 111
P a g e | 40 Nb, Ta, P and Ti anomalies, and relatively constant HFSE of the metagranodiorite and metadiorite. LREE enrichment relative to middle REE and relatively constant HREE of the metagranodiorite and metadiorite in the chondrite normalized REE
ip t
plot (Figs. 8d,f) also support volcanic arc affinity of the felsic to intermediate
cr
rocks.
us
In the TiO2-MnO-P2O5 triangular diagram (Fig. 10a) for basaltic rocks, the mafic granulite, magnesian metagabbro, and garnet charnockite (samples HC-1, HC-8,
an
HC-10, and HC-11) show island-arc tholeiite affinity, whereas the ferroan metagabbro (sample HC-7) and garnet amphibolite (sample KDG-2) are
M
classified as MORB and oceanic island alkali basalt, respectively. The arc
d
signature of the magnesian rocks is also supported by enrichment of LILE
te
elements in Fig. 8a, and LREE enrichment relative to middle REE and relatively constant HREE in Fig. 8d. On the other hand, continuous fractionation from
Ac ce p
LREE to HREE, high P2O5 content in TiO2-MnO-P2O5 plot, and high Nb content in Nb-Zr-Y plot support oceanic-island affinity of the garnet amphibolite. The MORB signature of ferroan metagabbro is supported by depleted in LILE and positive Nb and Ta anomalies in Fig. 8b, and slightly LREE-depleted and nearly flat HREE pattern in Fig. 8e.
4.4. Zircon U-Pb geochronology and Lu-Hf isotopes
4.4.1. U-Pb geochronology
Page 40 of 111
P a g e | 41 The zircon U-Pb analytical results are given in Supplementary Table 6, and zircon REE data are given in Supplementary Table 7. Representative CL images of the zircons from different rock types in this study are shown in Figs. 11-13, and
ip t
the age data are plotted in Figs. 14-17. Chondrite-normalized REE patterns are
cr
shown in Figs. 18 and 19.
us
Mafic granulite
Zircon grains in the mafic granulite sample HC-1 are translucent and colorless to
an
pale pink. Most of the grains are subhedral, displaying spherical to weakly elongate features with lengths ranging from 30 to 100 μm and an aspect ratio of
M
1.5:1 to 1:1. In CL images, the zircon grains are dark gray and structureless. A
d
total of 21 spots from 21 zircons were analyzed from this rock. Their Th, U and
te
Th/U values are in the range of 474-933 ppm, 1559-2280 ppm and 0.24-0.37, respectively (Supplementary Table 7). They exhibit fractionated REE patterns
Ac ce p
with LREE depletion and HREE enrichment, and weakly negative Eu anomalies (Fig. 18a). All of the analyzed spots form a coherent group on the concordia, yielding a weighted mean
206
Pb/238U age of 553.0 ± 3.2 Ma (MSWD = 2.4) (Fig
14a,b), which we interpret as the timing of metamorphism.
Metagabbro The zircon grains in metagabbro sample HC-7 are translucent, colorless or light brown. The euhedral to anhedral grains show ellipsoidal or prismatic morphology with a length of 50-200 μm and a length to width ratio of 2:1 to 1:1. In
Page 41 of 111
P a g e | 42 CL images, the grains show slight patchy zoning patterns or structureless domains with homogeneous gray color. A total of 15 spots from 15 zircons were analyzed from this rock. Their Th, U and Th/U contents show ranges of 5-95
ip t
ppm, 32-298 ppm, and 0.14-0.33, respectively (Supplementary Table 7). They exhibit slight fractionated REE patterns with LREE depletion and HREE
cr
enrichment, and variable negative Eu anomalies (Fig. 18b). All the spots are
us
distributed near the concordia line and can be divided into two groups (Fig 14c,d). Seven spot ages range from 555 ± 16 Ma to 589 ± 12 Ma and define a
an
weighted mean 206Pb/238U age of 576.8 ± 9.3 Ma (MSWD = 0.92) with Th/U ratios (0.15-0.33). Another eight spots form a coherent tight group on the concordia and 206
Pb/238U age of 523.0 ± 7.1 Ma (MSWD = 0.54) with
M
yield a weighted mean
Ac ce p
Garnet charnockite
te
distinct thermal events.
d
Th/U ratios (0.14-0.32). These two groups of ages are regarded to represent two
The zircon grains in garnet charnockite sample HC-10 are translucent,
colorless or light brown. The subhedral to anhedral grains show ellipsoidal or irregular morphology with a length of 50-300 μm and a length to width ratio of 3:1 to 1:1. In CL images, some grains possess dark inherited core and bright rim texture. A total of 29 spots from 29 zircons were analyzed from this rock. Their Th, U and Th/U contents show a range of 5-289 ppm, 36-758 ppm, and 0.021.42, respectively (Supplementary Table 7). The inherited cores of zircon exhibit distinct fractionated REE patterns and with negative Eu anomalies (Fig. 18b),
Page 42 of 111
P a g e | 43 whereas the bright rims of zircon shows slightly fractionated and even flat REE patterns with variable negative Eu anomalies (Fig. 18c). The majority of U-Pb data are distributed along the concordia or along the Pb loss line and yield a
ip t
lower intercept age of 533 ± 33 Ma (MSWD = 4.6) which is taken to indicate the timing of metamorphism of this rock. Twenty spots show ages ranging from 509.0
206
Pb/238U ages of 511.1 ±
us
be divided into three groups yielding weighted mean
cr
± 5.8 Ma to 624.0 ± 11.4 Ma and form a cluster near the lower intercept and can
5.9 Ma (MSWD = 0.13), 540.4 ± 6.0 Ma (MSWD = 1.5) and 579 ± 10 Ma (MSWD
an
= 0.13) (Fig 15a,b) with Th/U ratios 0.21-0.52, 0.24-1.42, and 0.02-0.24, respectively. These ages might represent three distinct thermal events of zircon
d te
Mafic granulite sill
M
growth.
The zircons from mafic-granulite sill sample HC-11 are colorless or light
Ac ce p
brown, transparent to translucent. The subhedral to anhedral grains show ellipsoidal, irregular or elongate morphology with a length of 40-300 μm and a length to width ratio of 3:1 to 1:1. In CL images, most zircons are bright with structureless or slight patchy zoning, and a few zircons show dark core and bright rim textures, suggesting their complex origin. A total of 22 spots from 22 zircons were analyzed from this rock. Their Th, U and Th/U contents show a range of 11-205 ppm, 47-720 ppm, and 0.09-0.91, respectively (Supplementary Table 7). Most spots exhibit slightly fractionated or flat REE patterns with variable negative Eu anomalies (Fig. 18d). Sixteen analyses of CL-bright, structureless
Page 43 of 111
P a g e | 44 domains exhibit coherent age and yield a weighted mean
206
Pb/238U age of 539.1
± 4.4 Ma (MSWD = 1.6) (Fig 15 c, d). Their Th/U ratios range between 0.14 and 0.91 with 10 of the 16 spot showing Th/U>0.3. The remaining six analyses of CL-
ip t
dark core or bright patchy domains show older ages between 563 ± 4 Ma and 665 ± 10 Ma. These spots have variable Th/U ratios from 0.09 to 0.35. The
cr
oldest core age of ca. 665 Ma might represent the timing of magmatic
us
emplacement of these rocks, and the weighted mean age of ca. 539 Ma from the
an
bright structureless zircons marks the metamorphic age.
M
Metadiorite
d
The zircons from metadiorite sample KDG-1 are colorless or light brown,
te
transparent to translucent. The euhedral to subhedral grains show near-spherical or prismatic morphology with a length of 40-200 μm and a length to width ratio in
Ac ce p
the range of 3:1 to 1:1. In CL images, most zircons show oscillatory-zoned core and bright thin rim texture. A total of 34 spots from 33 zircons were analyzed from this rock. Their Th, U and Th/U contents show a range of 119-806 ppm, 226-1417 ppm, and 0.13-1.06, respectively (Supplementary Table 7). They exhibit distinct fractionated patterns with variable negative Eu anomalies (Fig. 19a). All of the data when regressed define an upper intercept age of 916 ± 57 Ma and a lower intercept age of 462 ± 180 Ma (MSWD = 2.7) (Fig. 16 a,b). Among these, thirty spots exhibit a wide range of 206Pb/238U ages between 659 ± 5 Ma and 1086 ± 11 Ma suggesting variable Pb loss after the emplacement of the protolith in early
Page 44 of 111
P a g e | 45 Neoproterozoic. The remaining four spots form a coherent group near the lower intercept and yield a weighted mean
206
Pb/238U age of 532 ± 18 Ma (MSWD =
ip t
6.2), representing the metamorphic age.
cr
Garnet amphibolite
us
Zircons in the garnet amphibolite sample KDG-2 are colorless or light brown, transparent to translucent. The subhedral to anhedral grains show near-spherical
an
or irregular morphology with a length of 50-250 μm and a length width ratio of 3:1 to 1:1. Most grains are structureless and light gray in CL images. A total of 21
M
spots from 21 zircons were analyzed from this rock. Their Th, U and Th/U
d
contents show a range of 1-96 ppm, 37-195 ppm, and 0.03-0.60, respectively
te
(Supplementary Table 7). Their REE patterns can be divided into two types, the first type is characterized by LREE depletion and HREE enrichment, whereas the
Ac ce p
second type by flat HREE (Fig. 19b). Two spot ages show 923 Ma (Th/U=0.03) and 848 Ma (Th/U=0.07). Nineteen spots with
206
Pb/238U age ranging from 501 ±
9 Ma to 549 ± 14 Ma form a coherent tight group and yield a weighted mean 206
Pb/238U age of 520.7 ± 6.6 Ma (MSWD = 1.6) with Th/U ratios 0.32-0.60
(Supplementary Table 7, Fig. 16 c,d ). The Cambrian ages might mark the timing of metamorphism of this rock.
Meta-granodiorite
Page 45 of 111
P a g e | 46 Zircons in meta-granodiorite sample KDG-4 are colorless or light brown, and transparent to translucent. The euhedral to subhedral grains show near-spherical or prismatic morphology with a length of 30-200 μm and a length to width ration
ip t
of 3:1 to 1:1. In CL images, most zircons show clear oscillatory zoning and some of the grains display heterogeneous fractured domains. A total of 33 spots from
of
182-2242
ppm,
385-2485
ppm,
and
0.35-1.51,
respectively
us
range
cr
31 zircons were analyzed from this rock. Their Th, U and Th/U contents show a
(Supplementary Table 7), suggesting magmatic crystallization. Moreover, these
anomalies (Fig. 19c). Spot 4.4 yields
206
an
spots exhibit distinct fractionated REE patterns with strong negative Eu Pb/238U spot age of 538 ± 6 Ma, probably
M
corresponding to metamorphism. The remaining 32 spots are distributed along 207
Pb/206Pb age of 980 ± 16 Ma
d
the concordia and yield a weighted mean
this rock.
te
(MSWD = 1.13) (Fig. 16 e,f), which is taken to indicate the emplacement age of
Ac ce p
Zircons in the meta-granodiorite sample WC-2 are colorless or light brown
and translucent. The euhedral to subhedral grains show near-spherical or prismatic morphology with a length of 50-250 μm and a length to width ratio of 3:1 to 1:1. In CL images, most zircons show core-rim texture with weakly light rim and clear oscillatory zoning or show heterogeneous fractured core. A total of 37 spots from 37 zircons were analyzed from this rock. Their Th, U and Th/U contents show a range of 56-776 ppm, 55-1550 ppm, and 0.08-1.05, respectively (Supplementary Table 7). They have variable LREE contents but similar and fractionated HREE patterns with negative Eu anomalies (Fig. 19d). Six spots
Page 46 of 111
P a g e | 47 form a tight group (Group 1) with and a weighted mean
206
206
Pb/238U ages between 789 Ma and 822 Ma
Pb/238U age of 805 ± 12 Ma (MSWD = 4.0) with high
Th/U ratios 0.34-0.96. Six spots define the second group (Group 2) with a narrow
206
206
Pb/238U ages between 723 Ma and 745 Ma and a weighted mean
ip t
range of
Pb/238U age of 734.0 ± 4.6 Ma (MSWD = 1.1) (Th/U ratios 0.19-1.05) (Fig. 17). 206
Pb/238U ages
cr
Eighteen spots constitute another tight group (Group 3) with
us
between 524 Ma and 565 Ma and a weighted mean 206Pb/238U age of 546.0 ± 5.7 Ma (MSWD = 7.8) and relatively low Th/U ratios 0.08-0.48 (15 of the 18 spots
an
show Th/U<0.27, 12 of the 18 spots show Th/U<0.18). The result suggests magmatic emplacement of the protolith at ca. 805 Ma and another thermal event
Ac ce p
te
4.4.2 Zircon Lu-Hf isotopes
d
M
at ca. 734 Ma, followed by metamorphism ca. 546 Ma.
In situ Hf isotope compositions of the dated zircons are given in Table 8, and
plotted in Fig 20. In the eight samples analyzed, the fLu/Hf values of zircons vary from -0.92 to -1.0, which are obviously lower than the fLu/Hf values of mafic crust (0.34, Amelin et al., 2000) and sialic crust (-0.72, Vervoort and Patchett, 1996). Therefore, the two-stage model age is considered to be a better representation than the single stage model age to evaluate the time of source material extraction from the depleted mantle or the residence time of the source material in the crust (Blichert-Toft and Albarede, 1997).
