Scholarly article on topic 'Chemical compositions and precipitation timing of basement calcium carbonate veins from the South China Sea'

Chemical compositions and precipitation timing of basement calcium carbonate veins from the South China Sea Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Marine Geology
Keywords
{"Calcium carbonate veins" / "Low temperature basement carbonates" / " 87Sr/86Sr ratios" / "Precipitation timing" / "The South China Sea" / "IODP expedition 349"}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Weiwei Ding, Yifeng Chen, Zhen Sun, Zihua Cheng

Abstract Sixteen calcium carbonate veins (CCVs) within the igneous basement recovered from both East and Southwest Sub-basin, close to the fossil spreading ridge of the South China Sea during the Integrated Ocean Drilling Program (IODP) Expedition 349 were investigated. The CCVs are composed primarily of either calcite or aragonite, and some of mixed aragonite and calcite. The δ18O values of CCVs range from 25.5 to 31.8‰ VSMOW, indicating these are typical low temperature basement carbonates precipitated at temperatures of 12–40°C. The 87Sr/86Sr ratios of CCVs from Site U1431 show a strong negative correlation with δ18O-calculated temperatures, regardless of carbonate phases – calcite or aragonite, indicating CCVs with lower 87Sr/86Sr ratios have precipitated from moderately warmer and more geochemically evolved hydrothermal fluids, and reflecting that precipitation of CCVs might have occurred any time between 14.5 and 0Ma at Site U1431. The formation timing of CCVs at Site U1431 is consistent with the ongoing hydrothermal flow and circulation led by recharging seawater into the volcanic basement through the nearby outcropped seamount. The oldest ages of CCVs from Site U1433 at the Southwestern sub-basin of SCS were determined to be ~18–11Ma, based on basement age of 18.5Ma and the well-established seawater 87Sr/86Sr ratio curve. It indicates that the hydrothermal circulation at Site U1433 which is more distal to a recharging/discharging site was only active until ~11Ma. In consequence, the CCVs within basalts from Sites U1431 and U1433 provide more insights into the past hydrologic conditions and hydrothermal circulation along the fossil ridge flank in the SCS.

Academic research paper on topic "Chemical compositions and precipitation timing of basement calcium carbonate veins from the South China Sea"

Contents lists available at ScienceDirect

Marine Geology

journal homepage: www.elsevier.com/locate/margo

Chemical compositions and precipitation timing of basement calcium ■ HCrossMark

carbonate veins from the South China Sea

Weiwei Dinga,b, Yifeng Chenc,% Zhen Sund, Zihua Chengb

a Second Institute of Oceanography, State Oceanic Administration, Hangzhou 310012, China b School of Earth Science, Zhejiang University, Hangzhou 310027, China

c CAS Key Laboratory of Ocean and Marginal Sea Geology, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences (CAS), Guangzhou 510640, China d CAS Key Laboratory of Ocean and Marginal Sea Geology, South China Sea Institute of Oceanology, Chinese Academy of Sciences (CAS), Guangzhou 510301, China

ARTICLE INFO

ABSTRACT

Keywords:

Calcium carbonate veins

Low temperature basement carbonates

Sr/Sr ratios Precipitation timing The South China Sea IODP expedition 349

Sixteen calcium carbonate veins (CCVs) within the igneous basement recovered from both East and Southwest Sub-basin, close to the fossil spreading ridge of the South China Sea during the Integrated Ocean Drilling Program (IODP) Expedition 349 were investigated. The CCVs are composed primarily of either calcite or ara-gonite, and some of mixed aragonite and calcite. The S18O values of CCVs range from 25.5 to 31.8%o VSMOW, indicating these are typical low temperature basement carbonates precipitated at temperatures of 12-40 °C. The 87Sr/86Sr ratios of CCVs from Site U1431 show a strong negative correlation with S18O-calculated temperatures, regardless of carbonate phases - calcite or aragonite, indicating CCVs with lower 87Sr/86Sr ratios have precipitated from moderately warmer and more geochemically evolved hydrothermal fluids, and reflecting that precipitation of CCVs might have occurred any time between 14.5 and 0 Ma at Site U1431. The formation timing of CCVs at Site U1431 is consistent with the ongoing hydrothermal flow and circulation led by recharging seawater into the volcanic basement through the nearby outcropped seamount. The oldest ages of CCVs from Site U1433 at the Southwestern sub-basin of SCS were determined to be ~ 18-11 Ma, based on basement age of 18.5 Ma and the well-established seawater 87Sr/86Sr ratio curve. It indicates that the hydrothermal circulation at Site U1433 which is more distal to a recharging/discharging site was only active until ~ 11 Ma. In consequence, the CCVs within basalts from Sites U1431 and U1433 provide more insights into the past hydrologic conditions and hydrothermal circulation along the fossil ridge flank in the SCS.

1. Introduction

Circulation of seawater through the ocean crust along the flanks of mid-ocean ridges has profound effects on the composition of both seawater and crust, and is also a principal mechanism of heat loss from the Earth's interior (e.g. Staudigel et al., 1981; Brady and GíslasonJ 1997; Coggon et al., 2004, 2010). It causes approximately one third of convective heat loss with the crustal ages > 1 Ma (e.g. Sclater et al., 1980; Stein and Stein, 1994; Elderfield et al., 1999). Seawater-derived hydrothermal fluid circulation within the basement will lead to hydrothermal fluid-rock interaction, which causes the dissolution of primary igneous minerals and formation of secondary minerals. The secondary minerals replace the primary minerals and fill veins/fractures, vesicles and pore spaces within the basalts. Calcium carbonate veins (CCVs), composed of either calcite or aragonite, are typical low-temperature (< 100 °C) hydrothermal by-products and are common in the upper oceanic crust. The weathering of igneous minerals especially

Ca2+-rich plagioclase and clinopyroxene etc. in oceanic crust driven by the circulation of hydrothermal fluids, releasing Ca2+ to the fluid and generating alkalinity, likely promotes authigenic calcium carbonate precipitation in veins and vesicles within basalt (Coogan and Gillis, 2013). The CCVs within basalt in upper oceanic crust provide an archive to understand the chemistry and evolution of hydrothermal fluids and alteration of upper oceanic crust, the temperatures and the duration of hydrothermal activities (e.g. Elderfield et al., 1999; Coggon et al., 2004, 2010; Rausch et al., 2013; Li et al., 2014).

No oceanic basement drilling work has been carried out in the South China Sea (SCS) until 2014. International Ocean Discovery Program Expedition 349 (IODP 349) — South China Sea Tectonics drilled five sites in the SCS, three of which cored into oceanic basement near the fossil spreading ridge, i.e. Sites U 1431, U1433 and U1434 (Fig. 1). Fractures and veins occur throughout the basalt in all these three sites. Most of the veins are composed of secondary calcium carbonates, which provide an archive to explore the effects of hydrothermal circulation on

* Corresponding author. E-mail address: yfchen@gig.ac.cn (Y. Chen).

http://dx.doi.org/10.10167j.margeo.2017.08.021

Received 29 September 2016; Received in revised form 23 August 2017; Accepted 24 August 2017

Available online 04 September 2017

0025-3227/ © 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/BY-NC-ND/4.0/).

