Historical background of ophiolite complexes

Ophiolite is a suite of rocks that is interpreted to represent oceanic crust and the upper mantle lithological sequence. The rock suite is usually preserved to varying degrees within continental-continental, arc-continental collision, ridge-trench interaction and/or subduction-accretion setting (Miyashiro, 1977; Pearce et al. 1983;

Bodinier and Godard, 2003;, Dilek and Furnes, 2011).). The name of ophiolite is derived from the Greek root “ophi”, which means snake skin or serpent, and “lite”

means stone from the Greek root “lithos”. This word was first used by a French mineralogist Alexandre Brongniart in 1813 to describe some serpentinite mélanges as the sheared rock surface has a dark greenish shiny appearance similar to some snakes (Coleman, 1977; Dilek, 2003). The definition of an ophiolite was refined to include a suite of magmatic rocks (i.e. ultramafic rocks, gabbro, diabase, and volcanic rocks) occurring in the Apennines region (Brongniart, 1827). Mineralogists and petrologists in late 19^th and early 20^th centuries, however, arbitrarily changed, extended or limited the definition of an ophiolite to meet their requirements. The German petrologist Gustuv Steinmann introduced a new concept known as the

“Steinmann Trinity”, which included serpentinite, diabase-spilite (spilite is a kind of albitized, vesicular basaltic lava rock) and radiolarian chert found in the Alps. He further constrained the definition of ophiolite in his 1927 publication to include peridotite (serpentinite), gabbro, diabase, spilite and some associated deep-sea sedimentary rocks (Steinmann, 1927; Coleman, 1977; Dilek 2003). The important conceptual break-through was recognizing that the rock suite could share a common

4

petrogenetic evolution and was one of the earliest systematical descriptions of the magmatic evolution of ophiolite.

Interest in ophiolite waned between the 1930s and 1960s but was still investigated by Hess (1955) in “Crust of the Earth” and argued in favour of the relationship proposed by Steinmann (1927). Hess (1955) insinuated that serpentinite and other mafic and ultramafic rocks of alpine-type could be derived from an island arc orogenic system. Years later he discussed the formation of oceanic basins in which he used seismic velocity to distinguish the boundary of oceanic crust and upper mantle and constrain the thickness of oceanic crust (Hess, 1962). Hess’s (1962) article laid the foundation for the study of the slow-spreading oceanic crust found at mid-ocean ridge settings and ophiolite studies. It was not until 1967 that a modern view of ophiolite began to develop when a collection of 33 papers was published in a special volume edited by Wyllie (1967) entitled “Ultramafic and Related Rocks”.

The collection of papers provided new and interesting facts about the mafic and ultramafic rocks and the compositions of the rocks within an ophiolite were further discussed. Alpine-type peridotite (i.e. peridotites and serpentinite occurrences in mountain belts as plutonic intrusions into folded geosynclinal sedimentary rocks of orogenic systems) was, for the first time, separated from other kinds of ultramafic rocks (e.g. ultramafic rocks in layered intrusion, zoned ultramafic complexes, alkalic ultramafic rocks, ultramafic nodules, kimberlites) as an independent group being discussed (Wyllie, 1967). Thayer (1967) discussed the petrogenetic relationship between the mafic and ultramafic rocks from Alpine-type ophiolite and their same origination of the ultramafic magma evolution. In this article, Thayer (1967)

5

interpreted how the gabbro, diabase, and other leucocratic rocks in alpine-type peridotites could have originated from a single primary peridotitic magma. Jackson and Thayer (1972) classified the peridotites into: 1) harzburgite-type alpine peridotite and 2) lherzolite-type alpine peridotite. The classification was used as the basis for categorizing different types of ophiolite and inferring their petrogenetic development.

In their study, Jackson and Thayer (1972) pointed out that harzburgite-type alpine peridotite is more depleted and represented the uppermost mantle, whereas the lherzolite-type alpine peridotite was initiated from less depleted and represented the characteristics of subcontinental mantle or deeper oceanic mantle. Nineteen seventy-two was a watershed moment in ophiolite studies as a consensus was reached at the first Geological Society of America Penrose Conference on ophiolites (1972), in which the complete ophiolite sequence was defined (Anonymous, 1972; Dilek, 2003 ). The definition included tectonized peridotite, cumulated peridotite, intrusive isotropic gabbro, layered gabbro, sheeted dikes, pillow lava, and pelagic sediments (Fig 1). The Penrose Conference definition is still widely used today and is the basis of modern ophiolite studies (Anonymous, 1972).

