A New Genetic Model for the East Taiwan Ophiolite and Its Implications for Dupal Domains in the Northern Hemisphere

Earth and Planetary Science Letters, 109 (1992) 133-145 133 Elsevier Science Publishers B.V., Amsterdam

[CL]

A new genetic model for the East Taiwan Ophiolite and its

implications for Dupal domains in the Northern Hemisphere

S u n - L i n C h u n g a a n d S h e n - s u S u n ~

a Department of Geology, National Ta[wan UniversiO', 245 Choushan Road, Taipei 10770, Taiwan, ROC b Minerals and Land Use Program, Bureau o f Mineral Resources, Geology and Geophysics, G.P.O. Box 378, Canberra,

A C T 260l, Australia

Received June 17, 1991; revision accepted December 26. 1991

ABSTRACT

The Miocene East Taiwan Ophiolite (ETO) has been suggested to have been generated at an "abnormal" mid-ocean ridge along a transform fault in the South China Sea [1]. As a result of arc-continent collision, the ETO was fragmented and incorporated into the Pliocene Lichi Melange of the Coastal Range, eastern Taiwan. Based on the structure of the slow-spreading Mid-Atlantic Ridge and the Tethyan Ophiolites, a new genetic model is proposed for ETO: it was formed in a normal slow-spreading axis environment where serpentinized peridotites and gabbros had been exposed onto the ocean floor.

Further, we propose that ETO basalts which overlie pelagic red shales represent products of near-ridge seamounts erupted by the end of South China Sea spreading ( ~ 15 Ma ago). Basalt samples collected from small areas near Kuanshan in the Coastal Range, eastern Taiwan are characterized by simple mineralogy, general lack of hydrothermal alteration. abundance of commonly fresh glass (up to 95% by volume), and high MgO content averaging 8.8 wt.%. In addition, they have considerably heterogeneous incompatible trace element (e.g. La/Sm ratios) and Nd and Pb isotope compositions. These characteristics are consistent with a magma origin from near-ridge seamounts.

The Pb isotope compositions of ETO basalts and post-spreading seamount basalts from the South China Sea have Dupal anomaly characters, containing higher 2°spb/z°apb ratios (with A2°8/2°4pb = 31-74) than MORB of the Northern Hemi- sphere. The South China Sea, from which the ETO originated, is the only domain where EM2 Dupal-type Pb composition has been found in well-developed spreading centres of the Northern Hemisphere. Since volcanism in this region was not associated with mantle plume activity, the observed Dupal Pb property is probably a result of melting of convecting mantle which has been contaminated by thermal erosion and delamination of continental lithosphere with EM2 character prior to or during the formation of the South China Sea.

1. Introduction

T h e East T a i w a n O p h i o l i t e ( E T O ) is o n e of the y o u n g e s t o p h i o l i t e complexes in the world. It was f o r m e d at a b o u t 15 M a [1,2] a n d was era- p l a c e d t h r o u g h a r c - c o n t i n e n t collision at a b o u t 3 - 4 M a [3-5]. M o r e i m p o r t a n t l y , it provides the only c u r r e n t l y available m a t e r i a l for studying the S o u t h C h i n a Sea o c e a n floor.

T h e E T O displays some u n u s u a l geological, p e t r o l o g i c a l a n d g e o c h e m i c a l characteristics: (1) some basaltic lavas directly overlie p e r i d o t i t e s , a n d show chilled m a r g i n s a l o n g the c o n t a c t [6]; (2) pelagic red shales occur i n t e r c a l a t e d b e t w e e n the extrusive a n d p l u t o n i c s e q u e n c e s [6,7]; (3) the basalts c o n t a i n a b u n d a n t glass (up to 95%) a n d

c o m m o n l y have high M g O c o n t e n t s (up to 10.2 w t . % ) [2,7,8]; (4) s a m p l e s collected from a small area show c o n s i d e r a b l e v a r i a t i o n s of i n c o m p a t i b l e e l e m e n t a b u n d a n c e s a n d isotope c o m p o s i t i o n s [2,9]; (5) Pb isotope c o m p o s i t i o n s of the E T O basalts show a clear D u p a l - t y p e [10] character. T h e S o u t h C h i n a Sea, accordingly, a p p e a r s to b e the only region w h e r e D u p a l - t y p e Pb composi- tions have b e e n f o u n d in w e l l - d e v e l o p e d s p r e a d - ing c e n t r e s of the N o r t h e r n H e m i s p h e r e .

I n view of these p a r t i c u l a r p r o p e r t i e s , it is n e c e s s a r y to critically r e - e v a l u a t e the surprisingly large a m o u n t of p u b l i s h e d d a t a of the E T O for b e t t e r u n d e r s t a n d i n g its genesis. I n this p a p e r , they will be e v a l u a t e d in the light of some n e w

134 S L. ( 2 H U N G A N D S.-S. S U N

observations at slow-spreading ridges and the geochemical characteristics of near-ridge sea- mount basalts. Finally, the origin of the Dupal isotopic anomaly will be discussed.

2. Geotectonic framework

Taiwan is a late Cenozoic orogenic belt gener- ated by an oblique collision between the Luzon arc sitting on the Philippine Sea plate and the Eurasian continent (Fig. 1) [4,5 and references therein]. T h e Longitudinal Valley in eastern Tai- wan has b e e n considered as the boundary be- tween these two plates (Fig. 2). As a result of eastward subduction of the South China Sea, the Luzon arc was formed and the E T O was incorpo- rated- into the Lichi Melange as allochthonous blocks on the west side of the Coastal Range of

Taiwan. Being the northern extension of the Lu- zon arc, the Coastal Range was gradually ac- creted to the continental margin by northwest- ward motion of the Philippine Sea plate [11,12] and the attached Lichi Melange was tectonically emplaced through the a r c - c o n t i n e n t collision

[1,3,5].

T h e Cenozoic tectonic evolution of Southeast Asia has been closely related to the | n d i a - E u r a s i a collision commencing about 50 Ma ago. Opening of the South China Sea was consistent with the clockwise rotation of Indochina relative to South China and left-lateral m o v e m e n t along the Red River fault caused by this collision [13,141. No deep-sea drilling has yet been undertaken in this marginal sea, but the seafloor magnetic data have been used to reconstruct the spreading history of the South China Sea which spans the mid-Oligo-

Indochina

South China

Sea

f_.~,,,~,~ ETO RVuV, I"

/ f ~ l .