Fig. 1. Morphological map showing the locations of drilling sites near the fossil spreading ridge from which calcium carbonate veins (CCVs) were recovered during the IODP 349. Blue broken line shows the boundary between the East Sub-basin and the Southwest Sub-basin (referred from Franke et al., 2014). Yellow dashed line is the fossil spreading ridge (referred from Sibuet et al., 2016). The seamount closest to the Site U1431 has no name yet, though most seamounts around the fossil ridge have been named, such as ZBSM: Zhenbei Seamout; HYSM: Huangyan Seamount; DMSM: Daimao Seamount; XBSM: Xianbei Seamount; SXSM: Shixing Seamount; ZZSM: Zhangzhong Seamount; XNSM: Xiannan Sea-mount; LBSM: Longbei Seamount; LNSM: Longnan Seamount; ZNSM: Zhongnan Seamount.

the upper oceanic crust. Only two CCVs were collected from Site U1434, thus CCVs from this site will not be included here. Carbonate mineralogy, the Mg/Ca and Sr/Ca ratios, trace element contents, stable carbon and oxygen isotopic compositions and 87Sr/86Sr ratios of CCVs within the basalt from Sites U1431 and U1433 were investigated to address the hydrothermal fluid chemistry and evolution in the upper oceanic crust, during the ending of the seafloor spreading of SCS. In addition, the global seawater Sr-isotope stratigraphy was used to determine the possible timing of carbonate vein precipitation, which in turn provides some constraint to the duration of hydrothermal circulation within the ocean crust at the flank of the fossil ridge in SCS.

2. Regional setting and geology

The SCS is a western Pacific marginal sea situated at the junction of the Eurasian, Pacific, and Indo-Australian plates. It is composed of three main sub-basins - East Sub-basin, Southwest Sub-basin and Northwest Sub-basin (Fig. 1). It has been developed by latest Cretaceous to Pa-leogene magma-poor rifting, seafloor spreading in the Oligocene--middle Miocene, followed by southward subduction beneath Borneo during the Paleocene and middle Miocene, and eastward subduction under the Philippine Sea plate starting at early middle Miocene (i.e. Taylor and Hayes, 1983; Briais et al., 1993; Cullen et al., 2010; Franke et al., 2014; Li et al., 2015b; Ding and Li, 2016; Sibuet et al., 2016; Hayes and Nissen, 2005; McIntosh et al., 2013). The opening scenario of the SCS was first proposed by Taylor and Hayes (1983) and Briais

et al. (1993) based on the magnetic anomalies, indicating that seafloor spreading occurred between 32 and 16 Ma. The opening history was complicated by the southward ridge jump and southwest stepwise ridge propagation. The timing of seafloor spreading is still controversial due to lack of sedimentary and basement age constraints in the SCS. A different cessation time of 20.5 Ma was suggested by Barckhausen et al. (2014). However recent shipboard results of microfossil biostratigraphy and palaeomagnetic measurement during IODP exp. 349, combined with deep tow magnetic anomalies have shown that the seafloor spreading occurred between ~33Ma and 15 Ma, with a southward ridge jump at ~23.6 Ma in the East Subbasin (Li et al., 2014; Li et al., 2015a, 2015b). It supports the age models proposed by Taylor and Hayes (1983) and Briais et al. (1993).

Site U1431 (15°22.50' N, 117°0.00' E) lies ~15km north of the relict spreading ridge of the South China Sea in the East Sub-basin, with a water depth of 4250 m. Numerous seamounts occur around the fossil spreading ridge in the East Sub-basin. Most seamounts have been named, though the nearest seamount to the Site U1431 has no name yet (Fig. 1 and Fig. 4). The closest seamount is ~ 17 km away from Site U1431 (Li et al., 2015a). The basaltic basement of Site U1431 was covered by ~ 900 m thick sedimentary sequence deposited since ~ 13 Ma. The sedimentary sequence is mainly consisted of clay and/or claystone with silt and calcareous turbidites. Before reaching the basement crust, a distinctive layer of ~ 300 m thick volcaniclastic breccia occurs across 600-885 mbsf. It is mostly formed by multiple volcanic eruptions from the nearby seamount during a period of ~8-13 Ma (Yan et al., 2014), shortly after the cessation of spreading (Li et al., 2015a and b; Zhang et al., 2017). It is noted that there is ~9 m (962.5-972.0 mbsf) thick sequence of pelagic claystone and claystone breccia of early Miocene in age, occurring between two basalt flows (Li et al., 2015b). In general, sediment deposited rapidly shortly after the oceanic crust formed, with an average sedimentation rate of ~7.0 cm/ ka at Site U1431.

In total ~ 46.2 m thick basement was recovered by penetrating ~118m beneath the top of the igneous basement across 889.9-1007.9 mbsf at Site U1431. The basement is primarily comprised of large massive basalt with limited pillow basalt interbeds. The basalt is predominately aphyric, with grain size ranging from microcrystalline to fine grained. The main phase assemblage of plagioclase and clin-opyroxene in the groundmass, with 0.1-0.5 mm subhedral-euhedral olivine microphenocrysts in some igneous lithologic units, together with geochemical composition, indicates that basalt recovered at Site U1431 is tholeiitic, typical mid-ocean ridge basalt (MORB) (Li et al., 2015b). Fractures, veins and vesicles occur throughout the basalt, with veins mostly filled with white carbonates and/or reddish to brown iron oxyhydroxide/oxide (Fig. 2a). It indicates that basalt has mostly undergone moderate alteration due to oxidative seawater infiltration.

Site U1433 (12°55.1' N, 115° 2.8' E) is ~50km southeast of the relict ridge in the Southwest Sub-basin, with a water depth of 4390 m. The basement of Site U1433 is blanketed by ~800m sedimentary succession composed mainly of clay and claystone with frequent silt and biogenic carbonate turbidites deposited since last ~ 15 Ma. Overall the sediment deposited rapidly at an average sedimentation rate of ~5.3 cm/ka, just a little lower compared to Site U1431, shortly after the basement crust formed at ~ 18.5 Ma (see discussion in Section 5.1).