In the early 1970s Japanese geochemist Miyashiro (1973) pointed out that a suite of pillow basalt and sheeted dike rocks from Troodos ophiolite showed calc-alkaline characteristics and suggested that some ophiolites are created at a volcanic island arc setting rather than a mid-ocean ridge setting. Miyashiro (1973) was the first to propose the ophiolite-subduction zone relationship and this hypothesis led to tremendous debate and re-evaluation of the tectonic setting of many ophiolites.

Pearce (1975) suggested a marginal basin origin for the Troodos massif, which he

6

considered to have formed at the beginning of a submarine island arc suite. A few years later Pearce et al. (1984) introduced the concept of “suprasubduction zone (SSZ) type ophiolite” in which oceanic crust could form above a subduction zone setting as the overriding plate is extended (Fig. 2). This is one of the most important conceptual evolution of the ophiolite study. The SSZ type ophiolite concept led to a series of on land and marine geological and geophysical investigations within modern oceanic convergent margin setting, in particular drilling programs, seismic experiments and deep-water investigations were initiated around the Izu-Bonin-Mariana arc-trench system, which is one of the best documented suprasubduction zones in the world.

Nowadays more than 75% of the ophioolite is classified as SSZ type ophiolite or subduction related.

7

Figure 1. Typical ophiolite sequence (revised from Coleman, 1977).

Figure 2. Suprasubduction Zone (SSZ) type ophilite (revised from Dilek and Furnes, 2011).

8

9 1.2.2 Types of ophiolite

The Ophiolite was studied as a whole issue and was not being detailed classified. However, it is clear that there are a variety of ophiolite with different structural architecture, chemical fingerprints, and evolutionary paths, suggesting different tectonic environments of origin. A Japanese geologist Miyashiro (1975) proposed a classification by the rock series of the ophiolite, 1) ophiolite complexes containing both calc-akalic (CA) type and tholeiite (TH) type volcanic rocks, 2) Ophiolite complexes containing only TH type rocks, and 3) ophiolite complexes containing TH type and alkalic series rocks (Miyashiro, 1975).

Moores (1982) classified ophiolite as Tethyan type and Cordilleran type based upon the presence or absence of a continental substrate (i.e. passive margin of a continental plate or fragment), arc volcanic edifices, and/or accretionary mélanges and is still a widely recognized concept of ophiolite classification.

Nicolas (1989) proposed another classification based on the tectonic setting of the ophiolite emplacements. The three main categories are: 1) ophiolite tectonically resting on continental passive margins (e.g. Semail ophiolite in Oman, Papuan ophiolite in Papua-New Guinea), 2) ophiolite incorporated into the active continental margins of the Circum-Pacific belt (i.e. ophiolitic occurrences in the Franciscan Complex in California), and 3) suture zone ophiolite occurring in continent-continent or arc-continent collision zones (i.e. ophiolites in the Alpine-Himalayan orogenic system, Caledonian ophiolites, Hercynian and Uralian ophiolites).

10

Dilek and Furnes (2011) proposed a new classification partially based on the criteria of Nicholas (1989) and Dilek (2003) that divided ophiolite into two major groups and some subgroups: 1) subduction unrelated group and 2) subduction related group, which are in parallel with the Tethyan-type and Cordilleran-type of Moores (1982) (Moores, 1982; Dilek, 2003; Wakabayashi and Dilek, 2003; Dilek and Furnes, 2011). The subduction unrelated group, which was the basis of earlier work, is further subdivided by Dilek and Furnes (2011) into continental-margin type (CM type), mid-ocean ridge type (MOR type) and plume type (P type). The CM type ophiolite is derived from the Ligurian type ophiolite of Dilek (2003) and can be traced back to the Class-III type of Miyashiro (1975) (Miyashiro, 1975; Dilek, 2003; Wakabayashi and Dilek, 2003; Dilek and Furnes, 2011). It was formed during the early stage of the ocean basin evolution and is characterized by the widespread existence of largely serpentinized peridotites that are intruded and/or covered by small to moderate volumes of gabbros, local dikes, and pillow lavas. The MOR type ophiolite is close to the Class-II or Class-III type of Miyashiro (1975) based on the presence of tholeiitic and alkaline volcanic rocks (Miyashiro, 1975; Dilek, 2003; Dilek and Furnes, 2011). It generally represents the stratigraphy of the Panrose conference and shows N-MORB, E-MORB and/or contaminated(C) MORB geochemical characteristics (Dilek and Furnes, 2011). P type ophiolite is characterized for having thick plutonic and volcanic sequences and is subdivided into plume-proximal ridge and oceanic plateau subtypes (Dilek and Furnes, 2011).