Philippine

h : ~ ~

Sea

105°E 110°E 115°E 320°E 125°E

Fig. 1. Geotectonic framework of Southeast Asia. In the South China Sea, the oceanic crust (shaded area) is s u r r o u n d e d by microcontinental fragments, such as Paracel Bank (P.B.) and Macclesfield Bank (M.B.) to the north, and North Palawan and R e e d Bank to the south. The Scarborough s e a m o u n t s (A marked by S.S.) are located close to or within the extinct spreading axis of the South China Sea ( p o i n t - d a s h line). In Taiwan, the E T O is located on the west side of the Coastal Range, east to the Longitudinal

A NEW GENETIC MODEL FOR THE EAST TAIWAN OPHIOLITE 135

L EGE ND

~

huliaokeng

• ilien

•

~ Shuilien _{Con9 Iomerate} t

_~__ Futien Paliwan Member Formation Taiyuan Member ' ] ~ F a n s h u l i a o F o r m a t i o n ~ Lichi Melange

~

Fig. 2. Distribution of the E T O and the geologic m a p of the Coastal R a n g e of Taiwan (after Liou et al. [7] and T e n g [5], respectively).

136 S.-L. C H U N G A N D S.-S. S U N

cene through Early Miocene ( ~ 3 2 - 1 5 Ma) [15,16].

3. Origin of the ETO

Chai [17] proposed in 1972 that the mafic and ultramafic blocks in the Lichi Melange were parts of an ophiolite complex. In the same year Shih et al. [18] proposed the name "Coastal Range Ophi- olite of Taiwan" and suggested an origin at an oceanic spreading centre based on petrological and geochemical data. Since then, detailed field and laboratory studies have been carried out re- sulting in a consensus that the E T O originated from a marginal basin, most likely the South China Sea [1,2,7,19,20]. A schematic geologic col- umn has been constructed (Fig. 3) by Liou et al. [7] who coined the term " E a s t Taiwan Ophiolite".

3.1. Was the E T O generated at an " a b n o r m a l " M O R em,ironment?

Suppe et al. [1] argued that " E T O was not formed at a normal oceanic spreading centre but rather has an unusual stratigraphy composed of submarine scree deposits consisting of angular mafic and ultramafic plutonic clasts and overlying basaltic flows that accumulated below the calcite compensation depth along a fault scarp in dis- turbed, preexisting oceanic crust". They sug- gested the setting to be a leaky transform fault offsetting mid-ocean ridge with a crest form like the East Pacific Rise [21]. Since the E T O exists as chaotic fragments within the Lichi Melange, the preserved ophiolitic sequence was explained to be the most surficial parts of this "atypical" section of the oceanic crust.

As shown in Fig. 3, the E T O can be grouped into two units: the lower plutonic sequence com-

- v v v ~ v ~ V A v ZXv v v v v v v v v v layers o f ~ . . , . . : ,, . v v Vl i pelagic - - " ~ ~ [

redshale ~

~

Extrusive Sequence

Plutonic Sequence

t + .~ diabases, plagiogranites

DEcollement Fault

[ - '

~-~c=:Q~ (5:~--O--':5:3"

~=:Y~<:D

Lichi M~lange

Fig. 3. Schematic stratigraphic section of the ETO (after Liou et al. [7]). Note that the ETO consists of two igneous units and the lowermost red-shale layer contains calcareous rmnnofossils of zone NN5 ( ~ 15 Ma).

A N E W G E N E T I C M O D E L F O R T H E E A S T T A I W A N O P H I O L I T E 137

posed of serpentinized peridotites and gabbros and the upper extrusive sequence of basaltic lavas usually marked by pillow structure. The red shales, indicating an unconformable contact, oc- cur between these two units in many places. However, along the Chia-wu-chi (name of a brook), east of Kuanshan (Fig. 2), some basaltic lavas with a chilled margin directly overlie the peridotites. Accordingly, Wang [6] concluded that " t h e pillow sequence may have been poured out through submarine eruption on a sea floor where at least some basic and ultrabasic plutonics had been exposed". The red shales consist mainly of pelagic sediments with very little calcareous ma- terials, suggesting deposition below the calcite compensation depth [22]. They were thus pre- sumed to be deposited far from land in water depths greater than 4 kin. The overlying MORB- type basalts were considered as lavas that spilled over from the ridge crest to the older and deeper ocean floor across a transform fault [1].

T h e r e are problems in Suppe et al.'s model. First, the proposed ridge morphology is not ap- propriate to the South China Sea. It is now well established that variation of mid-ocean ridge to- pography correlates rather well with the spread- ing rate [23]. The slow-spreading ridges (e.g. the Mid-Atlantic Ridge, with half rate of 1-3 c m / y r , [23,24]) have a wide median valley bounded by rift mountains. Within the broad valley, there is a narrow inner valley, controlled by faults, in which the youngest volcanic activity occurs [25].

The South China Sea Basin is an "Atlantic- type" marginal basin bounded by passive conti- nental margins to the north and south, and opened slowly (2.2 to 3.0 c m / y r ) [15]. Conse- quently an Atlantic-type ridge morphology can be inferred. This suggestion is further supported by the bathymetric studies of the extinct spreading axis [26].

Additional inconsistencies of the earlier ge- netic model for the E T O include the following: (1) If the scree deposits were formed along an active transform fault, they should have incorpo- rated some fragments of the basaltic lavas. This is not the case since basaltic blocks are very rare in the scree deposits [1,6,7]. (2) If the suggested a g e - d e p t h relationships of the E T O is correct, a transform fault with at least 20 Ma offset (440-600 km) is required. This is very unlikely for a small

marginal basin with a short spreading history (32-15 Ma). (3) Some massive, not brecciated, peridotites existing along the Chia-wu-chi (i.e. Chia-wu brook) are directly capped by basaltic pillow lavas with chilled margin between the con- tact. (4) The chemical and isotopic data indicate that the basaltic lavas of E T O may not be typical MORB.

