Cold-Water Corals in Gas Hydrate Drilling Cores from the South China Sea: Occurrences, Geochemical Characteristics and Their Relationship to Methane Seepages

Abstract

Cold-water corals (CWCs) are frequently found at cold seep areas. However, the relationship between fluid seepage and CWC development is not clear. Here, for the first time, we report the occurrences, species identification, mineralogy, carbon and oxygen isotopes, as well as elemental compositions of fossil CWC skeletons from gas-hydrate-bearing sediment in drilling cores from the South China Sea (SCS). Three sites (GMGS-08, GMGS-09B, and GMGS-16) were investigated but CWCs were only found at one site (GMGS-09B). Interestingly, the CWCs were found in three horizons and they were all embedded with authigenic carbonates. Three genera of fossil CWCs (Crispatotrochus sp., Solenosmilia sp. and Enallopsammia sp.) were identified. The CWC fragments are predominantly aragonite. The CWCs exhibit δ¹³C values between −8.4‰ and −0.6‰ that are significantly higher than δ¹³C values of the associated seep carbonates (δ¹³C values with an average of −55.6‰, n = 19), which indicates a carbon source other than methane for the CWCs. It appears that authigenic carbonates provide a substratum for coral colonization. Bathymetric high points, appropriate water temperature and stronger bottom-water currents at site GMGS-09B might be crucial to keep conditions favorable for the growth of CWCs in the studied area. In addition, high trace-element concentrations of Cr, Ni, Pb, U, Ba, Th, and Sr suggest that the CWCs are influenced by strong fluid seepage that can reach the water-sediment interface, and associated microbial activity. Hence, it also becomes evident that CWCs in hydrocarbon-rich seepage areas not only provide a critical constraint on the impact of fluid emission on the bottom water chemistry, but also are likely to be very precise recorders of the end time of cold seep activity.

1. Introduction

Cold-water corals (CWCs), also termed deep-water corals, belong to the phylum Cnidaria. They include black corals (Antipatharia), stony corals (Scleractinia), hydrocorals (Stylasteridae), and soft corals (Octocorallia) [1,2]. These corals commonly occur in marine settings on continental margins, slopes, seamounts, and deep-sea basins in water depths of 30–1000 m, e.g., the moderately deep parts of the Southern Ocean and North Atlantic [3,4,5]. Rather than relying on symbiotic algae, the shallow-water coral’s food supply, CWCs feed on plankton, organic matter, and tiny organisms in the oceans [1,6]. As such, CWCs are able to live in environments without sunlight and can provide paleoceanographic records in regions where other fossils are usually sparse [2].

Previous studies of CWCs have focused mainly on the geochemical record of deep-ocean changes archived in their skeletons [2,7]. Accurate dating by uranium-series disequilibrium and ¹⁴C systems are facilitated by the high contents of U in their aragonitic skeletons [8,9,10,11,12,13].Various geochemical proxies, including isotopes of carbon, oxygen, and neodymium, rare-earth elements (REEs), trace elements, and Ba/Ca, Mg/Ca and Sr/Ca ratios of CWC skeletons are commonly used to reveal oceanic processes such as carbon cycling [14,15], past water-mass changes [16,17,18], hydrothermal activity [7], changes in water temperature [19], and ocean acidification [20].

It has been proposed that the development of CWCs is controlled by various oceanographic conditions, including salinity, temperature, substrate, tractive current, and dissolved oxygen [21], as well as hydrocarbon seepage [22]. CWCs have been reported in association with hydrocarbon seeps in Norwegian fjords [6], offshore Ireland [23], the Kristen hydrocarbon field [24], the Norwegian Margin [25], the Hikurangi Margin (New Zealand) [26,27], eastern central Atlantic [28], the Gulf of Mexico [29], and the Gulf of Cádiz [30]. However, the influence of fluid seepage on CWC growth is not yet fully understood. One theory, termed the “hydraulic theory”, suggests that the energy and nutrient supply of CWCs may be directly fueled by cold seepage [22,31,32]. However, the absence of DNA from methane- and sulfide-oxidizing microbes in CWCs from an active hydrocarbon-rich seepage area of the Gulf of Cádiz indicates that CWCs do not harbor the chemosynthetic symbionts required to consume the reduced compounds in hydrocarbon-rich fluids [33]. It is possible that CWCs use authigenic carbonates, formed through anaerobic oxidation of methane (AOM) mediated by a syntrophic consortium of anaerobic methanotrophic archaea and sulfate-reducing bacteria [34,35,36,37,38,39], as a hard substrate for coral larval settlement [26,33,40,41,42]. In addition, CWCs potentially benefit from fluid-seepage-related structures such as mud volcanoes and carbonate mounds, as these structures provide morphological barriers that direct bottom currents to deliver nutrients to the corals [33,43,44].