Page 47 of 111
P a g e | 48 Mafic granulite
Seventeen zircons were analyzed from mafic granulite sample HC-1 for Lu176
Hf/177Hf ratios in the range of 0.282365 to
ip t
Hf isotopes. The result show initial
0.282432. The εHf(t) values show a tight cluster from -2.2 to 0.1 with an average
cr
value of -1.1, falling close the line of the Chondrite Uniform Reservoir (CHUR)
us
(Fig. 20). The Hf depleted model ages (TDM) are between 1135 Ma and 1227 Ma (mean 1184 Ma), and Hf crustal model ages (TDMC) show a range of 1501-1651
an
Ma (mean 1582 Ma). The results suggest a mixed source from both juvenile
M
Neoproterozoic and reworked Mesoproterozoic components.
te
d
Metagabbro
Twelve zircons were analyzed from metagabbro sample HC-7 for Lu-Hf
Ac ce p
isotopes. The result shows initial
176
Hf/177Hf ratios ranging from 0.282221 to
0.282280. The εHf(t) values are in the range of -7.5 to -5.5 with an average value of -6.3. The Hf depleted model ages (TDM) are between 1338 Ma and 1418 Ma (mean 1372 Ma), and Hf crustal model ages (TDMC) in the range of 1847-1978 Ma (mean 1902 Ma). The predominantly negative εHf(t) values suggest that the source material involved reworked Paleoproterozoic crust.
Garnet charnockite
Page 48 of 111
P a g e | 49 Twenty-five zircons were analyzed from the garnet charnockite sample HC10 for Lu-Hf isotopes. The result show initial
176
Hf/177Hf ratios ranging from
0.281500 to 0.282144. The εHf(t) values are in the range of -33.3 to -10.5 with an
ip t
average value of -17.7. The Hf depleted model ages (TDM) are between 1522 Ma and 2401 Ma (mean 1809 Ma), and Hf crustal model ages (TDMC) varies 2156-
cr
3580 Ma (mean 2614 Ma averagely). The highly negative εHf(t) values suggest
us
that the source material for the charnockite was reworked Paleoproterozoic crust. Twenty zircons were analyzed from mafic granulite sample HC-11 for Lu-Hf 176
Hf/177Hf ratios in the range of 0.281857 to
an
isotopes. The result show initial
0.282095. The εHf(t) values are in the range of -20.5 to -12.1 with an average
M
value of -14.1. The Hf depleted model ages (TDM) are between 1590 Ma and
d
1922 Ma (mean 1666 Ma), and Hf crustal model ages (TDMC) varies from 2263 to
te
2790 Ma (mean 2386 Ma). The dominantly negative εHf(t) values suggest that the source material for the magma evolved from the Neoarchean-Paleoproterozoic
Ac ce p
crust.
Metadiorite
Twenty-six zircons were analyzed from metadiorite sample KDG-1 for Lu-Hf
isotopes. The result show initial
176
Hf/177Hf ratios ranging from 0.282270 to
0.282505. The εHf(t) values are in the range of -1.1 to 7.2 with an average value of 1.5. The Hf depleted model ages (TDM) are between 1031 Ma and 1360 Ma (mean 1259 Ma), and Hf crustal model ages (TDMC) varies from 1206 to 1733 Ma
Page 49 of 111
P a g e | 50 (mean 1569). The results suggest a mixed source from both juvenile and Proterozoic components.
ip t
Garnet amphibolite
176
Hf/177Hf ratios in the range of
us
for Lu-Hf isotopes. The result show initial
cr
Twenty zircons were analyzed from the garnet amphibolite sample KDG-2
0.281845 to 0.282058. The εHf(t) values are in the range of -21.3 to -13.8 with an
an
average value of -16.5. The Hf depleted model ages (TDM) are between 1641 Ma and 1931 Ma (mean 1744 Ma), and Hf crustal model ages (TDMC) varies from
M
2356 to 2828 Ma (mean 2526 Ma). The predominantly negative εHf(t) values
Ac ce p
Meta-granodiorite
te
Paleoproterozoic crust.
d
suggest that the source material was evolved from the Neoarchean-
Twenty-six zircons were analyzed from meta-granodiorite sample KDG-4 for
Lu-Hf isotopes. The result show initial
176
Hf/177Hf ratios ranging from 0.282230 to
0.282341. The εHf(t) values are in the range of 2.5 to 6.4 with an average value of 3.9. The Hf depleted model ages (TDM) are between 1266 Ma and 1417 Ma (mean 1365 Ma), and Hf crustal model ages (TDMC) varies from 1431 to 1679 Ma (mean 1592 Ma). The results suggest a mixed source from both juvenile and Paleo-Mesoproterozoic components. The dominantly positive εHf(t) values
Page 50 of 111
P a g e | 51 suggest that the source material was evolved from the Mesoproterozoic juvenile material. Thirty-two zircons were analyzed from metagranodiorite sample WC-2 for 176
Hf/177Hf ratios range from 0.281968 to
ip t
Lu-Hf isotopes. The result show initial
0.282283. The εHf(t) values are in the range of -15.0 to -3.9 with an average
cr
value of -6.2. The Hf depleted model ages (TDM) are between 1344 Ma and 1789
us
Ma (mean 1440 Ma), and Hf crustal model ages (TDMC) varies from 1799 to 2498 Ma (mean 1944 Ma). The dominantly negative εHf(t) values suggest that the
an
source material was evolved from reworked Paleoproterozoic crust.
d
M
5. Discussion
te
The samples analyzed in this study are petrographically classified as metagranodiorite (samples KDG-4 and WC-2; Pl + Kfs + Qtz + Bt + Hbl),
Ac ce p
metadiorite (sample KDG-1; Pl + Qtz + Bt + Hbl), mafic granulite (samples HC-1 and HC-11; Cpx + Opx + Pl + Hbl), metagabbro (samples HC-7 and HC-8; Grt + Cpx + Opx + Pl), garnet amphibolite (sample KDG-2; Grt + Hbl + Pl) and garnet charnockite (sample HC-10; Grt + Pl + Qtz + Opx + Cpx). Pressure and temperature conditions of the study area were estimated based on conventional geothermobarometry on mafic granulite, amphibolite, metagranodiorite, and metadiorite. The peak P-T conditions were estimated from garnet-clinopyroxeneplagioclase-quartz assemblages in samples HC-7 and HC-8 as 830-860°C and 9.4-9.8 kbar. Slightly lower temperature ranges of 700-720°C obtained for garnet-
Page 51 of 111
P a g e | 52 hornblende pairs in garnet amphibolite (sample KDG-2) and 720−780°C for hornblende-plagioclase pairs in metagranodiorite and metadiorite probably correspond to retrograde conditions. The peak metamorphic conditions of ~860
ip t
°C and ~9.8 kbar estimated in this study are consistent with previous P-T estimates from the Highland Complex (~850-900 °C and ~8-10 kbar; Schenk et
cr
al., 1988; Schumacher et al., 1990; Kriegsman, 1991; Hiroi et al., 1994, Raase
us
and Schenk, 1994; Schumacher and Faulhaber, 1994; Kriegsman and Schumacher, 1999; Dharmapriya et al., 2014). The UHT metamorphic conditions
an
of >900 °C or even >1100 °C estimated based on the stability of diagnostic mineral assemblages (e.g., sapphirine + quartz) and phase equilibria (e.g.,
M
Kriegsman and Schumacher, 1999; Osanai et al., 2006; Sajeev and Osanai,
te
Highland Complex.
d
2004a, 2004b; Sajeev et al., 2007) is mainly restricted to some localities in the
Ac ce p
The petrographic and geochemical features discussed in previous sections suggest distinct petrogenetic history for the protoliths of the meta-igneous suite from the crustal blocks of Sri Lanka. The REE and trace element patterns (Fig. 8) as well as Rb-Y-Nb and Rb-Yb-Ta diagrams (Fig. 9) reveal volcanic arc affinity for the metagranodiorite, metadiorite and garnet charnockite, suggesting that the protoliths of the rocks were derived from felsic to intermediate arc magmas. The two mafic granulites (HC-1 and HC-11) and magnesian metagabbro (HC-8) also suggest volcanic arc affinity. The garnet-free mafic granulite (sample HC-1) occurring within marble and magnesian metagabbro (sample HC-8) enclosed in
Page 52 of 111
P a g e | 53 quartzite shows evidence for subduction-related mafic magmatism in the crustal blocks of WC, KC and the HC.
ip t
In contrast, the protolith of sample HC-7 (ferroan metagabbro from the Highland Complex) is inferred as N-MORB, suggesting the accretion of remnant accreted
cr
oceanic lithosphere accreted along with the arc components and trench
us
sediments during the subduction-collision event. The garnet amphibolite (KDG-2)
an
displays a signature of oceanic island alkali basalt.
From the Kadugannawa Complex, a metadiorite (KDG-1), garnet amphibolite
M
(KDG-2), and a meta-granodiorite (KDG-4) were analyzed in this study. Zircons in
d
the metadiorite define an upper intercept age of 916 ± 57 Ma and a lower
te
intercept age of 462 ± 180 Ma (MSWD = 2.7). Metamorphic zircons in this rock form a coherent group and yield a weighted mean
206
Pb/238U age of 532 ± 18 Ma
Ac ce p
(MSWD = 6.2). Two zircon grains from the garnet amphibolite show spot ages of 923 Ma and 848 Ma (Th/U=0.07). The majority of zircons in this rock are of metamorphic origin and define a weighted mean
206
Pb/238U age of 520.7 ± 6.6
Ma (MSWD = 1.6). Zircons in the meta-granodiorite sample yield a weighted mean
207
Pb/206Pb age of 980 ± 16 Ma (MSWD = 1.13) marking the timing of
emplacement of the protolith. From the Wanni Complex, zircons in a metagranodiorite (WC-2) define three groups with a weighted mean
206
Pb/238U
age of 805 ± 12 Ma (MSWD = 4.0) (emplacement of the magmatic protolith) 734.0 ± 4.6 Ma (MSWD = 1.1) (Cryogenian thermal event) and 546.0 ± 5.7 Ma
Page 53 of 111
P a g e | 54 (MSWD = 7.8) (latest Neoproterozoic-Cambrian metamorphism). Three samples were analyzed from the Highland Complex. All of the analyzed spots in a mafic granulite (HC-1) form a coherent group on the 206
Pb/238U age of 553.0 ± 3.2 Ma (MSWD =
ip t
concordia, yielding a weighted mean
2.4) marking the timing of metamorphism. Zircons in a metagabbro sample (HC206
Pb/238U ages of 576.8 ± 9.3 Ma
cr
7) form two groups with weighted mean
us
(MSWD = 0.92) and 523.0 ± 7.1 Ma (MSWD = 0.54) suggesting two distinct thermal events. Zircons from a garnet charnockite (HC-10) show Pb loss with a
an
lower intercept age of 533 ± 33 Ma (MSWD = 4.6) marking the timing of metamorphism. Another group of zircons from this rock show spot ages in the
M
range of 509.0 ± 5.8 Ma to 624.0 ± 11.4 Ma with weighted mean
206
Pb/238U ages
d
of 511.1 ± 5.9 Ma (MSWD = 0.13), 540.4 ± 6.0 Ma (MSWD = 1.5) and 579 ± 10
te
Ma (MSWD = 0.13), suggesting different stages of zircon growth during the late Neoproterozoic. Zircons in a mafic granulite sill (HC-11) yield a weighted mean Pb/238U age of 539.1 ± 4.4 Ma (MSWD = 1.6). Another group with CL-dark
Ac ce p
206
core or bright patchy domains shows older ages between 563 ± 4 Ma and 665 ± 10 Ma, and a weighted mean metamorphic age of 539 Ma from the bright structureless zircons.
Zircons in the metadiorite (KDG-1) and metagranodiorite (KDG-4) from the
Kadugannawa complex show mostly positive εHf(t) values in the range of -1.1 to 7.2 and Hf crustal model ages (TDMC) varies from 1206 to 1733 Ma suggesting a mixed source from both juvenile and Paleo-Mesoproterozoic components. However, zircons in the garnet amphibolite (KDG-2) show dominantly negative
Page 54 of 111
P a g e | 55 εHf(t) values (mean -16.5) with TDMC in the range of 2356 to 2828 Ma suggesting reworked Neoarchean-Paleoproterozoic crustal source. Zircons in the Wanni Complex metagranodiorite (WC-2) also possess negative εHf(t) values (mean -
ip t
6.2) with TDMC in the range of from 1799 to 2498 Ma suggesting reworked Paleoproterozoic crust as the magma source.
cr
The Highland Complex rocks preserve distinct imprints of reworking of
us
older crust. The zircon εHf(t) values in mafic granulite show a tight cluster from 2.2 to 0.1 with TDMC in the range of 1501-1651 Ma suggesting a mixed source
an
from both juvenile Neoproterozoic and reworked Mesoproterozoic components. Zircons in the metagabbro (HC-7) from this complex show negative εHf(t) values
crust as the source.