3.2. Analogues of the ETO: modern and ancient

Recent studies of the slow-spreading Mid- Atlantic Ridge show that in many areas the basaltic lavas of layer 2 are very thin or even locally missing [cf. 27]. Extensive deep drilling, dredging and diving programmes in the Mid- Atlantic regions away from the transform fault or fracture zones have shown surprisingly that, in places, serpentinized peridotites a n d / o r gabbros are overlain by only a thin cover of basaltic lavas or sediments. Elsewhere they even crop out di- rectly on the ocean floor [27-29]. Similar rela- tionships can be seen in older Atlantic crust. This feature seems to be common at the slow-spread- ing centres, possibly related to limited, discontin- uous magma supply and long periods of tectonic extension. The plutonics might be exposed as a result of block faulting a n d / o r detachment fault- ing [30-32].

Furthermore, similar relationships are seen in many Mesozoic Tethyan, e.g. the A l p i n e - A p e n - nine, ophiolites [27,33]. Instead of previous "abnormal" ocean basin interpretation for gene- sis [e.g. 33], a slow-spreading, Atlantic-type origin has recently been argued for this ancient ana- logue of the E T O [27].

3.3. A new genetic model for the E T O

On the basis of new informations at slow- spreading ridges, we propose that the E T O was formed in the South China Sea at a normal Atlantic-type mid-ocean environment (Fig. 4).

The South China Sea was characterized by a broad median valley bounded with rift mountains predominantly consisting of exposed peridotites and gabbros, from which sedimentary scree de- posits were derived as a response to block fault- ing and landslides. By the end of spreading, sub- stantial subsidence of the South China Sea Basin

138 S . - L C H U N G A N D S. S. S U N

®

Fig. 4. Schematic reconstruction, after models for the Mid-Atlantic Ridge and the Mesozoic Tethys [27], of the South China Sea spreading axis during Miocene time (about 15 Ma): 1 = serpentinized peridotites; 2 = gabbros; 3 = N-MORB; 4 = ancient scree deposits composed of ultramafic and mafic breccias; 5 = pelagic red shales; 6 = isolated near-ridge seamounts erupted by the end of the South China Sea spreading. Note that the red shales, shown by exaggerated thickness, cover most parts of the broad median

valley and the seamount basalts directly overlie the serpentinized peridotites, gabbros and/or scree deposits.

was prevailing d u e to the onset of t h e r m a l con- traction [15,34]. Pelagic red shales, then, b e g a n to be d e p o s i t e d intermittently u p o n the m e d i a n val- ley d u r i n g periods of tectonic and igneous quies- cence. Only six collected f r o m the lowermost shale layer (Fig. 3) as c e m e n t i n g matrix of the plutonic breccias, out o f a total of 71 E T O red- shale samples studied by H u a n g et al. [22], con- tain c a l c a r e o u s nannofossils of solution-resistant g e n e r a b e l o n g i n g to the Early to Middle M i o c e n e nannofossil zone N N 5 (i.e. a b o u t 15 Ma). O t h e r s f r o m the b e t t e r - d e v e l o p e d u p p e r layers show no trace of c a l c a r e o u s materials [22]. This indicates that the red shales w e r e d e p o s i t e d on a subsiding e n v i r o n m e n t , c h a n g i n g f r o m above to below the calcite c o m p e n s a t i o n d e p t h ( ~ 4 km).

Isolated n e a r - r i d g e s e a m o u n t s e r u p t e d discon- tinuously in the b r o a d m e d i a n valley with axial d e p t h of a b o u t 4 km (see next section for f u r t h e r discussion). T h e y m a y overlie the red shales, scree deposits or even massive plutonic b a s e m e n t with chilled margins. T h e "classical" s e q u e n c e o f o c e a n i c crust is also drawn in Fig. 4, a l t h o u g h only in s u b o r d i n a t e position b e c a u s e of its scarcity

in the scree deposits. J a h n [2] has published o n e K - A r age for fresh E T O basalt glass (14.6 +_ 0.4 Ma) which is consistent with the fossil age ( ~ 15 Ma). Evidently, the near-ridge s e a m o u n t basalts were e r u p t e d during the very final stage of the South China Sea spreading.

4. The ETO basaits of near-ridge s e a m o u n t ori- gin

Most g e o c h e m i c a l and petrological studies [2,7,18,35-37] suggested the E T O basalts to be M O R B - t y p e m a g m a s . However, an island arc ori- gin was also p r o p o s e d [8,9,38]. To resolve this controversy, J a h n [2] r e p o r t e d additional d a t a a n d i n t e g r a t e d the overall g e o c h e m i c a l a n d iso- topic compositions o f the r e p r e s e n t a t i v e rock types. H e c o n c l u d e d that the E T O is truly " o c - eanic" in nature, and strongly a r g u e d for a mid- o c e a n or marginal basin origin which was f u r t h e r s u p p o r t e d by an immobile trace e l e m e n t study using discrimination diagrams [19]. M o r e o v e r , an association of N - t y p e ( L R E - d e p l e t e d ) and P-type ( L R E - e n r i c h e d ) E T O basalts within a small area

A NEW GENETIC MODEL FOR THE EAST TAIWAN OPH1OLITE

T A B L E 1

The m e a n major element compositions (wt.%)

139 E T O basalts M O R B N R S B L R E - d e p l e t e d E R E - e n r i c h e d N = 29 N = 2 N = 590 N = 45 SiO 2 49.34 (1.16) 49.88 (0.91) 50.53 49.56 TiO 2 0.94 (0.20) 1.09 (0.02) 1.56 1.19 A l e O 3 15.73 (0.65) 16.32 (0.25) 15.27 16.12 F e O t 10.10 (0.72) 9.72 (0.41) 10.46 9.31 M n O 0.15 (0.02) 0.15 (0.02) 0.16 0.15 M g O 8.87 (0.69) 7.70 (0.05) 7.47 8.29 C a O 11.14 (0.81) 10.79 (0.08) 11.49 12.48 N a 2 0 2.49 (0.30) 2.63 (0.20) 2.62 2.60 K 2 0 0.15 (0.08) 0.49 (0.11) 0.16 0.06 P205 0.10 (0.02) 0.20 0.13 0.13 M g # 63.6 61.3 58.8 64.0 N a s 0 2.82 2.43 C a O / A l 2 0 3 0.71 0.75

Inferred axial depth ~ 4 km ~ 3 km

T h e values between p a r e n t h e s e s represent a standard deviation of lo-.