CWCs in the South China Sea (SCS) were first discovered by Bassett-Smith [45] near the Zhongsha Islands and the Nansha Islands. Subsequently, many species of CWCs have been observed near the Xisha Islands and along the northern margin of the SCS [46]. Previous studies have focused mostly on species identification and ecological analysis with only limited reports of geochemical data. No evidence of CWCs being present in cold-seep areas of the SCS was reported until 2013, when abundant authigenic carbonates and cold-seep biota including CWCs were discovered in gas-hydrate-bearing sediments from cores GMGS2-08, GMGS2-09B, and GMGS2-16 obtained during China’s second gas-hydrate drilling expedition (GMGS-2) (Figure 1). Despite the near-identical oceanographic and sedimentological conditions of the three drill cores, CWCs were found only at core site GMGS2-09B (Figure 2). However, the exact environmental conditions that regulate the occurrence of CWCs in this region remain unresolved. These gas-hydrate drill cores represent excellent materials to study the environmental factors that affect CWC development in methane-rich settings.

The present study reports, for the first time, species identification, mineralogy, and geochemistry (carbon and oxygen isotopes and major and trace elements) of three CWC species (Crispatotrochus sp., Solenosmilia sp., and Enallopsammia sp.) retrieved from core GMGS-2 from a cold-seep area in the SCS (Figure 2). Our results provide insights into the relationship between fluid seepage and CWC development. Furthermore, the investigation helps to improve understanding of the factors that control coral growth and of interpretations of bottom-water chemistry recorded by CWC skeletons in fluid-seepage settings.

2. Geological Background

The northern SCS is an Atlantic-type passive continental margin, with extensive troughs, seamounts, and basins, including the Xisha Trough, Pearl River Mouth Basin, Beibuwan Basin, Qiongdongnan Basin, and Taixinan Basin [48]. The investigated area is located in the Dongsha area, which is situated in the eastern part of the Pearl River Mouth Basin of the SCS, at a water depth of ~664 m (Figure 1). Faults and mud diapirs, which are considered to be effective pathways for hydrocarbon migration, are common in the area [49]. The depths of bottom simulating reflectors are in the range of 160–220 m below the seafloor (mbsf) with an average depth of 180 mbsf [48]. Active cold-seep authigenic carbonates and benthic seep biota have been observed in the Dongsha area [49,50,51]. Moreover, gas hydrates have been recovered at this site during GMGS-2 in 2013 [48,52].

Gas hydrates were recovered during the GMGS-2 cruise conducted by the Chinese Geological Survey in coordination with Fugro and Geotek in 2013, with drill cores taken at sites GMGS2-08, GMGS2-09B, and GMGS2-16 (Figure 1). Authigenic carbonate nodules and chemosynthetic bivalve shells are common in the three cores. Three fossil CWC skeleton layers were identified in GMGS2-09B core samples, at depths of 3.00–4.47, 6.00–6.42, and 22.00–22.60 mbsf (Figure 2).

3. Samples and Methods

3.1. Samples

Drill core samples were collected at site GMGS2-09B. Gas hydrates containing >99% methane were recovered at depths ranging from 9 to 21 mbsf (Figure 2). There is an extraordinarily high sedimentation rate of ~73.3 cm/k.y. since 0.12 Ma at GMGS2-09B [48]. These samples exhibit both nodular and massive morphologies. The drilling depth reached ~105 mbsf in a water depth of 664 m. The dominant sediments were clayey silt and silt. Samples of fossil CWCs and authigenic carbonate nodules were obtained. The AMS¹⁴C (¹⁴C accelerator mass spectrometry) ages cluster of the CWCs obtained at 3.35–3.55 mbsf between 15.1 and 15.5 ka BP, and CWCs at 6.20–6.30 mbsf have an age of 17.1 ka BP (unpublished data). All samples were washed immediately after collection and placed in storage at 4 °C.