M
(mean -6.3) and TDMC of 1847-1978 Ma suggesting reworked Paleoproterozoic Zircons in the garnet charnockite (HC-10) also display
te
d
highly negative εHf(t) values (mean -17.7) and older TDMC) (mean 2614 Ma) suggesting reworked Paleoproterozoic crust. Zircons in the mafic granulite sill
Ac ce p
(HC-11) also show negative εHf(t) values (mean -14.1) and TDMC between 2263 to 2790 Ma suggesting that the source material for the magma evolved from the Neoarchean-Paleoproterozoic crust. The εNd(t) and εHf(t) isotopic data reported in previous studies and those
from the present study from the Sri Lankan crustal blocks provide evidence for magma generation from mixed sources involving both juvenile and reworked components. The Nd isotope data reported elsewhere for the KC and the WC show highly heterogeneous εNd(t) values, ranging from +2 to -10 (Willbold et al., 2004), and +4 to -8 (Millisenda et al., 1994). These data suggest variable crustal
Page 55 of 111
P a g e | 56 contamination of mantle-derived melts in a continental magmatic arc setting. However, in clear contrast, the HC rocks show dominantly negative εNd(t) values from -7 to -25 suggesting a different tectonic genetic setting from that of the KC
ip t
and the WC, with substantial contribution from older reworked crustal components.
cr
In summary the 916 to 980 Ma ages from the Kadugannawa Complex
us
represent arc magmatism during early Neoproterozoic. Zircon in the diorites from this complex show positive εHf(t) values suggesting juvenile addition. This was
an
followed by the 805 Ma tonalitic magma emplacement in the Wanni Complex. Multiple thermal events during Neoproterozoic are recorded from the Wanni and
M
Highland Complexes, with subduction-related signature, culminating in Cambrian
te
d
high-grade metamorphism (~550 Ma) that overprinted all the three complexes.
We propose a tectonic model where we envisage that the Wanni and Vijayan
Ac ce p
complexes as continental arcs underlain by Neoarchean – Paleoproterozoic basement (Fig. 21). The similarity in the arc-related geochemical features of the magmatic rocks emplaced broadly and coevally in both Wanni and Vijayan complexes prompt us to suggest a double-sided subduction of an oceanic lithosphere during early to mid Neoproterozoic. The broadly negative εNd(t) and εHf(t) values of these magmatic suites, together with the Neoarchean – Paleoproterozoic two stage Hf crustal model ages are also in accordance with melting of underlying older basement in both of these blocks, as well as the presence of older crust within the WC. The mafic/intermediate suites in the KC
Page 56 of 111
P a g e | 57 complex
(metadiorites)
show
input
of
juvenile
components
within
a
suprasubduction setting. In our model, the WC is envisaged as an accretionary suture with both oceanic and continental components, together with fragments of
ip t
older continental crust, analogous to the tectonic milieu in modern Phanerozoic accretionary belts such as the Central Asian Orogenic Belt (e.g., Xiao and The N-MORB signature of the garnet-bearing metagabbro
cr
Santosh, 2014).
us
(sample HC-7) and the oceanic alkali basalt chemistry of the garnet amphibolite (sample KDG-2) support our model of the accretion of remnants of the oceanic
an
lithosphere along with trench sediments onto the arc (continent). Occurrence of granitic-ring structures, serpentinites, anorthosites, and Cu-magnetite deposits in
M
association with chert bands and apatite in the VC at the boundary with the HC
d
(Pathirana, 1980; Cooray, 1984) indicates active continental margin volcanism
te
and supra-subduction zone magmatism at the junction of the HC and the VC. The submarine canyon with wall heights >1 km in the north east VC and hot
Ac ce p
springs are contiguous with the HC-VC contact zone. These features lend further support our subduction model.
Importantly, our study identifies juvenile magmatic additions during mid
Neoproterozoic (KDG-1, associated with subduction) and latest Neoproterozoic – Cambrian (HC-1; magmatism in post-collisional asthenospheric upwelling). The post-collisional slab-break off and asthenospheric upwelling is also envisaged as the principal contributor of heat that triggered widespread ultrahigh-temperature metamorphism within the HC, similar to the scenario envisaged for the formation
Page 57 of 111
P a g e | 58 of the UHT rocks elsewhere (e.g., Santosh et al., 2013).
As discussed in earlier sections, previous studies had identified the HC, WC and
ip t
VC as distinct tectonic units based largely on Nd model age mapping, combined with regional structural interpretations. The Nd model ages are as follows: HC=2-
cr
3.4 Ga; WC=1-2 Ga; VC=1-1.8Ga; and KC=1.4-1.8Ga. The zircon ages from
us
these three different tectonic units (either magma emplacement ages or depositional ages of sediments) are HC =~ 2.3Ga; WC=750-1.1Ga; VC=650Clearly the HC preserves records of the oldest
an
1.9Ga; and KC=0.8-1.0Ga.
sources of sedimentary units in Sri Lanka, and/or the oldest recycled continental
M
crust in this island. In the present study also, the oldest TDMC (2790 Ma) is Such older
d
obtained from zircons in a garnet charnockite within the HC.
te
components are common in many younger accretionary belts where ancient micro-continents or arcs are accreted and admixed during the final collision In the present case, the subduction-accretion continued until late
Ac ce p
stage.
Neoproterozoic and the final architecture of the HC was built during the Cambrian collisional event.
The ‘flower-style’ extrusion of the HC is clearly
marked by the thrust zone mapped in previous studies at the border of the HC with the VC; the structural break is also corroborated by geochronological studies that show a distinct boundary between the HC and the VC. However, the boundary is unclear on the western margin of the HC at its contact with the WC, where the large arc magma chamber of the KC is found emplaced at the inferred boundary. Away from the KC to the north and south-west also the HC-WC
Page 58 of 111
P a g e | 59 boundary is not clear. However, the doubly plunging synforms defined by the KC are considered to mark the structural boundary in the central region along the HC-WC inferred boundary. Such large-scale sheath folds are considered as the
ip t
hallmarks of major accretionary orogens elsewhere (e.g., Chetty et al., 2012). According our model, KC is not a discrete crustal block, but part of a disrupted
cr
huge arc magma chamber that was exhumed and transposed along the margin
us
of the WC.
an
Some of the existing models suggest that the WC, HC and VC were discrete terranes (Vitanage, 1972, 1985). The WC is proposed to have collided with HC
M
first, and subsequently, both WC and HC together as a unified block collided
d
with, and was thrust over, the VC (Vitanage, 1972; Voll and Kleinschrodt, 1991;
te
Kröner and Jaeckel, 1994; Kriegsman, 1995) during the amalgamation of Gondwana (Kriegsman, 1995; Kehelpannala 1997, 2003, 2004; Kröner et al.
Ac ce p
2003). However, there is no robust geochronologic and tectonic evidence to support this speculation, particularly the two discrete docking events, which should produce two separate metamorphic orogens with different ages along both margins of the HC. In fact, our data, coupled with available information clearly show that the WC and the VC were coeval Neoproterozoic arcs which were welded along the HC during late Neoproterozoic-Cambrian. The Cambrian collision event was accompanied by metamorphic overprint in all the rocks, formation of UHT metamorphic orogen along the HC, and post-collisional mantlederived magmatism within the HC (e.g. sample HC-1 in our study), all suggesting
Page 59 of 111
P a g e | 60 the HC as a collisional suture.
Whether the older continental basement
underlying WC and VC were contiguous prior to Neoproterozoic, or whether these bocks were derived independently from other continental blocks, remains
ip t
to be investigated in future studies.
cr
In previous studies, the magmatic episodes ranging from the Grenvillian
us
(ca. 1.1 Ga) to ca. 882 Ma in the KC (e.g. Kröner et al., 2003; Willbold et al., 2004) were considered to mark the beginning of the breakup of the Rodinia
an
supercontinent. However, the Neoproterozoic magmatic pulses in the Sri Lankan crustal complexes are clearly related to convergent margin setting as shown in
M
our study, as against the extensional setting required for Rodinia breakup. Except
d
for the Late Neoproterozoic-Cambrian mafic dykes (sample HC-1), and some
te
latest Neoproterozoic syn- and post-collisional granitoids in the HC, there are no other rock records of widespread extension-related mafic or bimodal magmatism
Ac ce p
in Sri Lanka during mid Neoproterozoic, and therefore a Rodinia break-up model to explain the magmatic suites is untenable. Accretion of oceanic components and arc magmatism in a convergent margin setting are also clearly indicated from our results. The global assembly of the Rodinia supercontinent with South China at its center is considered to have occurred at ca. 900 Ma (Li et al., 2008). Initial rifting of Rodinia started at ca. 860 with widespread breakup during 830800 Ma. From the Eastern Ghats Belt in India, Korhonen et al. (2011) reported SHRIMP ages of 980 Ma in monazites associated with garnet and cordierite from ultrahigh-temperature Mg–Al granulites, marking the timing of collisional
Page 60 of 111
P a g e | 61 amalgamation of this terrane within the Rodinia assembly. They also reported a range of
207
Pb/235U monazite ages (from ca. 1014 Ma to 959 Ma for one sample
and ca. 1043 Ma to 922 Ma for the second sample) suggesting protracted
ip t
monazite growth during the high-temperature retrograde evolution, which would broadly correspond with the global assembly of the Neoproterozoic Rodinia However, the data from crustal blocks in Sri Lanka indicate
cr
supercontinent.
us
protracted arc magmatism in a convergent margin setting, commencing in early Neoproterozoic, continuing through mid Neoproterozoic and culminating in
suture (HC).
an
regional metamorphism in all the blocks, with UHT imprint along the collisional The arc-related signature of these magmatic suites typical of
M
convergent margin magmas prompt us to suggest that the Neoproterozoic
d
magmatic events culminating in Cambrian collisional metamorphism might be
te
related to the prolonged Pacific-type orogeny associated with the assembly of Gondwana supercontinent, as also proposed for the Southern Granulite Terrane
Ac ce p
in India (Santosh et al., 2009).
Thus, our model deviates from the existing
concepts that link Sri Lanka with Rodinia rifting, and correlates the Neoproterozoic arc-related magmatic suites with convergent margin processes associated with the assembly of Gondwana.
8. Conclusions
Petrological and geochemical analyses and zircon U-Pb and Lu-Hf data from a suite of metamorphosed magmatic rocks from the major tectonic units in Sri
Page 61 of 111
P a g e | 62 Lanka, coupled with the available information from the crustal blocks, lead us to the following conclusions.
ip t
1. Regional metamorphism in the crustal bocks reached up to 860°C and 9.8 kbar, together with the extruded metamorphic orogen in the Highland
cr
Complex showing ultrahigh-temperature signature, which we correlate to
us
heat input from asthenospheric sources in a post-collisional slab-break off
an
setting.
2. Trace element and rare earth element patterns as well as Rb-Y-Nb and
M
Rb-Yb-Ta discrimination plots show volcanic arc affinity and protolith
te
d
crystallization from felsic to intermediate arc magmas.
3. The mafic magmatic units including those incorporated within the
Ac ce p
metasedimentary belt of Highland Complex show N-MORB signature, or oceanic island alkali basalt affinity, suggesting accretion of the remnants of oceanic lithosphere during the subduction-collisional event.
4. Zircon U-Pb ages ranging from 916 to 980 Ma from the Kadugannawa Complex represent arc magmatism during early Neoproterozoic, followed by the 805 Ma granodioritic magma emplacement in the Wanni Complex. Repeated thermal events during mid and late Neoproterozoic are also recorded from the Wanni and Highland Complexes, culminating in
Page 62 of 111
P a g e | 63 Cambrian high-grade metamorphism that reached ultrahigh-temperature conditions.
ip t
5. Zircon Lu-Hf data reveal mixed sources with the involvement of both juvenile and reworked Neoarchean – Paleoproterozoic basement. There
cr
is also clear indication of juvenile addition during both Neoproterozoic and
us
Cambrian tectonothermal events.
6. According to our data, the KC is part of a disrupted huge arc magma
an
chamber that was exhumed and transposed along the margin of the WC
d
simply a part of the WC.
M
which was previously regarded to be either an exotic crustal block or
te
7. A model of double-sided subduction during the Neoproterozoic, where the Wanni Complex to the west and the Vijayan Complex to the east represent
Ac ce p
continental arcs during the Neoproterozoic, culminating in collision along the Highland Complex during late Neoproterozoic-Cambrian is proposed. The occurrence of supra-subduction zone magmatic complexes and volcanic suites formed in a convergent margin setting along the contact of the VC with the HC lend support to our model.
8. The data presented in this study are consistent with a prolonged convergent margin setting from early to late Neoproterozoic, followed by final collisional amalgamation in the Cambrian, with the Highland Complex
Page 63 of 111
P a g e | 64 defining an accretionary belt as well as the collisional suture.
9. Both vertical (arc magmatic) and lateral (accretionary) growth of
ip t
continental crust during the Neoproterozoic is well recorded in the crustal
cr
blocks of Sri Lanka.
us
Acknowledgments
an
We thank Prof. Guochun Zhao, Editor and two referees for constructive and helpful comments. We extend our thanks to Qiongyan Yang and Xueming Teng at
M
the China University of Geosciences Beijing for help with zircon analyses. This study contributes to the Talent Award to M. Santosh from the Chinese
te
d
Government under the 1000 Talent Plan. Partial funding for this project was produced by a Grant-in-Aid for Scientific Research (B) from Japan Society for the
Ac ce p
Promotion of Science (JSPS) to Tsunogae (No. 26302009).
References
Amelin, Y., Lee, D.C., Halliday, A.N., 2000. Early-Middle Archaean crustal evolution deduced from Lu-Hf and U-Pb isotopic studies of single zircon grains. Geochim. Cosmochim. Acta 64, 4205-4225. Almond, D.C., 1991. Arena Gneiss and Kandy Gneiss-a proposed subdivision of the Highland Series around Kandy, and its significance. J. Geol. Soc. Sri Lanka 3, 41-50. Blichert-Toft, J., Albarede, F., 1997. The Lu-Hf isotope geochemistry of chondrites and the evolution of the mantle–crust system. Earth and Planetary Science Letters 148, 243-258.
