(Pl. = plagioclase, Hbl. = hornblende, same below.)
Picture 31. BSE image of sample TD01014.
(Ilm. = ilmenite).
68
Chapter 4
Results
4.1 Mineral Chemistry
4.1.1 Spinel from the ETO peridotites
The results of EPMA analysis are listed in appendix 1. The spinels are Cr-rich and classify as Cr-spinel. The Cr2O3 content ranges from a relatively low concentration of 35 wt% for TD01001, to an intermediate range between 38 and 41 wt% for TD01002, TD01003 and TD01004 and increases to a slightly higher concentration of 41 to 46 wt% for sample TD01007, TD01008 and TD01013. The mean Cr2O3 value is ~42 wt%. The mean aluminum content of all samples is ~26.5 wt%, while samples TD01001 and TD01003 have higher concentrations (31 and 29 wt%) and TD01008 and TD01013 has lower (25 wt%). The Cr and Al contents are used to calculate the Cr# (100Cr/ (Cr+Al)). The mean Cr number of all samples is
~51.6. Sample TD01001 has the lowest Cr number value (Cr# = 43), whereas all other samples have values >47. The highest Cr# is 54 from TD01008.
The MgO content of the spinels are relatively constant at ~15% but their overall range is from 12% to 17%. TD01001 and TD01003 have higher values than average, mostly above 15.5 wt% and up to 16.5 wt%. Except few analyses, the value of TD01004 and TD01007 concentrate at 14 to 15.5 wt%. The value of sample TD01008 and TD01013 are slightly lower than others, between 12.5 and 14.5 wt%.
Total iron content (detected as FeO) counts for all samples between 13 and 17 wt%.
The ratio of Mg and Fe, calculated as the Mg# (100Mg/ (Mg+Fe)) ranges between
69
60 and 66. The concentration of other elements in the analysis (SiO2, TiO2, MnO, NiO, CaO, Na2O and K2O) are lower than 0.5 wt%.
4.1.2 Gabbro Mineral Chemistry
The gabbro mainly consists of three minerals, hornblende, plagioclase and ilmenite. The EPMA results of hornblende give the SiO2 content between 44 and 49 wt%, FeO content between 16.5 and 19 wt%, MgO content between 10.5 and 14 wt%, CaO content between 9.5 and 11 wt%, Al2O3 content between 4.5 and 7.5 wt%, TiO2
content and Na2O content below 2 wt% and MnO and K2O under 0.2 wt%. The SiO2
content in plagioclase composition range between 64 and 68 wt%, Al2O3 between 19.5 and 20.5 wt%, CaO around 1 wt%. There is around 10 to 11 wt% Na2O in the plagioclase, while there’s merely no K2O in it (less than 0.2 wt%). And other elements are also close to detection limits. The ilmenite have ~47.5 to 50.7 wt % FeO and 47.9 to 49.3 wt% TiO2 with other element concentrations < 1 wt%.
70 4.2 In situ zircon U/Pb age dating result
Picture 32 to 34 show the CL images of the 20 zircons separated from the gabbro TD01014 used for LA-ICP-MS analysis. The grain size of the zircons are
~100 to ~250 μm in diameter, and most of the grains are broken. Some of the zircons have obvious zoning while some are more complicated.
The full U/Pb analysis data are shown in appendix 2. The average 206Pb/238U detection ratio is 0.0022 ± 0.00014, while the average 207Pb/235U detection ratio is 0.0134 ± 0.00361. The Concordia diagram, constructed using Isoplot 3.0 (Ludwig, 2003), is shown in figure 11 with 2 error ellipses of the individual spot analyses.
The weighted mean 206Pb/238U age is 14.1 ± 0.4 Ma on 20 individual zircon crystals and the MSWD (i.e. mean square of weighted deviation) is 4.2. Due to the young age of these zircons, the 207Pb/235U age has a larger uncertainty and thus is less meaning in this research.
71
Picture 32. Zircon sample CL image and LA-ICP-MS testing spot.
(The individual zircon age is listed in the picture, same below.)
72
Picture 33. Zircon sample CL image and LA-ICP-MS testing spot.
73
Picture 34. Zircon sample CL image and LA-ICP-MS testing spot.