In total 29.0 m basalt was recovered at Site U1433, by penetrating 60.8 m into the igneous basement at an interval of 796.7-857.5 mbsf. The basement is composed of ~ 37.5 m thick pillow basalt on the top and ~23.3 m massive basalt toward the bottom of the site. The basement basalt at Site U1433 is tholeiitic in composition, similar to typical MORB with plagioclase and clinopyroxene being the groundmass assemblage (Li et al., 2015b). Alteration veins are abundant at the top but decrease sharply below ~ 830 mbsf, indicating limited downward flow of seawater. It is consistent with fewer fractures with increasing depth. Alteration along fractures and veins produced 1 -2 cm wide yellow to brown halos parallel to fractures/veins (Fig. 2b). The alteration

intensity is significantly lower at Site U1433 than at Site U1431. Vein filling secondary minerals within basalts in Site U1433 are predominantly composed of carbonate, with minor celadonite, iron oxide/ oxyhydroxide and saponite, suggesting low-temperature alteration of basalts due to seawater-derived hydrothermal fluids (Li et al., 2015b).

3. Sampling and analytical methods

In total 16 basalt samples, each containing at least one calcium carbonate vein, were chosen for investigation. Ten samples are from Site U1431, and six from Site U1433 (Table 1). The selected veins cover a depth interval from 20.1 m to 114.2 m below the top of the basement at Site U1431, and from 12.66 m down to 58.5 m at Site U1433. The carbonates within veins were obtained by either micro-drilling or handpicking from crushed basalt samples.

The carbonate mineralogy, the stable carbon and oxygen isotopes of CCVs were analyzed at the Second Institute of Oceanography, SOA, China. The bulk carbonate mineralogy was determined by X-ray diffraction. The XRD patterns were obtained from 3° to 70° 20 at a low scan speed of 0.02° per second. The d(101) peak of quartz was used as internal standard.

The stable carbon and oxygen isotopes were determined using a Finnigan MAT 253 triple collector isotope ratio mass spectrometer with a Kiel IV Carbonate preparation system. The carbonate samples were reacted with orthophosphoric acid at ~ 70 °C to release CO2 for stable carbon and oxygen isotopic analyses. The stable carbon and oxygen isotopic values of CCVs were averaged of 3-5 measurements of each sample (Table 1). The carbon isotopic compositions are reported relative to Vienna Pee Dee Belemnite (VPDB) standard, while the oxygen isotope values relative to VSMOW. The analytical accuracy was better than 0.03% for S13C and 0.06% for S18O, respectively.

The Ca, Mg, Sr, Fe and Mn contents of CCVs were measured at the ALS Minerals, Guangzhou, China. About 100 mg powdered samples were digested in 5 ml concentrated HNO3 at 130 °C. The solution was centrifuged, dried up and collected with 1 ml 8 M HNO3. An aliquot of this solution was diluted with 2% v/v HNO3 by 100-200 fold, and spiked with 2.5 ppb Be, In, Re for internal standardization. The solution was measured for Ca, Mg, Sr, Fe and Mn in a high-precision mode by inductively coupled plasma-optical emission spectroscopy (ICP-OES). The analytical precision was better than 3% for Ca, 5% for Mg, 1% for Sr, 4% for Fe, 2.5% for Mn, respectively.

Sr isotopic ratios were determined at the ALS Scandinavia AB, Sweden. Approximately 0.5 g of sample was roasted in a muffle furnace at 850 °C for 5 h, before it was dissolved in concentrated HNO3, evaporated to dryness and re-dissolved in 3 ml of 3 M HNO3. The re-dissolved samples were loaded into small columns with 50 ml Sr-specific resins and eluted with 3 M HNO3 to remove major cations and Rb. Then Sr was collected by eluting with 0.05 M HNO3. The 87Sr/86Sr ratios were measured by ICP-SFMS (ELEMENT XR, ThermoScientific). The 87Sr/86Sr ratios are fractionation corrected during the measurements by normalizing the 86Sr/88Sr ratio to a convention value of 0.1194. The average value of 87Sr/86Sr for NIST SRM 987 was 0.710245 ± 0.000017 during the analyses. The precision of analyses was reported in 1o (Table 1).

4. Results

X-ray diffraction shows the carbonate phases in CCVs are composed primarily of either calcite or aragonite. Six of the CCVs are composed solely of calcite, seven of them are aragonite, and the remaining three are calcite and aragonite mixtures, as shown in Table 1.

4.1. Stable carbon and oxygen isotopes

In total 15 CCVs were obtained for stable carbon and oxygen isotopic values, due to the CCV of 349_U1433B_68R3W is of too little

Marine Geology 392 (2017) 170-178 Fig. 2. Typical CCVs developed at Sites U1431 (a) and U1433 (b).

amount to be measured. The S13C compositions are + 0.94 — + 3.37% VPDB (mean = 2.11 ± 0.66% VPDB, n = 15) (Table 1), similar to those of seawater dissolved inorganic carbon, indicating that those CCVs have mostly precipitated from seawater dominated fluids. All S13C values are > 0% VPDB, indicating no oxidized organic carbon has been incorporated into CCVs.

The S18O values of CCVs range from 25.8 to 31.8% VSMOW (mean = 29.2 ± 1.5% VSMOW, n = 15). The S18O values of arago-nite range from + 28.5 to + 30.7% VSMOW (mean = 29.8 ± 0.8%, n = 6), while calcite from +25.8 to +31.8% VSMOW (mean = 28.2 ± 2.0%, n = 6), showing that aragonite is enriched in 18O compared to calcite. It is mostly caused either by their different fractionation factors, or by different precipitation conditions of temperature and the fluid S18O. Within the individual site, the S18O values of CCVs fall into a narrow range: 27.4-31.8% VSMOW (mean = 29.7 ± 1.3%, n = 10) at Site U1431 in the East Sub-basin, 25.5-29.2% VSMOW (28.2 ± 1.4%, n = 5) at Site U1433 in the Southwest Sub-basin, respectively.

The 87Sr/86Sr ratios of CCVs within the basaltic basement from SCS range from 0.70741 to 0.70886 (n = 16), falling into the range between the two end-members of modern seawater value of 0.70918 (Veizer, 1989) and basalt value of ~ 0.7025 (Palmer and Edmond, 1989). The 87Sr/86Sr ratios of CCVs from Site U1431 vary from 0.70741 to 0.70886, averaging 0.70837 ± 0.00050 (n = 10), while the 87Sr/86Sr ratios of CCVs from Site U1433 from 0.70826 to 0.70886, averaging 0.70858 ± 0.00020 (n = 6). It seems that the CCVs at Site U1433 have more or less homogenous 87Sr/86Sr ratios compared to Site U1431. However there is no systematic differences in 87Sr/86Sr ratios between calcite and aragonite.