Page 64 of 111
P a g e | 65
Baur, N, Kröner, A, Todt, W, Liew, T.C., Hofman, A.W., 1991. U-Pb isotopic systematics of zircons from prograde and retrograde transition zones in high grade orthogneisses, Sri Lanka. J. Geol. 99, 527-545.
ip t
Berger, A.R., Jayasinghe, N.R., 1976. Precambrian structure and chronology in the Highland Series of Sri Lanka. Precambrian Res. 3, 559-576.
cr
Burton, K.W., O’Nions, R.K., 1990. The timescale and mechanism of granulite formation at Kurunegala, Sri Lanka. Contrib. Mineral. Petrol. 106, 66-89.
us
Cawood, P.A., Buchan, C., 2007. Linking accretionary orogenesis with supercontinent assembly. Earth-Science Reviews 82, 217-256.
an
Chetty, T.R.K., Yellappa, T., Mohanty, D.P., Nagesh, P., Sivappa, V.V., Santosh, M., Tsunogae, T., 2012. Mega Sheath Fold of the Mahadevi Hills, Cauvery Suture Zone, Southern India: Implication for Accretionary Tectonics. J. Geol. Soc. India 80, 747-758.
M
Condie, K.C., Aster, R.C., 2013. Refinement of the supercontinent cycle with Hf, Refinement of the supercontinent cycle with Hf, Nd and Sr isotopes: Geoscience Frontiers 4, 667-680.
te
d
Condie, K.C., Kröner, A., 2013. The building blocks of continental crust: Evidence for a major change in the tectonic setting of continental growth at the end of the Archean: Gondwana Res. 23, 394-402.
Ac ce p
Cordani, U., Cooray, P.G., 1990. Rb-Sr ages of granulites and gneisses from the Precambrian of Sri Lanka. J. Geol. Soc. Sri Lanka 2, 35-43. Cooray, P.G., 1962. Charnockites and their associated gneisses in the Precambrian of Ceylon. J. Geol. Soc. London 118, 239– 273. Cooray, P.G., 1984. An introduction to the geology of Sri Lanka. National Museum Sri Lanka Publication, 340. Cooray, P.G., 1994. The Precambrian of Sri Lanka: a historic review. Precambrian Res. 66, 3-18. De Maesschalck, A.A, Oen, I.S., Hebeda, E.H., Verschure, R., Arps, E.S., 1990. Rb-Sr whole rock ages of Kataragama and Pottuvil Charnockites and east Vijayan gneiss: indication of a 2 Ga metamorphism in the highlands of south east Sri Lanka. J. Geol. 98, 772-779. Dahanayake, K., Jayasena, H.A.K., 1983. General geology and petrology of some Precambrian crystalline rocks from the Vijayan Complex of Sri Lanka.
Page 65 of 111
P a g e | 66 Precambrian Res. 19, 301–315. Dahanayake, K., 1982. Structural and petrological studies on the Precambrian Vijayan Complex of Sri Lanka. Revista Brasiliano Geociencias 12, 89–93.
cr
ip t
Dharmapriya, P.L., Malaviarachchi, S.P.K., Galli, A., Su, B., Subasinghe, N.D., Dissanayake, C.B., Nimalsiri, T.B., Zhu, B., 2014. P-T evolution of a spinel + quartz bearing khondalite from the Highland Complex, Sri Lanka: Implications for non-UHT metamorphism, J. Asian Earth Sci., Doi: http://dx.doi.org/10.1016/j.jseaes.2014.05.003.
us
Ellis, D.J., Green, D.D., 1979. An experimental study of the effect of Ca upon garnet-clinopyroxene Fe-Mg exchange equilibria. Contrib. Mineral. Petrol 71, 13-22.
an
Faulhaber, S., Raith, M., 1991. Geothermometry and geobarometry of high grade rocks: a case study on garnet pyroxene granulites in southern Sri Lanka. Min. Mag. 55, 33-56.
M
Ganguly, J., Cheng, W., Tirone, M., 1996. Thermodynamics of aluminosilicate garnet solid solution: new experimental data, an optimized model, and thermometric applications. Contrib. Mineral. Petrol. 126, 137-151.
te
d
Graham, C.M., Powell, R., 1984. A garnet-hornblende geothermometer: calibration, testing, and application to Pelona Schist, Southern California. J. Metamorph. Geol. 2, 13-31.
Ac ce p
Green, T.H., Ringwood, A.E., 1967. An experimental investigation of the gabbro to eclogite transformation and its petrological applications. Geochim. Cosmochim. Acta 31, 767-833. Hansen, E.C., Janardan, A.S., Newton, R.C., Prame, W.K.B.N., Ravindrakumar, G.R., 1987. Arrested charnockite formation in southern India and Sri Lanka. Contrib. Mineral. Petrol. 196, 66-89. Hapuarachchi, D.J.A.C., 1968. Cordierite and wollastonite-bearing rocks of south-west of Ceylon. Geol. Mag. 105, 317-324. Harley, S.L., 1989. The origin of granulites: a metamorphic perspective. Geol. Mag. 126, 215-247. Hatherton, T., Pattiarachchi, D.B., Ranasinghe, V.V.C., 1975. Gravity map of Sri Lanka, 1:1,000,000. Geol. Surv. Dept., Sri Lanka, Prof. Paper 3, 39. Hiroi, Y., Ogo, Y., Namba, L., 1994. Evidence for prograde metamorphic evolution of Sri Lankan politic granulites and implications for the development of
Page 66 of 111
P a g e | 67 continental crust. Precambrian Res. 66, 245-263.
ip t
Hiroi, Y., Yanagi, A., Kato, M., Kobayashi, T., Prame, B., Hokada, T., SatishKumar, M., Ishikawa, M., Adachi, T., Osanai, Y., Motoyoshi, Y., Shiraishi, K., 2014. Super cooled melt inclusions in lower-crustal granulites as a consequence of rapid exhumation by channel flow. Gondwana Res. 25, 226234.
cr
Holland, T., Blundy, J., 1994. Non-ideal interactions in calcic amphiboles and their bearing on amphibole-plagioclase thermometry. Contrib. Mineral. Petrol. 116, 433-447.
us
Hölzl, S., Köhler, H., Kröner, A., Jaeckel, P., Liew, T.C., 1991. Geochronology of the Sri Lankan basement. Geological Survey Department, Sri Lanka, Professional Paper 5, 237-257.
an
Hölzl, S., Hofmann, A.W., Todt, W., Köhler, H., 1994. U–Pb Geochronology of the Sri Lankan basement. Precambrian Res. 66, 123–149.
M
Hu, Z.C., Gao, S., Liu,Y.S., Hu, S.H., Chen, H.H., Yuan, H.L., 2008. Signal enhancement in laser ablation ICP-MS by addition of nitrogen in the central channel gas. Journal of Analytical Atomic Spectrometry 23, 1093–1101.
Ac ce p
te
d
Hu, Z.C., Liu,Y.S., Chen, L., Zhou, L., Li, M., Zong, K.Q., Zhu, L., Gao, S., 2011. Contrasting matrix induced elemental fractionation in NIST SRM and rock glasses during laser ablation ICPMS analysis at high spatial resolution. Journal of Analytical Atomic Spectrometry 26, 425–430. Kagami, H., Owada, M., Osanai, Y., Hiroi, Y., 1990. Preliminary geochronological study of Sri Lankan rocks. Interim Report of Japan-Sri Lanka Joint Research, p. 55-70. Karunaratne, D.N., Perera, L.R.K. and Fernando, G.W.A.R., 2002. The missing link between Prograde decompression (?) and Retrograde Isobaric Cooling in the Highland Complex Granulites of Sri Lanka. Proc. 18th Ann. Technical Sessions of the Geol. Soc. of Sri Lanka, p. 3 (Abstract). Kehelpannala, K.V.W., 1991. Structural evolution of high-grade terrains in Sri Lanka with special reference to the areas around Dodanhaslanda and Kandy. Geological Survey Department, Sri Lanka, Professional Paper 5, 69-88. Kehelpannala, K.V.W., 1997. Deformation of a high-grade Gondwana fragment, Sri Lanka. Gondwana Res. 1, 47–68. Kehelpannala, K.V.W., 1999. Shear-zone controlled arrested charnockitization,
Page 67 of 111
P a g e | 68 retrogression and metasomatism of high-grade rocks. Gondwana Res. 2, 573-577. Kehelpannala, K.V.W., 2003. Structural evolution of the middle to lower crust in Sri Lanka—a review. J. Geol. Soc. Sri Lanka 11, 45–86.
ip t
Kehelpannala, K.V.W., 2004. Arc accretion around Sri Lanka during the assembly of Gondwana. Gondwana Res. 7, 1323–1328.
cr
Kleinschrodt, R., Voll, G., Kehelpannala, W., 1991. A layered basic intrusion, deformed and metamorphosed in granulite facies of the Sri Lanka basement. Geol. Rundschau 80, 779-800.
us
Kleinschrodt, R., 1994. Large-scale thrusting in the lower crustal basement of Sri Lanka. Precambrian Res. 66, 39-57.
an
Köhler, H., Dratch, V., Holzl, S., Fehr, T., 1991. Geochronology of Sri Lanka basement. Part II: Rb-Sr and Sm-Nd dating on mineral and whole rock samples. Terra Abstracts 3, 504.
te
d
M
Koizumi, T., Tsunogae, T., Santosh, M., Tsutsumi, Y., Chetty, T.R.K., Saitoh, Y., 2014. Petrology and zircon U-Pb geochronology of metagabbros from a mafic-ultramafic suite at Aniyapuram: Neoarchean to Early Paleoproterozoic convergent margin magmatism and Middle Neoproterozoic high-grade metamorphism in southern India. J. Asian Earth Sci., doi: 10.1016/j.jseaes.2014.04.013.
Ac ce p
Korhonen, F.J., Saw, A.K., Clark, C., Brown, M., Bhattacharya, S., 2011. New constraints on UHT metamorphism in the Eastern Ghats Province through the application of phase equilibria modelling and in situ geochronology. Gondwana Res. 20, 764-781. Kriegsman, L.M., 1991. Structural geology of the Sri Lankan basement – a preliminary review. Geological Survey Department, Sri Lanka, Professional Paper 5, 52-68. Kriegsman, L., 1994. Evidence for a fold nappe in the high-grade basement of central Sri Lanka: terrane assembly in the Pan-African lower crust? Precambrian Res. 66, 59-76. Kriegsman, L., 1995. The Pan-African events in East Antarctica: a review from Sri Lanka and the Mozambique Belt. Precambrian Res. 75, 263–277. Kriegsman, L.M., Schumacher, J.C., 1999. Petrology of sapphirine-bearing and associated granulites from central Sri Lanka. J. Petrol. 40, 1211–1239.
Page 68 of 111
P a g e | 69 Kroner, A and Jaeckel, P., 1994. Zircon ages from rocks of the Wanni Complex, Sri Lanka. J.Geol. Soc. Sri Lanka 5, 41-57.
ip t
Kröner, A., Williams, I.S., Compston, W., Baur, N., Vithanage, P.W., Perera., L.R.K., 1987. Zircon ion microprobe dating of high grade rocks in Sri Lanka. J. Geol. 95, 775-791.
cr
Kröner, A., Williams, I.S., Compston, W., Baur, N., Vithanage, P.W., Perera., L.R.K., 1987. Zircon ion microprobe dating of high grade rocks in Sri Lanka. J. Geol. 95, 775-791.
an
us
Kröner, A., Cooray, P.G., Vitanage, P.W., 1991. Lithotectonic subdivision of the Precambrian basement in Sri Lanka. Geological Survey Department, Sri Lanka, Professional Paper 5, 5-21. Kröner, A., Williams, I.S., 1993. Age of metamorphism in the high-grade rocks of Sri Lanka. J. Geol. 101, 513–521.
te
d
M
Kröner, A., Jaeckel, P., Williams, I.S., 1994. Pb-loss patterns in zircons from a high-grade metamorphic terrain as revealed by different dating methods: UPb and Pb-Pb ages for igneous and metamorphic zircons from northern Sri Lanka. Precambrian Res. 66, 151-181.
Ac ce p
Kröner, A., Kehelpannala, K.V.W., Hegner, A., 2003. Ca. 750–1100 Ma magmatic events and Grenville-age deformation in Sri Lanka: relevance for Rodinia supercontinent formation and dispersal, and Gondwana amalgamation. J. Asian Earth Sci. 22, 279-300. Kröner, A., Rojas-Agramonte, Y., Kehelpannala, K. V. W., Zack, T., Hegner, E., Geng, H. Y., Wong, J., Barth, M., 2013. Age, Nd–Hf isotopes, and geochemistry of the Vijayan Complex of eastern and southern Sri Lanka: A Grenville-age magmatic arc of unknown derivation, Precambrian Res. 234, 288-321. Li, Z.X., Bogdanova, S.V., Collins, A.S., Davidson, A., De Waele, B., Ernst, R.E., Fitzsimons, I.C.W., Fuck, R.A., Gladkochub, D.P., Jacobs, J., Karlstrom, K.E., Lu, S., Natapov, L.M., Pease, V., Pisarevsky, S.A., Thrane, K., Vernikovsky, V., 2008. Assembly, configuration, and break-up history of Rodinia: a synthesis. Precambrian Res. 160, 179–210. Liu, Y.S., Hu, Z.C., Gao, S., Gunther, D., Xu, J., Gao, C.G., Chen, H.H., 2008. In situ analysis of major and trace elements of anhydrous minerals by LA-ICPMS without applying an internal standard. Chemical Geology 257, 34–43.