Figure 11. 206Pb/238U - 207Pb/235U concordia plot.
0.000 0.004 0.008 0.012 0.016 0.020 0.024 0.028
U-Pb Concordia Age
Figure 12. 206Pb/238U mean age.
9 10 11 12 13 14 15 16 17 18
206
Pb/
238U Age
Mean = 14.13 ± 0.37, 95% conf.
N = 20
data-point error symbols are 2
75
76
4.3 Trace elements, Rare Earth Elements (REEs) and Sr/Nd isotopes
The trace elements in the peridotite samples are quite depleted except the transitional metals (i.e. Sc, Ti, V, Cr, Mn, Co, Ni, Cu, Zn). The concentration of the transition metals range from several ppm (e.g. Sc) to tens ppm (e.g. Ti, V, Zn) to hundreds ppm (Mn, Co) to thousands ppm (e.g. Cr, Ni). Sr concentration in TD01002 (~219 ppm) is several times higher than other samples (0 ~ 43 ppm). Sample TD01004, though still quite depleted (< 3 ppm), is more abundant in other trace elements (i.e. Ga, Rb, Y, Zr, Nb, Cs, Ba, Hf, Ta, W, Pb, Th) than other samples (< 1 ppm or under detection).
The rare earth elements (REEs) of all serpentinized peridotites except TD01004 are quite depleted (< 0.1 ppm). TD01004 has much higher concentration in REEs than other samples (~ 10 times). These serpentinized peridotites have a slight U-shape pattern (i.e. slightly higher LREE and HREE) and a negative Eu anomaly.
If normalized to chondrite values (McDonough and Sun, 1995), the REEs concentration except TD01004 are about 0.1 times of C1 chondrite concentration, while TD01004 is 1 to 2 times chondrite concentration.
The measurement of 87Sr/86Sr isotope ratio of four peridotite samples (i.e.
TD1001, TD01002, TD01003, and TD01004) ranges between 0.70883 and 0.70908.
The initial 87Sr/86Sr values cannot be calculated due to the extremely low Rb content (i.e. below detection limit) but given their likely age (i.e. ~14 Ma) the measured value will not be significantly different from the initial values (i.e. ± 0.00001). The Sm and Nd concentration of most peridotites is too low to measure the 143Nd/144Nd ratio;
however, sample TD1004 has sufficient concentration and has a measured
77
143Nd/144Nd ratio of 0.513106. The initial ratio based on the age of the gabbro is 0.513083 and corresponds to an Nd(T) value of +9.1 using a CHURtoday value of 0.512638.
Gabbro sample TD01014 has a wide range of transitional metals, which is abundant in Ti (~15040 ppm) and V (~860 ppm) while depleted in Cr (~6 ppm) and Cu (~8 ppm). The REEs concentrations in TD01014 are generally about 10 times more abundant to the chondrite concentration, and it has a similar pattern as N-type (normal) MORB. The measured 87Sr/86Sr isotope ratio of 0.70503 corresponds to an ISr value of 0.70501. The measured 143Nd/144Nd isotope ratio is 0.513226 with an
Nd(T) value of +11.4.
78
Trace elements concentration ( in ppm)
TD01001 TD01002 TD01003 TD01004 TD01004-2 TD01007 TD01008 TD01013 TD01014
P 66 78 73 84 114 55 59 61 260
Table 2. Trace elements concentration of the East Taiwan Ophiolite.
79
REE concentration ( in ppm)
Chondrite TD01001 TD01002 TD01003 TD01004 TD01004-2 TD01007 TD01008 TD01013 TD01014
La 0.237 0.001 0.001 0.100 0.300 0.500 0.001 0.100 0.001 2.400
TD01001 TD01002 TD01003 TD01004 TD01004-2 TD01007 TD01008 TD01013 TD01014
La 0.004 0.004 0.422 1.266 2.110 0.004 0.422 0.004 10.127 normalization value based on the CI chondrite from McDonough and Sun (1995).
Figure 13. Sample REE concentration spider diagram. (Model by Sun and McDonough, 1989).
80
Figure 14. Sample trace elements spider diagram (Model by McDonough and Sun, 1995).
