J. Cent. South Univ. (2012) 19: 1728-1740

DOI: 10.1007/s11771-012-1200-5

Carbon and oxygen isotope characteristics of foraminiferan from northern South China Sea sediments and their significance to

late Quaternary hydrate decomposition

LEI Huai-yan(雷怀彦)¹, CAO Chao(曹超)^{1, 2}, OU Wen-jia(欧文佳)¹,

GONG Chu-jun(龚楚君)¹, SHI Chun-xiao(史春潇)¹

1. State Key Laboratory of Marine Environment Science (Xiamen University), Xiamen 361005, China;

2. Third Institute of Oceanography, State Oceanic Administration, Xiamen 361005, China

? Central South University Press and Springer-Verlag Berlin Heidelberg 2012

Abstract: Carbon and oxygen isotope and dating analyses of foraminiferan in sediment cores collected from three different areas of the northern slope of the South China Sea were conducted, in order to examine the records of the gas hydrate decomposition events since the late Quaternary under the conditions of methane seepage. The results show that: 1) the δ¹³C values of the benthic foraminiferan Uvigerina spp. (size range of 0.25-0.35 mm) are from -0.212% to -0.021% and the δ¹⁸O values of the planktonic foraminiferan Globigerinoides ruber (size range of 0.25-0.35 mm) are from -0.311% to -0.060%; 2) three cores (ZD2, ZD3 and  ZS5) from the bottom of a hole are aged for 11 814, 26 616 and 64 090 a corresponding to the early oxygen isotope stage (MIS) I, III and IV final period, respectively; 3) a negative-skewed layer of carbon isotope corresponds to that of MIS II (cold period), whose degree of negative bias is -0.2‰; and 4) the δ¹³C compositions of foraminiferans are similar to those of the Blake Ridge and the Gulf of Mexico sediments of the late Quaternary. According to the analysis, the reasons for these results are that the studied area is a typical area of methane seep environment in the area during MIS II due to the global sea-level fall and sea pressure decrease. Gas hydrate is decomposed and released, and a large number of light carbon isotopes of methane are released into the ocean, dissolved to inorganic carbon (DIC) pool and recorded in the foraminiferan shells. A pyrite layer developed in the negative bias layers of the foraminiferans confirms that the δ¹³C of foraminiferans is more affected by methane and less by the reduction of marine productivity and early diagenesis. The use of foraminiferan δ¹³C could accurately determine late Quaternary hydrate release events and provide evidence for both reconstructing the geological history of methane release events and exploring natural gas hydrate.

Key words: northern South China Sea; foraminiferan; carbon and oxygen isotopes; natural gas hydrate

1 Introduction

In recent years, more and more negative bias phenomena of foraminiferan carbon isotopes in late Quaternary sediments have been identified from the Okinawa Trough, the Gulf of Mexico, Hydrate Ridge, Oregon continental margin, Norwegian Sea, Okhotsk Sea, the northern South China Sea continental slope and other methane seeps. The negative bias of foraminiferan carbon isotope ranges from -0.15% to -2.2% and near cold methane seep it is more obvious [1-8]. The absorbed capacity of ¹²C in the sediments diminishes because of low productivity, and the ¹³C content of the sediment decreases [9]. Foraminiferans devour methanogens in a strong reducing environment, and a light carbon isotope is recorded in calcareous foraminiferan shells as a result of biochemical processes [10-11].

Early diagenesis can post-reform the carbon elements of the foraminiferans, which results in a negative bias of the foraminiferan carbon isotope [12-13]. These factors do not clearly explain the basic reason why foraminiferan carbon and oxygen isotopes have shows significantly different compositions under different climatic conditions since the late Quaternary. DICKENS et al originally put forward the theory that large amounts of methane gas from the decomposition of instable gas hydrate can lead to a rapid global negative carbon isotope shift [14]. Later, many scholars linked a numbers of important events in geological history with this gas hydrate, for instance, the idea that decomposition of a large amount of gas hydrate led to the Fransnian/ Famennian boundary of the negative carbon isotope shift and extinction events [15-17]. Therefore, changes of the global climate and environment are closely related to the release of methane gas from hydrate decomposition.

The South China Sea is the largest semi-closed marginal sea of the western Pacific, which is influenced by global climate change. Meanwhile, the northern slope of the South China Sea is the potential zone for gas hydrate, and methane seeps and effusive mouths are widely distributed in this zone [18]. These provide a good environment for searching for evidence of decomposition of natural gas hydrate in the northern South China Sea and studying the connection between changes of foraminiferan carbon and oxygen isotopes and gas hydrate. Analysis of the carbon and oxygen isotope composition of benthic and planktonic foraminiferans in sediments from different areas and data of AMS ¹⁴C can be used to distinguish gas hydrate release events recorded in the foraminiferans from the late Quaternary. From this, the methane release history during geological time can be reconstructed, and the degree of influence of foraminiferan carbon isotope because of early diagenesis and marine productivity can be probed. All of these provides scientific useful information for the exploration of natural gas hydrate in this area.