3.2. Analytical Methods

Well-preserved (without apparently crushed) fossil CWCs and authigenic carbonate samples were selected and further cleaned with deionized water using an ultrasonic cleaner in the laboratory. The shell fragments were separated from adhered sediments under a microscope to remove possible contaminants. CWCs were investigated using Stereo microscope (Discovery V20) and 3D X-ray microscope using an Xradia520 Versa instrument (Zeiss, Oberkochen, Germany) at the Guangzhou Marine Geological Survey and at Carl Zeiss, Shanghai Co., Ltd. (Shanghai, China), respectively. Subsamples from aragonite skeletons of CWCs were collected for geochemical analyses, and all samples were crushed and pulverized to pass through a 200 mesh.

X-ray Powder Diffraction analysis (XRD) was used for semi-quantitative characterizations of CWC mineralogy using a Rigaku D/MAX 2500PC Diffractometer (Tokyo, Japan) equipped with a diffracted beam graphite monochromator and using Cu Kα radiation at the Guangzhou Marine Geological Survey. Scans were conducted from 2.5° to 65° (2θ) at 0.01°/s, using a 300 mA current and 40 kV acceleration voltage. The relative proportions of carbonate minerals were quantified by whole pattern fitting and Rietveld refinement of the MDI Jade 6 program (V6.1, MDI, CA, United States).

Carbon and oxygen isotope values of CWCs and authigenic carbonates were analyzed at the Guangzhou Marine Geological Survey laboratory, following the procedure described by Chen et al. [53]. Samples were processed with 100% phosphoric acid at 80 °C to release CO₂ for measurement using a Kiel IV online carbonate preparation line connected to a MAT 253 mass spectrometer (Thermo Scientific, Waltham, MA, United States). The isotope values are expressed using δ notation relative to the Vienna Pee Dee Belemnite (VPDB) standard. Precisions based on repeated measurements of carbonate standards are in the order of 0.1‰ (2σ) for both δ¹³C and δ¹⁸O.

Major- and trace-element analysis was performed with HF–HNO₃ solutions following the procedure established by Hu et al. [54]. Major elements were analyzed via inductively coupled plasma–optical emission spectrometry (ICP–OES) (Optima 8300, PerkinElmer, Waltham, MA, United States) and trace elements were measured via ICP–mass spectrometry (MS) (X Series2, Thermo Fisher Scientific, Waltham, MA, United States) at the Guangzhou Marine Geological Survey. The analytical precision of element contents is better than 5%.

4. Results

4.1. Distribution and Species of CWCs

The abundance of CWCs decreases with depth, with distinct layers of higher abundance at 3.00–4.47, 6.00–6.42, and 22.00–22.60 mbsf (Figure 3). Most of the CWCs are fragments 0.063–5 cm in length. The coral fragments’ preservation, based on integrity of skeletal structure, ranges from poor to moderate throughout the core. Three genera and two unknown genera were identified, including the solitary corals Crispatotrochus sp., Solenosmilia sp., and Enallopsammia sp. (Figure 4). The most commonly observed CWC fragments were classified as Solenosmilia sp. Authigenic carbonates, mainly occurring as gray to gray-white nodules, are found in association with these CWC fragments. These carbonates display a brecciform structure and have lengths of ~0.2–1 cm.

On the basis of microstructures revealed by stereo microscope and micro-CT, three coral species were identified: including Crispatotrochus sp., Solenosmilia sp., and Enallopsammia sp. (Figure 4).

Crispatotrochus Tenison-Woods, 1878

This genus is a sessile ceratoid solitary coral, with prominent costae, no paliform lobe, and bunchy columella, and does not have a symbiotic relationship with zooxanthellae. The species distribution is Pleistocene to present. Samples: GMGS2-09B at 3.45–3.55 mbsf, incomplete (Figure 4A,B).