81
Sample No. Rock Type Rb (ppm) Sr (ppm) 87Sr/86Sr Sm (ppm) Nd (ppm) 143Nd/144Nd εNd(T)
TD01001 Peridotite 0 43 0.708891 ± 9 0 0 - -
TD01002 Peridotite 0 219 0.708833 ± 7 0 0 - -
TD01003 Peridotite 0 9 0.709076 ± 8 0 0 - -
TD01004 Peridotite 0 20 0.709035 ± 7 0.3 0.83 0.513106 ± 6 + 9.1
TD01007 Peridotite 0 0 - 0 0 - -
TD01008 Peridotite 0 0 - 0 0 - -
TD01013 Peridotite 0 0 - 0 0 - -
TD01014 Gabbro 3 219 0.705030 ± 7 3.1 8.51 0.513226 ± 6 + 11.4
Table 4. Sr-Nd isotopic ratio of the peridotite and gabbro.
82
83 4.4 Whole rock geochemistry
The peridotite samples have 37.9 to 39.9 wt% of SiO2, 35.8 to 38.5 wt% of MgO, 6.7 to 8.2 wt% Fe2O3, and < 1.0 wt% Al2O. TiO2, MnO, Na2O, K2O, P2O5
contents are close to the detection limit and commonly < 0.01 wt%. CaO content is more variable and is likely due to the presence of secondary calcite veins in the rock (c.f. Pic. 19 and 20). Samples TD01001, TD01002, TD01003 and TD01004 have CaO content > 0.5 wt% whereas samples TD01007, TD01008 and TD01013 have ≤ 0.05 wt%. The Mg# (100 Mg2+/ (Mg2+ + Fe2+)) of the serpentinized peridotite samples are between 90 and 92. The loss on ignition (LOI) of the peridotites is very high (i.e. 12 wt% to 13 wt%) and is consistent with highly serpentinized nature of the rocks. The gabbro has 48.4 wt% of SiO2, 2.6 wt% TiO2, 12.1 wt% Al2O3, 18.1 wt% Fe2O3, 0.2 wt% MnO, 5.9 wt% MgO, 8.1 wt% CaO, 3.6 wt% Na2O, 0.3 wt%
K2O, 0.1 wt% P2O5 and a LOI value of 0.9 wt%.
Content (wt %) TD01001 TD01002 TD01003 TD01004 TD01007 TD01008 TD01013 TD01014
Table 5. Bulk rock composition of the serpentinized peridotites (TD01001 to TD01013) and hornblende gabbro sample (TD01014).
84
85
Chapter 5
Discussion
5.1 Age of the East Taiwan Ophiolite
The first attempt to determine the age of the ETO was made by Huang et al.
(1979). Their study identified foraminifera species Sphenolithus heteromorphus and Calcidiscus macintyrei within red shales from nanofossil zone NN5 that constrains
the deposition age of the sediments to lower and middle Miocene (Fig. 15).
The first radioisotopic ages were reported by Jahn (1986) on two types of basaltic rock (i.e. glass and crystalline), plagiogranite and gabbro using the K-Ar method. The results produced four different ages: 33 ± 5 (plagiogranite), 14.6 ± 0.4 Ma (basaltic glass), 11 ± 4 Ma (gabbro) and 8.1 ± 0.9 Ma (crystalline basalt).
Consequently, Jahn (1986) interpreted the likely age of the ETO to be 14.6 ± 0.4 Ma based on mutual consistency with the paleontological interpretations of Huang et al.
(1979).
Recently, U-Pb age dates were reported by Shao (2015) on gabbro and diorite from the north branch of the Chiawu creek. The concordant ages of the gabbros are
~17.5 Ma (i.e. 17.5 ± 0.2 and 17.4 ± 0.2) whereas the diorites are 14.3 ± 0.5 Ma and plagiogranites are 14.1 ± 0.2 Ma. Shao (2015) did not provide an explanation for the 3 million years gap between the ages of the gabbros and the diorites. The mean zircon U-Pb date from the hornblende gabbro (i.e. 14.1 ± 0.4 Ma) in this study is within error of the K-Ar age of the basaltic glass reported by Jahn (1986), the diorite group of Shao (2015) and interpreted age of the red shale layers suggesting that the magmatism in the ETO was active during Langhian stage (within NN5) of the
86
Miocene. The radiometric age of the rocks from the ETO is as a reference of the duration of the active magmatism of the South China Sea.