On the other hand, the subduction related group, which representing ~75% of all ophiolites, is subdivided into suprasubduction zone type (SSZ type) and volcanic

11

arc type (VA type). The SSZ type is subdivided into three subtypes: 1) backarc to forearc environment, 2) forearc setting, and 3) oceanic/continental backarc subtype (Dilek, 2011). The SSZ type ophiolite may correspond to the Class-I type of Miyashiro (1975) and the combination of Mediterranean type and Chilean type of Dilek (2003), which are the products of high degrees of melting of depleted, harzburgitic mantle (Miyashiro, 1975; Dilek, 2003; Dilek, 2011). In general, the SSZ type ophiolite structurally have a complete Penrose sequence, and it represents the mid-ocean ridge basalt (MORB), island arc tholeiite (IAT) and boninite geochemical characteristics (Dilek, 2003; Dilek, 2011). VA type ophiolite is similar to SSZ type but differ for having thicker and more developed arc crust which is representing tholeiitic to calc-alkaline composition (Dilek, 2003; Dilek, 2011).

12 1.2.3 Global distribution of ophiolite

There are many exposed ophiolite outcrops around the world, and they are usually distributed near major collisional zones or accretionary orogenic belts (Fig.

3). There are three major orogenic belts which preserved ophiolite: 1) Alpine - Himalayan Orogenic Belt, 2) Appalachian - Caledonian - Hercynian - Uralian and Central Asian belts and 3) Western Pacific and Cordilleran Orogenic Belts.

The earliest studied ophiolites are around the Mediterranean Sea, which is a part of the Alpine - Himalayan Orogenic Belt. The Alpine - Himalayan Orogenic Belt stretches nearly half of the globe, from East Australia, Indochina, through Tibet, South Central Asia all the way west through the Arabian Peninsula, Turkey, Mediterranean and Alps, and even across the Atlantic Ocean into the eastern Caribbean. Some very well-known ophiolite occurrences in this belt includes:

Ligurian Ophiolite (Apennines, Italy), Troodos Ophiolite (Troodos, Cyprus) and Semail Ophiolite (Oman and The United Arab Emirates). The Internal Ligurian Ophiolite of the Apennines Mountains was the first ophiolite described by Brongniart (1827) in which he used the term “ophiolite” and later was the inspiration of the

“Steinmann Trinity”. The Jurassic Ligurian Ophiolite is interpreted as a continental margin type ophiolite, which consists of exhumed subcontinental lherzolite overlain by basaltic lava and intruded by some small gabbroic plutons and mafic dikes (Dilek and Furnes, 2011). A part of the same Alpine - Himalayan belt, the Late Cretaceous Troodos ophiolite of Cyprus is considered to be one of the best known ophiolite complex in the world. The Troodos ophiolite is thought to be a preserved portion of oceanic crust that formed during the Mesozoic development of the Tethyan Seaway

13

at a SSZ (Coleman, 1977; Staudigel et al., 1999; Dilek and Furnes, 2009). The Late Cretaceous Semail ophiolite is located along the eastern margin of the Arabian Peninsula and is the largest ophiolite complex (30,000 km³) and the best exposed ancient oceanic section in the world (Coleman, 1977; Searle and Cox, 1999). These three ophiolites are amongst the best studied and most complete ophiolites in the world.

The second orogenic belt is Appalachian-Caledonian-Hercynian-Uralian and Central Asian belts, which preserved ophiolite in Norway (Solund-Stavfjord), Greenland (Isua), Newfoundland (Bay of Islands) and New York State (Stark's Knob).

The most famous ophiolite in the Appalachians is the Bay of Islands Ophiolite in western Newfoundland. The Bay of Islands ophiolite was originally interpreted to be the remnants of the proto-Atlantic oceanic lithosphere generated at a spreading center or marginal basin (Coleman, 1977; Pearce et al., 1984; ), but is now classified as a supra subduction zone forearc type (Varfalvy et al., 1997; Dilek and Furnes, 2011).