Page 69 of 111
P a g e | 70
Liu, Y.S., Gao, S., Hu, Z., Gao, C., Zong, K., Wang, D., 2010. Continental and oceanic crust recycling-induced melt-peridotite interactions in the Trans-North China Orogen: U-Pb dating, Hf isotopes and trace elements in zircons of mantle xenoliths. Journal of Petrology 51, 537–571.
ip t
Ludwig, K.R., 2003. ISOPLOT 3.00: A Geochronological Toolkit for Microsoft Excel. Berkeley Geochronology Center, Special Publications 1a.
cr
Malaviarachchi, S., Takasu, A. 2005. Petrographic study on the high-grade metamorphic rocks from the Highland and Kadugannawa Complexes, central Sri Lanka. Geoscience Reports of Shimane University, Japan, 24, 31-46.
an
us
Malaviarachchi, S.P.K., Takasu, A., 2011a. Petrology of metamorphic rocks from the Highland and Kadugannawa Complexes, Sri Lanka. J. Geol. Soc. Sri Lanka 14 103-122. Malaviarachchi, S.P.K., Takasu, A., 2011b. Electron Microprobe dating of Monazites from Sri Lanka. J. Geol. Soc. Sri Lanka 14, 81-90.
d
M
Mathavan, V, W. K. B. N., Prame, W.K.B.N., Cooray, P.G., 1999. Geology of the high grade Proterozoic terrains of Sri Lanka and the assembly of Gondwana: an update on recent developments, Gondwana Res. 2, 237-250.
te
Mathavan, V., Fernando G.W.A.R., 2001. Reactions and textures in grossularwollastonite-scapolite calc-silicate granulites from Maligawila, Sri Lanka. Lithos, 59, 217-232.
Ac ce p
McDonough, W.F., Sun, S.-S., 1995. The composition of the Earth. Chemical Geology 120, 223-253. Milisenda, C.C., 1991. Compositional characteristics of the Vijayan Complex. Geological Survey Department, Sri Lanka, Professional Paper 5, 135-140. Milisenda, C.C., Pohl, J.R., Hofmann, A.W., 1991. Charnockite formation at Kurunegala, Sri Lanka. Geological Survey Department, Sri Lanka, Professional Paper 5, 141-149. Milisenda, C.C, Leiw, T.C., Hofmann, A.W., Kröner, A., 1988. Isotopic mapping of age provinces in Precambrian high grade terrains: Sri Lanka. J. Geol. 96, 608-615. Milisenda, C.C., Leiw, T.C., Hofmann, A.W., Köhler, H., 1994. Nd isotopic mapping of the Sri Lanka basement: update, and additional constraints from Sr isotopes. Precambrian Res. 66, 95-110.
Page 70 of 111
P a g e | 71 Moecher, D.P., Essene, E.J., Anovitz, L.M., 1988. Calculation and application of clinopyroxene-garnet-plagioclase-quartz geobarometers. Contrib. Mineral. Petrol. 100, 92–106.
ip t
Osanai, Y., Sajeev, K., Owada, M., Kehelpannala, K.V.W., Prame, W.K.B. Nakano, N., Jayatileke, S., 2006. Metamorphic evolution of ultrahightemperature and high-pressure granulites from Highland Complex, Sri Lanka. J. Asian Earth Sci. 28, 20-37.
cr
Pathirana, H.D.N.C., 1980. Geology of Sri Lanka in relation to plate tectonics. J. National Science Council Sri Lanka. 8, 75-85.
us
Pearce, J.A., Harris, N.B.W., Tindle, A.G., 1984. Trace element discrimination diagrams for the tectonic interpretation of granitic rocks. J. Petrol. 25, 956– 983.
an
Perchuk, L.L., Safonov, O.G., Gerya, T.V., Fu, B., Harlov, D.E., 2000. Mobility of components in metasomatic transformation and partial melting of gneisses: an example from Sri Lanka. Contrib. Mineral. Petrol. 140, 212-232.
M
Perkins, D., Newton, R.C., 1981. Charnockite geobarometers based on coexistence garnet-pyroxene-plagioclase-quartz. Nature 292, 144–146.
te
d
Perera, L.R.K., 1983. The origin of pink granites of Sri Lanka – another view. Precambrian Res. 20, 17-37.
Ac ce p
Perera, L.R.K., 1984. Co-existing cordierite-almandine-a key to the metamorphic history of Sri Lanka. Precambrian Res. 25, 349-364. Perera, L.R.K., 1987, Petrogenesis of granulite-facies metamorphic rocks in Sri Lanka. University of Peradeniya (Mphil. thesis) (unpubl.). Perera, L.R.K., Kagami, H., 2011. Centimetre- and metre-scale Nd and Sr isotopic homogenization in Kadugannawa Complex, Sri Lanka. J. Geol. Soc. Sri Lanka 14, 129-141. Pohl, J.G., Emmermann, R., 1991. Chemical composition of the Sri Lanka Precambrian Basement. Geological Survey Department, Sri Lanka, Professional Paper 5, 94-123. Prame, W.K.B.N, 1991a. Metamorphism and nature of the granulite facies crust in south western Sri Lanka: Characterization by pelitic/ psammopelitic rocks and associated granulites. Geological Survey Department, Sri Lanka, Professional Paper 5, 188-200. Prame, W.K.B.N., 1991b. Petrology of the Kataragama Complex, Sri Lanka:
Page 71 of 111
P a g e | 72 evidence for high pressure granulite facies metamorphism and subsequent isobaric cooling. Geological Survey Department, Sri Lanka, Professional Paper 5, 200-224.
ip t
Prame, W.K.B.N., 1997. Geochemistry and genesis of granitoid rocks from southern Sri Lanka. Min. Petrol. 60, 245–265.
cr
Raase, P., Schenk, V., 1994. Petrology of granulite facies metapelites of the Highland Complex, Sri Lanka: implication for the metamorphic zonation and the P-T path. Precambrian Res. 66, 265-294.
an
us
Saitoh, Y., Tsunogae, T., Santosh, M., Chetty, T.R.K., Horie, K., 2011. Neoarchean high-pressure metamorphism from the northern margin of the Palghat-Cauvery Suture Zone, southern India: petrology and zircon SHRIMP geochronology. J. Asian Earth Sci. 42, 268-285.
M
Sajeev, K., Osanai, Y., 2004a. Ultrahigh-temperature metamorphism (1150° C,12 kbar) and multistage evolution of Mg-, Al-rich granulites from the Central Highland Complex Sri Lanka. J. Petrol. 45, 1821–1844.
te
d
Sajeev, K., Osanai, Y., 2004b. 'Osumilite' and 'spinel+quartz' from Highland Complex, Sri Lanka: a case of cooling and decompression after ultrahightemperature metamorphism. J. Min. Petrol. Sci. 99, 320–327.
Ac ce p
Sajeev, K., Osanai, Y., 2005. Thermal gradients in the Sri Lankan granulite terrane: a garnet–biotite thermometric approach, J. Metamorph. Geol. 23, 383-397. Sajeev, K., Osanai, Y., Suzuki, S., Kagami, H., 2003. Geochronological evidence for multistage metamorphic events in ultra-high temperature granulites from central Highland Complex, Sri Lanka. Polar Geosci. 16, 137-148. Sajeev, K., Osanai, Y., Connolly, J.A.D., Suzuki, S., Ishioka, J., Kagami, H., Rino, S., 2007. Extreme Crustal Metamorphism during a Neoproterozoic Event in Sri Lanka: A study of dry mafic granulites. J. Geol. 115, 563–582. Sajeev, K., Windley, B.F., Connolly, J.A.D., Kon, Y., 2009. Retrogressed eclogite (20 kbar, 1020ºC) from the Neoproterozoic Palghat-Cauvery suture zone, southern India. Precambrian Res. 171, 23-36. Sajeev, K., Williams, I.S., Osanai, Y., 2010. Sensitive high-resolution ion microprobe U–Pb dating of prograde and retrograde ultrahigh-temperature metamorphism as exemplified by Sri Lankan granulites. Geology 38, 971– 974.
Page 72 of 111
P a g e | 73
Sandiford, N., Powell, R., Martin, S.F., Perera, L.R.K., 1988. Thermal and baric evolution of garnet granulites from Sri Lanka. J. Metamorph. Geol. 6, 351364.
cr
ip t
Santosh, M., Yoshida, M., Nanda-Kumar, V., 1991. Fluid characteristics across a gneiss-charnockite reaction front in Sri Lanka: Implication for granulite formation in Gondwana deep crust. J. Mineral. Petrol. Econnomic Geol. 86, 27-44.
us
Santosh, M., Tsunogae, T., Shimizu, H., Dubessy, J., 2010. Fluid characteristics of retrogressed eclogites and mafic granulites from the Cambrian Gondwana suture zone in southern India. Contrib. Mineral. Petrol. 159, 349-369.
an
Santosh, M., Maruyama, S., Sato, K., 2009. Anatomy of a Cambrian suture in Gondwana: Pacific-type orogeny in southern India? Gondwana Res. 16, 321341.
d
M
Santosh, M., Yang, Q.Y., Shaji, E., Tsunogae, T., Ram Mohan, M., Satyanarayanan, M., 2013a. An exotic Mesoarchean microcontinent: The Coorg Block, southern India. Gondwana Res., DOI: 10.1016/j.gr.2013.10.005
Ac ce p
te
Santosh, M., Liu, S.J., Tsunogae, T., Li, J.H., 2013b. Paleoproterozoic ultrahightemperature granulites in the North China Craton: Implications for tectonic models on extreme crustal metamorphism. Precambrian Res. 222-223, 77106. Schenk, V., Raase, P., Schumacher, R., 1988. Very high temperatures and isobaric cooling before tectonic uplift in the Highland Series of Sri Lanka. Terra Cognita 8, 265. Schenk, V., Raase, P., Schumacher, R., 1991. Metamorphic zonation and P–T history of the Highland Complex in Sri Lanka. Geological Survey Department, Sri Lanka, Professional Paper 5, pp. 150–163. Schumacher, R., Shenk, V., Raase, P., Vithanage, P.W., 1990. Granulite facies metamorphism of metabasic and intermediate rocks in the Highland Series of Sri Lanka. In: Ashworth, J.R., Brown, M. (Eds.), High temperature metamorphism and Crustal Anatexis. The Mineralogical Society series 2, 235271. Schumacher ,R., Faulhaber, S., 1994. Summary and discussion of P-T estimates from garnet-pyroxene-plagioclase-quartz bearing granulite facies rocks from Sri Lanka. Precambrian Res. 66, 295-308.
Page 73 of 111
P a g e | 74
Shiraishi, K., Ellis, D.J. ,Hiroi, Y., Fanning, C.M., Motoyoshi, Y., Nakai, Y., 1994. Cambrian orogenic belt in east Antarctica and Sri Lanka: implications for Gondwana assembly. J. Geol. 102, 47-65.
ip t
Stosch, H.G., 1991. Geochemistry of the mafic intrusion in the synforms of the Kandy area. Geological Survey Department, Sri Lanka, Professional Paper 5, 125-134.
us
cr
Sun, S.-S., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In Saunders, A.D., Norry M.J. (Eds.), Magmatism in the Ocean Basins. Geol. Soc. London, Spec. Publ. 42, 313-345
an
Tani, Y., Yoshida, M., 1996. The structural evolution of the Arena Gneisses and its bearing on Proterozoic tectonics of Sri Lanka. J. Southeast Asian Earth Sci. 14, 309-329.
M
Vervoort, J.D., Patchett, P.J., Gehrels, G.E., Nutman, A. P., 1996. Constraints on early Earth differentiation from hafnium and neodymium isotopes. Nature 379, 624-627.
te
d
Voll, G., Evangelakakisb, C., Kroll, H., 1994. Revised two-feldspar geothermometry applied to Sri Lankan feldspars. Precambrian Res. 66, 351367.
Ac ce p
Voll, G., Kleinschrodt, R., 1991. Sri Lanka: Structural, magmatic and metamorphic development of a Gondwana fragment. Geological Survey Department, Sri Lanka, Professional Paper 5, 22-52. Vitanage, P.W., 1972. Post-Precambrian uplift and regional neotectonic movements in Ceylon. 24th Int. Geol. Congr., Montreal, Sect. 3, 642-654. Vitanage, P.W., 1985. Tectonics and mineralization in Sri Lanka. Bull. Geol. Soc. Finland 57, 157-168. Wiedenbeck, M., Alle, P., Corfu, F., Griffin, W.L., Meier, M., Oberli, F., Quadt, A.V., Roddick, J.C., Spiegel, W., 1995. Three natural zircon standards for U-Th-Pb, Lu-Hf, trace element and REE analyses. Geostandards and Geoanalytical Research 19, 1-23. Willbold, M., Hegner, E., Kleinschrodt, R., Stosch, H.-G., Kehelpannala, K.V.W., Dulski, P., 2004. Geochemical evidence for a Neoproterozoic magmatic continental margin in Sri Lanka-relevence for the Rodinia-Gondwana
Page 74 of 111
P a g e | 75 supercontinental cycle. Precambrian Res. 130, 185-198. Xiao, W.J., Santosh, M., 2014. The western Central Asian Orogenic Belt: A window to accretionary orogenesis and continental growth. Gondwana Res. 25, 1429-1444.
an
us
cr
ip t
Yoshida, M., Funaki, M., Vitanage, P.W., 1992. Proterozoic to Mesozoic East Gondwana: The juxtaposition of India, Sri Lanka, and Antarctica. Tectonics 11, 381-391.
M
Figure captions
Fig. 1 Generalized geological and tectonic framework of Sri Lanka showing the
d
major crustal blocks and their boundaries (after Cooray, 1994). The study area is
te
shown in the Fig. 2 is marked by box. Sample localities are shown by open
Ac ce p
circles. (Sanjeewa)
Fig. 2 Detailed geological map and sample locations of the study areas (Figure modified after Kröner et al. 2003).