87
Figure 15. The foraminifera timescale NN2 to NN6 correspond to the geological timescale.
(Timescale data from Mandur, 2009 and Marzouk, 2009)
88
5.2 Rare earth elements, trace elements and isotopic composition of the ETO mantle
Rare earth elements (REEs) of the rock shows the characteristics of the original composition of the magma and the surrounding environment during the rock forming process. In this research, the REEs of all the samples were analyzed, but several elements in the serpentinized peridotite samples are too low and cannot be detected, for instance, La, Ce and Lu. The REE concentration abundance is very low in peridotites (i.e. 0.001-0.1 ppm, or ~ 0.01-0.5 times of chondrite content) except sample TD01004. In the chondrite-normalized REE pattern graph (Fig. 16), the serpentinized peridotites show U shapes, where LREE and HREE are slightly abundant. The U-shaped REE patterns have been reported from harzburgite and dunite and associated boninite in several Phanerozoic ophiolites, including the Izu-Bonin-Mariana, Troodos, Urals, New Caledonia, Trinity, Pyrenean, Betts Cove, and Cuban ophiolites (Jahn, 1986; Gruau et al., 1998; Zhou et al., 2005; Kapsiotis, 2013).
The negative Eu anomalies are believed to be irrelevant to their magmatic processes;
instead, they could be induced by alteration/serpentinization (Fig. 16).
The REE patterns of the hornblende gabbro, on the other hand, are sub-parallel those of the peridotites at higher concentration levels. The slight depletion of LREE in hornblende gabbro (i.e. La/Yb ≦ 1) is very similar to of N-MORB (Fig. 17).
Figure 16. REE pattern of the peridotites show slightly U-shape and negative Eu anomaly.
89
Figure 17. Hornblende gabbro TD01014 REE pattern shows the same pattern of N-MORB.
90
91
The early studies have reported anomalously high 87Sr/86Sr values in serpentinized peridotite and concluded that they were genetically unrelated to oceanic crust (Bonatti et al., 1970; Menzies and Murthy, 1978); however, more recent studies in abyssal peridotites and in the plagioclase peridotite of the Zabargad Island (Red Sea) demonstrated that the high 87Sr/86Sr values measured in oceanic peridotites may result from seawater alteration processes (Kempton and Stephens, 1997; Gruau et al., 1998). Information conveyed by mantle xenoliths indicates that stable sub-continental lithosphere is dominated by refractory peridotites that are enriched in HIE and LREE and often have acquired an enriched isotopic signature. A striking feature of several orogenic peridotites is the existence of a negative correlation between Nd–
Sr isotopic enrichment and peridotite fertility. Bodinier et al. (1991) and Downes et al. (1991) suggest the involving of enriched melts in porous-flow channels would interact with the peridotites and produce more enriched compositions. At Lanzo, this process is tentatively ascribed to a ‘‘pre-rift’’ stage of the opening of the Liguro-Piemontese ocean basin (Bodinier et al., 1991). Secondary alteration processes are strongly responsible for element deviations, particularly the strong depletion of calcium and sodium relative to aluminum in some abyssal peridotites, as well as part of the scattered variations of silicon, magnesium, calcium, and sodium. Serpentinized, orogenic and ophiolitic peridotites may be strongly depleted in calcium and sodium as well.
Figure 18. Initial 87Sr/86Sr vs. εNd value.
92
93
5.3 The tectonic setting of the East Taiwan Ophiolite
The nature and tectonic setting of the East Taiwan Ophiolite is a topic that is still under debate (Suppe et al., 1981; Jahn, 1986; Chung and Sun, 1992; Shao, 2015).