4.3. Trace elemental contents

Trace element contents of CCVs show distinct Sr and Mg concentrations for aragonite and calcite (Table 1; Fig. 3). Calcite is characterized by relatively low Sr contents with a range of 84.6-253.7 ppm (an average = 169.1 ± 59.8 ppm, n = 5). In consequence, the Sr/Ca ratios in calcite are low, ranging from 0.17 to 0.44, with an average of 0.35 ± 0.11 (n = 5). Whereas Sr contents are much higher in aragonite, between ~ 3805.0 and 9893.1 ppm (average = 5868.2 ±

2469.3 ppm, n = 5). The Sr/Ca ratios (mmol/mol) in aragonite are > 5, ranging from 5.38 to 14.00, averaging of 9.04 ± 3.54, n = 5). While a sample of mixed calcite and aragonite has a Sr/Ca ratio of 4.75, falling between those of calcite and aragonite. In contrast, the aragonite samples have an average Mg concentration of ~ 3100 ppm (a range of ~ 2050-4600 ppm), much lower than the average of ~ 15,000 ppm (ranging ~ 12,400-19,400 ppm) in calcite. On average, the Mg/Ca ratios (mmol/mol) of ~ 119.2 in calcite are nearly about one order magnitude higher than those of ~ 17.8 in aragonite in CCVs. The higher Mg and lower Sr contents in calcite compared to those in ara-gonite are in agreement with previous trace element analyses on CCVs (Coggon et al., 2004, 2010; Rausch et al., 2013).

The significant distinction between Sr and Mg contents of aragonite and calcite is mostly controlled by the crystal chemsitry (Deer, 1992). The rhombohedral calcite accommodates Ca2+ (ionic radius 1.00 A) as well as minor and trace elements having an ionic radius not > 1.00 A. The orthorhombic aragonite accommodates Ca2+ together with minor and trace elements having radii greater that 1.00 A (Deer, 1992). Mg is an element with small ionic radius, while Sr is a large ionic radius element, therefore Mg is present at higher concentrations in calcite than in aragonite, while Sr is concentrated in aragonite.

The Fe contents in carbonates are 0.5-3.0 wt%, with a mean of 1.6 ± 0.8 wt% (n = 11), while Mn of 0.03-0.5 wt%, averaging 0.2 ± 0.2 wt% (n = 11) (Table 1), indicating that these carbonates were precipitated from the hydrothermal circulation of the seawater. The much higher Fe contents of 0.5 to 3.0 wt% in CCVs from SCS compared to those of 24-1510 ppm (mean = 380 ± 350 ppm, n = 27) Fe in CCVs within basement basalt from eastern flank of the Juan de Fuca Ridge, are mostly caused by the widespread iron oxy-hydroxide/oxide within the CCVs from SCS, occurring as the reddish to brown staining within the CCV as shown in Fig. 2a.

5. Discussion

5.1. Carbonate precipitation temperatures and the ridge-flank hydrothermal flow in SCS

A recent age model of the SCS based on deep tow magnetic anomalies together with the biostratigraphy of IODP 349 (Li et al., 2014), reveals the Site U1431 situated near the older part of magnetic anomaly C5Br, corresponding to a basement age of ~ 16 Ma, while Site U1433 projected to the older part of magnetic anomaly C5En indicates

4.2. 87Sr/86Sr ratios

Table 1

Geochemistry of calcium carbonate veins.

Sample ID Depth Depth within Carbonate 813С (%о 8180 (%о Tfluíd СС) Ca (wt%) Mg (ppm) Fe (ppm) Mn (ppm) Sr (ppm) Sr/Ca Mg/Ca 87Sr/86Sr lo

(mbsf) basement (m) phases VPDB) VSMOW) (mmol/mol) (mmol/mol)

Site U1431 (-16 Ma)

349_U1431E_39R1W 910.1 20.1 A 1.90 30.4 22 32.2 2593 12,272 697 9893 14.07 13.30 0.70886 0.000027

349_U1431E_41R2W (15) 925.08 35.08 С 1.89 28.5 27 22.4 12,421 16,607 774 169 0.35 91.59 0.70852 0.000021

349_U1431E_41R2W (20) 925.13 35.13 С 2.50 27.4 32 18.3 12,783 13,146 465 169 0.42 115.24 0.70741 0.000032

349_U1431E_41R5W 928.6 38.6 A 2.00 30.2 23 26.9 2774 21,117 387 6511 11.06 16.98 0.70871 0.000024

349_U1431E_42R4W (16) 937.6 47.6 С 1.55 31.8 12 30.4 15,376 5279 852 254 0.38 83.30 0.70883 0.000022

349_U1431E_43R2W 944.7 54.7 С 2.55 29.0 24 17.4 17,728 27,620 5033 169 0.44 167.67 0.70812 0.000019

Unit X: claystone 962.5-972.0 mbsf

349_U1431E_47R5W 987.3 97.3 A 3.37 30.0 24 25.4 4583 30,103 2710 4820 8.69 29.79 0.70855 0.000033

349_U1431E_49R2W 993.9 103.9 A 2.45 30.7 21 32.2 2050 12,237 4259 3805 5.41 10.51 0.70851 0.000016

349_U1431E_50R2W 1003.5 113.5 AC 2.90 30.0 \ 30.2 4824 12,062 1704 3129 4.75 26.38 0.70858 0.000029

349_U1431E_50R3W 1004.2 114.2 A 1.73 28.5 30 \ \ \ \ \ \ 0.70763 0.000022

Site U1433 (-18.5 Ma)

349_U1433B_66R4W (12) 809.36 12.66 CA 2.12 28.3 \ \ \ \ \ \ \ \ 0.70864 0.000025

349_U1433B_66R4W(21/ 809.45 12.75 CA 2.61 28.6 \ \ \ \ \ \ \ \ 0.70864 0.000027

22) 349_U1433B_66R4W (26) 809.5 12.8 A 1.01 29.1 27 31.7 3558 11,992 3794 4312 6.22 18.49 0.70886 0.000024

349_U1433B_68R3W 823.82 27.12 С \ \ \ 23.08 19,356 11,538 310 85 0.17 138.29 0.70859 0.000034

349_U1433B_75R1W 854.88 58.18 С 0.94 29.2 23 \ \ \ \ \ \ \ 0.70846 0.000027

349_U1433B_75R2W 855.13 58.43 с 2.06 25.8 40 \ \ \ \ \ \ \ 0.70826 0.000022

Note: Carbonate phases are determined by XRD. A = aragonite, C = calcite, AC = aragonite-calcite mixture, \: no data or not determined. Temperatures are calculated according to oxygen isotopic fractionation equations for calcite (Friedman and O'Neil, 1977) or aragonite (Hudson and Anderson, 1989) by assuming precipitating fluids S180 = 0%o SMOW. lo represents the internal error of 87Sr/86Sr ratio measurement runs.

o £ o E E

- o o Aragonite

♦ Calcite

+ Aragonite-calcite

— mixture

i i i 1 i i i !♦ ♦ i .♦I . * . 1 «1

40 80 120

Mg/Cacarb (mmol/mol)

Fig. 3. The Mg/Ca and Sr/Ca ratios of CCVs from both Sites U1431 and U1433.

a basement age of ~ 18.5 Ma. In consequence, these CCVs must have precipitated < 20 Ma. It is well known that carbonate S18O value is controlled by both the S18O and the temperature of water from which it precipitates. The oxygen isotopic composition of fluids from which CCVs precipitated can be assumed to be ~ 0% VSMOW as applied for the carbonate veins of < 20 Ma by Coggon et al. (2010). The temperatures of fluids from which CCVs precipitated were estimated following the oxygen isotope fractionation equations of Friedman and O'Neil (1977) for calcite, and of Hudson and Anderson (1989) for aragonite. The precipitating temperatures of samples with mixed calcite and aragonite cannot be calculated, due to the exact contribution from calcite and aragonite were not determined for a mixed CCV.