2 Geological setting

The South China Sea is the largest marginal sea in the western Pacific, where the intersections between the Eurasian, Pacific and Indian plates are located. The South China Sea has a unique structural feature because of the movement by these three restricting plates. The eastern margin is of a convergent-type and the northern and western margins are of discrete-types [19]. The north continental slope of the South China Sea ranges from the continental shelf front to the southern deep water region of the Dongsha and Shenhu uplifts. The overall direction is NE, it is 900 km in total length, with its width gradually narrowing from west to east, and covers an area of about 21×10⁴ km². The general trend in topography is largely sloping ground from northwest to southeast, with the isobaths approximately parallel to the coastline, and with a water depth ranging from 200 to   3 000 m (Fig. 1). Sedimentary basins, such as the Southwest Taiwan, Pearl River Mouth, Xisha Sea Trough, Southeast Qiong and Yingge Sea Basins, are well developed on the continental slope. The maximum sediment thickness is more than 10 km. A diapiric fold structure and tensile fractures are developed in the slope area, particularly in the developed area or at the edge of the site, and many early fault formations are still active, which cut through the Quaternary sediments to the seabed surface, and support gas migrates from deep to shallow strata. In addition, we also found a series of geophysical and geochemistry criteria, such as bottom simulating reflector marks, decreases in the pore water concentrations of Cl^-, satellite thermal infrared anomalies and abnormal levels of methane, all of which show that the north continental slope of the South China Sea area is a most promising gas hydrate prospect area [20-21].
 

Fig. 1 Location of ZD2, ZD3 and ZS5 cores

3 Materials and methods

Three cores were collected by National Basic Research Program of China Team in Shenhu and Dongsha sea areas of the northern slope of the South China Sea at the ZD2 (20°03.3555′N, 117°25.3285′E), ZD3(19°27.3272′N, 116°17.0249′E) and ZS5 (19°52.135′N, 115°09.6138′E) sites (Fig. 1) with a gravity piston tube. Water depth at the three sites ranges from 1 300 to 2 115 m. The three core lengths are all about 8 m. Their lithology is mainly gray-green clay silt with sandy silt at local layers, which has turbidite deposit facies and calcareous nannofossil development at the core. There is also yellow-brown silt layer with relatively strong oxidation (2-4 cm). In the laboratory, samples were cut in every 10 cm, and detailed segments were taken at special lithological layers. Once the samples have been marinated for at least 2 d, they were wet sieved using 63-250 μm mesh sieves. The 250 μm fraction of the sediment was used for this work while the 63-250 μm fraction was stored.

Calcareous benthic and planktonic foraminiferans without any sign of chemical or physical alterations  were selected for stable isotope analyses. We were careful to use specimens of comparable size to avoid any isotopic effects due to ontogenetic variation. Each specimen was subjected to two steps of cleaning following procedures outlined in Ref. [22]. The first step involves mechanically cleaning the specimen by repeatedly rinsing in distilled water and reagent grade methanol followed by ultrasonication to remove adhering detrital material [23-24]. The second step includes removal of the organic matter by soaking the specimens in 15% hydrogen peroxide for 15 min followed by rinsing with methanol [25]. Globigerinoides ruber and Uvigerina spp. from the northern slope of the South China Sea were broken using a metal probe and the isotopic numbers for these specimens were reported as an average of two analyses. Cleaned foraminiferans were then treated with anhydrous phosphoric acid at 73 °C and analyzed using a Kiel III device connected to a Finnigan MAT 252 isotope ratio mass spectrometer in the Institute of Geology and Geophysics, Chinese Academy Sciences. Depending on the size of the tests, broken tests or single tests were included in the analyses. Data were reported with respect to the PDB standard. Precision of the technique was measured using an internal standard of Carrera Marble calibrated with NSB-19, and found to be ±0.004% for δ¹⁸O and ±0.008% for δ¹³C. The samples chosen from the same layer with foraminiferans collected from ZD2, ZD3 and ZS5 were rinsed with deionized water to remove the residual micro-organisms and dried at a constant temperature (500 °C). Then they were sealed in glass containers at 2-4 °C. About 5 g sediment of each sample was crushed to powder with agate grinding bowl, through 180 μm sieve after being baked for 24 h at lower temperature (below 40 °C), in which the inorganic carbonate minerals were then removed by ultrasonic wave for 30 min in hydrochloric acid solution. Elemental analyses of carbon were performed with Vario EL III CHNOS Elemental Analyzer. Analytical precision for the experimental procedure was estimated to be ±2% for total organic carbon (TOC) [26].

Planktonic foraminiferan (Neogloboquadrina dutertrei) samples taken at four, three and two layers from the ZD2, ZD3 and ZD5 cores were used as control points for AMS ¹⁴C dating analysis. The laboratory common code was XA and the equipment used was a HVEE 3.0 MV tandem accelerator mass spectrometer. AMS ¹⁴C data were conversed to the calendar year using the Fairbanks processing method (Table 1). A global carbon reservoir of 400 a between the atmosphere and the sea water was corrected automatically by the program [27-28].

4 Results

4.1 Sedimentary rate and stratigraphic framework

Modeling the ZD2 core involves the application of four age data of AMS ¹⁴C as control points, while ZD3 core modeling applies three age data of AMS ¹⁴C as control points (Fig. 2). Within and beyond these control points, the linear interpolation and linear extrapolation methods were applied to establishing a chronological framework. The δ¹⁸O distribution curve of the planktonic foraminiferan G. ruber for the ZS5 core is generally consistent with the SO49-8KL curve (Fig. 3) [29]. Furthermore, by comparing the oxygen isotopes in the late Quaternary period with the SPECMAP δ¹⁸O curve [30], oxygen isotopes of period I-IV and a series of the sub-period could be identified. This is in accordance with the AMS ¹⁴C dating results, thus further confirming the reliability of the stratigraphic framework.