Suppe et al. (1981) studied the stratigraphy of the Lichi formation of the Coastal Range of East Taiwan and suggested that the East Taiwan Ophiolite is a submarine scree deposit consisting of angular mafic and ultramafic plutonic blocks that formed at a ‘leaky’ transform fault offsetting (Liou et al., 1977, Suppe et al., 1981). Jahn (1986) presented geochemical and isotopic data of glassy basalts, gabbros and plagiogranites and considered that the East Taiwan Ophiolite is derived from a spreading ridge of an open ocean or marginal basin. Furthermore, Chung and Sun (1992) proposed that the East Taiwan Ophiolite formed at a slow spreading ridge environment. Shao (2015), however, compared the Hf isotope of gabbros, diorites and plagiogranites of the East Taiwan Ophiolite to similar rocks from other ophiolites and proposed that the East Taiwan Ophiolite is probably the result of upwelling of a fore-arc subduction zone.
The rock sequence of East Taiwan Ophiolite is described by previous studies (Liou et al., 1977, Liou and Ernst, 1979, Suppe et al., 1981, Chung and Sun, 1992) (Fig. 19). The red shale layers between plutonic rocks and the extrusive rocks indicate that it was exposed under the carbonate compensation depth (CCD) for a period of time, perhaps a deep ocean setting at great distance from land. If this is the case, a mid-ocean ridge setting seems to be more promising than the original ‘leaky olistostrome’ interpretation of Suppe et al. (1981). On the other hand the absence of boninite, an important mafic volcanic rock that is rich in both Mg and Si and
94
commonly found at subduction zones, suggests the East Taiwan Ophiolite is unlikely to related to a suprasubduction zone (SSZ) type ophiolite or to any subduction-related setting. The relatively high εNd(T) values of the gabbroic ETO ophiolitic rocks (i.e. >
+9) also suggests that the fore-arc subduction interpretation is vulnerable
The bulk rock geochemistry and mineral geochemistry of peridotites can be used to distinguish between different tectonic settings (Dick and Bullen, 1984;
Bonatii and Michael, 1989; Arai, 1992, 1994; Kamenetsky et al., 2001; Hebert et al., 2003). Some major and trace elements such as Mg, Cr and Al content, that are relatively immobile during serpentinization, can retain their original concentration or ratio and thus be used for interpreting the tectonic setting. Dick and Bullen (1984) compared a wide range of abyssal and Alpine-type spinel bearing peridotite and categorized the rocks by Cr# as an indicator of the degree of the depletion of the mantle source. Bonatti and Michael (1989) have demonstrated that whole rock Mg#
and Al content, and Cr# of spinel can be effective in distinguishing peridotites derived at ridge settings, passive margin settings, pre-oceanic rift and oceanic trench settings. Arai (1994) reviewed and discussed the Fo-Cr# trends in peridotites in different conditions (e.g. ocean-floor, back-arc basin, fore-arc, oceanic hot-spot, island arcs, subcontinental, subcratonic peridotite).
In this study the classification scheme of Bonatti and Michael (1989) is applied to the bulk rock geochemistry and spinel chemistry of the ETO peridotites.
The Mg# of the peridotites from this study and previous work range between 90 and 93 which are typical of a mid-ocean ridge setting like the North Atlantic (Fig. 20).
The bulk rock aluminum content of the serpentinized peridotites, although somewhat
95
depleted, is consistent with the Mg# and an oceanic ridge setting (Fig. 21). The Cr#
of spinel is a particularly robust indicator for the tectonic setting of peridotites because it is relatively unaffected by hydrothermal conditions commonly experience by oceanic peridotites (Dick and Bullen, 1984; Bonatii and Michael, 1989; Arai, 1994). The mineral chemistry of the spinels as shown in Figure 22 is consistent with the previous results of bulk aluminum and Mg#. Therefore the peridotite whole rock and spinel geochemistry indicate that the ETO was formed at a mid-ocean ridge setting which is consistent with the interpretations of Liou et al. (1977) and Jahn (1986). However, it is uncertain if the ETO was a slow-spreading ridge as proposed by Chung and Sun (1992).
.
96
Figure 19. Rock sequence of East Taiwan Ophiolite (revised from Liou, 1977).
97
Figure 20. Whole rock Mg number (revised from Bonatti, 1988).
98
Figure 21. Whole rock aluminum content (revised from Bonatti, 1988).
99
Figure 22. Spinel chrome number (revised from Bonatti, 1988).