Fig. 3 Representative field photographs of locations KDG-4, KDG-1, KDG-2 from the Kadugannawa Complex and WC-2 from the Wanni Complex.
(a)
Metagranodiorite;
(d)
(b)
metadiorite;
(c)
garnet-bearing
amphibolite;
metagranodiorite with patches of incipient charnockite. See text for discussion.
Page 75 of 111
P a g e | 76 Fig. 4 Representative field photographs of locations HC-1, HC-11, HC-7 and HC10 from the Highland Complex. (a) Blocks and boudins of mafic granulite within
garnet-bearing charnockite. See text for discussion.
ip t
marble; (b) mafic sills in metasediments; (c) garnet-bearing metagabbro; (d)
cr
Fig. 5. Photomicrographs showing textures of representative samples. (a) A
us
typical granoblastic texture of hornblende + biotite + quartz + plagioclase + Kfeldspar in metagranodiorite (sample WC-2). (b) Metagranodiorite from the
an
Kadugannawa Complex (sample KDG-4). (c) Orthopyroxene-bearing and Kfeldspar-absent metadiorite (sample KDG-1). (d) Porphyroblastic garnet in the
M
matrix of fine-grained hornblende and plagioclase in garnet amphibolite (sample
d
KDG-2). (e) Medium-grained foliated pyroxene + hornblende + plagioclase
te
assemblage in mafic granulite (sample HC-1). (f) Equigranular texture of
Ac ce p
pyroxene + hornblende + plagioclase in mafic granulite (sample HC-11).
Fig. 5 (continued). (g) Orthopyroxene + plagioclase symplectite present as grayish spots in sample HC-11 as an evidence of decomposition of garnet during exhumation. (h) Coarse-grained massive metagabbro with subidioblastic garnet, clinopyroxene, and minor orthopyroxene (sample HC-7). Plagioclase and ilmenite fill the matrix of the minerals. (i) Medium-grained garnet and quartz which are separated by orthopyroxene + plagioclase corona in metagabbro (sample HC-8). (j) Granoblastic plagioclase + clinopyroxene + orthopyroxene + hornblende with symplectitic aggregates of orthopyroxene + plagioclase garnet charnockite
Page 76 of 111
P a g e | 77 (sample HC-10).
Fig. 6. Compositional diagrams showing chemistry of representative minerals. (a) (b)
Compositional
diagrams
showing
Ca/(Fe+Mg+Ca+Mn)
and
ip t
and
Mn/(Fe+Mg+Ca+Mn) versus XMg of garnet. (c) Wollastonite-enstatite-ferrosilite
cr
diagram showing compositions of clinopyroxene and clinopyroxene. (d) Anorthite-
us
albite-orthoclase diagram showing compositions of plagioclase. (e) Si (pfu) versus XMg diagram showing compositions of calcic amphibole. (f) XMg versus
an
TiO2 (wt.%) diagram showing biotite chemistry. GR: metagranodiorite, GD:
M
metadiorite, GA: garnet amphibolite, MG: mafic granulite, MGB: metagabbro.
d
Fig. 7. Major element variation diagrams showing concentrations of major
te
elements in the examined samples.
Ac ce p
Fig. 8. (a) to (c) Primitive mantle-normalized multi-element variation diagrams for magnesian (a), ferroan (b), and felsic (c) groups. Normalizing values are from Sun and McDonough (1989). (d) to (f) Chondrite-normalized REE spider plots for magnesian (d), ferroan (e), and felsic (f) groups. Normalizing values are from McDonough and Sun (1995).
Fig. 9. Discrimination diagrams for granites based on Rb-Y-Nb and Rb-Yb-Ta variations after Pearce et al. (1984). (a) Nb-Y diagram. (b) Ta-Yb diagram. (c) RbY+Nb diagram. (d) Rb-Yb+Ta diagram. VAG: volcanic-arc granites, syn-COLG:
Page 77 of 111
P a g e | 78 syn-collisional granites, WPG: within plate granites, ORG: ocean-ridge granites
Fig. 10. Triangular diagrams showing compositions of mafic granulites and
ip t
amphibolite. (a) TiO2-MnOx10-P2O5x10 diagram. OIT: oceanic-island tholeiite or seamount tholeiite, OIA: oceanic-island alkali basalt or seamount alkali basalt,
cr
CAB: island-arc calc-alkaline basalt, IAT: island-arc tholeiite, Bon: boninite. (b)
us
2Nb-Zr/4-Y diagram. AI: within-plate alkali basalt, AII: within-plate alkali basalt and within-plate alkali tholeiite, B: E-type MORB, C: within-plate tholeiite and
an
volcanic-arc basalt, D: N-type MORB and volcanic-arc basalt. (c) Y/15-La/10Nb/8 diagram. 1A: calc-alkali basalt, 1C: volcanic arc tholeiite, 1B: an area of
M
overlap between 1A and 1C. 2A: continental basalt, 2B: back-arc basin basalt,
d
3A: alkali basalt from intercontinental rift, 3B: enriched E-MORB, 3C: weakly
te
enriched E-MORB, 3D: N-MORB.
Ac ce p
Fig. 11 Cathodoluminescence (CL) images of representative zircons in samples HC -1, HC- 7, and HC- 10. Analytical points (white circles for U-Pb and red circles for Lu-Hf), ages in Ma and εHf (t) values are also shown.
Fig. 12 Cathodoluminescence (CL) images of representative zircons in samples HC-11, KDG-1, and KDG-2. Analytical points (white circles for U-Pb and red circles for Lu-Hf), ages in Ma and εHf (t) values are also shown.
Fig. 13 Cathodoluminescence (CL) images of representative zircons in samples
Page 78 of 111
P a g e | 79 KDG-4 and WC-2. Analytical points (white circles for U-Pb and red circles for LuHf), ages in Ma and εHf (t) values are also shown.
ip t
Fig. 14 Zircon U-Pb concordia plots, age data histograms and probability curves
cr
for samples HC-1, HC-7.
us
Fig. 15 Zircon U-Pb concordia plots, age data histograms and probability curves
an
for sample HC-10, HC-1.
Fig. 16 Zircon U-Pb concordia plots, age data histograms and probability curves
d
M
for samples KDG-1, KDG-2, KDG-4.
te
Fig. 17 Zircon U-Pb concordia plots, age data histograms, probability curves and bar charts for sample WC-2.
Ac ce p
Fig. 18 REE patterns of zircons in samples HC-1, HC- 7, HC-10, HC-11. Fig. 19 REE patterns of zircons in samples KDG-1, KDG-2, KDG-4, WC-2. Fig. 20 εHf (t) versus
207
Pb/206Pb age diagram of zircons from the magmatic
suites of Sri Lanka.
Fig. 21 Plate tectonic cartoons illustrating the model of Neoproterozoic doublesided subduction (a) culminating in Cambrian collision (b) proposed in this study. See text for details.
Page 79 of 111
P a g e | 80 Table captions
Table 1 Location, rock type and mineralogy of samples from the crustal blocks in
ip t
Sri Lanka analysed in this study.
Table 8 LA-MC-ICPMS Lu-Hf isotope data on zircons from the magmatic suites
Ac ce p
te
d
M
an
us
cr
of Sri Lanka.
Page 80 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Graphical Abstract (for review)
Page 81 of 111
Table-1
HC-11 WC-2
cr
metagabbro
Cpx Grt Opx Pl Ilm Mag Hbl Zrn
o
o
metagabbro
Pl Cpx Grt Qtz Opx Ilm Mag Rt Zrn
o
o
garnet charnockite
Pl Qtz Cpx Opx Hbl Ilm Py Cal Zrn Bt
o
o
mafic granulite
Pl Cpx Opx Hbl Ilm Zrn
o
o
metagranodiorite
Pl Kfs Qtz Bt Hbl Ilm Zrn Ap
N 7 15' 59.9"; E 80 37' 58.4"
Hantana
N 7 15' 00.1"; E 80 37' 59.0"
Hantana
N 7 12' 22.3"; E 80 47' 54.9"
Hakurutale 2km SE from Kivullinda JCT Udadigana
Pl Cpx Hbl Opx Ilm Mag Zrn
o
us
HC-10
mafic granulite
o
N 7 14' 21.0"; E 80 43' 48.1" N 7 30' 09.3"; E 80 20' 43.3"
an
HC-8
Pl Kfs Hbl Bt Qtz Mag Zrn
o
M
HC-7
metagranodiorite
o
N 7 18' 07.4"; E 80 44' 12.1"
Diagana
Hbl Pl Grt Ilm Mag Ap Zrn Qtz
ed
HC-1
garnet amphibolite
pt
KDG-4
o o Mahaberiyatenna Airport site N 7 16' 36.8"; E 80 45' 34.3" N 7o 22' 47.2"; E 80o 31' 51.8" Maussawa
Ac ce
KDG-2
ip t
Table 1 Location, rock type and mineralogy of samples from the crustal blocks in Sri Lanka analysed in this study. U-Pb geochronology and Lu-Hf isotope analysis Sample No. Locality Co-ordinates Rock type Assemblage o o N 7 16' 44.1"; E 80 40' 49.1" KDG-1 Kundasale metadiorite Pl Hbl Bt Qtz Ilm Opx Py Ap Zrn
Page 82 of 111
Yb/177Hf
176
Lu/177Hf
176
Hf/177Hf
WC-2-01
608.0
0.034894
0.001226
0.282199
WC-2-02
608.0
0.021266
0.000788
0.282249
WC-2-03
608.0
0.020237
0.000744
0.282257
WC-2-04
608.0
0.018627
0.000675
0.282168
WC-2-05
608.0
0.027675
0.000978
WC-2-06
608.0
0.016021
WC-2-07
608.0
WC-2-08 WC-2-09
1s
176
Hf/177Hfi
εHf(0)
εHf(t)
TDM (Ma)
TDMC (Ma)
fLu/Hf
0.282185
-20.3
-7.4
1495
2018
-0.96
0.000011
0.282240
-18.5
-5.4
1409
1895
-0.98
0.000011
0.282249
-18.2
-5.1
1395
1875
-0.98
0.000010
0.282161
-21.3
-8.2
1516
2071
-0.98
0.282292
0.000009
0.282281
-17.0
-4.0
1355
1803
-0.97
0.000569
0.282262
0.000012
0.282256
-18.0
-4.9
1382
1860
-0.98
0.015693
0.000541
0.281985
0.000012
0.281979
-27.8
-14.7
1762
2475
-0.98
608.0
0.027203
0.000977
0.282197
0.000012
0.282186
-20.3
-7.3
1487
2014
-0.97
608.0
0.028383
0.001019
0.282255
0.000010
0.282243
-18.3
-5.3
1409
1888
-0.97
WC-2-10
608.0
0.028674
0.000987
0.282172
0.000011
0.282161
-21.2
-8.2
1523
2070
-0.97
WC-2-12
608.0
0.018053
0.000683
0.282233
0.000010
0.282225
-19.1
-6.0
1427
1928
-0.98
WC-2-13
608.0
0.035602
0.001250
0.282257
0.000011
0.282243
-18.2
-5.3
1414
1888
-0.96
WC-2-14
608.0
0.014895
0.000568
0.282273
0.000008
0.282267
-17.6
-4.5
1367
1834
-0.98
WC-2-15
608.0
0.045200
0.001550
0.282143
0.000014
0.282125
-22.2
-9.5
1587
2149
-0.95
WC-2-16
608.0
0.019193
0.000718
0.282256
0.000012
0.282248
-18.2
-5.1
1396
1876
-0.98
WC-2-17
608.0
0.029196
0.001037
0.282291
0.000011
0.282279
-17.0
-4.0
1359
1807
-0.97
WC-2-18
608.0
0.020072
0.000761
0.282205
0.000010
0.282196
-20.1
-7.0
1469
1993
-0.98
608.0
0.033156
0.001157
0.282253
0.000011
0.282240
-18.4
-5.4
1417
1895
-0.97
608.0
0.019062
0.000709
0.282288
0.000010
0.282280
-17.1
-4.0
1351
1805
-0.98
608.0
0.015912
0.000597
0.282251
0.000009
0.282244
-18.4
-5.3
1398
1885
-0.98
608.0
0.030750
0.001084
0.282255
0.000010
0.282243
-18.3
-5.3
1410
1887
-0.97
608.0
0.013215
0.000543
0.282282
0.000009
0.282276
-17.3
-4.2
1354
1814
-0.98
WC-2-24
608.0
0.018073
0.000667
0.282273
0.000011
0.282265
-17.6
-4.5
1371
1838
-0.98
WC-2-25
608.0
0.013856
0.000551
0.282289
0.000009
0.282283
-17.1
-3.9
1344
1799
-0.98
WC-2-26
608.0
0.021998
0.000825
0.282281
0.000012
0.282272
-17.4
-4.3
1365
1824
-0.98
WC-2-20 WC-2-21 WC-2-22 WC-2-23
d
ep te
Ac c
WC-2-19
an
0.000009
M
Age (Ma)
us
176
No.