Solenosmilia Duncan, 1873

This species exhibits dendritic or bush-shaped complexes, thin walls, and several septa with three complete rounds, or four or five incomplete rounds. The septa of the first and second rounds are long and twisted, with a central connection. Columella are spongy or unapparent. This species does not have a symbiotic relationship with zooxanthellae. Samples: GMGS2-09B at 3.45–3.55 mbsf (Figure 4C,F).

Enallopsammia Michelotti in Sismonda, 1871

This species exhibits dendritic complexes and has thin walls covered by costae. The septa are regularly arranged. The columella is mastoid. This species does not have a symbiotic relationship with zooxanthellae. The species distribution is Pleistocene to present. Samples: GMGS2-09B at 6.30–6.45 mbsf (Figure 4G,H), 6.00–6.15 mbsf (Figure 4I,L), and 3.45–3.55 mbsf (Figure 4K,M).

4.2. CWC Mineralogy

XRD results show that the CWC samples are 95–99% aragonite, and 1–5% calcite (

Table A1. Mineral compositions of cold-water coral fragments from the South China Sea.

Depth (cmbsf)	Sample	Ara	Cal
335–345	Cri.	99	1
345–355	Cri.	97	3
345–355	Sol.	99	1
345–355	Sol.	95	5
345–355	Sol.	98	2
355–365	Sol.	99	1
355–365	Ena.	97	3
630–645	Ena.	99	1

Note: Cri.—Crispatotrochus species; Sol.—Solenosmilia species; Ena.—Enallopsammia species; Ara—aragonite; Cal—calcite.; cmbsf—cm below seafloor.

), and authigenic carbonate samples are dominated by aragonite, high-Mg calcite, low-Mg calcite and dolomite (unpublished data).

4.3. Carbon and Oxygen Isotopes of CWCs and Carbonates

Stable isotope values for CWCs and carbonates are listed in

Table A2. Stable carbon and oxygen isotopes of cold-water corals and associated authigenic carbonates from the gas hydrate-bearing sediments of the South China Sea.