Since the cooling and rapid expansion of the Antarctic continental ice-sheets in the earliest Oligocene, seawater S18O changed due to the ice volume on the earth. The amplitude of seawater S18O has been estimated to be ~0.6 to 1.0% (Zachos et al., 2001; Lear et al., 2000), which introduces the uncertainties in estimated temperatures up to ~ ± 4 °C. Accordingly, these CCVs from Sites U1431 and U1433 precipitated at temperatures from 12 to 40 °C, averaged of 25 ± 7 °C (n = 12), confirming these carbonates precipitated at low temperatures (< 100 °C). Specifically, the precipitation temperatures of the CCVs from Site U1431 are calculated to be from 12 to 32 °C, averaging

24 ± 6 °C (n = 9), and Site U1433 from 23 to 40 °C, with a mean of 30 ± 9 °C (n = 3), respectively. It indicates that temperatures of the uppermost crust at Sites U1431 is a little colder than those at Site U1433 during the period that these CCVs precipitated. However we should also notice that there are much less data available from Site U1433.

Presently, the seafloor is blanketed by hundreds of meters continuous and thick sediments, with similar water depths of ~ 4300 m, at both Sites U1431 and U1433. The marine sediments has a permeability much lower than that of the upper volcanic crust and prevents vertical fluid flow at thermally significant rates (e.g. Spinelli et al., 2004; Fisher and Wheat, 2010). Consequently the bottom seawater temperatures at both sites are the same, ~2.5°C. Based on the present geothermal gradients of 14.8 °C/km for Site U1431 and 78 °C/km for Site U1433 (Li et al., 2015b), the temperatures within the uppermost crust are ~20 °C and ~60°C at Sites U1431 and U1433, respectively. Thus the temperatures of the uppermost oceanic crust at Site U1431are presently more or less similar to those when these CCVs precipitated in the past. In contrast, the temperatures are much higher at present than those when the CCVs formed in the past in the uppermost crust at Site U1433.

The heat flow at these two sites is also quite different, with values of 17.2 mW/m2 and 83 mW/m2, respectively at Sites U1431 and U1433 (Li et al., 2015b). These heat flow values were estimated based on laboratory-determined thermal conductivities and the APC-3 downhole temperature measurements during IODP 349 (Li et al., 2015b). The heat flow at Site U1431 is much lower than those of ~ 90-100 mW/m2 at the central SCS, while Site U1433 is more or less close to the local heat flow (Shi et al., 2003; Li et al., 2010). The differences in heat flow and geothermal conditions at these two sites must have been caused by different geological and hydrologic conditions.

The basement at Site U1431 is buried by ~900 m sedimentary sequence, which includes ~ 300 m thick sequence of volcaniclastic breecia overlying directly on the basaltic basement. The thick volca-niclastic breccia sequence is part of an apron of an outcropped sea-mount nearby, and wedged in between sediments above and basalt below, as shown in Fig. 4. The seamount is only ~ 17 km away from the Site 1431, and was active during the middle to late Miocene, shortly after the cessation of spreading (Yan et al., 2014; Li et al., 2015b). It is known that seamounts around the oceanic ridges can serve as recharging sites, where cold seawater flows downward so rapidly to minimize heating during descent, and lead to low heat flow around recharge sites (Fisher et al., 2003; Spinelli and Fisher, 2004; Fisher and Wheat, 2010). The pore-water sulfate concentration reversal to the seawater value at the bottom of Hole U1431D (Li et al., 2015b) indicates that seawater likely enters the volcaniclastic breccia and oceanic crust through the seamount, and flows laterally within the crustal rock below the thick

Fig. 4. Schematic illustration of the ridge-flank hydrothermal flow and circulation at Site U1431 near the fossil ridge in the East Sub-basin, South China Sea, based on the modeling of Fisher and Wheat (2010). The outcropped seamount closest to the Site U1431 serves as a recharging zone, while a small seamount tens of kilometers away with moderately high heat flow (~120 mW/m2) possibly serves as a discharging zone. The cold seawater enters the basaltic basement through the seamount and flows laterally within the crustal rock below the thick sediments, causing the extremely low heat flow (~ 17.2 mW/m2) at Site U1431 in present. The interpreted seismic profile is revised from Li et al. (2015a).

Fig. 5. The 87Sr/86Sr ratios of CCVs versus S18O calculated formation temperatures, showing (a) a strong negative correlation for the Site U1431, where dash lines represent 95% confidence belts; and (b) some correlation for Site U1433. The symbols are the same as for Fig. 3.

sediments (Fig. 4). In consequence, the seamount close to Site U1431 mostly serves as a recharging zone for cold seawater to infiltrate into the volcanic basement and cool down the temperatures within the uppermost crust since the cessation of SCS spreading. That is, the CCVs within Site U1431 precipitated possibly under the hydrologic conditions similar to those at present. It is possibly the reason that the temperatures CCVs precipitated in Site U1431 are lower than those in Site U1433.

On the contrary, Site U1433 is located much further (~ 50 km) away from the fossil spreading ridge axis, without any seamounts within ~50km radius (Fig. 1). The volcanic basement is buried by ~800m thick, typical marine sediments, mainly of clay and claystone deposited since the late Miocene. The typical marine sediment has a permeability much lower than that of the volcanic materials, such as volcanicclastic breccia and basalt basement (Spinelli et al., 2004). Once the sediment thickness exceeds 10-20 m, the reduced fluid flow through sediment cannot extract a significant fraction of lithospheric heat (Fisher and Wheat, 2010). Thus the ~800 m thick sediments at Site U1433 must have significantly reduced the fluid circulation between sediment and lithospheric basement since the late Miocene, and basement has been warmed up in response to reduced heat extraction led by less percolating of cold seawater through sediment into the volcanic basement at Site U433. In consequence, the temperatures at uppermost volcanic basement are much higher at present than in the past when CCVs precipitated at Site U1433.

In summary, the precipitation temperatures of CCVs within the basalt in SCS also give some insights on the past hydrothermal flow and circulation at the ridge-flank in the SCS.