100 5.4 Implications for the South China Sea
The East Taiwan Ophiolite, representing the rock sequence of the oceanic crust and uppermost mantle, is considered to be a preserved remnant of oceanic crust from either the South China Sea (Liou et al. 1977; Suppe and Liou, 1979; Liou, 1979;
Suppe et al., 1981; Jahn, 1986; Chung and Sun, 1992). The spinel chemistry, Nd isotopes and whole rock Mg# and Al2O3 content from the peridotites indicates that the rocks were likely formed at a Mid-Ocean Ridge setting. Therefore the ETO is more likely to be a remnant of the South China Sea because there was no other active spreading center located within the region during the Miocene.
The tectonic development of the South China Sea is a highly debated subject.
The duration of spreading, in particular, is a one of a number of issues that has yet to be fully resolved. Originally Taylor and Hayes (1980, 1983) proposed a magnetic anomaly chronology for South China Sea basin which is still widely accepted (11 to 5d, Taylor and Hayes, 1980, 1983; revised to 11 to 5c, Briais et al., 1993). They identified an E-W oriented spreading center between the Macclesfield Bank and Reed Bank (Fig. 23), where the symmetric anomalies 5d through 6c have been modeled to the north and south. Beyond the Macclesfield Bank and close to the South China were anomalies 7 to 11. The spreading ridge has a conjunction from NW-SE direction to E-W direction, and it is said to be a result of a ridge jump that occurred at anomaly 7 (~24.8-25 Ma) (Brais et al. 1993; Barckhausen and Roeser, 2004; Barckhausen et al. 2014; Li et al. 2014)
101
Figure 23. South China Sea spreading ridge and magnetic anomalies (revised from Taylor and Hayes, 1983; Briais et al., 1993).
102
Currently the duration of tectonic/magmatic activity of the South China Sea is widely accepted to be 32-15.5 Ma (Taylor and Hayes, 1980, 1983; Briais et al., 1993), however Barckhausen and Roeser (2004) suggested the South China Sea was active for only 12 Ma from 32 to 20.5 Ma, about 4-5 million years earlier than previous researches. Barckhausen et al. (2014) supported the interpretation by proposing a new spreading rate of the South China Sea before and after the ridge jump, suggesting a 56 mm/yr rate in the early stages and increasing to 72 mm/yr after the ridge jump in the central and NE sub-basin, and increasing to 80 mm/yr in the SW sub-basin. Li et al. (2014), based on the new deep tow magnetic anomalies and IODP Expedition 349 core, pulled back the age of spreading to 33-15 Ma and suggested the ridge jump occurred at ~23.6 Ma. Li et al. (2014) also disagreed the spreading rate that proposed by Barckhausen et al. (2014) as they pointing out that the spreading rate decreased from ~50 mm/yr to ~35 mm/yr at the later stage of the spreading instead of increasing to 72 mm/yr after the ridge jump. Chang et al. (2015) also commented on Barckhausen’s saying that Barckhausen and Roeser (2004) and Barckhausen et al. (2014) had neglected the radioactive age dating (Jahn, 1986) and nanofossil assemblage from the ETO (Hunag et al., 1979) and re-announced the slow-spreading ridge is the more plausible ocean spreading motion of the South China Sea.
If the East Taiwan Ophiolite is really a remnant of the South China Sea, the new age result dated in this study and by Shao (2015) suggest that magmatism and thus extensional tectonism of the South China Sea lasted until at least the mid Miocene (i.e. ~14 Ma) and is ~1.5 million years later than the accepted age now.
103
Jahn (1986) predicted that the East Taiwan Ophiolite had a life span of < 10 Ma based on a spreading rate of ~10 cm/yr indicating that it traveled a maximum distance of 1000 km. However, the ~10 cm/yr rate may be too fast and the ~1000 km distance may be too large but a rate of 5-8 cm/yr and maximum distance of 500 km
Jahn (1986) predicted that the East Taiwan Ophiolite had a life span of < 10 Ma based on a spreading rate of ~10 cm/yr indicating that it traveled a maximum distance of 1000 km. However, the ~10 cm/yr rate may be too fast and the ~1000 km distance may be too large but a rate of 5-8 cm/yr and maximum distance of 500 km