Core Number	Depth (cmbsf)	Sample Type	δ13C (‰)	δ18O (‰)
09B-2M-1a	325–335	Grey nodular carbonate	−48.5	3.6
09B-2M-1a	335–345	Grey nodular carbonate	−52.4	3.4
09B-2M-1a	345–355	Grey nodular carbonate	−52.0	3.9
09B-2M-1a	355–365	Grey nodular carbonate	−56.0	3.9
09B-2M-1a	600–615	Grey nodular carbonate	−59.5	4.4
09B-2M-1a	620–630	Grey nodular carbonate	−57.0	4.5
09B-2M-1a	620–630	Opalescent nodular carbonate	−53.2	4.6
09B-2M-1a	630–645	Grey nodular carbonate	−52.6	6.0
09B-2M-1a	630–645	Opalescent nodular carbonate	−60.4	4.3
09B-2M-1a	830–840	Opalescent nodular carbonate	−53.8	4.2
09B-2M-1a	830–840	Grey nodular carbonate	−55.5	4.7
09B-2M-1a	840–845	Grey nodular carbonate	−60.5	4.7
09B-5H-1	2200–2215	Grey nodular carbonate	−54.3	4.5
09B-5H-1	2220–2230	Grey nodular carbonate	−53.0	4.3
09B-5H-1	2230–2240	Opalescent nodular carbonate	−60.9	4.4
09B-5H-1	2230–2240	Grey nodular carbonate	−53.1	4.9
09B-5H-1	2240–2250	Opalescent nodular carbonate	−62.3	4.5
09B-5H-1	2240–2250	Grey nodular carbonate	−56.4	5.0
09B-5H-1	2250–2260	Grey nodular carbonate	−54.9	4.1
09B-2M-1a	325–335	Crispatotrochus species	−5.6	1.8
09B-2M-1a	325–335	Crispatotrochus species	−7.4	1.6
09B-2M-1a	325–335	Crispatotrochus species	−5.3	2.0
09B-2M-1a	335–345	Crispatotrochus species	−6.6	1.5
09B-2M-1a	335–345	Crispatotrochus species	−8.4	1.0
09B-3M-1	630–645	Crispatotrochus species	−4.6	2.2
09B-5H-1	2230–2240	Crispatotrochus species	−6.0	2.0
09B-2M-1a	345–355	Crispatotrochus species	−4.5	2.0
09B-5H-1	2200–2215	Enallopsammia species	−3.1	2.1
09B-5H-1	2200–2215	Enallopsammia species	−6.5	1.9
09B-5H-1	2230–2240	Enallopsammia species	−3.9	2.0
09B-5H-1	2220–2230	Enallopsammia species	−4.2	2.7
09B-3M-1	630–645	Enallopsammia species	−2.4	3.1
09B-3M-1	630–645	Enallopsammia species	−1.7	4.5
09B-2M-1a	2220–2230	Enallopsammia species	−4.6	0.9
09B-2M-1a	335–345	Solenosmilia species	−1.3	2.7
09B-2M-1a	335–345	Solenosmilia species	−4.5	1.0
09B-2M-1a	335–345	Solenosmilia species	−6.5	1.1
09B-2M-1a	335–345	Solenosmilia species	−7.6	1.2
09B-2M-1a	335–345	Solenosmilia species	−7.5	1.5
09B-2M-1a	345–355	Solenosmilia species	−5.9	1.8
09B-2M-1a	345–355	Solenosmilia species	−6.0	1.3
09B-2M-1a	345–355	Solenosmilia species	−6.8	2.6
09B-2M-1a	335–345	Solenosmilia species	−6.4	2.0
09B-2M-1a	345–355	Solenosmilia species	−1.7	2.4
09B-2M-1a	345–355	Solenosmilia species	−3.8	1.6
09B-2M-1a	345–355	Solenosmilia species	−6.3	1.9
09B-2M-1a	345–355	Solenosmilia species	−3.7	1.7
09B-2M-1a	345–355	Solenosmilia species	−6.5	1.5
09B-2M-1a	335–345	Solenosmilia species	−5.1	2.1
09B-2M-1a	345–355	Solenosmilia species	−6.5	1.0
09B-2M-1a	345–355	Solenosmilia species	−8.4	1.8
09B-2M-1a	345–355	Solenosmilia species	-−7.9	0.7
09B-2M-1a	355–365	Solenosmilia species	-−2.0	3.3
09B-2M-1a	355–365	Solenosmilia species	−7.8	0.6
09B-3M-1	600–615	Solenosmilia species	−3.8	2.9
09B-3M-1	600–615	Solenosmilia species	−3.4	1.7
09B-3M-1	630–645	Solenosmilia species	−2.4	2.4
09B-5H-1	2200–2215	Solenosmilia species	−3.6	1.8
09B-5H-1	2230–2240	Solenosmilia species	−4.3	3.0
09B-5H-1	2230–2240	Solenosmilia species	−6.4	1.0
09B-5H-1	2220–2230	Solenosmilia species	−7.5	1.0
09B-5H-1	2220–2230	Solenosmilia species	−6.8	1.7
09B-2M-2a	345–355	Solenosmilia species	−6.8	1.3
09B-2M-1a	610–615	Solenosmilia species	−5.7	1.5
09B-2M-1a	335–345	coral (species undefined)	−7.2	1.1
09B-2M-1a	345–355	coral (species undefined)	−7.3	0.7
09B-2M-1a	620–630	coral (species undefined)	−0.6	3.4

and presented in Figure 5. CWC δ¹³C values range from −8.4‰ to −0.6‰ (mean = −5.3‰, n = 48). The mean δ¹³C values of Crispatotrochus sp. and Solenosmilia sp. are −6.1‰ and −5.4‰ respectively, but Enallopsammia sp. displays a slight ¹³C enrichment compare to the above corals, with a mean value of −3.8‰. δ¹⁸O values for CWCs vary from 0.6‰ to 4.5‰ (mean = 1.9‰, n = 48). Crispatotrochus sp. and Solenosmilia sp. exhibit similar δ¹⁸O values that range from 0.6‰ to 3.3‰ (mean = 1.7‰, n = 38). Enallopsammia sp. yields higher positive δ¹⁸O values (mean = 2.5‰, n = 7).

