1 Introduction

Methane seepage is common along worldwide continental boundaries and is usually associated with natural gas hydrates (Dickens, 2001; Bayon et al., 2011; Peketi et al., 2012; Suess, 2014). Due to external environmental factors (such as temperature rise and pressure decrease), natural gas hydrates are unstable and readily dissociate. This releases large amounts of methane along the sea floor or into the atmosphere, having a significant impact on both ocean and terrestrial environments, and has been observed in many geological surveys (Dickens, 2001; Mienert et al., 2005; Them et al., 2018). Methane seepages are associated with a shallow sulfate-methane transition zone (SMTZ) in the sediments, an important redox boundary hosting microbial consortia performing anaerobic oxidation of methane (AOM) (Suess, 2014; Feng et al., 2018b); this process leaves a variety of geochemical signals in the sediment, which provide an opportunity and a window for us to study past methane seepage (Bayon et al., 2011; Chen et al., 2016; Feng et al., 2018b). Therefore, it is necessary and important to identify methane seepage activity signals correctly by means of reliable geochemical proxies to accurately locate the position of modern and paleo-SMTZs.

Many studies have focused on interpreting methane seeping activity signals based on carbon, sulfur isotopic and trace elemental systematics and the enrichment and diffusion mechanisms of trace elements (Chen et al., 2016, 2019; Li et al., 2018). Those researches mainly focused on the study of inorganic geochemical characteristics of authigenic carbonate (e.g., Bayon et al., 2011; Feng et al., 2018b; Wei et al., 2022), authigenic pyrite (e.g., Borowski et al., 2013; Li et al., 2018) and bulk sediment inorganic (e.g., Feng et al., 2018b; Wei et al., 2019) as well as organic geochemical characteristics (e.g., biomarker, Aloisi et al., 2002; Knittel et al., 2005). Studies have shown that due to methane seepage, the enrichment of large amounts of authigenic carbonate and pyrite in SMTZ sediments results in many signals that differ from the sediments from the normal marine sedimentary environments, such as the depletion of δ¹³C and enrichment of δ¹⁸O in authigenic carbonate (Chen et al., 2019), enrichment of ³⁴S in pyrite (Borowski et al., 2013; Lin et al., 2016), increase of total sulfur (TS) levels and the total organic carbon (TOC) (TS/TOC) ratio in sediments, and the enrichment of special trace elements (Mo) (Chen et al., 2016; Miao et al., 2021a). Also, anaerobic methanotrophic archaea (ANME, which comprises three subpopulations: ANME-1, ANME-2, ANME-3) and sulfate-reducing bacteria (SRB, comprising two sub-populations: Desuilfosarcina and Desulfococcus) are the dominant species in methane-rich environments and control the methane biogeochemical cycle in sediments (Boetius et al., 2000; Peckmann and Thiel, 2004). Moreover, the biomarkers they produced have significantly negative δ¹³C values (Zhang et al., 2002), which differ significantly from the δ¹³C levels observed in general marine sediment lipids. Therefore, they are viewed as effective indicators of methane seepage activities.

The degradation of organic matter begins with early diagenesis of the entire sediment (Dale et al., 2019), but few works on δ¹⁵N_TN and δ¹³C_TOC characteristics in methane seepage fluid and its isotopic compositions in sediments have been reported, and the comprehensive characteristics of δ¹⁵N_TN and δ¹³C_TOC in sediments in which methane seepage activity occurs have rarely been demonstrated (Hu et al., 2020; Yang et al., 2020). A recent study reported the contribution of deep methane fluid on the sediment organic carbon content, which had a significant impact on its isotopic composition (Coffin et al., 2015). Additional experiments suggested that methane oxidizing archaea directly absorb CH₄ as the sources of carbon and its symbiotic sulfate-reducing bacteria are autotrophic, and confirmed that electron transfer between the two was conducted in a redox fashion (Wegener et al., 2008). Around the same time, studies in the CH₄ fluid active zone of the Eel River Basin (California, USA) found the nitrogen-fixing effect of anaerobic methanotrophic archaea in deep-sea sediments (Dekas et al., 2009). This provided a new opportunity to study carbon and nitrogen isotopic compositions during methane seepage and the balance of global carbon and nitrogen cycles. In this paper, we analyzed sediments obtained from the 'Haima seep' in Qiongdongnan Basin, margin of the South China Sea in detail. The aim is to reveal the coupling response of carbon and nitrogen isotopes in sediments under the background of methane seepage. This work is important to reveal the historical evolution of natural gas hydrate dissociation and its influence on the deep-sea carbon and nitrogen pool.