Sources: the E T O basalts [2,8,35,36]; M O R B [48]; N R S B (near-ridge s e a m o u n t basalts) [45].

F e O t represents total F e O determination. M g # is calculated by 100 M g / ( M g + F e ) assuming F e O = 0.9FeO t [45]. Nas. 0 and inferred axial depth calculations are according to Klein and L a n g m u i r [46].

T A B L E 2

Summary of geochemical and petrographic characteristics of the E T O basalts in comparison with M O R B and near-ridge s e a m o u n t basalts (NRSB) Geochemical Petrographic M g O ( L a / S m ) n ENd P h e n o . / m a t r i x Effect of hydroth. alter. E T O basalts 8.77 M O R B 7.60 N R S B 8.20 0.46 + 13.3 to to 2.02 + 8.7 5 - 1 8 % microphenocrysts (95-82% glass) composed of Ol(Fo = 86.2-84.6), Plag, Cr-Sp and absence of Cpx 0.40 + 13 2 0 - 3 0 % phenocrysts composed to to of OI, Cr-Sp, Plag, (Cpx, 3.04 + 7 Fe-Ti oxides) rare c o m m o n

0.31 + 12 1 - 2 0 % phenocrysts composed rare

to to of O1, Plag, Cr-Sp and

2.81 + 5 absence of Cpx

Sources: E T O basalts [2,8,9,35-37]; M O R B [47-50]; N R S B [24,41-45].

* ( L a / S m ) n due to range of different segments for M O R B (both N- and E-type) but a single s e a m o u n t for N R S B ( s e a m o u n t 6, E P R , [411).

140 S.-L. C H U N G A N D S.-S. S U N

near Kuanshan [2,9] was considered as a mixing result of a highly depleted asthenosphere and an enriched plume-type or hot-spot source [2].

It is important to note the result of the meta- morphic history studies of the E T O [7,39,40]. The brecciated plutonic rocks have been subjected to two stages of post-magmatic recrystallization, namely the ocean-ridge metamorphism and the later off-axis ocean-floor metamorphism (mainly attained to greenschist facies and zeolite facies mineral assemblages, respectively). However, the overlying basalts have been subjected only to zeolite facies metamorphism. In view of the sup- posed time gap (perhaps over 10 Ma [40]), the E T O basalts may not be genetically related to the plutonics.

The petrological and geochemical characters of E T O basalts show strong affinities with the near-ridge seamount magmatism which has re- cently become better-understood as a result of numerous studies [e.g. 24,41-45]. The " s e a m o u n t effect" [42] is capable of sampling small-scale heterogeneity in a veined or plum-pudding-type mantle source of MORB. Therefore, near-ridge seamount basalts are characterized by chemically heterogeneous and less fractionated compositions

compared to M O R B originated from the same mantle source.

The E T O basalts have predominantly O1-Hy normative compositions [2,7,8]. They contain higher average M g # of 63.6 for L R E - d e p l e t e d and 61.3 for LRE-enriched basalts (Table 1) than average M O R B ( M g # = 58.8), but in good agree- ment with near-ridge seamount basalts ( M g # = 64.0). More strikingly, the E T O basalts are char- acterized by a large proportion of glass, enclosing microlites a n d / o r microphenocrysts of olivine (Fo = 84.6-86.2), plagioclase (An = 70-72) and minor chrome spinel, whereas clinopyroxene is always lacking (Table 2). Except for pervasive palagonization along cracks and veins, the glass portions are usually very fresh and lack evidence of hydrothermal alteration [7,8,35]. Moreover, as mentioned, samples from a small area display a considerable heterogeneity in abundances of in- compatible elements, L a / S m ratios and Nd and Pb isotope compositions [2,9,37]. All these char- acteristics, combined with the specific field occur- rence and eruption age, lend strong supports to show that E T O basalts are products of near-ridge seamounts.

The major element chemistry of ocean ridge

O 1 ( 0 704 " 0 ~ • ) D-10: Scarborough seamount AB 100 D8-4: Scarborough seauount TH BJ-106: LRE-enriohed ~ ETO Basalt -~N I 0 ' - - ~ B J - I I 3 : LRE-depleted ETO Basalt

z t " N - J 0 R a D-1O (3.5 Ma) 264 0.61 14.2 43.1 30.2 D8-4 (13.9 Ma) 233 0.83 15.7 40.5 29.3 BJ-106(~ 15 Ma) 212 0.59 18.5 -- 32.4 BJ-II3(~ 15 Ma) 311 1.85 14.2 -- 24.2 ,1 I I L I I I I [ L I I I I I / I I i i Rb Ba Th Nb K La Ce Sr Nd Zr HI Sm Eu Gd Tb Dy Y Er Yb

Fig. 5. The primitive-mantle-normalized patterns for some representative ETO and Scarborough seamount basalts generated from the spreading centre of the South China Sea. The formation ages and some elemental ratios are shown for reference. The normalizing values and comparative patterns (e.g. N- and E-MORB and OIB) are from Sun and McDonough [51]. Note that the OIB is of a non-Dupal type with STSr/~6Sr ~ 0.7035 while the Scarborough alkali basalt (D-10) has Dupal character with

A NEW GENEI'IC MODEL FOR THE EAST TAIWAN OPHIOI, IT[" 141

basalts show a global correlation with the axial depth [46]. Although seamount magmas may have anomalous chemistry, the average composition of the E T O basalts appears to be of significance (Table 1). The E T O basalt chemistry (e.g. Nas. 0 = 2.82, C a O / A l 2 0 3 = 0.71) suggests that these basalts did originate from an axis region with a depth of about 4 km [46], consistent with the presumed South China Sea axial depth based on the red-shale deposition.

The primitive-mantle-normalized patterns for some representative LRE-depleted and enriched E T O basalts show good internal consistency, ex- cept for several mobile elements (e.g. Rb and Th). They are shown by BJ-113 and BJ-106, re- spectively, in Fig. 5. Patterns for the N- and E-type M O R B are also shown for comparison. The E T O basalts have S r / N d ratios similar to M O R B and intraplate basalts, close to the chon- dritic ratio ( S r / N d = 15.5, [51]). No detectable relative HFSE-depletions, and thus no subduc- tion-related component can be observed.