Based on the stratigraphic framework, the sedimentary rates of the three cores could be estimated, and the average rates of the ZD2, ZD3 and ZS5 cores are 0.68, 0.28 and 0.13 m/ka, respectively. Although the three cores are all taken from the same slope, due to the differences in the slopes and sedimentary environments [31], sedimentary thicknesses per unit time is not the same. The locations of the three core samples are close, respectively, to the ODP184 voyage 1144 stations, the 1146 stations and the BY3 Shenhu drilling location. The average sedimentary rates are 0.87, 0.35 and 0.17 m/ka and, therefore, the average rates of the cores are consistent with those of these stations [32-33]. There are no specific comments or discussions of slumping and turbidity on the shallow surface of stations 1144, 1146 and the BY3 Shenhu drilling location. Dating the sample data does not show any inverted or confused time series, and the samples are mainly composed of clayey silt. The particle size is uniform, and there are no significant mutation layers. All these demonstrate that the samples could accurately reflect the normal sedimentary sequence of the region and the depositional environment. So the sedimentary system is relatively reliable. From the perspective of sedimentary rate in each period, the ZD2 core reveals only the oxygen isotope period I, and the sedimentary rate of this period is significantly higher than that of the other two stations. In the oxygen isotope period II, the sedimentary rate of the DZ3 core is 2-3 times higher than that of the ZS5 core, and the sedimentary rates of the oxygen isotope period I and III are also higher than those of the oxygen isotope periods II and IV. This shows that from the late Quaternary, the substance deposition flux ratio of the northern Dongsha is higher than that of the southern Dongsha, the sedimentary rate of the Dongsha sea area is significantly higher than that of the Shenhu sea area, and the substance deposition flux of the glacial is significantly lower than that of the interglacial. It is noted, therefore, that the sedimentary environments of different blocks in the northern South China Sea continental slope are greatly different.

Table 1 ¹⁴C age of foraminiferans from ZD2, ZD3 and ZS5 cores from northern South China Sea

Fig. 2 Stratigraphic framework and sedimentary rate of ZD2 (a), ZD3 (b) and ZS5 (c) cores from northern South China Sea

Fig. 3 G. ruber δ¹⁸O contrast curves of SO49-8KL (a) and ZS5 (b) cores from northern South China Sea

4.2 Stable isotopic composition of foraminiferans

From the foraminiferan oxygen isotope curve of the study area, it can be seen that carbon isotopic values of the benthic foraminifera Uvigerina spp. range from -0.021% to -0.054% and the oxygen isotopic values of the planktonic foraminiferan Globigerinoides ruber range from -0.078% to -0.218% on the ZD2 core from the Dongsha sea area (Table 2). The negative bias layer is not significant. The 180 cm depth corresponds to oxygen isotope 1.1, and at the 550 cm depth, the age of the lowest δ¹⁸O is about 7 000 a, which corresponds to the time of maximum sea level in the last Glacial [34-35]. The age of the core bottom is 11 814 a, revealed as the early I stage of the oxygen isotope (the age boundary of the oxygen isotope stages I and II is 12 ka). The carbon isotopic values of the benthic foraminiferans range from -0.041% to -0.196% and the oxygen isotopic values of the planktonic foraminiferans range from -0.060% to -0.263% from the ZD3 core in the Dongsha sea area (Table 2). The 0-220 cm δ¹³C value is above -0.1%, and the δ¹³C values of the 220-660 cm layers have a significant negative bias. The lowest value is at the   320 cm depth, which has a negative bias of -0.18%. There is a negative bias layer from 250 to 660 cm, and the negative bias value is up to -0.18%. The age of the core bottom is 26 616 a, revealed as the end of the oxygen isotope III stage (the age boundary of the oxygen isotope stages II and III is 24 ka). Similar to the carbon isotope, the δ¹⁸O value is also divided into two sections: the δ¹⁸O values at the 0-220 cm depth are from -0.2% to -0.3%, and the δ¹⁸O values at the 220-680 cm depth are from -0.15% to -0.08%. This section shows oxygen isotope features of sea water during a cold period. The carbon isotopic value of the benthic foraminiferans ranges from -0.041% to -0.212% and the oxygen isotopic values of the planktonic foraminiferans range from -0.096% to -0.311% in the ZS5 core from the Shenhu sea area (Table 2). Significant differences between the warm period and the cold period indicate that the oxygen isotopes of the planktonic foraminiferans are strongly affected by climate and the sea oxygen isotope composition. There are two negative bias layers from 160 to 360 cm and from 800 to 810 cm. The age of the core bottom is 64 090 a, revealing the end of the oxygen isotope IV stage (the age boundary of the oxygen isotope stages III and IV is 59 ka). The compositions of the carbon isotopes and oxygen isotopes in the three cores do not show any regional differences, but show great differences between the warm and cold periods, indicating that while the sea water changes with climate changes in the three blocks, there is universality and homogeneity of the oxygen isotope composition in the three blocks [37-38].

5 Discussion

5.1 Relationship between climate change and oxygen isotopes of foraminiferans

The δ¹⁸O value within planktonic foraminiferans is mainly controlled by water temperature and salinity, but is also affected by the volume of the polar glaciers. The foraminiferan Globigerinoides ruber lived in the surface water mixed layer in a period of geological history when the basic temperature and salinity were stable. Oxygen isotope equilibrium with the surrounding water environment only occurred in particular geological history, such as the alternation between the glacial and interglacial epochs, when changes in temperature were large. The polar ice was melted and aggregation occurred, which had great impact on the oxygen isotope composition of the seawater. SHACKLETON and OPDYKE believed that the δ¹⁸O value of seawater became larger when the global sea level decreased, because the glacial polar ice caps increased during the glacial epoch. In the interglacial epoch, the ice-melt water poured into the ocean, so the composition of oxygen isotopes in the sea water became smaller [39]. Changes of the δ¹⁸O in sea water are completely recorded in foraminiferan shells, with the temperature change being superimposed in the same direction. Namely, when temperatures increased during the interglacial epoch, theδ¹⁸O values of foraminiferans became negative; when temperatures decreased during the glacial epoch, the δ¹⁸O values of foraminiferans became larger; so the global marine oxygen isotope events were simultaneous with the global scale of mixing in the sea water. In other words, the oxygen isotope composition of the foraminiferans provides detailed events for stratigraphic contrasting.