2 Regional Setting

The South China Sea is the largest marginal low latitude sea and contains rich reserves of oil and gas. The Qiongdongnan Basin is a key exploration area of the northern South China Sea. Recently, a series of comprehensive geological and geophysical investigations have shown some geophysical evidences and geochemical markers of many methane fluid activities, such as bottom simulating reflectors (BSR), mud volcanoes, mud diapirs, and gas chimneys (Wang et al., 2008; Hui et al., 2016). From 2015 – 2018, scientists obtained large amounts of authigenic carbonate deposits from the Qiongdongnan Basin, which proved that methane seepage had occurred in this area for a long period (Liang et al., 2017). In 2018, the Guangzhou Marine Geological Survey conducted the 5th China gas hydrate drilling expedition (GMGS5) in this basin and discovered massive gas hydrates (Wei et al., 2019; Ye et al., 2019). The study found those natural gases were primarily biological in origin and came from organic fermentation (Feng et al., 2018a). Therefore, the Qiongdongnan Basin is an ideal laboratory for the study of natural gases.

3 Materials and Methods

Between April and May in 2019, the Guangzhou Marine Geological Survey obtained a series of survey samples along the northern slope of the South China Sea that included the Q6 core. The core Q6 (water depth: 1400 m; length: 282 cm long) was recovered from the edge sedimentary area of the 'Haima seep' in the Qiongdongnan Basin (Fig.1). Then a series of geochemical tests were conducted, including major element compositions, and contents of TOC and TS (Miao et al., 2021a, 2021b, 2022).

The carbon isotope (organic carbon) and nitrogen isotope analyses were conducted by using an elemental analyzer (EA) and a stable isotopic mass spectrometer (IRMS) in the Key Laboratory of Submarine Geosciences and Prospecting Techniques. The ¹³C and ¹⁵N isotope concentration were measured separately in a water trap, where carbon dioxide was removed from the sample flow by the NaOH adsorbent. Based on the carbon and nitrogen levels in the samples, additional samples were prepared. At first, approximately 2 – 3 mg or 25 – 30 mg of sediment samples, for carbon and nitrogen analyses, respectively, were weighed and tightly wrapped in tin cups. International standard materials IAEA600, USGS64, USGS40, LA-R006, and EMA-B2153 were used as reference working standards. An analysis sequence of reference substance-sample to be tested reference substance was performed, and six samples were analyzed between two groups of reference substances. The test results were standardized by using a multi-point correction method, the test accuracies were within ±0.08‰ for δ¹³C and 0.1‰ for δ¹⁵N. Finally, the results were converted to δ values of international standard isotopes.

4 Results and Discussion 4.1 Evidence for Methane Release

AOM releases H₂S which reacts with an active iron component to form pyrite, thus generating geochemical anomalies in the sediment (Xie et al., 2013). The enrichment of authigenic pyrite increases the total sulfur (TS) concent in the sediment and the TS/TOC ratio (Liang et al., 2017; Feng et al., 2018b; Miao et al., 2021a). Previous works from our laboratory reported the methane seepage events based on TS levels and TS/TOC ratios of core sediments (Miao et al., 2021a). Those results showed unusual increases in TS levels in the 90 – 124 cm, 144 – 162 cm, and 254 – 282 cm layers of Q6 (Figs.2 – 3), accompanied by the increase in the TS/TOC ratios (all > 0.36) (Miao et al., 2021a, 2021b, 2022). This was due to high hydrogen sulfide production by AOM linked to the gas hydrate decomposition, which led to enhanced pyrite accumulation at the SMTZ (Boetius et al., 2000; Peketi et al., 2012). Therefore, we believe that TS levels and TS/TOC ratios are related to methane seepage activities, and the strata from 90 – 124 cm, 144 – 162 cm, and 254 – 282 cm correspond to the active strata of ancient methane seepage (Miao et al., 2021a).