The third orogenic belt is the Western Pacific and Cordilleran Orogenic Belts, or also known as the circum-Pacific orogenic belt. The two orogenic belts cover the largest area and stretch along the Circum-Pacific Seismic Belt (Fig. 3). The best preserved ophiolites around the Circum-Pacific belt include: Papuan Ophiolite (Papua New Guinea), Zambales Ophiolite (west Luzon, Philippines), Coast Range Ophiolite (California, United States), Olivos Ophiolite (Chihuahua, Mexico), Rocas Verdes Ophiolite (Patagonian Andes, Chile) and the East Taiwan Ophiolite (Taiwan).

The ophiolites in the Circum-Pacific orogenic belt are mostly suprasubduction zone (SSZ) type or volcanic arc (VA) type ophiolite. The Papuan ophiolite extends ~400

14

km discontinuously along the Owen Stanley mountain range in Papua New Guinea.

And the Izu-Bonin-Mariana belt (I-B-M belt) is the most famous supra-subduction zone ophiolite belt, which extends from Japan to Guam, a distance of ~2800 km, and includes the Izu Islands, Bonin Islands, and Mariana Islands (Stern et al., 2003, Plank et al., 2007).

15

Figure 3. Global distribution of some major ophiolites.

The information of the ophiolite is listed in Table 1 (revised from Yakubchuk et al., 1994; Dilek, 2003; Dilek and Furnes, 2011).

16

Table 1. Simple information of important ophiolite around the world.

The location of the ophiolites list here are shown in Figure 3.

No. Ophiolite name Location Age (Ma) Type

1 Ligurian Italy 200 Continental Margin

2 Tihama Red Sea, Saudi Arabia 20 Continental Margin

3 Macquarie Island Australia 10 Mid Ocean Ridge

4 Taitao S. Chile 10 Mid Ocean Ridge

5 Peri-Caribbean Caribbean Sea 105 Plume

6 Zambales Philippines 40-44 SSZ

7 Troodos Cyprus 92-94 SSZ

8 Semail Oman 92-95 SSZ

9 Kizildag Turkey 92-94 SSZ

10 Xigaze Tibet, China 120-126 SSZ

11 Yakuno Japan 270-280 SSZ

12 Magnitogorsk S. Urals, Russia 385-400 SSZ

13 Bay of Islands Canada 484 SSZ

14 Itogon Philippines 30 Volcanic-arc

15 Coast Range California, US 140 Volcanic-arc

17 1.2.4 Ophiolite of circum-Pacific

The circum-Pacific orogenic belt is referred here as the western Pacific island belts (e.g. Japan islands, Ryukyu islands, Taiwan island, Philippine islands, Indonesian islands, Papua New Guinea islands, east Australia and New Zealand islands), the Cordilleran belts of western North America, and the western South America which extends along a great circle for more than 25,000 km in the Pacific Rim (Ishitawari, 1991, 1994; Dilek, 2003; Furnes et al., 2014). Ophiolites of the circum-Pacific belts are characterized by a ranges of ages that cover the entire Phanerozoic (Ishitawari, 1994; Dilek, 2003). The circum-Pacific ophiolites are usually suprasubduction zone type or volcanic arc type ophiolite and commonly show the supra-subduction zone type ophiolite characteristics (i.e. structurally and/or geochemically heterogeneous to the crustal components attesting to the progressive evolution of their mantle melt sources) (Ishitawari, 1994; Dilek, 2003; Dilek and Polat, 2008). The circum-Pacific ophiolites are found mostly on or among the accreted oceanic and trench sediments that characterizing active continental margin , and thus it contrasts with the Tethyan ophiolite occurrence, where ophiolites are mostly emplaced on passive continental margins through continental collision events (Ishitawari, 1994). The East Taiwan Ophiolite, on the other hand, is unusual amongst the circum-Pacific ophiolites as it represent a highly depleted mantle composition that differs from the suprasubduction zone type ophiolite and is more similar to a mid-ocean ridge origin and the opening of a marginal sea basin.

18 1.3 Geological Background

1.3.1 Geological setting and tectonic evolution of the South China Sea

South China Sea is a marginal sea of Eurasia and borders the South China Block, Indochina Block, Luzon arc, Palawan. The opening of South China Sea is considered to have occurred during the Oligocene and is one of the youngest ocean basins in the world (Ludwig, 1970; Ben-Avraham and Uyeda, 1973; Briais et al., 1993; Lee and Lawver, 1995,). Based on the magnetic anomaly mapping in 1979 (anomalies 11 to 5d, Taylor and Hayes, 1980, 1983; revised to anomalies 11 to 5c, Briais et al., 1993), the opening lasted from the late Oligocene to the early Middle Miocene (32 to 15 Ma). The tectonic process responsible for the opening of the South China Sea is debated.