5.2. Geochemical constraints for rock-fluid interaction

During the hydrothermal circulation, strontium is easily mobilized and leached from basaltic basement. The basalts are characterized with a Sr concentration of ~ 100 ppm, which is similar to seawater Sr concentrations > 38 ppm in Cretaceous (Coogan, 2009), but much higher than ~7.8 ppm in modern oceans (Veizer et al., 1999; Veizer, 1989). It is well known that the 87Sr/86Sr ratios of basalt (~ 0.7025) (e.g., Palmer and Edmond, 1989; Elderfield et al., 1999) and seawater are distinct. The 87Sr/86Sr ratio of seawater has varied through time, with values of 0.7068-0.7092 through Phanerozoic time, and a standard secular curve of seawater 87Sr/86Sr ratios has been well established for the last 500 Ma based on marine biogenic carbonates (McArthur and Howarth, 2004). Therefore Sr isotope ratios (87Sr/86Sr) are useful to trace the hydrothermal fluid-rock interactions. Meanwhile Bohm et al. (2012) confirmed that the natural inorganic calcium carbonates precipitated very slowly due to the limitation of seawater dissolved inorganic carbon (DIC) within the basalt of the upper oceanic crust, leading the Sr isotopes of calcium carbonate and fluid achieved equilibrium. That is, there is negligible Sr isotope fractionation between inorganic calcium carbonate and the fluid it precipitated within the basalt. In consequence, the 87Sr/86Sr ratios of CCVs within the basalts will faithfully record the characteristics of hydrothermal fluids and hydrothermal

fluid evolution, that is, the hydrothermal fluid-rock interactions.

The 87Sr/86Sr ratios of carbonate veins from SCS range from 0.70741 to 0.70886 (0.70845 ± 0.00415, n = 16), which are much higher than that (~ 0.7025) of basalt. It reveals that carbonates must have precipitated from seawater-derived hydrothermal fluids which exchanged Sr with igneous basement to some lesser extent. The 87Sr/86Sr ratios of aragonite vary in a range of 0.70763-0.70886 (mean = 0.70852 ± 0.00046, n = 6), while those of calcite from 0.70741 to 0.70883 (mean = 0.70831 ± 0.00046, n = 7). At both sites, though aragonite is characterized with the highest 87Sr/86Sr ratio of 0.70886 among all the CCVs, the highest 87Sr/86Sr ratio in calcite is 0.70883, which is insignificantly lower than the highest value of 0.70886 in aragonite. That is, there is no systematic difference in 87Sr/86Sr ratios between calcium carbonate phases of aragonite and calcite. The variation in 87Sr/86Sr ratio of CCVs mostly reflects the evolution of hydrothermal fluids from which they precipitated.

The 87Sr/86Sr ratios of CCVs and the S18O-calculated precipitation temperatures are negatively correlated, regardless of carbonate phases of calcite or aragonite at both Sites of U1431 (Fig. 5a) and U1433 (Fig. 5b). The linear correlation between 87Sr/86Sr ratios of CCVs and the S18O-calculated precipitation temperatures at the Site U1431 is significant (p = 0.009), with a coefficient (R) of 0.79. While the regression for the Site U1433 is not significant (p = 0.6), which may be caused by less S18O data available from the Site U1433. Relatively high temperatures will increase the extent and intensity of seawater-basalt reactions, leading more Sr leached from the upper crust basalt into the evolved hydrothermal fluids (e.g. Coggon et al., 2004; Fisher and Wheat, 2010). Basalt is characterized with less radiogenic Sr (~ 0.7025), causing the hydrothermal fluids of high temperature with relative low 87Sr/86Sr ratios. As a result, CCVs within basalt precipitated from relatively hotter fluids are characterized with much lower 87Sr/86Sr ratios, and vice versa.

5.3. Precipitation timing of CCVs

The strong linear regression of 87Sr/86Sr ratios and the precipitation temperatures of CCVs from the Site U431 (Fig. 5a), indicates that sea-water 87Sr/86Sr ratios during the precipitation of these CCVs can be estimated by following the approach of Coggon et al. (2010). As the CCVs formed in the deep sea, the bottom seawater temperatures were likely only slightly above 0 °C. Presently the bottom seawater temperature of the Site U1431 is ~2.5 °C, and the basement formed at ~16Ma (see discussion in Section 5.1). Accordingly, the contemporaneous bottom seawater temperatures can be assumed to be 2.5 °C. The contemporaneous seawater 87Sr/86Sr ratios for the timing of CCVs formation are calculated to be in a range of 0.7088-0.7109, at 95% confidence level (Fig. 5a), by extrapolating the linear regression of 87Sr/86Sr ratio - temperature to an assumed bottom seawater temperature of 2.5 °C. Based on the well-established standard curve of seawater 87Sr/86Sr ratios through last 16 Ma (McArthur and Howarth, 2004), the CCVs within the uppermost basalt from the Site U1431 have precipitated any time between 14.5 and 0 Ma. This is consistent with

Fig. 6. The 87Sr/86Sr ratios of CCVs in Site U1433 versus the host basement age, and the curve is the seawater 87Sr/86Sr ratios of last 20 Ma (McArthur and Howarth, 2004). The oldest precipitated ages of Site U1433 were determined from the samples with higher 87Sr/86Sr ratios than the contemporaneous seawater of ~ 18.5 Ma. The symbols are the same as for Figs. 3 and 5.

the ongoing hydrothermal flow and circulation at the Site U1431 as being a recharging site triggered by the nearby outcropped seamount (see discussion in Section 5.1, Fig. 4).

However there is no significant linear correlation between 87Sr/86Sr ratios and temperatures of CCVs from the Site U1433 (Fig. 5b), the extrapolating approach cannot be applied to this Site. The ages of CCVs will be estimated from samples that have higher 87Sr/86Sr ratios than that of contemporaneous seawater by using the well-established standard curve of seawater 87Sr/86Sr ratios (McArthur and Howarth, 2004), as applied by Coggon et al. (2010) and Rausch et al. (2013). The age of basement basalt of the Site U1433 is estimated to be ~18.5 Ma (see discussion in Section 5.1), providing an oldest age constraint for the CCVs. This approach is straightforward for CCVs of last 18.5 Ma, as 87Sr/86Sr ratios of seawater has increased monotonically since last 38 Ma (McArthur and Howarth, 2004; Fig. 6). The three higher 87Sr/86Sr ratios (0.70859, 0.70864, 0.70886) of CCVs from the Site U1433 than that of contemporaneous seawater of 18.5 Ma demonstrate that CCVs precipitated after ~ 18.0 Ma, 17.4 Ma and 11 Ma (Fig. 6). These ages of CCVs of the Site U1433 represent the maximum or oldest timing when the CCVs precipitated, as these CCVs precipitated from the seawater-derived hydrothermal evolved fluids having incorporated Sr leached from basement basalts to some extent as discussed in Section 5.2. In turn, it must have lowered 87Sr/86Sr ratio of CCVs relative to the contemporaneous seawater, and causes CCVs having some older ages than their real ages in a period of last 38 Ma. However the minimum age of the CCVS in Site U1433 cannot be predicted because the exact Sr contribution from the basalt is hard to know. Nevertheless the estimated ages indicate that CCVs of the Site U1433 precipitated ~ 0.5-7.5 Ma after the basalt eruption. This is consistent with that most veining and alteration takes place within the first 10 Ma of oceanic crustal formation (e.g. Staudigel et al., 1981). In consequence, the hydrothermal circulation had been active until ~ 11 Ma at the Site U1433.