4.2 Effects of Methane Seepage on Organic Carbon and Nitrogen Isotopes in Sediments

Marine sediments represent one of the major worldwide sinks for carbon and nitrogen. Approximately (1.3 – 2.3) × 10¹⁴ g of carbon and nitrogen enters marine sediments each year, and more than 90% of that gets deposited on continental shelfs and slopes (Brooks et al., 1991). These organic carbon and nitrogen deposits are important carriers for the study of carbon and nitrogen 'sources-sink' processes and paleoenvironmental evolution (Hu et al., 2013). However, the methane seepage from hydrate dissociation and the development of the corresponding palaeontological bacterial community will certainly affect the composition of organic carbon and nitrogen isotopes in the sediments and change their evolutionary trajectory (Joye et al., 2004; Dekas et al., 2009, 2014; Cao et al., 2010; Yu et al., 2013; Feng et al., 2018a).

4.2.1 δ¹⁵N_TN and δ¹³C_TOC

The vertical variations in the TS content and TS/TOC ratio in sediments of core Q6 were reported in a previous work (Miao et al., 2021a, 2021b, 2022). Those results showed that TS levels and the TS/TOC ratio increased in the 90 – 124 cm, 144 – 162 cm and 254 – 282 cm strata (Miao et al., 2021a) (Figs.2 – 3). In contrast, our results show that δ¹⁵N_TN and δ¹³C_TOC have similar vertical variation trends, with low levels in the 90 – 124 cm, 144 – 162 cm, and 254 – 282 cm horizons, and the lowest δ¹⁵N_TN and δ¹³C_TOC values all occurred in those three horizons. Values of δ¹³C_TOC ranged from −23.6‰ to −20.6‰ PDB; δ¹⁵N_TN of the same sedimentary samples ranged from 1.8‰ to 5.3‰. Compared with the sediment intervals unaffected by AOM (δ¹³C_TOC ≈ 21.0‰, δ¹⁵N_TN ≈ 4.5‰), δ¹³C_TOC and δ¹⁵N_TN displayed different degrees of negative excursions, with maximums of 2.6‰ and 2.5‰, respectively (Fig.3).

4.2.2 Effect of methane seepage on nitrogen isotopic compositions

Stable nitrogen isotope signals in marine sediments contain key biogeochemical information and are important in the identification of marine nitrogen sources and to understand the marine nitrogen cycle (Zheng et al., 2015). The nitrogen isotope signals in marine sediments can be used to trace the biogeochemical cycles of marine systems and the geological evolution of the marine environment. The nitrogen isotopic composition of marine sediments depends on the isotopic composition of bioavailable nitrogen in seawater, the fractionation that occurs during the assimilation and a series of subsequent transformations in water bodies and sediments (Robinson et al., 2012). Therefore, when we discuss the coupling relationship between the ocean nitrogen cycle, environmental change and deposition, we need to consider many factors such as temporal and spatial transformations of the ocean environment, the ocean nitrogen cycle, and material sources.

The largest nitrogen reservoir on earth occurs in the atmosphere (δ¹⁵N = 0); the nitrogen fixation introduces N₂ into the marine biogeochemical cycle and provides nitrogen for marine organisms. The oceanic nitrogen cycle is complex and variable, and involves a series of microbe-mediated biological processes, such as nitrogen fixation, digestion and denitrification, and assimilation (Zhu et al., 2020). Conversion of nitrogen from one form to another inevitably leads to the isotopic fractionation that changes the bioavailable nitrogen isotopes in seawater and affects the composition of nitrogen isotopes in marine sediments (Stüeken et al., 2016). Studies have shown that nitrate assimilation, nitrogen fixation, and denitrification generally keep in equilibrium for modern oxidized oceans (Junium et al., 2018). Nitrogen isotopes in seawater are relatively stable with little change, and δ¹⁵N levels in South China Sea water remain relatively steady at 4‰ – 5‰ (Yang et al., 2017). This implies the nitrogen isotope composition of sediments will not drastically change because of the change of seawater nitrogen isotope compositions in the normal marine sedimentary environment of the South China Sea.