Authigenic carbonates from site GMGS2-09B exhibit extremely negative δ¹³C values ranging from −62.3‰ to −48.5‰ (mean = −55.6‰, n = 19). δ¹⁸O values range from 3.4‰ to 6.0‰ (mean= 4.4‰, n = 19).

4.4. CWC Element Contents

Contents of major and trace elements in the CWCs are presented in

Table A3. Major and trace elements contents of cold-water corals from the South China Sea.

Depth	(cmbsf)	345–355	345–355	610–615	2220–2230
Al	(%)	0.021	0.028	0.076	0.036
Fe	(%)	0.003	0.011	0.026	0.002
Mn	(%)	0.0003	0.001	0.002	0.006
P	(%)	0.009	0.023	0.012	0.006
Ti	(%)	0.0001	0.0002	0.0003	0.0005
Ca	(%)	37.89	37.09	36.34	38.82
Mg	(%)	bl	bl	0.046	bl
Na	(%)	1.52	1.47	1.22	2.06
K	(%)	0.021	0.045	0.048	0.1
Cr	(μg/g)	2.5	2.6	2.9	5.3
Co	(μg/g)	0.7	0.8	0.9	1.4
Ni	(μg/g)	9.3	10.6	10.2	12.0
Cu	(μg/g)	1.3	3.1	2.7	3.8
Zn	(μg/g)	5.0	8.0	7.0	12.3
Cd	(μg/g)	0.1	0.1	0.1	0.1
Pb	(μg/g)	0.8	1.0	1.1	3.3
U	(μg/g)	4.6	4.4	6.9	7.0
Ba	(μg/g)	16.2	23.4	20.4	29.5
Th	(μg/g)	0.1	0.1	0.2	0.1
Sr	(μg/g)	9218.0	9248.0	9706.0	9853.0

Note: bl-Below detection limit.

. Mean calcium contents are 37.89% in Crispatotrochus sp., 36.72% in Solenosmilia sp., and 38.82% in Enallopsammia sp. All samples exhibit similarities in major-element abundances with Ca > Na > K > Al. Contents of Mg, Mn, and Fe are very low, and less than 0.03%.

All samples exhibit similar trace-element contents in the order Ba > Ni > Zn > U > Cr > Cu > Pb >Co > Cd (Table A3). Only Ba, Zn, and Ni show moderate to strong content in the CWC samples, with mean values of 22.36 μg/g for Ba, 9.06 μg/g for Zn, and 10.51 μg/g for Ni. Contents of Cr (2.51–5.25 μg/g), Cu (1.30–3.84 μg/g), Pb (0.79–3.26 μg/g), Co (0.73–1.40 μg/g), and Cd (0.082–0.093 μg/g) show little variation.

5. Discussion

5.1. CWCs Distribution and Factors Affect Their Growth

The factors determining CWC formation at cold seep areas are largely unknown. Hovland and Thomsen [22] established the “hydraulic theory” on the basis of the presence of cold seeps. These authors posited that hydrocarbon seeps directly fertilize CWCs, a premise supported by the fact that all the studied CWCs are embedded in authigenic carbonates. However, despite the almost identical oceanographic and sedimentological environments of the hydrate drilling area, CWCs were only observed at site GMGS-09B (Figure 2). Even at site GMGS-09B, not all the authigenic carbonate layers contain CWCs. Unlike bivalves and other seep-dwelling animals that directly use the chemical energy produced by cold-seep processes, CWCs may benefit indirectly from cold seeps through the formation of associated authigenic carbonates [33]. Authigenic carbonates potentially create a hard substrate for CWC larvae settlement [41]. Although cold-seep activity was widely observed in the research area, observations made by a remotely operated vehicle revealed that authigenic carbonates occurred in seabed only at site GMGS2-09B.

Oceanographic processes have also been regarded as an important factor in CWC development [33,55,56]. Bathymetric highs form areas where eddy currents and internal waves result in increased concentrations of nutrients and a higher food flux for CWCs [56,57]. The shallowest bathymetry recorded during the GMGS-2 survey occurred at site GMGS2-09B (water depth of ~664 m; Figure 2), and the raised seafloor might have resulted in an increased nutrient supply to drive CWC growth [55].