6. Conclusion

The hydrothermal fluid-rock interaction and the precipitation timing of CCVs were studied through the chemical and isotopic compositions of CCVs within the igneous basement basalt recovered near the fossil spreading ridge of SCS. The CCVs are primarily composed of either calcite or aragonite, and some of mixed calcite and aragonite. The S18O-calculated precipitated temperatures of CCVs range from ~ 12 to ~ 40 °C demonstrating that these CCVs are typical low-temperature basement carbonates.

The 87Sr/86Sr ratios of CCVs from both Sites U1431 and U1433 are

between 0.70741 and 0.70886, revealing these CCVs have preciptated from seawater-derived hydrothermal fluids which exchanged Sr with basement basalt to some lesser extent. Independent of the carbonate phases, 87Sr/86Sr ratios of CCVs at the Site U1431 have a strong negative correlation with the S18O-calculated precipitated temperatures showing that the 87Sr/86Sr ratios of CCVs were controlled by degree of hydrothermal fluid-rock interaction, and also reflecting CCVs from the Site U1431 have precipitated any time between 14.5 and 0 Ma. The timing of CCVs from the Site U1431 is consistent with the ongoing hydrothermal circulation here, due to a recharging zone triggered by the outcropped seamount close to the Site U1431.

Based on basement age of the Site U1433, and the well-established seawater 87Sr/86Sr ratio curve, the precipitated timing of CCVs within the uppermost basalt recovered from the Site U1433 were determined to be younger than 18.0-11.0 Ma. In consequence, the hydrothermal circulation was only active until ~ 11 Ma at the ridge-flank at the Southwestern Sub-basin.

The differences in CCVs from Sites U1431 and U1433 are mostly led by the different hydrologic conditions. The CCVs from Site U1431 precipitated at or near the recharging zone triggered by a nearby sea-mount, and those from Site U1433 formed more distal to a recharging/ discharging zone.

Acknowledgement

We are grateful to two anonymous reviewers for their constructive suggestions and fruitful comments for the manuscript. Discussion on the geology of SCS with Prof. Chun-Feng Li at Zhejiang University is appreciated. This research used samples provided by the International Ocean Discovery Program (IODP). Funding for this research was provided by the Dream Project of Ministry of Science and Technology of China (No. 2016YFC0600402), the National Programme on Global Change and Air-Sea interaction, SOA (No. GASI-GEOGE), the National Natural Science Foundation of China (No. 41676037, No. 41176040), and IODP China.

References

Barckhausen, U., Engels, M., Franke, D., Ladage, S., Pubellier, M., 2014. Evolution of the South China Sea: revised ages for breakup and seafloor spreading. Mar. Pet. Geol. 58, 599-611.

Bohm, F., Eisenhauer, A., Tang, J., Dietzel, M., Krabbenhoft, A., Kisakurek, B., Horn, C., 2012. Strontium isotope fractionation of planktic foraminifera and inorganic calcite. Geochim. Cosmochim. Acta 93, 300-314.

Brady, P.V., Gislason, S.R., 1997. Seafloor weathering controls on atmospheric CO2 and global climate. Geochim. Cosmochim. Acta 61 (5), 965-973.

Briais, A., Patriat, P., Tapponnier, P., 1993. Updated interpretation of magnetic anomalies and seafloor spreading stages in the South China Sea: implications for the tertiary tectonics of Southeast Asia. J. Geophys. Res. Solid Earth 98 (B4), 6299-6328.

Coggon, R.M., Teagle, D.A.H., Cooper, M.J., Vanko, D.A., 2004. Linking basement carbonate vein compositions to porewater geochemistry across the eastern flank of the Juan de Fuca ridge, ODP leg 168. Earth Planet. Sci. Lett. 219, 111-128.

Coggon, R.M., Teagle, D.A.H., Smith-Duque, C.E., Alt, J.C., Cooper, M.J., 2010.

Reconstructing past seawater Mg/Ca and Sr/Ca from mid-ocean ridge flank calcium carbonate veins. Science 327, 1114-1117.

Coogan, L.A., 2009. Altered oceanic crust as an inorganic record of paleoseawater Sr concentration. Geochem. Geophys. Geosyst. 10 (4), 111-116.

Coogan, L.A., Gillis, K.M., 2013. Evidence that low-temperature oceanic hydrothermal systems play an important role in the silicate-carbonate weathering cycle and long-term climate regulation. Geochem. Geophys. Geosys. 14 (6), 1771-1786.

Cullen, A., Reemst, P., Henstra, G., Gozzard, S., Anandaroop, R., 2010. Rifting of the South China Sea: new perspective. Pet. Geosci. 16, 273-282.

Deer, W.A., 1992. An introduction to the rock-forming minerals. Longman Sci. Tech. 66 (25), 509-517.

Ding, W.W., Li, J.B., 2016. Propagated rifting in the Southwest Sub-basin, South China Sea: insights from analogue modelling. J. Geodyn. http://dx.doi.org/10.1016/jjog. 2016.02.004.

Elderfield, H., Wheat, C.G., Mottl, M.J., Monnin, C., Spiro, B., 1999. Fluid and geo-chemical transport through oceanic crust: a transect across the eastern flank of the Juan de Fuca Ridge. Earth Planet. Sci. Lett. 172 (1), 151-165.

Fisher, A.T., Wheat, C.G., 2010. Seamounts as conduits for massive fluid, heat, and solute fluxes on ridge flanks. Oceanography 23 (1), 74-87.

Fisher, A.T., Davis, E.E., Hutnak, M., Spiess, V., Zuhlsdorff, L., Cherkaoui, A., Christiansen, L., Edwards, K.M., Macdonald, R., Villinger, H., et al., 2003.

Hydrothermal recharge and discharge across 50 km guided by seamounts on a young ridge flank. Nature 421, 618-621.

Franke, D., Savva, D., Pubellier, M., Steuer, S., Mouly, B., Auxietre, J.L., Meresse, F., Chamot-Rooke, N., 2014. The final rifting evolution in the South China Sea. Mar. Pet. Geol. 58, 704-720.

Friedman, I., O'Neil, J.R., 1977. Compilation of stable isotope fractionation factors of geochemical interest. In: Fleischer, M. (Ed.), Data of Geochemistry, U.S. Geological Survey Professional Paper 440-KK, 6th edition. (Reston).

Hayes, D.E., Nissen, S.S., 2005. The South China sea margins: implications for rifting contrasts. Earth Planet. Sci. Lett. 237, 601-616.