5. The post-spreading seamount basalts in the South China Sea

The E T O is the best material currently avail- able for investigating the structure and chemistry of the ocean floor of the South China Sea. Ophio- lites from Mindoro and Palawan are two other possible candidates. They are, however, not as well preserved and studied. In addition, there are a few surveys at seamounts within the South China Sea [e.g. 15,52]. Some dredged samples from these post-spreading seamounts have been studied for chemical and isotopic compositions which show typical intraplate OIB-type character- istics [53-55].

The post-spreading seamounts were mostly erupted adjacent to the continental fragments scattered along the margins of the South China Sea Basin [15,52]. A few exceptions, such as the Scarborough seamounts, are located close to or within the extinct spreading axis (Fig. 1). The Scarborough basalts yield average K-Ar and Af- Ar ages ranging from 13.9 to 3.5 Ma [56]. During this period, it is significant to note that the mag- mas gradually change from OIB-type olivine tholeiitic to transitional and finally to alkalic basalts.

A simple binary mixing model for a veined or plum-pudding-type mantle source can explain the origin of near-ridge seamount lavas changing from depleted tholeiitic to relatively enriched alkalic basalts [e.g. 42,44]. The continuous chemical vari- ation, from the depleted and enriched E T O basalts to the Scarborough tholeiitic and alkali basalts (Fig. 5), seems to imply their genetic con-

0.5133 Z 0.5131 'ID Z 0.5129 15.6 ,,~ 13. o" c,a 15.5 15.4 39.0 t~ o ~ o~ I~ 38.0 37.0 17.5 18.0 18.5 2 0 6 p b / 2 0 4 [ ~ 0 19.0

Fig. 6.2°~pb/2°4pb (a), 2°7pb/2°4pb (b), and 143Nd/144Nd (c) vs. Z°6Pb/Z°4pb, for the E T O basalts (filled symbols) and Scarborough s e a m o u n t basalts (open symbols) originated from the South China Sea. Data sources for the E T O include: Chou et al. [9], Jahn [2] and Sun [37] presented as triangles, circles and squares, respectively. The values for the Scarbor- ough tholeiitic (squares) and alkali basalts (triangles) are taken from Tu et al. [55]. The pillow basalts from the Mindoro (m) and Palawan ( p ) ophiolites, both from [55], are also plotted. For comparison fields of the Central Indian Ridge M O R B (CIRB) [63] and the Circum-SCB basalts (CSB) [65] are also drawn. Fields for isotopic data from the North Atlantic, Pacific and Indian O c e a n N - M O R B are shown by hatched areas. The Northern H e m i s p h e r e Reference Line

142 S.-L. C H U N G A N D S.-S. S U N

nection and reinforces the validity of binary mix- ing as the genetic process of the South China Sea basalts.

This binary mixing hypothesis is also supported by the isotope systematics (Fig. 6). The isotope compositions of E T O and Scarborough basalts demonstrate a similar variation corresponding to their chemical characteristics. These chemical and isotopic variations could be caused by different degrees of partial melting from a veined or plum-pudding-type heterogeneous mantle source. The enriched component (i.e. the plum, enriched in L R E and other incompatible elements as well as radiogenic isotopes) is supposed to contain a higher volatile content, and therefore to have a lower solidus t e m p e r a t u r e than that of the de- pleted component (i.e. the pudding or N - M O R B mantle source). Compared to the E T O basalts, the Scarborough basalts were derived from smaller degrees of melting due to gradual cooling beneath the extinct axis region, which would ac- count for the more enriched chemical and iso- topic characteristics in the younger post-spread- ing South China Sea basalts.

Studies on the southeast China (i.e. Fujian- Taiwan region) basalts and mantle xenoliths con- tained therein show that the southeast China lithosphere has been severely stretched, as a poor-developed counterpart of the South China Sea spreading, in association with asthenosphere upwelling [57]. Decompressional melting of a plum-pudding-type convecting mantle, generated by thermal erosion or delamination of the litho- spheric mantle, can explain the spatial chemical and isotopic variations of southeast China basalts [57]. The same processes might have taken place prior to or during the formation of South China Sea [58].

6. Dupal-type d o m a i n s in the Northern Hemi- sphere

The Pb isotope compositions of the E T O and Scarborough basalts, both originated from the South China Sea axis region at about 15°N, are distinctive by having positive AZ°8/2°4pb values (from 31 to 56 for the former and 45 to 74 for the later, respectively) above the Northern Hemi- sphere Reference Line ( N H R L ) consistent with the EM2 Dupal character (Fig. 6a). The South

China Sea appears to be the only province of the Northern Hemisphere where EM2 Dupal-type Pb compositions have been found in the well-devel- oped mid-oceanic regions. Understanding its ori- gin is critical to further evaluate the following controversies. (1) Is the Dupal anomaly limited to the Southern Hemisphere as originally suggested by Hart [10]? (2) Is it of deep convecting mantle or shallow continental lithospheric mantle origin, or could it be both? (3) What is the mechanism(s) for producing such an isotopic anomaly [e.g. 10,51,59-621?

Strontium isotope analysis has been attempted by several authors [e.g. 9 and Shih, unpublished] on the E T O basalts and leached powders. How- ever, no success has been achieved to totally eliminate all effects of palagonization due to sea-water-rock interaction. Therefore, Sr isotope data will not be discussed in this paper.

In comparison with the 2°spb/2°4pb ratios, the 2°Tpb/2°4pb ratios of E T O basalts are not as distinctive (Fig. 6b). This feature is very similar to that of the M O R B glasses from the Central In- dian Ridge [63]. In addition, two ophiolitic basalts from Mindoro and Palawan [55], probably also originating from the South China Sea spreading [20,64], show similar Dupal-type Pb isotope con- figurations undistinguishable from the depleted E T O basalts (Fig. 6). This finding confirms the existence of a presumed ancient Dupal compo- nent [65] in the convecting mantle of the South China Sea.

Mukasa et al. [66] first recognized the pres- ence of a Dupal-type isotopic anomaly in the Philippine arc volcanics and pointed out that the Dupal anomaly may not be restricted to the Southern Hemisphere mantle. Later, Flower et al. [58] proposed that the Philippine volcanics share the same isotopic characteristics of Ceno- zoic intraplate basalts from the South China Sea (SCS) post-spreading seamounts, Hainan Island, and southeast China provinces including Fujian, Penghu Islands and northern Taiwan. They fur- ther defined a circum-SCS EM2 Dupal-type man- tle domain [55,65]. In the absence of geophysical evidence for a mantle plume, an origin of shallow lithospheric mantle, overprinted by recent sedi- ment subduction, was postulated [65].