Additional environmental factors favoring CWC growth at site GMGS2-09B are oceanic temperatures and the prevailing hydrodynamic regime [56]. The bottom seawater temperature at site GMGS2-09B was measured as 5.35 °C, which is within the preferred temperature range of 4–13 °C reported for CWCs along continental margins worldwide [1]. In addition, the research area is located in an area where deep Pacific water flows through the Luzon Strait to enter the SCS [47]. This presence of a strong and deep seawater flow should deliver increased nutrients that drive CWC growth. Furthermore, the strong bottom current and the steep and hard seafloor structure prevent deposition of large amounts of debris that could bury CWCs. These favorable oceanographic conditions are likely a significant factor favoring the growth of CWCs in the study area.

5.2. Bottom-Water Chemistry and Fluid Seepage Archived in CWCs

5.2.1. Comparison of C and O Isotope Signatures

As CWCs use HCO₃⁻ from both ambient seawater and internally produced carbon dioxide for skeleton biomineralization, δ¹³C signatures of coral skeletons can be used to trace external carbon sources [33,58,59]. Food sources will also affect CWC stable isotope values [60]. CWCs may capture live zooplankton or detrital particulate organic matter from the upper water column [1]. Alternatively, it is also possible that they feed on bacterial mats or microorganisms associated with methane fluids [6]. Cold seeps are characterized by widespread chemosynthesis-based ecosystems and are commonly associated with high biomass, fueled by microbial chemosynthesis [61].

However, the data from 48 CWC samples in this study show a linear trend, with obviously high δ¹³C values (−8.4‰ to −0.6‰; Figure 5) compared to those of authigenic carbonates. These values are significantly different from those of the co-occurring authigenic carbonates. The highly ¹³C-depleted authigenic carbonates (mean = −55.6‰, n = 19) at the four intervals (3.25–3.65, 6–6.45, 8.30–8.45, and 22–22.6 mbsf) in core GMGS2-09B indicate a methane-derived isotopic signature, indicating the presence of past methane seepages at the times of deposition of these intervals (Figure 5). Biogenic methane typically exhibits δ¹³C values in the range of −110‰ to −50‰ [62] and thermogenic methane in the range of −50‰ to −30‰ [63]. The extremely negative δ¹³C values of the carbonates (−62.3‰ to −48.5‰) from core GMGS2-09B indicate that the dominant source of carbon fueling the system is biogenic methane.

A summary of δ¹³C and δ¹⁸O values of CWCs, coral reefs, and carbonates in the SCS and other locations are shown in Figure 6. The values exhibited by the CWCs and carbonates of the present study differ markedly from those from shallow coral reefs at the Xisha Islands [64] and Sanya Bay [65] in the SCS. Carbonates from hydrothermal sites from Mariana Forearc [66] and Lost City [67] show higher δ¹³C values relative to the authigenic carbonates and CWCs from site GMGS2-09B, indicating that the research area is not influenced by low-temperature hydrothermal processes. Our data are comparable with the ranges of δ¹³C and δ¹⁸O values in CWCs from the North Atlantic [68] and the Japan Sea [69], particularly those from the Dongsha area in the SCS [70,71], which does not have any cold-seep activity. As such, our results do not support the idea that CWCs take up methane-derived carbon at fluid-seepage areas. We therefore postulate that the primary food sources for CWCs are phytoplankton, zooplankton, and dissolved organic matter, as proposed by other studies [26,72]. However, more work on living coral skeletons is needed to verify this conclusion.

The strong linear trend of δ¹⁸O and δ¹³C from CWCs may be exploited for reconstructing temperature of past climates [1]. However, δ¹⁸O appear to have a relatively weak covariance with respect to δ¹³C in this study (Figure 6). The most reasonable explanation is that this observation is a result of gas hydrate dissociation. Decomposition of gas hydrate could liberate ¹⁸O-rich water with methane [35,50,53]. During an intense methane seepage, this fluid is transported from the sedimentary subsurface to the seabed and into the water column, resulting in alteration of δ¹⁸O in porewater and bottom seawater [51,54]. This environment would favor heavier δ¹⁸O values of CWCs during skeleton precipitation. The variance of oxygen isotopes in CWCs caused by gas hydrate dynamics is also supported by occurrence of gas hydrate resources at depth of 9–21 mbsf in GMGS2-09B (Figure 2).