Hudson, J.C., Anderson, T.F., 1989. Ocean temperatures and isotopic compositions through time. In: Clarkson, E.N.K., Curry, G.B., Rolfe, W.D.I. (Eds.), Environments and Physiology of Fossil Organisms. Trans. R. Soc. Edinburgh, Earth Science, pp. 183-192.

Lear, C.H., Elderfield, H., Wilson, P.A., 2000. Cenozoic deep-sea temperatures and global ice volumes from Mg/Ca in benthic foraminiferal calcite. Science 287 (5451), 269-272.

Li, C.-F., Shi, X., Zhou, Z., Li, J., Geng, J., Chen, B., 2010. Depths to the magnetic layer bottom in the South China Sea area and their tectonic implications. Geophys. J. Int. 182 (3), 1229-1247. http://dx.doi.org/10.1111/j.1365-246X.2010.04702.x.

Li, C.F., Xu, X., Lin, J., Sun, Z., Zhu, J., Yao, Y.J., Zhao, X.X., Liu, Q.S., Kulhanek, D.K., Wang, J., Song, T.R., Zhao, J.F., Qiu, N., Guan, Y., Zhou, Z., Williams, T., Bao, R., Briais, A., Brown, E., Chen, Y., Clift, P., Colwell, F., Dadd, K., Ding, W., Almeida, I., Huang, X., Hyun, S., Jiang, T., Koppers, A., Li, Q., Liu, C., Liu, Z., Nagai, R., Peleo-Alampay, A., Su, X., Tejada, M., Trhnh, H., Yeh, Y., Zhang, C., Zhang, F., Zhang, G., 2014. Ages and magnetic structures of the South China Sea constrained by deep tow magnetic surveys and IODP expedition 349. Geochem. Geophys. Geosyst. 15 (12), 4958-4983.

Li, C.F., Li, J.B., Ding, W.W., Franke, D., Yao, Y., Shi, H., Pang, X., Cao, Y., Lin, J.,

Kulhanek, D.K., Williams, T., Bao, R., Briais, A., Brown, E., Chen, Y., Clift, P., Colwell, S., Dadd, K., Almeida, I., Huang, X., Hyun, S., Jiang, T., Koppers, A., Li, Q., Liu, C., Liu, Q., Liu, Z., Nagai, R.H., Peleo-Alampay, A., Su, X., Tejada, M., Trhnh, H., Yeh, Y., Zhang, C., Zhang, F., Zhang, G., 2015a. Seismic stratigraphy of the central South China Sea basin and implications for neotectonics. J. Geophys. Res. (Solid Earth) 120 (3), 1377-1399.

Li, C.F., Lin, J., Kulhanek, D.K., Expedition 349 Scientists, 2015b. Proceedings of the Intergrated Ocean Drilling Program: College Station, TX (International Ocean Discovery Program). http://dx.doi.org/10.14379/iodp.proc.349.2015.

McArthur, J.M., Howarth, R.J., 2004. In: Gradstein, F., Ogg, J., Smith, A. (Eds.),

Strontium isotope stratigraphy. In: A Geologic Time Scale. Cambridge University

Press, Cambridge, pp. 96-105. McIntosh, K., van Avendonk, H., Lavier, L., Lester, W.R., Eakin, D., Wu, F., Liu, C.-S., Lee,

C.-S., 2013. Inversion of a hyper-extended rifted margin in the southern Central Range of Taiwan. Geology 41 (8), 871-874.

Palmer, M.R., Edmond, J.M., 1989. The strontium isotope budget of the modern ocean.

Earth Planet. Sci. Lett. 92, 11-26. Rausch, S., Böhm, F., Bach, W., Klügel, A., Eisenhauer, A., 2013. Calcium carbonate veins in ocean crust record a threefold increase of seawater Mg/Ca in the past 30 million years. Earth Planet. Sci. Lett. 362, 215-224. Sclater, J., Jaupart, C., Galson, D., 1980. The heat flow through oceanic and continental

crust and the heat loss of the Earth. Rev. Geophys. 18 (1), 269-311. Shi, X., Qiu, X., Xia, K., Zhou, D., 2003. Characteristics of surface heat flow in the South

China Sea. J. Asian Earth Sci. 22 (3), 265-277. Sibuet, J.C., Yeh, Y.C., Lee, C.S., 2016. Geodynamics of the South China Sea.

Tectonophysics. http://dx.doi.org/10.1016/j.tecto.2016.02.022. Spinelli, G.A., Fisher, A.T., 2004. Hydrothermal circulation within topographically rough basaltic basement on the Juan de Fuca ridge flank. Geochem. Geophys. Geosyst. 5, Q02001. http://dx.doi.org/10.1029/2003GC000616. Spinelli, G.A., Giambalvo, E.R., Fisher, A.T., 2004. Sediment permeability, distribution, and influence on fluxes in oceanic basement. In: Elderfield, H., Davis, E. (Eds.), Hydrogeology of the Oceanic Lithosphere. Cambridge Press, UK, pp. 151-188. Staudigel, H., Hart, S.R., Richardson, S.H., 1981. Alteration of the oceanic crust: processes

and timing. Earth Planet. Sci. Lett. 52 (2), 311-327. Stein, C.A., Stein, S., 1994. Constraints on hydrothermal heat-flux through the oceanic

lithosphere from global heat-flow. J. Geophys. R. 99, 3081-3095. Taylor, B., Hayes, D.E., 1983. Origin and history of the South China Sea basin. In: Hayes,

D.E. (Ed.), The Tectonic and Geologic Evolution of the Southeast Asian Seas and Islands: Part 2. Geophys. Monogr., AGU, Washington, D. C. 27. pp. 23-56.

Veizer, J., 1989. Strontium isotopes in seawater through time. Annu. Rev. Earth Planet. Sci. 17, 141-167.

Veizer, J., Al, D., Azmy, K., Bruckschen, P., Buhl, D., Bruhn, F., Carden, G.H.F., Diener, A., Ebneth, St, Godderis, Y., Jasper, t., Korte, C., Pawellek, F., Podlaha, O.G., Strauss, H., 1999. 87Sr/86Sr, 813C and S18O evolution of Phanerozoic seawater. Chem. Geol. 161 (1-3), 59-88.

Yan, Q.S., Shi, X.F., Castillo, P., 2014. The late Mesozoic-Cenozoic tectonic evolution of

the South China Sea: a petrologic perspective. J. Asian Earth Sci. 85, 178-201. Zachos, J., Pagani, M., Sloan, L., Thomas, E., Billups, K., 2001. Trends, rhythms, and

aberrations in global climate 65 Ma to present. Science 292 (5517), 686-693. Zhang, G.L., Chen, L.H., Jackson, M.G., Hofmann, A.W., 2017. Evolution of carbonated melt to alkali basalt in the South China Sea. Nat. Geosci. 10 (3), 229-235.