Impact of water depth on the distribution of iGDGTs in the surface sediments from the northern South China Sea: applicability of TEX₈₆ in marginal seas

Abstract

The paleothermometer on the base of isoprenoid glycerol dialkyl glycerol tetraethers (iGDGTs) has been widely applied to various marine settings to reconstruct past sea surface temperatures (SSTs). However, it remains uncertain how well this proxy reconstructs SSTs in marginal seas. In this study, we analyze the environmental factors governing distribution of iGDGTs in surface sediments to assess the applicability of paleothermometer in the South China Sea (SCS). Individual iGDGT concentrations increase gradually eastwards. Redundancy analysis based on the relative abundance of an individual iGDGT compound and environmental parameters suggests that water depth is the most influential factor to the distribution of iGDGTs, because thaumarchaeota communities are water-depth dependent. Interestingly, the SST difference (ΔT) between derived temperature and remote-sensing SST is less than 1°C in sediments with water depth>200 m, indicating that was the robust proxy to trace the paleo-SST in the region if water depth is greater than 200 m.

Keywords

iGDGTs / distribution / South China Sea (SCS) / sea surface temperature / water depth

Introduction

Phylogenomic analysis showed that Crenarchaeota and Euryarchaeota are two kingdoms, Crenarchaeota (Group I) and Euryarchaeota (Group II), within the domain of the Archaea. Recently, the Group I Archaea and their close relatives from terrestrial environments were reclassified as a separate phylum. Thaumarchaeota are likely the most ancient Archaea. Many studies confirm that Thaumarchaeota are among the most abundant organisms in the ocean, approximately 40% of all cells in the bathypelagic (e.g., Karner et al., 2001), whereas the Group II Euryarchaeota, in general, are less abundant and shown near the surface ocean (Pernthaler et al., 2002; Herndl et al., 2005).

Glycerol dialkyl glycerol tetraethers (GDGTs) are characteristic membrane lipids of Archaea and bacteria. GDGTs have been reported in almost every type of environment, such as soils (e.g., Weijers et al., 2007; Yang et al., 2014), lacustrine water columns and sediments (e.g. Powers et al., 2010; Weber et al., 2015), and marine water columns and sediments (e.g., Schouten et al., 2002; Basse et al., 2014). GDGTs include mainly two types: isoprenoidal GDGTs (iGDGTs) and branched GDGTs (bGDGTs) (Fig. 1). Isoprenoidal GDGTs are produced from two groups of Archaea: Thaumarchaeota (formerly marine group I Crenarchaeota) and Euryarchaeota. Cyclopentane moieties increase with growth temperature (De Rosa et al., 1980; Gliozzi et al., 1983). The TEX₈₆ (TetraEther index of tetraethers consisting of 86 carbon atoms) was thus constructed to trace the paleo-sea surface temperature, SST (Schouten et al., 2002), which has been recalibrated to improve the accuracy of reconstructing the paleo-SSTs in different environments (e.g., Sluijs et al., 2006; Kim et al., 2008; Liu et al., 2009; Kim et al., 2010). Kim et al. (2010) concluded that the {\mathrm{TEX}}_{\mathrm{86}}^H and {\mathrm{TEX}}_{\mathrm{86}}^L are the most promising proxies.

Branched glycerol dibiphytanyl glycerol tetraethers (bGDGTs) were reported to be produced mainly from the terrestrial bacteria in soil or peat environments although some in situ bGDGTs may exist in marine and lacustrine environments (Peterse et al., 2009; Sinninghe Damsté et al., 2009; Lü et al., 2014; Zell et al., 2015). Hopmans et al. (2004) proposed a BIT index (Branched Isoprenoid Tetraether Index) that can be used to estimate the relative contribution of terrestrial and aquatic organic matter to marine sediments (i.e., soil- and/or river-derived organic matter input). BIT value>0.8 likely corresponds to a strong soil OM footprint while BIT value<0.2 may be for a relatively low soil OM input (Walsh et al., 2008; Zhang et al., 2012).

The South China Sea (SCS) is the largest marginal sea in Southeast Asia with annual SST varying from 25°C to 29°C. Its climatic variations in the atmosphere and in the upper ocean are primarily dominated by the East Asian monsoon (Liu et al., 2002). The significant correlation between the sea surface temperature anomaly in summer and the equatorial sea surface temperature anomalies in the preceding winter suggest that the sea surface temperature anomaly could be considered as an indicator for the monsoon variations and El Niño and southern oscillation (ENSO) (Ose et al., 1997). Therefore, accurate SST reconstruction appears to be critical to study the inter-feedback between climate change and oceanic physical and chemical parameters. Study on SST reconstruction with GDGTs was conducted in the SCS during the last several years and results show that the TEX₈₆-derived temperature, used to trace the subsurface temperature or winter SST, was lower than the remote-sensing SST(Wei et al., 2011; Jia et al., 2012; Zhou et al., 2014). However, Shintani et al. (2011) argued that the TEX₈₆ -derived temperature corresponded to the warm season temperature in the open northern SCS. The different findings by those researchers may be due to complex sources of sediments and insufficient knowledge about Thaumarchaeota. Up to date application of TEX₈₆ to marginal seas is still limited, likely due to that (i) terrestrial input makes sources of GDGTs complicated, and (ii) spatial variability of environmental factors (such as nutrient levels, salinity, pH, and water depth) is significant.