The palaeomagnetic and palaeontologic data [e.g. 67,68] suggested that South China was a part

A NEW GENETIC MODEL FOR THE EAST TAIWAN OPHIOIJTE 143

of Gondwanaland and drifted since the late Palaeozoic to the Northern Hemisphere. This Gondwana connection may explain the northward extension of the Dupal region in the South China Sea, i.e. it was transported from the Southern Hemisphere either through northward migration of a Dupal-type convecting mantle [10] a n d / o r delamination of a Gondwana-type lithospheric mantle [63]. However, this is not a unique solu- tion. If the EM2 anomaly has indeed originated from modification of lithospheric mantle by sub- duction of sediments, it could have taken place anywhere on the Earth. A Gondwana connection is not required.

In addition to the South China Sea of the western Pacific in the Northern Hemisphere, the existence of an EMl-type Dupal anomaly in the Japan Sea and NE China has also been suggested [51,69]. Nakamura et al. [70] first reported a Dupal-type Pb isotope composition with very high A2°s/Z°4Pb (about +150) in a leucitite from Ulungdo Island in the Japan Sea. These leucitites have 87Sr/~6Sr=0.705 and end---- --2 [71]. This finding was recently confirmed by Tatsumoto [69]. A mantle plume origin for the volcanism in Japan Sea was proposed by Nakamura et al. [70,71]. However, Sun and McDonough [51] and Tat- sumoto [69] favour a lithospheric mantle connec- tion for the observed EM1 Dupal anomaly.

As a concluding remark, we would like to emphasize that although a deep mantle plume connection for EMl-type ocean islands in the Southern Hemisphere Dupal anomaly regions has been well established, the origin of Dupal anomaly domains in the Northern Hemisphere as observed in the South China Sea and the Japan Sea deserves further detailed study. Their exis- tence in the Northern Hemisphere raise some questions. Why should the Dupal anomaly re- gions be mainly concentrated in the Southern Hemisphere [10]? Is it related to the fundamental lower mantle convection patterns [61] or a fre- quent near-equatorial aggregation of superconti- nent throughout the geological time?

Acknowledgements

Constructive comments and discussions with B.M. Jahn, U. Knittel, L.S. Teng, and Y. Wang are greatly appreciated. We also acknowledge

Kan Tu for permission to quote his works in advance to publication. S i . Chung is most grate- ful to Prof. C-H. Chert for his moral support and continual encouragement. S.-s. Sun acknowledges financial support from the National Science Council (R.O.C.) for his three-month visit to the Institute of Geology, National Taiwan University in 1990. Most ideas presented in this paper were fermented during this visit. Editorial reviews by R. Batiza and an anonymous reviewer provided substantial improvements. J.C. Lu is thanked for his help in computer editing. S.-s. Sun publishes with permission of the Executive Director of Bu- reau of Mineral Resources, Geology and Geo- physics, Australia.

References

1 J. Suppe, J.G. Liou and' W.G. Ernst, Paleogeographic origins of the Miocene East Taiwan Ophiolite, Am. J. Sci. 281,228-246, 1981.

2 B.M. Jahn, Mid-ocean ridge or marginal basin origin of the East Taiwan Ophiolite: chemical and isotopic evi- dence, Contrib. Mineral. Petrol. 92, 194-206, 1986. 3 B.M. Page and J. Suppe, The Pliocene Lichi melange of

Taiwan: its plate tectonic and olistostromal origin, Am. J. Sci. 281, 193-227, 1981.

4 E. Barrier and J. Angelier, Active collision in eastern Taiwan: the coastal range, Tectonophysics 125, 39-72, 1986.

5 L.S. Teng, Geotectonic evolution of late Cenozoic a r c - continent collision in Taiwan, Tectonophysics 183, 57-76, 1990.

6 Y. Wang, Some geologic observations in the Coastal Range, eastern Taiwan, Proc. Geol. Soc. China 9, 86-93, 1966. 7 J.G. Liou, C.Y. Lan, J. Suppe and W.G. Ernst, The East

Taiwan Ophiolite: its occurrence, petrology, metamor- phism and tectonic setting, Mining Research and Service Organization Special Report, No 1,212 pp, 1977. 8 V.C. Juan, H.J. Lo and C-H. Chen, Crystallization-differ-

entiation of Taiwanite, Proc. Geol. Soc. China 19, 87 97, 1976.

9 C.L. Chou, H.J. Lo, J.H. C h e n and V.C. Juan, Rare earth element and isotopic geochemistry of K u a n s h a n igneous complex, Taiwan, Proc. Geol. Soc. China 21, 13-24, 1978. 10 S.R. Hart, A large-scale isotope anomaly in the southern

hemisphere, Nature 309, 753 757, 1984.

11 R. Seno, The i n s t a n t a n e o u s rotation vector of the Philip- pine Sea plate relative to the Eurasian plate, Tectono- physics 42, 209-226, 1977.

12 R. Seno and S. M a r u y a m a , Paleogeographic reconstruc- tion and origin of the Philippine Sea, Tectonophysics 102, 53 84, 1984.

13 P. Tapponnier, G. Peltzer and R. Armijo, On the mechan- ics of the collision between India and Asia, in: Collision

144 S.-L. C H U N G A N D S.-S. S U N

Tectonics, M.P. Coward and A.C. Ries, eds., Geol. Soc. London Spec. Publ. 19, pp. 115-157, 1986.

14 P. Tapponnier, R. Lacassin, R.H. Leloup, U. Scharer, D.L. Zhong, H.W. Wu, X.H. Liu, S.C. Ji, L.S. Z h a n g and J.Y. Zhong, The Ailao S h a h / R e d River m e t a m o r p h i c belt: Tertiary left-lateral s h e a r between Indochina and South China, Nature 343, 431-437, 199/).

15 B. Taylor and D.E. Hayes, Origin and history of the South China Basin, in: T h e Tectonic and Geologic Evolution of Southeast Asian Seas and Islands, Part 2, D.E. Hayes, ed., A G U Geophys. Monogr. 27, pp. 23 56, 1983.