Table 2 Isotope characteristics and age of foraminiferans and TOC content of ZD2, ZD3 and ZS5 cores at different horizons

The oxygen isotope curves of the planktonic foraminiferan Globigerinoides ruber from the ZD2, ZD3 and ZS5 cores can be seen in Figs. 4-6. The global climate became warmer during oxygen isotope stage III (in the interstadial of the last Glacial); the sea level slowly increased and stabilized at about the level of  -50 m; the proportion of sea water increased in the continental shelf water; the δ¹⁸O curve presented low values that ranges from -0.094% to -0.168%. In oxygen isotope stage II (the heyday of the last Glacial), the sea level dropped to the lowest position of the last glaciation, which was about -150 m; the proportion of water in the continental shelf water dropped to a minimum point; the δ¹⁸O value depended largely on oxygen isotopic composition of the fresh water from the land, which ranged from -0.060% to -0.150%. During oxygen isotope stage I, the value of oxygen isotopes rapidly became lighter. This period was the end of the last Glacial, the climate was warm, sea level rose, and sea water poured into the shelf, causing the δ¹⁸O values of foraminiferans to rapidly decrease to a range from -0.078% to -0.311%. The climate of the South China Sea has been affected mainly by the western Pacific warm pool since the late Quaternary, and the climate change curve was jagged and showed characteristics of slow heating and rapid cooling. This agrees with global climate change [40-42]. This response relationship between planktonic oxygen isotope variations and changes in global climate and sea level had a similar trend to the global sea-level curve obtained by CHAPPELL and SHACKLETON at the Huon Peninsula, Papua New Guinea, in the southern Pacific Ocean    (Fig. 7) [43].

Fig. 4 Isotope characteristics of foraminiferans and TOC content of ZD2 core

Fig. 5 Isotope characteristics of foraminiferan and TOC content of ZD3 core

Fig. 6 Isotope characteristics of foraminiferan and TOC content of ZS5 core

5.2 Hydrate decomposing events affecting carbon isotope composition of foraminiferans

Currently, the major view holds that changes in foraminiferan carbon isotopes are mainly affected by primary productivity. Photosynthesis would integrate more ¹²C in the organism, which would result in the δ¹³C dissolved in sea water and recorded in the foraminiferan shell. On the contrary, the δ¹³C of foraminiferans shows a smaller value with low primary productivity [44-45]. Combined with the carbon cycle theory, during the end of the interglacial period, the highest primary productivity should appear and the carbon isotopes of foraminiferans should show their largest value [46]. Therefore, the carbon isotope curve of foraminiferans from the ZD3 and ZS5 cores show that the carbon isotope is larger in the warm period (oxygen isotope stages I and III) than in the cold period (oxygen isotope stage II), but does not appear to be the smallest value at the end of oxygen isotope stage III on the curve. This indicates that there could be other factors affecting the variation of carbon isotopes of the foraminiferans. The life of benthic foraminiferans is limited to a few centimeters from the sea surface and the seabed, and only living foraminiferans and their gametes can reflect the sedimentary environment at that time. When dead foraminiferans are buried, the capacity of their shells to exchange ions with the environment is weakened, and their chemical composition is less changed. The carbonate content curve (Fig. 8) of the marine sediments in the BY3 core of the Shenhu sea area shows a consistent trend with the foraminiferan carbon isotope curve of the ZS5 core, which indicates that the carbon isotope composition of the benthic foraminiferans is mainly affected by the inorganic carbon pool with a very small range from the seabed surface to the overlying water layer, and these effects are mainly recorded by living foraminiferans in geological history [32].