cr
ip t
Table-8
Page 83 of 111
ip t 0.001009
0.282174
WC-2-28
608.0
0.025933
0.000933
0.281979
WC-2-29
608.0
0.017038
0.000656
0.282261
WC-2-31
608.0
0.016344
0.000586
0.282266
WC-2-32
608.0
0.024456
0.000898
0.282258
WC-2-33
608.0
0.018576
0.000708
0.282240
HC-11-01
539.1
0.000998
0.000031
HC-11-02
539.1
0.000446
0.000013
HC-11-04
539.1
0.000477
0.000014
HC-11-05
539.1
0.000516
HC-11-06
539.1
HC-11-07
0.000011
cr
0.027214
0.282163
us
608.0
-21.1
-8.2
1521
2067
-0.97
0.281968
-28.0
-15.0
1789
2498
-0.97
0.000010
0.282254
-18.1
-4.9
1387
1864
-0.98
0.000010
0.282259
-17.9
-4.8
1378
1852
-0.98
0.000015
0.282247
-18.2
-5.2
1400
1878
-0.97
0.000012
0.282232
-18.8
-5.7
1417
1912
-0.98
0.282058
0.000012
0.282057
-25.3
-13.4
1641
2346
-1.00
0.282028
0.000010
0.282028
-26.3
-14.5
1681
2412
-1.00
0.282059
0.000010
0.282059
-25.2
-13.4
1639
2343
-1.00
0.000015
0.282043
0.000010
0.282043
-25.8
-13.9
1660
2377
-1.00
0.001572
0.000054
0.282058
0.000013
0.282057
-25.3
-13.4
1642
2346
-1.00
539.1
0.000503
0.000015
0.282061
0.000010
0.282060
-25.2
-13.3
1636
2339
-1.00
HC-11-08
539.1
0.000441
0.000013
0.282057
0.000010
0.282057
-25.3
-13.4
1641
2346
-1.00
HC-11-09
539.1
0.000124
0.000003
0.282004
0.000010
0.282004
-27.2
-15.3
1712
2464
-1.00
HC-11-10
539.1
0.000987
0.000031
0.282073
0.000012
0.282072
-24.7
-12.9
1621
2313
-1.00
HC-11-11
539.1
0.000487
0.000017
0.282081
0.000010
0.282081
-24.4
-12.6
1608
2293
-1.00
HC-11-12
539.1
0.000689
0.000020
0.282071
0.000010
0.282071
-24.8
-12.9
1622
2315
-1.00
HC-11-14
539.1
0.000309
0.000009
0.282057
0.000009
0.282057
-25.3
-13.4
1641
2346
-1.00
539.1
0.000939
0.000029
0.282067
0.000010
0.282067
-24.9
-13.1
1628
2324
-1.00
539.1
0.000741
0.000021
0.282079
0.000010
0.282079
-24.5
-12.7
1611
2298
-1.00
539.1
0.000060
0.000002
0.281999
0.000010
0.281999
-27.3
-15.5
1719
2475
-1.00
539.1
0.000902
0.000028
0.282095
0.000011
0.282095
-23.9
-12.1
1590
2263
-1.00
539.1
0.000130
0.000004
0.282024
0.000011
0.282024
-26.5
-14.6
1686
2421
-1.00
HC-11-20
539.1
0.001070
0.000035
0.281947
0.000011
0.281947
-29.2
-17.3
1791
2591
-1.00
HC-11-21
539.1
0.000737
0.000022
0.282072
0.000010
0.282072
-24.8
-12.9
1621
2314
-1.00
HC-11-22
539.1
0.009487
0.000325
0.281860
0.000010
0.281857
-32.2
-20.5
1922
2790
-0.99
HC-11-16 HC-11-17 HC-11-18 HC-11-19
d
ep te
Ac c
HC-11-15
an
0.000012
M
WC-2-27
Page 84 of 111
ip t 0.000496
0.282002
HC-10-03
533.0
0.000208
0.000006
0.282106
HC-10-04
533.0
0.004298
0.000152
0.282137
HC-10-05
533.0
0.004251
0.000164
0.282015
HC-10-06
533.0
0.000561
0.000015
0.282145
HC-10-07
533.0
0.004830
0.000186
HC-10-08
533.0
0.006191
HC-10-09
533.0
0.004883
HC-10-11
533.0
0.000161
0.000005
HC-10-12
533.0
0.012997
HC-10-13
533.0
HC-10-14
0.000019
cr
0.013494
0.281997
us
533.0
-27.2
-15.7
1737
2483
-0.99
0.282106
-23.6
-11.8
1575
2242
-1.00
0.000013
0.282136
-22.4
-10.8
1537
2175
-1.00
0.000010
0.282013
-26.8
-15.1
1705
2447
-1.00
0.000011
0.282144
-22.2
-10.5
1522
2156
-1.00
0.281993
0.000011
0.281991
-27.6
-15.9
1736
2497
-0.99
0.000241
0.281993
0.000011
0.281990
-27.6
-15.9
1738
2498
-0.99
0.000178
0.281933
0.000019
0.281931
-29.7
-18.0
1816
2629
-0.99
0.282017
0.000011
0.282017
-26.7
-15.0
1694
2438
-1.00
0.000450
0.281765
0.000022
0.281760
-35.6
-24.1
2059
3007
-0.99
0.002461
0.000099
0.282142
0.000018
0.282141
-22.3
-10.6
1529
2163
-1.00
533.0
0.008187
0.000293
0.282114
0.000018
0.282111
-23.3
-11.6
1575
2229
-0.99
HC-10-15
533.0
0.000393
0.000011
0.282120
0.000010
0.282119
-23.1
-11.4
1556
2212
-1.00
HC-10-16
533.0
0.006434
0.000236
0.281929
0.000015
0.281927
-29.8
-18.2
1824
2639
-0.99
HC-10-17
533.0
0.004681
0.000178
0.281927
0.000011
0.281925
-29.9
-18.3
1825
2644
-0.99
HC-10-18
533.0
0.034608
0.001208
0.281676
0.000013
0.281664
-38.8
-27.5
2222
3218
-0.96
HC-10-19
533.0
0.005998
0.000237
0.281502
0.000016
0.281500
-44.9
-33.3
2401
3580
-0.99
HC-10-20
533.0
0.007711
0.000309
0.281694
0.000013
0.281691
-38.1
-26.5
2147
3160
-0.99
533.0
0.030751
0.001051
0.281673
0.000011
0.281663
-38.9
-27.5
2217
3221
-0.97
533.0
0.008653
0.000329
0.281826
0.000012
0.281823
-33.5
-21.9
1969
2870
-0.99
533.0
0.018805
0.000679
0.281850
0.000011
0.281844
-32.6
-21.1
1953
2823
-0.98
533.0
0.002831
0.000107
0.281977
0.000010
0.281976
-28.1
-16.5
1754
2531
-1.00
HC-10-22 HC-10-23 HC-10-24 HC-10-25
d
ep te
Ac c
HC-10-21
an
0.000011
M
HC-10-01
533.0
0.004084
0.000161
0.281972
0.000010
0.281970
-28.3
-16.6
1763
2543
-1.00
HC-10-26
533.0
0.003430
0.000131
0.282035
0.000010
0.282034
-26.0
-14.4
1675
2401
-1.00
HC-10-27
533.0
0.003593
0.000141
0.282029
0.000011
0.282028
-26.3
-14.6
1684
2415
-1.00
HC-7-01
543.0
0.001018
0.000031
0.282260
0.000011
0.282260
-18.1
-6.2
1366
1892
-1.00
Page 85 of 111
ip t 0.000022
0.282279
HC-7-05
543.0
0.001113
0.000033
0.282251
HC-7-06
543.0
0.000665
0.000018
0.282254
HC-7-07
543.0
0.000549
0.000016
0.282262
HC-7-08
543.0
0.001852
0.000056
0.282280
HC-7-09
543.0
0.000577
0.000017
0.282255
HC-7-11
543.0
0.000873
0.000024
HC-7-13
543.0
0.000688
0.000020
HC-7-14
543.0
0.000992
0.000030
HC-7-15
543.0
0.001226
HC-7-16
543.0
HC-1-01
0.000010
cr
0.000762
0.282279
us
543.0
-17.4
-5.5
1340
1850
-1.00
0.282251
-18.4
-6.5
1378
1912
-1.00
0.000010
0.282254
-18.3
-6.4
1374
1905
-1.00
0.000011
0.282262
-18.0
-6.1
1363
1888
-1.00
0.000013
0.282279
-17.4
-5.5
1340
1849
-1.00
0.000011
0.282254
-18.3
-6.4
1373
1904
-1.00
0.282280
0.000010
0.282280
-17.4
-5.5
1338
1847
-1.00
0.282255
0.000010
0.282255
-18.3
-6.4
1372
1903
-1.00
0.282233
0.000012
0.282232
-19.1
-7.1
1403
1953
-1.00
0.000040
0.282236
0.000012
0.282236
-19.0
-7.0
1399
1946
-1.00
0.000874
0.000026
0.282221
0.000012
0.282221
-19.5
-7.5
1418
1978
-1.00
553.0
0.005983
0.000268
0.282368
0.000014
0.282365
-14.3
-2.2
1227
1651
-0.99
HC-1-02
553.0
0.006039
0.000256
0.282374
0.000010
0.282371
-14.1
-2.0
1218
1637
-0.99
HC-1-03
553.0
0.006708
0.000282
0.282373
0.000016
0.282370
-14.1
-2.0
1220
1639
-0.99
HC-1-04
553.0
0.007236
0.000298
0.282375
0.000014
0.282372
-14.1
-2.0
1218
1636
-0.99
HC-1-05
553.0
0.005477
0.000231
0.282380
0.000015
0.282378
-13.8
-1.8
1208
1621
-0.99
HC-1-06
553.0
0.005387
0.000228
0.282395
0.000014
0.282392
-13.3
-1.3
1188
1589
-0.99
HC-1-07
553.0
0.003809
0.000138
0.282417
0.000013
0.282416
-12.5
-0.4
1155
1537
-1.00
553.0
0.005521
0.000238
0.282390
0.000015
0.282388
-13.5
-1.4
1195
1600
-0.99
553.0
0.005921
0.000250
0.282393
0.000011
0.282390
-13.4
-1.3
1192
1594
-0.99
553.0
0.004044
0.000142
0.282416
0.000015
0.282414
-12.6
-0.5
1157
1540
-1.00
553.0
0.006902
0.000294
0.282427
0.000011
0.282424
-12.2
-0.1
1146
1519
-0.99
HC-1-09 HC-1-10 HC-1-11 HC-1-12
d
ep te
Ac c
HC-1-08
an
0.000012
M
HC-7-04
553.0
0.005637
0.000229
0.282404
0.000011
0.282401
-13.0
-0.9
1176
1569
-0.99
HC-1-13
553.0
0.007241
0.000308
0.282435
0.000012
0.282432
-11.9
0.1
1135
1501
-0.99
HC-1-14
553.0
0.006267
0.000270
0.282387
0.000012
0.282384
-13.6
-1.5
1200
1607
-0.99
HC-1-15
553.0
0.006233
0.000268
0.282407
0.000012
0.282404
-12.9
-0.8
1172
1562
-0.99
Page 86 of 111
ip t 0.000243
0.282429
HC-1-18
553.0
0.005871
0.000241
0.282399
KDG-4-01
980.0
0.045121
0.001538
0.282369
KDG-4-02
980.0
0.030845
0.001091
0.282290
KDG-4-03
980.0
0.012870
0.000482
0.282338
KDG-4-04
980.0
0.030585
0.001150
KDG-4-05
980.0
0.042430
KDG-4-06
980.0
0.030300
KDG-4-07
980.0
0.024680
0.000915
KDG-4-08
980.0
0.035258
KDG-4-09
980.0
KDG-4-10
0.000015
cr
0.005770
0.282426
us
553.0
-12.1
-0.1
1142
1513
-0.99
0.282396
-13.2
-1.1
1183
1580
-0.99
0.000014
0.282341
-14.3
6.4
1266
1431
-0.95
0.000013
0.282270
-17.1
3.9
1363
1590
-0.97
0.000012
0.282329
-15.4
6.0
1275
1458
-0.99
0.282272
0.000011
0.282251
-17.7
3.2
1389
1631
-0.97
0.001499
0.282335
0.000014
0.282307
-15.5
5.2
1314
1506
-0.95
0.001114
0.282278
0.000011
0.282257
-17.5
3.5
1380
1618
-0.97
0.282246
0.000011
0.282230
-18.6
2.5
1417
1679
-0.97
0.001267
0.282295
0.000012
0.282271
-16.9
4.0
1362
1586
-0.96
0.027611
0.001038
0.282276
0.000013
0.282257
-17.5
3.5
1379
1618
-0.97
980.0
0.033971
0.001216
0.282278
0.000018
0.282255
-17.5
3.4
1384
1622
-0.96
KDG-4-11
980.0
0.051090
0.001764
0.282350
0.000017
0.282317
-14.9
5.6
1301
1483
-0.95
KDG-4-12
980.0
0.025094
0.000901
0.282274
0.000012
0.282257
-17.6
3.5
1378
1618
-0.97
KDG-4-13
980.0
0.035424
0.001379
0.282270
0.000012
0.282245
-17.7
3.0
1401
1645
-0.96
KDG-4-14
980.0
0.010098
0.000429
0.282289
0.000011
0.282281
-17.1
4.3
1340
1564
-0.99
KDG-4-15
980.0
0.029547
0.001075
0.282261
0.000012
0.282242
-18.1
2.9
1402
1653
-0.97
KDG-4-16
980.0
0.030448
0.001091
0.282273
0.000012
0.282253
-17.7
3.3
1386
1628
-0.97
980.0
0.021464
0.000856
0.282307
0.000010
0.282291