16 G. Pautot and C. Rangin, Subduction of the South China Sea axial ridge below Luzon (Philippines), Earth Planet. Sci. Lett. 92, 57-69, 1989.

17 B.H.T. Chai, Structure and tectonic evolution of Taiwan, Am. J. Sci. 272, 389-442, 1972.

18 C.Y. Shih, S.-s.Sun, J.G. Liou, T.P. Yen, J.M. R h o d e s and 1.C. Hsu. Petrology and geochemistry of the Coastal R a n g e ophiolite of Taiwan, EOS 53, 535, 1972.

19 T.F. Yui and M.H. Yang, Trace element implication of the paleotectonic environment of pillow basalts from the East Taiwan Ophiolite, Proc. Geol. Soc. China 31, 11-23, I988. 20 J. Suppe, Tectonics of a r c - c o n t i n e n t collision on both sides of the South China Sea, Acta Geol. Taiwanica 26, 1-18, 1988.

21 J.G. Sclater, R.N. A n d e r s o n and M.L. Bell, The elevation of ridges and evolution of the central Eastern Pacific, J. Geophys. Res. 71, 7888 7915, 1971.

22 T.C. Huang, M.P. C h e n and W.R. Chi, Calcareous nanno- fossils from the red shale of the ophiolite-melange com- plex, eastern Taiwan, Mere. Geol. Soc. China 3, 131-138,

1979.

23 R. Hekinian, Petrology of the Ocean Floor, 393 pp., Else- vier, A m s t e r d a m , 1982.

24 P. Castillo and R. Batiza, Strontium. n e o d y m i u m and lead isotope constraints on near-ridge s e a m o u n t production beneath the South Atlantic, Nature 342, 262-265, 1989. 25 R.D. Ballard and Tj.H. Van Andel, Morphology and tec-

tonics of the inner rift valley at Lat. 36°50'N on the Mid-Atlantic Ridge, Bull. Geol. Soc. Am. 88, 507-530, 1977.

26 J. Mammerickx, R.L. Fisher, F.J. E m m e l and S.M. Smith, Bathymetry of the east and southeast Asian seas, Map Chart Ser. MC-17, Geol. Soc. Am. Boulder, Colo., 1976. 27 V. Lagabrielle and M. Cannat, Alpine Jurassic ophiolites

resemble the modern central Atlantic basement, Geology 18, 319 322, 1990.

28 J.A. Karson et al., Along axis variations in sea-floor spreading in the M A R K area, Nature 328, 681-685, 1987. 29 L.P. Z o n e n s h a i n , M.I. Kusmin, A.P. Lisitsin, Y.A. Bog- danov and B.V. Baranov, Tectonics of the Mid-Atlantic rift valley between the T A G and M A R K areas (26-24°N): evidence for vertical tectonism, Tectonophysics 159, 1-23, 1989.

30 G.D. Harper, Tectonics of slow spreading mid-oceanic ridges and c o n s e q u e n c e s of a variable depth to the brittle/ductile transition, Tectonics 4, 395-409, 1985. 31 J.A. Karson, Seafloor spreading on the Mid-Atlantic

Ridge: implications for the structure of ophiolites and oceanic lithosphere produced in slow-spreading environ-

ments, Proc. Symp. Ophiolites and Oceanic Lithosphere, Troodos, 547-555, 1987.

32 C. Mevel, M. Cannat, P. Gente, E. Marion, J.M. A u z e n d e and J.A. Karson, E m p l a c e m e n t of deep crustal and mantle rocks on the west median valley wall of the M A R K area ( M A R , 23°N), Tectonophysics 190, 31 53, 1991.

33 M. Lemoine, G. Boillot and P. Tricart, Ultramafic and gabbroic ocean floor of the Ligurian Tethys (Alps, Corsica, Apennines): in search of a genetic model, Geology 15, 622-625, 1987.

34 K. Ru and J.D. Piggot, Episodic rifting and subsidence in the South China Sea, Am. Assoc. Pet. Geol. Bull. 70, 1136-1155, 1986.

35 J.G. Liou, B.M. Jahn and T.P. Yen, Petrology of East Taiwan Ophiolites, Pet. Geol. Taiwan 13, 59-82, 1976. 36 S.-s. Sun, R.W. Nesbitt and A.Y. Sharaskin, Geochemical

characteristics of mid-ocean ridge basalts, Earth Planet. Sci. Lett. 44, 119-138, 1979.

37 S.-s. Sun, Lead isotopic study of young volcanic rocks from mid-ocean ridges, ocean islands and island arcs, Philos. Trans. R. Soc. London 297, 409-455, 1980.

38 V.C. Juan, H.J. Lo and C-H. Chen, Genetic relationship and the e m p l a c e m e n t of the exotic basic rocks enclosed in the Lichi melange, Coastal Range, east Taiwan, Proc. Geol. Soc. China 23, 56-68, 1978.

39 J.G. Liou, Zeolite facies m e t a m o r p h i s m of basaltic rocks from the East Taiwan Ophiolite, Am. Mineral. 64, 1-14. 1979.

40 J.G. Liou and W.G. Ernst, Oceanic ridge m e t a m o r p h i s m of the East Taiwan Ophio[ite, Contrib. Mineral. Petrol. 68, 335-348, 1979.

41 R. Batiza and D. Vanko, Petrology of young Pacific seamounts, J. Geophys. Res. 89, 11235-11260, 1984. 42 A. Zindler, H. Staudigel and R. Batiza, Isotope and trace

element geochemistry of young Pacific seamounts: implica- ~tions for the scale of u p p e r mantle heterogeneity, Earth Planet. Sci. Lett. 70, 175 195, 1984.

43 D.J. Fornari, M.R. Perfit, J.F. Allan, R. Batiza, R. Hay- mon, A. Barone, W.B.F. Ryan, T. Smith, T. Simkin and M.A. Luckman, Geochemical and structural studies of the Lamont seamounts: s e a m o u n t s as indicators of mantle processes, Earth Planet. Sci. Lett. 89, 63 83, 1988. 44 D.W. G r a h a m , A. Zindler, M.D. Kurz, W.J. Jenkins, R.