Interestingly, negative bias layers of foraminiferan carbon isotopes are at 250-660 cm in the ZD3 core and 160-360 cm and 810 cm in the ZS5 core of the studied area. These layers correspond to the oxygen isotope II and IV stages (cold period). A similar phenomenon is found in the late Quaternary sediments of the Gulf of Mexico, Blake Ridge, Okhotsk Sea, Okinawa Trough and other hydrate regions, where the degree of negative bias is -0.2% [47-49]. During oxygen isotope stage II, the sea level is 150 m lower than it is now, and submarine pressure is reduced by 1.5×10⁶ Pa. This breaks the balance of the hydrate stability field, which results in gas hydrate decomposition to methane. Large amounts of methane migrated into the sulfate methane interface (SMI) and reacted with  in the pore water by anoxic oxidation of methane: CH₄+→+ HS^-+H₂O. So the δ¹³C of methane changed into  in the pore water (Fig. 9). Methane carrying light carbon isotopes containing  migrated into the inorganic carbon pool of the surrounding seabed environment, and was taken by the living foraminiferans and transformed into their calcium shell in a physiological process, in which light carbon was also recorded. Light carbon  was directly recorded by the living benthic foraminiferans during diagenesis of calcium in the calcareous shell [50-53]. The concentration of  in the sediments was high enough so that large amounts of methane consumed   and then the methane that carried with  made lighter carbon isotopes break the SMI and the zone of anaerobic methane oxidation into the seabed (Fig. 10). LU et al pointed out that in order to make the δ¹³C value of the global seawater dissolve to inorganic carbon (DIC) pool to reach 0.05%-0.15% negative bias, about (3.5-8)×10¹⁷ g carbon must be provided. The decomposition of organic matter producing methane alone is not far from this amount, but gas hydrate also releases methane and is maintained for a long time, which could provide enough carbon source [54]. In 2004, the Guangzhou Marine Geological Survey found the great Kowloon methane reef in the Dongsha sea area (Fig. 11), which formed 55 000 years ago and corresponded with oxygen isotope stage IV (cold period). This shows that methane release was sustained in the area during the cold period [55]. Using scanning electron microscopy, a large number of spherical pyrite aggregates are observed in sediment samples located   in the 350-500 cm layers of the ZD3 core and the 200-300 cm layers of the ZS5 core (Fig. 12). This indicates that the submarine environment shows strong reduction, and HS^- concentration is large during the cold period. Large HS^- concentration is related with hydrate releasing methane that reduces  concentration. In the Shenhu hydrate region, which is very close to the ZS5 site, free methane concentration is 5-200 μg/L in the sediments by field test and methane isotopes are about -6%, which display a biogenic methane type. This indicates that there is a favorable carbon isotope source [18]. From the TOC curve of the three blocks, it can be seen that most of the TOC contents are between 0.8% and 1.2% during the cold period, with an average TOC content in the cold period twice than that in the warm period. A large amount of matters of terrestrial origin was transported into the sea and provided great parent material for the organic carbon isotopes of the sediments and pore water.
 

Fig. 7 Sea-level and oxygen isotope curves: (a) Sea-level change curve and oxygen isotope curve for benthic foraminiferans in Huon Peninsula, Papua New Guinea; (b) Sea- level change curve of South China Sea 150 thousands years ago [43]

Fig. 8 Relationship between carbon isotopes of foraminiferans and carbonate content in Shenhu sea area: (a) ZS5; (b) Shenhu area 

Fig. 9 Chemical reactions at ulfate methane interface [50]

Fig. 10 Relationship between methane flux and SMI [51]

Fig. 11 Carbonate incrustation from Kowloon methane reef in Dongsha area [55]
 

Terrestrial higher plants utilize atmospheric CO₂ (δ¹³C= -0.7%), while marine planktonic plants and animals utilize dissolved CO₂ (δ¹³C=0%) [56]. Therefore, organic carbon isotopes also indirectly affect foraminiferan carbon isotopes, which are important factors in carbon negative bias.

Although the biochemical principles and fractionation mechanisms of negative bias on benthic foraminiferan δ¹³C are not fully understood, it is an indisputable fact that the negative bias extent of benthic foraminiferan δ¹³C is related to the concentration of methane, and the higher the concentration of methane, the more the negative bias of benthic foraminiferan δ¹³C (Table 1). This phenomenon is verified in other hydrate areas and methane leak environments, such as the Blake Ridge, Okhotsk Sea and Okinawa Trough, where δ¹³C average values of benthic foraminiferans are from -0.128% to -0.564%, but are only from -0.081% to -0.085% in environments with little leakage of methane [51-52]. South China Sea hydrate spreads in the sediments, and methane concentration is relatively low [32], with little influence on the δ¹³C of benthic foraminiferans in the modern sedimentary environment, which is the main reason that benthic foraminiferan δ¹³C negative bias is at a low level in methane leakage environments and the δ¹³C values of benthic foraminiferans is up to -2.2% near cold seeps.

Early diagenesis is one of the factors for foraminiferan isotopic composition. In buried foraminiferans, there is a concentration difference between the ions of the surrounding pore water in the sediments and foraminiferan shell ion, so that the calcareous shells of foraminiferans were dissolved and exchanged with the pore water by ion, and the foraminiferan shell isotopes were transformed to different degrees [57]. OLIVER et al analyzed foraminiferan carbon isotopes of strata presented on the northern Antarctic continent for 70 Ma, and found that for the carbon isotopes of the same foraminiferans, there is not much difference between the top and bottom of a borehole, and the carbon isotope loss is less than -0.1%, which suggests that diagenesis of foraminiferan isotopes mainly affects oxygen isotopes, while has less effect on carbon isotopes [58]. The sample depth in the seabed of this work is more than 8 m, and early diagenesis and mineralization are not very strong. We do not observe a significant foraminiferan recrystallization phenomenon using an optical microscope (Fig. 13), which indicates that early diagenesis has limited impact on the δ¹³C of benthic foraminiferans in the Dongsha and Shenhu sea areas.

Fig. 12 Micrographs of pyrite

Fig. 13 Micrographs of foraminiferans: A-Neogloboquadrina dutertrei; B-Globigerinoides ruber (red); C-Globigerinoides ruber; D-Uvigerina asperula; E-Uvigerina proboscidea; F-Uvigerina ampullaceal; G-Uvigerina schwageri

6 Conclusions

1) The carbon isotope values of the benthic foraminiferan Uvigerina spp. range from -0.212% to -0.021% and the oxygen isotope values of the planktonic foraminiferan Globigerinoides ruber range from -0.311% to -0.060% in the Shenhu and Dongsha sea areas, which indicates the foraminiferan isotope composition of the methane leakage environment.

2) Three cores, ZD2, ZD3 and ZS5, are taken from the bottom of holes aged 11 814, 26 616 and 64 090 a. Compared with the SPECMAP δ¹⁸O curve corresponding to the early MIS I, III and IV final period, the curve of foraminiferan oxygen isotope and the curve of global sea-level changes are consistent, indicating that the sea level of the South China Sea is significantly influenced by global climate changes since the late Quaternary, and the negative bias layer of the carbon isotope corresponds to that of MIS II (cold period).