-16.5
4.7
1330
1542
-0.97
980.0
0.044247
0.001600
0.282279
0.000014
0.282250
-17.4
3.2
1396
1635
-0.95
980.0
0.043887
0.001672
0.282303
0.000012
0.282272
-16.6
4.0
1365
1584
-0.95
980.0
0.033846
0.001309
0.282305
0.000013
0.282281
-16.5
4.3
1349
1565
-0.96
KDG-4-18 KDG-4-19 KDG-4-20 KDG-4-21
d
ep te
Ac c
KDG-4-17
an
0.000012
M
HC-1-16
980.0
0.028177
0.001111
0.282271
0.000011
0.282250
-17.7
3.2
1390
1633
-0.97
KDG-4-22
980.0
0.033561
0.001188
0.282263
0.000012
0.282241
-18.0
2.9
1404
1653
-0.96
KDG-4-23
980.0
0.024356
0.000897
0.282253
0.000013
0.282236
-18.4
2.7
1407
1665
-0.97
KDG-4-24
980.0
0.075508
0.002571
0.282345
0.000013
0.282297
-15.1
4.9
1338
1529
-0.92
Page 87 of 111
ip t 0.001550
0.282287
KDG-4-26
980.0
0.021552
0.000854
0.282265
KDG-2-01
520.7
0.002531
0.000086
0.281977
KDG-2-02
520.7
0.000244
0.000006
0.282023
KDG-2-03
520.7
0.001201
0.000042
0.282020
KDG-2-04
520.7
0.000201
0.000005
KDG-2-05
520.7
0.003280
KDG-2-06
520.7
0.001328
KDG-2-08
520.7
0.001784
0.000070
KDG-2-09
520.7
0.004295
KDG-2-10
520.7
KDG-2-12
0.000017
cr
0.041104
0.282259
us
980.0
-17.1
3.5
1383
1615
-0.95
0.282249
-17.9
3.2
1389
1636
-0.97
0.000016
0.281976
-28.1
-16.7
1753
2538
-1.00
0.000012
0.282022
-26.5
-15.1
1687
2435
-1.00
0.000014
0.282019
-26.6
-15.2
1693
2442
-1.00
0.281996
0.000011
0.281996
-27.4
-16.0
1723
2493
-1.00
0.000136
0.281896
0.000011
0.281895
-31.0
-19.6
1864
2718
-1.00
0.000048
0.281933
0.000011
0.281933
-29.7
-18.3
1810
2634
-1.00
0.281892
0.000013
0.281892
-31.1
-19.7
1866
2725
-1.00
0.000162
0.281996
0.000015
0.281995
-27.4
-16.1
1730
2497
-1.00
0.000281
0.000007
0.281994
0.000012
0.281994
-27.5
-16.1
1726
2498
-1.00
520.7
0.001944
0.000065
0.281997
0.000013
0.281996
-27.4
-16.0
1725
2493
-1.00
KDG-2-13
520.7
0.000934
0.000030
0.282022
0.000011
0.282022
-26.5
-15.1
1689
2436
-1.00
KDG-2-14
520.7
0.001584
0.000048
0.282058
0.000016
0.282058
-25.2
-13.8
1641
2356
-1.00
KDG-2-15
520.7
0.001525
0.000047
0.282006
0.000009
0.282005
-27.1
-15.7
1712
2473
-1.00
KDG-2-16
520.7
0.002683
0.000108
0.281846
0.000015
0.281845
-32.7
-21.3
1931
2828
-1.00
KDG-2-17
520.7
0.000766
0.000021
0.282039
0.000013
0.282039
-25.9
-14.5
1666
2399
-1.00
KDG-2-18
520.7
0.000173
0.000004
0.282012
0.000011
0.282012
-26.9
-15.4
1701
2458
-1.00
520.7
0.003531
0.000133
0.281983
0.000013
0.281981
-27.9
-16.5
1747
2526
-1.00
520.7
0.000428
0.000011
0.281989
0.000013
0.281989
-27.7
-16.3
1733
2510
-1.00
755.0
0.003763
0.000131
0.282389
0.000012
0.282387
-13.5
3.0
1193
1471
-1.00
755.0
0.002483
0.000095
0.282474
0.000010
0.282473
-10.5
6.1
1075
1279
-1.00
755.0
0.003083
0.000120
0.282408
0.000011
0.282406
-12.9
3.7
1167
1429
-1.00
KDG-1-05
755.0
0.015456
0.000596
0.282344
0.000011
0.282336
-15.1
1.2
1270
1586
-0.98
KDG-1-06
755.0
0.028044
0.001131
0.282370
0.000014
0.282353
-14.2
1.9
1252
1547
-0.97
KDG-1-07
755.0
0.003021
0.000115
0.282507
0.000011
0.282505
-9.4
7.2
1031
1206
-1.00
KDG-2-20 KDG-1-01 KDG-1-03 KDG-1-04
d
ep te
Ac c
KDG-2-19
an
0.000011
M
KDG-4-25
Page 88 of 111
ip t 0.000677
0.282322
KDG-1-09
755.0
0.024695
0.000976
0.282313
KDG-1-10
755.0
0.013257
0.000540
0.282341
KDG-1-11
755.0
0.025974
0.001031
0.282348
KDG-1-12
755.0
0.020893
0.000864
0.282342
KDG-1-13
755.0
0.010112
0.000385
0.282353
KDG-1-14
755.0
0.014504
0.000592
KDG-1-15
755.0
0.014082
0.000570
KDG-1-16
755.0
0.021951
0.000844
KDG-1-17
755.0
0.017256
KDG-1-18
755.0
KDG-1-19
0.000014
cr
0.017363
0.282313
us
755.0
-15.9
0.4
1303
1638
-0.98
0.282300
-16.2
-0.1
1325
1667
-0.97
0.000011
0.282334
-15.2
1.2
1272
1591
-0.98
0.000013
0.282333
-15.0
1.1
1279
1592
-0.97
0.000012
0.282330
-15.2
1.0
1281
1599
-0.97
0.000013
0.282348
-14.8
1.6
1251
1560
-0.99
0.282337
0.000013
0.282329
-15.4
1.0
1279
1602
-0.98
0.282278
0.000013
0.282270
-17.5
-1.1
1360
1733
-0.98
0.282340
0.000012
0.282328
-15.3
0.9
1284
1605
-0.97
0.000737
0.282341
0.000012
0.282330
-15.3
1.0
1279
1599
-0.98
0.018840
0.000774
0.282336
0.000012
0.282325
-15.4
0.9
1287
1610
-0.98
755.0
0.007692
0.000315
0.282320
0.000014
0.282315
-16.0
0.5
1294
1633
-0.99
KDG-1-20
755.0
0.022287
0.000900
0.282337
0.000015
0.282324
-15.4
0.8
1290
1613
-0.97
KDG-1-21
755.0
0.015511
0.000630
0.282291
0.000012
0.282282
-17.0
-0.7
1344
1705
-0.98
KDG-1-22
755.0
0.025541
0.001046
0.282350
0.000014
0.282335
-14.9
1.2
1276
1587
-0.97
KDG-1-23
755.0
0.019186
0.000755
0.282370
0.000014
0.282360
-14.2
2.1
1238
1533
-0.98
KDG-1-24
755.0
0.017021
0.000682
0.282355
0.000012
0.282345
-14.7
1.6
1257
1565
-0.98
KDG-1-25
755.0
0.013953
0.000561
0.282291
0.000013
0.282283
-17.0
-0.6
1342
1704
-0.98
755.0
0.005558
0.000218
0.282343
0.000012
0.282340
-15.2
1.4
1259
1578
-0.99
755.0
0.021774
0.000862
0.282363
0.000012
0.282351
-14.5
1.7
1253
1553
-0.97
KDG-1-27
d
ep te
Ac c
KDG-1-26
an
0.000014
M
KDG-1-08
Page 89 of 111
Ac ce p
te
d
M
an
us
cr
ip t
Figure-1
Page 90 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-2
Page 91 of 111
Ac ce p
te
d
M
an
us
cr
ip t
Figure-3
Page 92 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-4
Page 93 of 111
Ac ce p
te
d
M
an
us
cr
ip t
Figure-5 part I
Page 94 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-5 part 2
Page 95 of 111
Figure-6
(b)
0.26
rim
KDG-1(GD) KDG-2 (GA) KDG-4 (GR) HC-1 (MG) HC-7 (MGB) HC-8 (MGB) HC-11 (MG) WC-2 (GR)
core
0.24
0.22
0.06
0.05
0.04
ip t
(a)
mantle
core
0.03 rim rim
0.02 rim
0.18 0.15
0.2
core
0.25
core
0.01 0.15
0.3
(d)
M
50
augite
Wo
endiopside
d
pigeonite
0.25
an
diopside hedenbergite 50
0.2
core
0.3
Mg/(Fe+Mg)
Mg/(Fe+Mg)
(c)
rim
rim
us
core
cr
0.2
mantle
50
50
Or
orthopyroxene 50
(e)
Ac ce pt e
En
Fs
(f)
0.65
50
Ab
An
6
CaB > 1.5, (Na+K)A > 0.5
5.5
0.6
0.55
0.5
0.45
0.4
5
pargasite
edenite
4.5
kaersutite
ferro-edenite
6.8
6.6
4
ferro-pargasite 6.4
Si (pfu)
6.2
6
3.5
0.4
0.45
0.5
0.55
0.6
0.65
Mg/(Fe+Mg)
Page 96 of 111
Figure-7
Al2O3
16
2
14
1
40
25
45
50
55
60
65
70
(c)
75
40
15
45
50
55
60
65
70
(d)
0.3
Fe2O3
20
0
MnO
0.2
10
10
45
50
55
60
65
70
75
0
40
14
(e)
MgO
8
45
50
8
40
45
50
55
60
65
75
2
40
45
5
(g)
4
70
55
60
(h)
Na2O
4
Ac ce pt e
3
3
50
70
75
65
70
75
70
75
70
75
K2O
d
0
4
M
6
2
65
CaO
10
4
60
(f)
12
6
55
an
40
us
0.1 5 0
75
cr
12
TiO2
(b)
3
ip t
(a)
18
2
2
1
1
40
2
45
(i)
50
55
Metagranodiorite
1.5
Metagabbro
HC-7 (ferroan)
KDG-4 (felsic)
HC-8 (magnesian)
70
75
P2O5
0
40
45
(j)
4
Mafic granulite
50
55
Grt amphibolite
Fe2O3/MgO
Ferroan
Felsic
Grt charnockite
2
HC-10 (magnesian)
Magnesian
1
40
45
50
65
HC-11 (magnesian)
KDG-2 (ferroan)
0
60
3
HC-1 (magnesian)
KDG-1 (magnesian)
0.5
65
WC-2 (felsic)
Metadiorite
1
60
55
60
SiO2 (wt.%)
65
70
75
40
45
50
55
60
65
SiO2 (wt.%)
Page 97 of 111
Figure-8
10 2
10 3 (d) Magnesian
10 2
1
10
10
Sr
K
Rb Ba Th Ta Nb Ce
P
Zr
Hf Sm Ti
Y
Yb
2
1
Ce
Pr
Nd Sm Eu Gd Tb
(e) Ferroan
10 2
Metagabbro
1
Grt amphibolite
Rb Ba Th Ta Nb Ce
P
Zr
Hf Sm Ti
Y
1
M
KDG-2 K
Er Tm Yb
Lu
10
HC-7
Sr
Dy Ho
an
10
10
La
10 3 (b) Ferroan
0.1
HC-8
us
0.1
Grt charnockite
HC-1 HC-10 HC-11 Metadiorite KDG-1 Metagabbro
ip t
10
Mafic granulite
cr
(a) Magnesian
Yb
3
La
Ce
Pr
Nd Sm Eu Gd Tb
Dy Ho
Er Tm Yb
Lu
Pr
Nd Sm Eu Gd Tb
Dy Ho
Er Tm Yb
Lu
10 3
(c) Felsic
(f) Felsic
Metagranodiorite
10
1
Sr
K
Ac ce pt e
10
d
WC-2 KDG-4
2
Rb Ba Th Ta Nb Ce
P
Zr
Hf Sm Ti
Y
Yb
10 2
10
1
La
Ce
Page 98 of 111
Figure-9
100
1000
Metagranodiorite WC-2 KDG-4
WPG
VAG + syn-COLG
syn-COLG 1
10
ORG
1
10
Y (ppm)
100
1000
0.1
0.1
1
1000
1000
syn-COLG WPG
10
10
100
an 1
1
10
100
Yb + Ta (ppm)
d
Y + Nb (ppm)
1000
ORG
Ac ce pt e
1
VAG
M
ORG
100
us
VAG
10
10
WPG
100
100
Yb (ppm)
cr
syn-COLG
1
ORG
VAG
ip t
1
WPG
Metadiorite KDG-1
10
100
Page 99 of 111
TiO2
ip t
Figure-10
(a)
cr
2Nb
MORB
M an
AI OIT
3D 1C
1B
2B
AII
3C
3B
B
IAT OIA
1A
C
Bon CAB
2A
3A
Y
La/10
Nb/8
ce pt
P2O5x10 Zr/4
ed
D
Ac
MnOx10
Y/15
(c)
us
(b)
Mafic granulite HC-1 (magnesian) HC-11 (magnesian) Metagabbro HC-7 (ferroan) HC-8 (magnesian) Grt amphibolite KDG-2 (ferroan) Grt charnockite HC-10 (magnesian)
Page 100 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-11
Page 101 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-12
Page 102 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-13
Page 103 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-14
Page 104 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-15
Page 105 of 111
Ac ce p
te
d
M
an
us
cr
ip t
Figure-16
Page 106 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-17
Page 107 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-18
Page 108 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-19
Page 109 of 111
Ac
ce
pt
ed
M
an
us
cr
i
Figure-20
Page 110 of 111
Ac ce p
te
d
M
an
us
cr
ip t
Figure-21 revised
Page 111 of 111