Batiza and H. Staudigel, He, Pb, Sr and Nd isotope constraints on m a g m a genesis and mantle heterogeneity beneath young Pacific seamounts, Contrib. Mineral. Petrol. 99, 446 463, 1988.

45 J.F. Allan, R. Batiza, M.R. Perfit, D. Fornari and R.O. Sack, Petrology of lavas from the L a m o n t s e a m o u n t chain and adjacent East Pacific Rise, 10°N, J. Petrol. 30, 1245- 1298, 1989.

46 E.M. Klein and C.H. Langmuir, Global correlations of ocean ridge basalt chemistry with axial depth and crustal thickness, J. Geophys. Res. 92, 8089-8115, 1987. 47 W.G. Melson, T.L. Vallier, T.L. Wright, G. Byerly and J.

Nelen, Chemistry diversity of abyssal volcanic glass erupted along Pacific, Atlantic and Indian Ocean sea-floor spread- ing centers, in: T h e Geophysics of the Pacific O c e a n Basin and Its Margin. A G U Geophys. Monogr. 19, 351-367, 1976.

A NEW GENETIC MODEL FOR THE EAST TAIWAN OPHIOLITE 145 48 BVSP, Basaltic Volcanism on the Terrestrial Planets, 1286

pp., Pergamon Press, New York, N.Y., 1981.

49 W.B. Bryan, Systematics of modal phenocryst assemblages in submarine basalts: petrologic implications, Contrib. Mineral. Petrol. 83, 62 74, 1983.

50 E. Ito, W.M. White and C. Gopel, The O, St, Nd and Pb isotope geochemistry of MORB, Chem. Geol. 62, 157-176, 1987.

51 S.-s. Sun and W.F. McDonough, Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes, in: Magmatism in the Ocean Basins, A.D. Saunders and M.J. Norry, eds., Geol. Soc. London Spec. Publ. 42, 313-345, 1989.

52 H.R. Kuddras, M. Wiedicke, P. Cepek, H. Kreuzer and P. Muller, Mesozoic and Cainozoic rocks dredged from the South China Sea (Reed Bank area) and Sulu Sea and their significance for plate-tectonic reconstruction, Mar. Petrol. Geol. 3, 19-30, 1986.

53 K. Tu, M.F.J. Flower and R.W. Carlson, Isotopic evidence for the Dupal anomaly in post-spreading magmas from the South China Basin, Chem. Geol. 70, 57, 1988.

54 K. Tu, M.F.J. Flower, G.H. Xie and R.W. Carlson, Decou- pied isotope and magmaphile element enrichment in the South China Basin magmas, Abstr., Symp. Geology and Geophysics of the South China Sea, Guangzhou, China, 1989.

55 K. Tu, M.F.J. Flower, R.W. Carlson, G.H. Xie and C.Y. Chen, Magmatism in the South China basin, 1. Isotopic and trace element evidence for an endogenous Dupal mantle component, Chem. Geol. in press, 1991.

56 M. Wang, D. Wu, I. Liang and A. Yin, Some geochemical characteristics of basalts from the South China Sea, Geochimica 4, 332-340, 1984.

57 S.L. Chung, S.-s. Sun, K. Tu and C-H. Chen, Generation of the late Cenozoic basalts in SE China: consequence of continental stretching and asthenosphere upwelling in re- sponse to India-Eurasia collision, in prep.

58 M.F.J. Flower, K. Tu, M. Zhang and G.H. Xie, The South China Basin domain: a common source for recent mid- plate and convergent margin magmas, Abstr. Symp. Geol- ogy and Geophysics of the South China Sea, Guangzhou, China, 1989.

59 C.J. Hawkesworth, M.S.M. Mantovani, P.N. Taylor and Z.

Palacz, Evidence from the Parana of South Brazil for a continental contribution to Dupal basalts, Nature 322, 356-358, 1986.

60 A. Zindler and S.R. Hart, Chemical geodynamics, An. Rev. Earth and Planet. Sci. 14, 493-571, 1986.

61 P.R. Castillo, The Dupal anomaly as a trace of upwelling lower mantle, Nature 336, 667-670, 1988.

62 S.R. Hart, Heterogeneous mantle domains: signatures, genesis and mixing chronologies, Earth Planet. Sci. Lett. 90, 273-296, 1988.

63 J.J. Mahoney, J.H. Natland, W.M. White, R. Poreda, S.H. Bloomer, R.L. Fisher and A.N. Baxter, Isotopic and geo- chemical provinces of the western Indian Ocean spreading centers, J. Geophys. Res. 94, 4033-4052, 1989.

64 C. Rangin, J.F. Stephan and C. Muller, Middle Oligocene oceanic crust of South China Sea jammed into Mindoro collision zone (Philippines), Geology 13, 425-428, 1985. 65 K. Tu, M.F.J. Flower, R.W. Carlson, M. Zhang and G.H.

Xie, Sr, Nd and Pb isotopic compositions of Hainan basalts (South China): implications for a subcontinental litho- sphere Dupal source, Geology 19, 567 569, 1991. 66 S.B. Mukasa, R. McCabe and J.B. Gill, Pb isotopic compo-

sitions of volcanic rocks in the west and east Philippine arcs: presence of the Dupal isotopic anomaly, Earth Planet. Sci. Lett. 84, 153-164, 1987.

67 J.L. Lin, M. Fuller and W.Y. Zhang, Preliminary Phanero- zoic polar wander paths for the North and South China blocks, Nature 313, 444-449, 1985.

68 K.J. Hsu, J.L. Li, H.H. Chen, Q.C. Wang, S. Sun and A.M.C. Sengor, Tectonics of South China: key to under- standing West Pacific geology, Tectonophysics 183, 9-39, 1990.

69 M. Tatsumoto, Dupal anomaly in the Sea of Japan: a plate recycling model of OIB sources, Caltech Plume Symp. Abs., May 2-4, 1991.

70 E. Nakamura, I.H. Campbell and S.-s. Sun, The influence of subduction processes on the geochemistry of Japanese alkaline basalts, Nature 316, 55-58, 1985.

71 E. Nakamura, I.H. Campbell, M.T. McCulloch and S.-s. Sun, Chemical geodynamics in a back-arc region around the Sea of Japan: implications for the genesis of alkaline basalts in Japan, Korea and China, J. Geophys. Res. 94, 4634-4654, 1989.