5.2.2. CWC Trace-Element Contents

Trace-element contents in skeletal corals have been used to reconstruct oceanographic conditions [73], but there is very limited information on baseline levels of trace elements in CWCs [7], particularly in cold-seep areas. This study provides the first reported trace-element contents in the fossil CWCs Crispatotrochus sp., Enallopsammia sp., and Solenosmilia sp., collected at SCS cold seeps.

Metal contents within CWC skeletons are controlled by two main factors: the external environment and vital effects [74]. It is considered that the uptake and accumulation of elements in corals during metabolic and physiological processes are influenced by the ambient environmental conditions, including organic matter availability, salinity, production cycle, and seawater temperature [75]. Trace element contents in CWCs that live in non-seep settings tend to be fairly low under the conditions of their typical habitat without an additional source of metal elements [7,76].

In general, contents of Cr, Co, Ni, Cu, Zn, Cd, Pb, U, Ba, Th, and Sr in the samples from core GMGS2-09B are similar to or higher than those observed in modern CWCs from non-seep marine environments (Figure 7). Crocket et al. [77] suggested that such high trace-element contents could be derived from contaminant phases. However, the absence of a systematic relationship between Fe, Mn, Al, and other trace elements (Figure S1) indicates that ambient contamination is not likely the primary influence on CWC trace-element contents. High metal contents are presumably related to the existence of an additional source of trace elements. The contents of Ni and Pb are in good agreement with those found at a hydrothermal site (Azores Islands) (Figure 7), supporting this interpretation. The fluids found in cold seeps contain high concentrations of dissolved trace elements (e.g., Ba), resulting in elemental enrichments in bottom seawater and surface sediments [78,79]. However, the presence of free H₂S produced by intense AOM tends to scavenge trace elements from ambient porewater or seawater, causing subsequent enrichment of redox-sensitive elements (e.g., U and Mo) at cold seeps [53,80]. This is further supported by the obvious enrichment of trace elements (e.g., Mo, U, Pb, Zn and Cu) in seep-impacted aragonites that forming close to the sediment-water interface with free H₂S in sediment pore water [81]. Our results suggest that CWCs might record the influence of biogeochemical processes of fluid seepage on bottom water chemistry.

The geochemical and isotopic signatures and mineralogy of authigenic carbonates are widely used to trace the composition of ascending fluids, seepage intensities, and associated environmental parameters [78,82,83,84,85]. The CWCs in seep environments can record rapid changes in ocean chemistry, providing an important constraint on bottom water environmental conditions [26]. As such, the combined study of authigenic carbonates and associated CWCs in seep environments is potentially useful to calibrate the geochemical signatures within authigenic carbonates to aid in reconstruction of past conditions (Figure 8). In addition, U–Th dating of authigenic carbonates is a powerful tool that can also contribute to our understanding of the geological factors and processes controlling seabed methane release [86,87,88,89,90,91]. However, accurate dating of authigenic carbonates is difficult owing to contamination by detrital terrestrial material [84]. Furthermore, CWC geochronology can potentially provide information about the cessation of cold seep activity, as CWCs use authigenic carbonates as a substrate for colonization.

6. Conclusions

A combined study of three gas-hydrate-bearing sediment cores in the SCS reveal that CWCs were only found at one core and embedded with authigenic carbonates. The close relationship between CWCs and authigenic carbonates is due to the CWCs using authigenic carbonates as a substratum for colonization. Stable isotope studies suggest the source of carbon for CWCs is not directly from fluid seepage. It is suggested that the development of CWCs also require a high seabed topography, appropriate water temperature and stronger bottom-water current. However, in the case of strong fluid seepage, a large amount of nutrient elements will release into the bottom water. This information will be recorded by the coral skeletons. Therefore, CWCs from cold seep area could record the period of the latest cold seep activity, and the detailed geochemical study of CWCs will reveal environmental information during their growth period.