3) The differences between the carbon and oxygen isotopes and between the cold period and warm period are larger, and the foraminiferan oxygen isotopic compositions are similar to those of the Blake Ridge, the Gulf of Mexico, Okhotsk Sea and Okinawa Trough sediments of the late Quaternary, indicating that during MIS II and IV, due to global sea-level fall, sea pressure decreases, gas hydrate is decomposed and released, and a large number of light carbon isotopes of methane are released into the sea DIC pool and recorded in the foraminiferan shell.

4) The maximum value of the carbon isotopes does not appear during the period of maximum primary productivity. Because the sampling depth is shallow, early diagenesis is short, and mineralization is not very strong, no significant foraminiferan recrystallization phenomenon is found under the optical microscope, demonstrating that the effects of primary productivity decline and early diagenesis of carbon isotopic composition is generally minor to the foraminiferans.

5) A pyrite layer and authigenic carbonate minerals developed in the negative bias layers of the foraminiferans confirm that methane released from hydrate decomposition is recorded in the late Quaternary sediments, further indicating that the light methane (methane hydrate release) isotope is the main cause of the negative bias of the foraminiferan carbon isotope.

6) Using foraminiferan carbon isotopes could thus accurately identify hydrate release events during the late Quaternary, and it could also help to reconstruct hydrate release geological history and provide a scientific method for hydrate exploration.

Acknowledgements

The authors gratefully acknowledge the National Basic Research Program of China (973) Project Team during the 2009-NGH Cruise to the South China Sea for their assistance in collecting the sediment cores. Thanks to Associate Professor LI Chao of Xiamen University for distinguishing foraminiferan species and Senior Technician YANG Tao for helping with the stable isotope analysis. Professor John HODGKISS is thanked for editing the English in this manuscript.

References

[1] BROOKS J M, FIELD N E, KENNICUTT M C. Observations of gas hydrates and in marine sediments, offshore northern California [J]. Marine Geology, 1991, 96(1): 103-109.

[2] BOROWSKI W S, PAULL C K, UESSLER III W. Global and local variations of interstitial sulfate gradients in deep water, continental margin sediments: Sensitivity to underling methane and gas hydrates [J]. Marine Geology, 1999, 159: 131-154.

[3] GINSBURG G D, SOLOVIEV V A, CRANSTON R E. Gas hydrate from the continental slope offshore Sakhalin Island, Okhotsk Sea [J]. Marine Geology, 1993, 13(1): 41-48.

[4] CHAPPELLAZ J, BLUNLER T, RAYNAUD D. Synchronous changes in atmospheric CH₄ and Greenland climate between 40 and 8 ka BP [J]. Nature, 1993, 366: 443-445.

[5] LEVIN L A, WHITCRAFT C R, MENDOZA G F, GONZALEZ J P, COWIE G L. Oxygen and organic matter thresholds for benthic faunal activity on the Pakistan Margin oxygen minimum zone (700- 1 100 m) [J]. Deep-Sea Research II, 2009, 56: 449-471.

[6] OSCAR S, LAURA S, MIGUEL A M. Effects of sample pre-treatment on the δ¹³C and δ¹⁸O values of living benthic foraminifera [J]. Chemical Geology, 2008, 257: 218-220.

[7] SENGUPTA B K, PLATON E, BERNHARD J M. Foraminiferal colonization of hydrocarbon-seep bacterial mats and underlying sediment, Gulf of Mexico slope [J]. Journal of Foraminifera Research, 1997, 20: 292-300.

[8] PAULL C K, UESSLER III W, BOROWSKI W S. Methane-rich plumes on the Carolina continental rise: Associations with gas hydrate [J]. Geology, 1995, 23(1): 89-92.

[9] CORLISS B H. Microhabitats of benthic foraminifera within deep-sea sediments [J]. Nature, 1985, 314: 435-438.

[10] BERNHARD J M. Potential symbionts in bathyal foraminifera [J]. Science, 2003, 299: 861.

[11] BREUER E, LAW G, WOULDS C, COWIE G L, PEPPE O, SCHWARTZ M, SHIMMIELD G, McKINLAY S. Sedimentary oxygen consumption and micro distribution at sites across the Arabian Sea Oxygen Minimum Zone (Pakistan Margin) [J]. Deep- Sea Research II, 2009, 56: 296-304.

[12] TORRES M E, MIX C, KINPORTS K. Is methane venting at the seafloor recorded by δ¹³C of benthic foraminiferan shells [J]. Paleoceanography, 2003, 18(3): 13.

[13] LEVIN L A, MENDOZA G F. Community structure and nutrition of deep methane seep macro benthos from the North Pacific (Aleutian) margin and the Gulf of Mexico (Florida Escarpment) [J]. Marine Ecology-An Evolutionary Perspective, 2007, 28: 131-151.

[14] DICKENS G R, O’NEIL J R, REA D K. Dissociation of oceanic methane hydrate as a cause of the carbon excursion at the end of the Paleoceane [J]. Paleoceanography, 1995, 10: 965-971.

[15] LAMOLDA M A, PERYT D, ION J. Planktonic foraminiferal bioevents in the Coniacian/Santonian boundary interval at Olazagutia, Navarra province, Spain [J]. Cretaceous Research, 2007, 28(1): 18-29.