During early diagenesis, selective degradation of organic matter and the dominant metabolism of primary producers both alter the nitrogen isotopic compositions of sediments. The results showed that nitrogen-containing compounds had the lowest resistance to degradation, so they initially reduced into an inorganic form in the overlying water and selectively released nitrogen rich in ¹⁴N, which resulted in the heavy δ¹⁵N isotopic compositions in the sediments (Meyers, 1997; Freudenthal et al., 2001). However, δ¹⁵N levels in the 90 – 124 cm, 144 – 162 cm, and 254 – 282 cm strata were significantly lighter and inconsistent with the above analysis. By studying the nitrogen isotopes of sediments from the cores close to the study area, Jia and Li (2011) discovered the variation of nitrogen isotopes in the South China Sea had not changed much in the past 25 kyr, and the variation range was less than 1‰. A similar trend was also observed in the normal sediments of Q6 (0 – 90 cm and 162 – 252 cm) (Fig.3). This indicated that in a normal marine sedimentary environment, sulfatereducing bacteria dominated by organic sulfate reduction in the SMTZ have a certain potential role for nitrogen fixation but little influence on the overall δ¹⁵N composition. In addition, Kienast (2000) found little difference in δ¹⁵N values between luff layers and the topmost sediments in the South China Sea, which indicated a negligible diagenetic overprint of nitrogen stable isotope compositions at the sediment-water interface. Therefore, the differential degradation of organic matter and the change of primary productivity were not the main factors that led to lower levels of δ¹⁵N in the horizon.

Similarly, clay minerals also capture NH₄⁺ in seawater and bury it in sediments. During this process, the fractionation of δ¹⁵N is small and can be ignored (Freudenthal et al., 2001). At the same time, organic nitrogen levels adsorbed by clay minerals are low in continental margin sediments (Freudenthal et al., 2001). Moreover, according to the analysis of the grain size characteristics of core Q6 sediments, we found that in the negative bias horizon of δ¹⁵N and δ¹³C in the sediments, the median grain size of sediments increased significantly, and the clay content was low (Fig.4). Therefore, we also believe that the nitrogen isotope fractionation associated with clay minerals have little influence on the organic nitrogen isotope levels in sediments.

In addition to the authigenic nitrogen of marine organisms, the input of terrigenous organic nitrogen also affects the nitrogen isotopic composition of marine sediments. Levels of δ¹⁵N in terrestrial plants are generally low due to the lack of ¹⁵N in the atmospheric nitrogen utilized during photosynthesis by nitrogen-fixing bacteria, so the δ¹⁵N of marine sediments decreases as the terrestrial inputs increase (Meyers, 1997). However, δ¹⁵N levels of sediments from the continental shelf to the deep sea in the South China Sea remain relatively constant, ranging from 4.2‰ to 6.0‰ (Gaye et al., 2009), and the inputs of terrestrial organic nitrogen are not enough to cause such low levels of δ¹⁵N in the three layers above (the lowest δ¹⁵N was 2.0‰). Meanwhile, the concentrations of elements re-presenting terrigenous sources (e.g., Al, Ti, Fe) did not show an obvious increase (Fig.4) and proved that terrigenous organic nitrogen did not significantly increase at the moments.

In the areas where denitrification was absent but nitrogen fixation was strong, the variation of nitrogen isotopes on a geological timescale occurs primarily through the changes in nitrogen fixation pathways (Meckler et al., 2007). Studies in the Carriaco Basin of the Atlantic Ocean showed that nitrogen isotopes during the interglacial period were significantly lower than those during the glacial period, and sediment δ¹⁵N values decreased by 3‰ from the Last Glacial Maximum to the Holocene (Meckler et al., 2007). The South China Sea is also a typical area of strong nitrogen fixation (Kienast, 2000). The significantly low δ¹⁵N_TN levels in core Q6 sediments were generally considered as the sign of nitrogen fixation. However, according to the chronological framework established by the AMS-¹⁴C in Q6 (Miao et al., 2021a), we found the period with enhanced nitrogen fixation did not overlap with the interglacial period. Previous studies have also found that δ¹⁵N_TN in South China Sea sediments show little variation during glacial-interglacial cycles (Kienast, 2000), which is considered to be the large fractionation signal of denitrification that suppressed nitrogen isotope changes caused by nitrogen fixation or benthic denitrification. Therefore, we believe the selective degradation of organic matter, the change of primary productivity during early diagenesis, terrigenous input, and glacial-interglacial nitrogen fixation are not responsible for low levels of nitrogen isotopes in core Q6.