There are three principal models which describe the formation of the South China Sea. 1) The South China Sea is an inactive marginal basin and opened as a consequence of back-arc extension due to eastward subduction of the Eurasia plate (Karig, 1971; Ben-Avraham and Uyeda, 1973, Zhang, 1984). 2) The South China Sea is opened mainly due to the mantle lateral flow or mantle plume (Watanabe et al., 1977; Taylor and Hayes, 1980; Flower et al., 1998; Zhang et al., 2001). 3) The South China Sea opened as a consequence of regional displacement in SE Asia due to the collision of India-Eurasia at ~50 Ma (Tapponnier et al., 1982, 1986, 1990; Lee and Lawver, 1995).

Karig (1971) proposed that the South China Sea is perhaps an inactive marginal basin with higher than normal heat flow value than traditional marginal basin. Ben-Avraham and Uyeda (1973) mentioned the opening of the South China

19

Sea may be related to the subduction of the Pacific plate and the compression of the China Sea basin. Zhang (1984) proposed that South China Sea as well as East China Sea are the continental margin under tension and the subduction of the plate led to the consequence of post-arc pull-apart basin. This interpretation was challenged when Taylor interpreted the magnetic anomalies. The magnetic anomalies reveal the E-W direction expansion axis of the South China Sea, which is unreasonable to be a backarc basin of the Luzon arc with the nearly N-S direction subduction axis (Taylor and Hayes, 1980).

Watanabe et al. (1977) published the heat flow of the South China Sea with the value 85 ± 8 mW/m² at north and 112 ± 16 mW/m² at south of the basin. In this paper (Watanabe et al., 1977), they also predicted an age of 14-36 Ma (lower Miocene to lower Oligocene) for the South China Basin. Taylor and Hayes (1980, 1983) proposed the magnetic anomaly of the spreading axis of South China Sea, and interpreted the spreading of the South China Sea between ~32 Ma and ~17 Ma (mid Oligocene through early Miocene). Flower et al. (1998) proposed a model explaining the dispersed volcanism and DUPAL-like asthenosphere in East Asia and West Pacific. Zhang et al. (2001) discussed the mantle lateral flow and the tectonic regime relationship and suggested that the lateral flow was the dominant factor of the thinning of the lithosphere and the expansion of the South China Sea. However, Cui et al. (2005) and Xia (2005) remodeled the convection of the mantle flow and suggested that the upwelling of the hot mantle material had an impact on the thinning of the lower lithosphere but less influence on the crust thinning and thus taking much longer time for the opening.

20

Tapponnier et al. (1982, 1986, 1990) proposed that the collision of the India block and Eurasia had a large impact on the East China, South China and Indochina blocks. The collision pushed out the East China and South China block and dominated the left-lateral shearing of the Ailao Shan-Red River metamorphic belt.

They proposed that the penetration of India block into the Eurasian continent pushed Indochina southeastward with clockwise rotation, and the over 500 km left-lateral movement directly influenced the Oligocene-Miocene opening of the South China Sea. Lee and Lawver (1995) used the pole of rotation to reconstruct the position of tectonic block of the Southeast China. Xia (2005) suggested that this tectonic pull apart by the rotation of Indochina should be supported by the mantle upwelling mentioned previously.

The tectonic evolution of the Southeast China and South China Sea began at the early Cenozoic (~60 Ma), at when Indochina was about 500 km NW of its present position. During the early Eocene (about 55-50 Ma), the northeastward moving Greater India began to collide with the southern margin of Eurasia and closed the Tethys Ocean. The collision of India and Eurasia forced the Indochina peninsula southeastwards along the left-lateral Ailao Shan-Red River fault and induced clockwise rotation of ~25° between 40 Ma and 22 Ma (Tapponnier et al., 1982, 1990, Lee and Lawver, 1995). The rotation of the Indochina peninsula led to the stretching and thinning of the continental shelf of South China block. The opening of the South China Sea started at the northwest part of the basin at ~32 Ma. The rheology of the South China block, Indochina and Sundaland blocks permitted the initial N-S directed extension of the South China Sea. The spreading direction changed during