[16] GALLEMI J, LOPEZ G, MARTINEZ R. Macrofauna of the Cantera de Margas section, Olazagutia: Coniacian/Santonian boundary, Navarro-Cantabrian Basin, northern Spain [J]. Cretaceous Research, 2007, 28(1): 5-17.

[17] LAMOLDA M A, PAUL R C. Carbon and oxygen stable isotopes across the Coniacian/Santonian boundary at Olazagutia, northern Spain [J]. Cretaceous Research, 2007, 28(1): 37-45.

[18] WU Neng-you, ZHANG Guang-xue, YANG Sheng-xiong. Preliminary discussion on natural gas hydrate reservoir system of Shenhu area, north slope of South China Sea [J]. Natural Gas Industry, 2007, 9: 1-7.

[19] CHEN Zhong, YANG Hua-ping. Diagenetic environment and implication of seep carbonate precipitations from the southwestern Dongsha Area, South China Sea [J]. Geo-science, 2008, 22(3): 382-389.

[20] ZHU You-hai, ZHANG Guang-xue, LU Zhen-quan. Accumulation and potential exploration of gas hydrate in South China Sea [J]. Petroleum, 2001, 22: 6-10.

[21] ZHANG Guang-xue, HUANG Yong-yang, ZHU You-hai. Prospect of gas hydrate resources in the South China Sea [J]. Marine Geology and Quaternary Geology, 2002, 22(1): 75-81.

[22] RATHBURN A E, PEREZ M E, MARTIN J B. Relationships between the distribution and stable isotopic composition of living benthic foraminifera and cold methane seep biogeochemistry in Monterey Bay, California [J]. Geochemistry Geophysics Geosystems, 2003, 4(12): 1106.

[23] BERNHARD J M, HABURA A, BOWAER S S. An endobiont- bearing allogromiid from the Santa Barbara Basin: Implications for early diversification of foraminifera [J]. Journal of Geophysical Research, 2006, 111: 1-10.

[24] LARKIN K E, GOODAY A J. Foraminiferal faunal responses to monsoon-driven changes in organic matter and oxygen availability at 140 m and 300 m water depth in the NE Arabian Sea [J]. Deep-Sea Research II, 2009, 56: 403-421.

[25] MARTIN R A, NESBITT E A, CAMPBELL K A. The effects of anaerobic methane oxidation on benthic foraminiferal assemblages and stable isotopes on the Hikurangi Margin of eastern New Zealand [J]. Marine Geology, 2010, 272: 270-284.

[26] SCHUMACHER S, JORISSEN F J, DISSARD D, LARKIN K E, GOODAY A J. Live (Rose Bengal stained) and dead benthic foraminifera from the oxygen minimum zone of the Pakistan continental margin (Arabian Sea) [J]. Marine Micropaleontology, 2007, 62: 45-73.

[27] FAIRBANKS R G, MORTLOCK R A, CHIU T C, CAO L, KAPLAN A, GUILDERSON T P, FAIRBANKS T W, BLOOM A L. Marine radiocarbon calibration curve spanning 100 000 to 50 000 years B.P. based on paired ²³⁰Th/²³⁴U/²³⁸U and ¹⁴C dates on pristine corals [J]. Quaternary Science Review, 2005, 24: 1781-1796.

[28] YEAL B S, JEAN C P. Carbon and oxygen isotopic composition of planktonic foraminifera from laboratory culture, plankton tows and recent sediment: implications for the reconstruction of paleoclimatic conditions and of the global carbon cycle [J]. Journal of Foraminiferan Research, 1985, 15(4): 302-320.

[29] QIAN Jian-xing. A study of Paleoceanography in the South China Sea during the Late Quaternary [M]. Beijing: Science Press, 1999: 20-21. (in Chinese)

[30] MARTINSON D G, IMBRIE T C, SHACHLETON N J. Age dating and the orbital theory of the ice ages: Development of a high- resolution 0 to 300000 year chronostratigraghy [J]. Quaternary Research, 1987, 27: 1-29.

[31] WU Shi-guo, QIN Yun-shan. The research of deepwater depositional system in the northern South China Sea [J]. Sedimentology, 2009, 27(5): 922-930.

[32] CHEN Fang. Benthic foraminifera and stable isotopic composition of gas hydrate-bearing sediments from Shenhu in the northern South China Sea [J]. Marine Geology and Quaternary Geology, 2010, 30(2): 1-8.

[33] LI Xue-jie. Preliminary study of gas hydrate in the northern slope of the South China Sea: Evidences from ODP [J]. South China Sea Geology Research, 2004, 16-28.

[34] MACAVOY S E, MORGAN E, CARNEY R S, MACKO S A. Chemoautotrophic production incorporated by heterotrophs in Gulf of Mexico hydrocarbon seeps: An examination of mobile benthic predators and seep residents [J]. Journal of Shellfish Research, 2008, 27: 153-161.

[35] MEYER K M, YUB M, JOST A B, KELLEY B M, PAYNE J L. δ¹³C evidence that high primary productivity delayed recovery from end-Permian mass extinction [J]. Earth and Planetary Science Letters, 2011, 302: 378- 384.

[36] CHEN Fang, SU Xin, LU Hong-feng. Carbon stable isotopic composition of benthic foraminifera from the north of South China Sea: Indicator of methane-rich environment [J]. Marine Geology and Quaternary Geology, 2007, 27(4): 1-7.

[37] GOODAY A J, NOMAKI H, KITAZATO H. Modern deep-sea benthic foraminifera: A brief review of their biodiversity and trophic diversity [C]// AUSTIN W E N, JAMES R H. Biogeochemical Controls on Palaeoceanographic Environmental Proxies. London: Special Publications, 2009, 303: 97-119.