Interestingly, low δ¹⁵N_TN levels in the core Q6 are always accompanied by methane seepage activities. Therefore, we speculate that decreases of δ¹⁵N in Q6 may be related to methane seepage activities. Recent isotopic labeling experiments showed that ANME-2 has good nitrogen fixation ability. Dekas et al. (2009, 2014) detected the gene pattern of nifH in the active zones of methane seepage fluid, which proved that ANME-2 was a nitrogen-fixing organism. When nitrogen fixation occurs, ANME or AOM bacterial aggregates preferentially incorporate ¹⁴N rather than ¹⁵N, which leads to a light δ¹⁵N value in organic matters (Hu et al., 2020). In addition, many signs of the ANME activity were found in the active methane seepage layer of the 'Haima seep' deposition area, and the main microorganisms involved in CH₄ metabolism were ANME-1, ANME-2ab, and ANME-2c (Niu et al., 2017). Therefore, we believe that methane seepage can change the nitrogen isotopic composition of bulk sedimentary organic matters.

4.2.3 Effect of methane seepage on organic carbon isotopic composition

The organic carbon isotopes in sediments were not affected by the sediment grain size and are often used to indicate the potential provenance distribution and environmental changes of organic matter (Hu et al., 2013). In general, δ¹³C values in sediments are primarily controlled by the photosynthesis processes and the isotopic composition of carbon sources. Different photosynthetic processes affix organic carbon isotopic (δ¹³C_TOC) levels in different ways. The δ¹³C values of terrestrial C₃ plants and C₄ plants are approximately −27‰ PDB and −14‰ PDB, respectively. The δ¹³C levels in marine plankton generally range from −19‰ – −22‰ PDB (average −20.5‰ PDB) (Lamb et al., 2006; Gaye et al., 2009; Hu et al., 2013). Therefore, the isotopic difference between terrestrial C₃ plants and marine plankton is 5‰ – 7‰ PDB and can be used to distinguish the sources of the organic matter very well (Meyers, 1997). Fig.3 (δ¹³C, −20.6‰ – −23.6‰) shows that the sediment organic matter in Q6 comes primarily from marine plankton, and a small amount comes from terrigenous organic matter. Interestingly, however, in the 90 – 124 cm, 144 – 162 cm, and 254 – 282 cm segments (Fig.3), we found obvious negative excursions of carbon isotopes (up to 2.5‰). This may indicate the increase of terrigenous organic matter, but as mentioned above, no obvious signal of the enhanced terrigenous input was found during this period (Fig.4). At the same time, previous research has shown that the terrigenous input of organic matter in sediments far from the continental shelf was less than 14%, and the organic carbon isotope levels in surface sediments of the Qiongdongnan Basin were approximately −20.8‰ (Chen et al., 2012). Therefore, we believe that the decrease of organic carbon isotopes in these horizons was not related to the input of terrigenous organic matter.

During early diagenesis, the δ¹³C of organic matter in seafloor sediments systematically changes with the depth and the δ¹³C level usually decreases after the sediments are buried (Meyers, 1997; Freudenthal et al., 2001). This is mainly due to the preferential degradation of the organic components with relatively heavy isotopic compositions (protein and carbohydrates), resulting in the depletion of δ¹³C (up to 2‰). In core Q6, the maximum depletion of δ¹³C in TOC was 2.6‰; since it surpassed 2.0‰, we concluded the selective degradation of organic matter during early diagenesis was not responsible for the depletion of δ¹³C in Q6.

Recent studies have shown that the presence of anaerobic methanotrophic archaea and sulfate-reducing bacteria affect the carbon isotopic composition of organic carbon in the sedimentary environment involved the methane-rich fluids (Peckmann and Thiel, 2000). First, organic carbon isotopes increase at some methane seep sites, such as the Hydrate Ridge, Dongsha and Shenhu of the South China Sea (Yu et al., 2006; Cao et al., 2010; Yang et al., 2020). Strengthening the AOM resulted in more efficient microbial utilization of organic matter to produce methane, and also resulted in the preferential utilization of organic matter rich in ¹²C as well as the increase of δ¹³C of residual organic matters in sediments (Yang et al., 2020). This does not fully explain the depletion of organic carbon isotopes in Q6. Hoehler et al. (1994) proposed that AOM was the reverse reaction of methane production, in which the carbon derived from methane was oxidized during a biochemical reaction, mediated by methane-oxidizing archaea at first and sulfate reducing bacteria subsequently, and finally the carbon was enriched in the membrane lipids of archaea and bacteria. The low δ¹³C of microbial membrane lipids was due to the relatively low δ¹³C of leaking methane and the fractionation of carbon isotopes was caused by microbial interaction during AOM. Therefore, low δ¹³C levels of sediment organic carbon are the signals of methane seepage activities. Recent studies have found the obvious negative excursions of organic carbon isotopes in the northeastern sediments of the South China Sea and the Shenhu (Yu et al., 2013; Xiong et al., 2020). Therefore, we believe the negative excursions of carbon isotopes in Q6 was primarily due to methane seepage activities.

4.3 Coupling of Negative Excursions to δ¹⁵N_TN and δ¹³C_TOC

Anaerobic methanotrophic archaea and sulfate-reducing bacteria living in the SMTZ control the entire methane biogeochemical cycle in sediments. By directly absorbing carbon from CH₄ sources, they caused a negative bias in organic carbon isotopes in the sediments (Yu et al., 2013; Xiong et al., 2020). Also, during nitrogen-fixation of anaerobic methanotrophic archaea, the ANME bacterial assemblages preferentially utilized ¹⁴N, which resulted in significant negative excursions of nitrogen isotopes in the sediments (Hu et al., 2020; Fig.5). Therefore, the analysis of the sources and production pathways of TOC and TN strongly suggested the coupling negative excursions of carbon and nitrogen isotopes in Q6 were closely related to methane seepage activities. More interestingly, previous studies reported that the contents of TOC and TN, especially TN, decreased during methane seepage activities (Miao et al., 2021a). This may be due to the decrease in organic matter contents due to the change in sediment particle size compositions, or it may be due to the continuation of OSR in SMTZ (Jørgensen et al., 2019). In this case, we think it is more related to the microbial nitrogen fixation. Indeed, Dekas et al. (2009) found that the nitrogen fixation of archaea was readily conducted. In order to compensate for the energy load of nitrogen fixation, the growth rate of archaea would decrease as it maintained the methane oxidation while fixing nitrogen. This conclusion is speculative and needs additional exploration to confirm.

5 Conclusions

The core Q6 was taken from the margin of the 'Haima seep' sedimentary area in the Qiongdongnan Basin, South China Sea. We previously identified three methane release events in core Q6 based on TS levels and TS/TOC ratios and identified specific horizons of methane seepages. The organic carbon isotopes (δ¹³C_TOC) and nitrogen isotopes (δ¹⁵N_TN) were determined and simultaneous negative excursions of δ¹³C_TOC and δ¹⁵N_TN were found in the horizons of methane seepages, with the maximum negative excursions of 2.6‰ and 2.5‰, respectively. We believe these negative excursions are due to the combined effect of ANME and SRB in SMTZ. This indicates the δ¹³C_TOC and δ¹⁵N_TN levels in sediments correspond well to and effectively indicate the methane seepage activities.

Acknowledgements

We would like to thank the staff of Haiyang 6 (the Guangzhou Marine Geological Survey) for their efforts. We would also like to thank Prof. D. Feng (Shanghai Ocean University) and Prof. Y. Hu (Shanghai Ocean University) for their help during the writing process. The study was supported by the National Key R & D Program of China (No. 2017Y FC0306703) and the Key Special Project for Introduced Talents Team of Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou) (No. GML2019 ZD0201).