[38] COWIE G L, LEVIN L A. Benthic biological and biogeochemical patterns and processes across an oxygen minimum zone (Pakistan Margin, NE Arabian Sea) [J]. Deep-Sea Research II, 2009, 56: 261- 270.

[39] SHACKLETON N J, OPDYKE N D. Oxygen isotope and palaeomagnetic stratigraphy of equatorial Pacific core V28-238: Oxygen isotope temperature and ice volumes on a 100000 year and 1000000 year scale [J]. Quaternary Research, 1973, 3: 39-55.

[40] WANG Pin-xian. Western Pacific in glacial cycles: Seasonality in marginal seas and variabilities of warn pool [J]. Science China Series D, 1998, 28(1): 1-6.

[41] CHEN Mu-hong, TU Xin, ZHENG Feng. Relations between sedimentary sequence and paleoclimatic change during last 200 ka in South China Sea [J]. Chinese Science Bulletin, 2000, 45(5): 542- 548.

[42] THURBER A R, KROGOR K, NEIRA C, WIKLUND H, LEVIN L A. Stable isotope signatures and methane use by New Zealand cold seep benthos [J]. Marine Geology, 2010, 272: 260-269.

[43] CHAPPELL W S, SHACKLETON N J. Oxygen isotopes and sea level [J]. Nature, 1986, 324: 137-140.

[44] ROLLION C, EREZ J, ZILLBERMAN T. Intra-shell oxygen isotope ratios in the benthic foraminifera genus Amphistegina and the influence of seawater carbonate chemistry and temperature on this ratio [J]. Geochimica et Cosmochimica Acta, 2008, 72: 6006-6014.

[45] LEVIN L A. Ecology of cold seep sediments: Interactions of fauna with flow, chemistry, and microbes [J]. Oceanography Marine Biology, 2005, 43: 1-46.

[46] WOULDS C, COWIE G L. Sedimentary pigments on the Pakistan margin: Controlling factors and organic matter dynamics [J]. Deep-Sea Research II, 2009, 56: 347-357.

[47] ROBINSON C A, BERNHARD J M, LEVIN L A. Surficial hydrocarbon seep infauna from the Blake Ridge (Atlantic Ocean, 2150 m) and the Gulf of Mexico (690-2 240 m) [J]. Marine Geology, 2004, 25(4): 313-336.

[48] GOODAY A J, LEVIN L A, SILVA A. Faunal responses to oxygen gradients on the Pakistan margin: A comparison of foraminiferans, macrofauna and megafauna [J]. Deep-Sea Research II, 2009, 56: 488-502.

[49] SILVA A, GOODAY A J. Large organic-walled Protista (Gromia) in the Arabian Sea: Density, diversity, distribution and ecology [J]. Deep-Sea Research II, 2009, 56: 422-433.

[50] BOROWSKI W S, HOEHLER T M, ALPERIN M J. Significance of anaerobic methane oxidation in methane rich sediments overlying the Blake Ridge gas hydrate [C]// Proceedings of the Ocean Drilling Program, Scientific Results. College Station, Texas: Ocean Drilling Program, 2000: 179-191.

[51] JORGENSEN B B, WEBER A, ZOPFI J. Sulfate reduction and anaerobic oxidation in Black Sea sediments [J]. Deep-sea Research PartⅠ, 2001, 48: 2097-2120.

[52] ROLLION C, MANGIN D, CHAMPENOIS M. Development and applications of oxygen and carbon isotopic measurements of biogenic carbonates by ion microprobe [J]. Geostandards and Geoanalytical Research, 2007, 31: 39-50.

[53] DATTAGUPTA S, ARTHUR M A, FISHER C R. Modification of sediment geochemistry by the hydrocarbon seep tubeworm Lamellibrachia luymesi: A combined empirical and modeling approach [J]. Geochimica et Cosmochimica Acta, 2008, 72: 2298-2315.

[54] LU Miao-an, MA Zhi-jie, CHEN Mu-hong. Rapid carbon isotope negative excursion events during the penultimate deglaciation in western Pacific marginal sea areas and their origins [J]. Quaternary Science, 2002, 22(4): 349-358.

[55] HUANG Yong-yang, SUESS E, WU Neng-you. Methane and natural gas hydrate geology in the northern of South China Sea [M]. Beijing: Geology Press, 2008, 67-71. (in Chinese)

[56] HAYWARD B W, GRENFELL H R, SABAA A T, KAY J. Using foraminiferal faunas as proxies for low tide level in the estimation of Holocene tectonic subsidence, New Zealand [J]. Marine Micropaleontology, 2010, 76: 23-36.

[57] MILLO C, SARNTHEIN M, ERLENKHEUSER H. Methane- induced early diagenesis of foraminiferal testsin the southwestern Greenland Sea [J]. Marine Micropaleontology, 2005, 58: 1-12.

[58] OLIVER F, GERHARD S, HELMUT E. Stable isotope composition of Late Cretaceous benthic foraminifera from the southern South Atlantic: Biological and environmental effects [J]. Marine Micropaleontology, 2006, 58: 135-157.

(Edited by HE Yun-bin)

Foundation item: Project(40976035) supported by the National Natural Science Foundation of China; Project(2009CB219501) supported by the National Basic Research Program of China; Project(908-ZC-I-07) supported by the Special Program of Comprehensive Survey and Assessment Offshore China Sea

Received date: 2011-04-12; Accepted date: 2011-07-22

Corresponding author: CAO Chao, PhD; Tel: +86-13646029904; E-mail: caochao123.student@sina.com