In this study, we analyzed the iGDGTs in the surface sediments from the coastal northern SCS, and performed the redundancy analysis (RDA) to distinguish the controlling factors that influence spatial distribution of iGDGTs and further assessed the applicability of {\mathrm{TEX}}_{\mathrm{86}}^Has a temperature proxy in this region.

Material and methods

Study area and sampling sites

The South China Sea is the largest marginal sea in Southeast Asia, which connects to the Pacific Ocean through the Luzon Strait. The East Asian monsoon plays an important role in the SCS currents and the water temperature (Shen and Lau 1995; Hu et al., 2000; Qu, 2001). In winter the SCS is dominated by the strong northeast monsoon, with an intense coastal current flowing from northeast to southwest along the coast of Guangdong and the east coast of Hainan Island. The strong northeast monsoon transports cold water from the north and causes a distinctly lower water temperature. In summer, the wind becomes weak and reverses direction to southwest (Hu et al., 2000).

A total of 28 surface sediments were collected from coastal to deep water (2‒505 m) in the north SCS (Fig. 2) and then refrigerated at ‒20°C prior to being further analyzed.

Lipid analysis

After frozen and ground, about 15 g of sediments were extracted ultra-sonically with DCM/MeOH (9:1, v/v) for 15 min and were extracted at least 3 times. Supernatants from each extraction were collected and combined after centrifugation. The total lipid extracts (TLEs) were concentrated to 1‒2 mL with a rotary evaporator and dried with N₂ in a 2 mL vial. The TLEs were separated into apolar-fraction (F1) and polar-fraction (F2) using an activated Al₂O₃ column eluted with hexane and MeOH, respectively. The F2 fraction was condensed to 1‒2 mL with N₂ and filtered through a 0.45 mL PTFE filter. The filtered F2 fraction was dried with N₂ (Yang et al., 2014), then dissolved in 300 ml hexane/isopropanol (99:1, v/v), with a C₄₆ glycerol trialkyl glycerol tetraether (C₄₆GTGT) added as internal standard (Huguet et al., 2006). GDGTs were analyzed using an Agilent 1200 series liquid chromatograph and 6460A triple quadrupole mass spectrometer equipped with an autosampler and ChemStation manager software. An aliquot of sample (10‒30 µL) was injected and separation was achieved with an Alltech Prevail Cyano column (150 mm×2.1 mm, 3 µm; Grace, Deerfield, IL, USA). The elution gradient followed that of Schouten et al. (2007) with minor modifications. GDGTs were eluted isocratically in the first 5 min with A/B 9:1 (v/v), where A= hexane and B=hexane/isopropanol (9:1, v/v). The following linear gradient was then used: 90/10 A/B to 82/18 A/B from 5 to 45 min, followed by 100% B (10 min) to wash the column and then 90/10A/B to equilibrate it. GDGTs were detected using selected ion monitoring (SIM), targeting m/z 1302,1300, 1298, 1296, 1292, 1050, 1048, 1046, 1036, 1034, 1032, 1022,1020, 1018, 653, and 744. Relative abundance was determined from the peak area integration of [M+H]⁺ in the extracted ion chromatograms. Unless noted specifically, the fractional abundance of an individual GDGT compound is proportional to the total iGDGTs or bGDGTs (Yang et al., 2014). The final concentration should be considered semi-quantitative because we did not determine the response factors for GDGTs vs. the C₄₆ GTGT standard.

Environmental parameters and statistical analysis

Environmental parameters for each sampling site were obtained from NOAA (http://www.esrl.noaa.gov/) and the Alfred Wegener Institute (http://odv.awi.de/en/data/ocean/). SST and mean annual temperatures at 20 m and 50 m for each sampling site were retrieved from NOAA (http://www.esrl.noaa.gov/psd/), and annual mean values of total alkalinity (TAlk) and NO₃^‒ were obtained from literature. 10-yr average values of the annual and seasonal SST and mean annual salinity SST data were used to explore the relationship of the iGDGT distributions with environmental parameters (Table 1) because longer times may be needed for suspended particles to settle down to the bottom. The sedimentation rates in the South China Sea are about 0.19‒0.47 cm/yr (Li, 1988; Li and Yuan, 1991). Therefore, 10-yr average values were used to reduce such bias caused by ephemeral environmental parameters.

Relative abundance of each iGDGT component was obtained by normalizing the peak area to the summed area of all iGDGTs (Table 2). Redundancy analysis (RDA) was used to assess the correlation between iGDGTs distribution and environmental parameters. {\mathrm{TEX}}_{\mathrm{86}}^Hand {\mathrm{TEX}}_{\mathrm{86}}^H-derived temperature (abbreviated as {\mathrm{TEX}}_{\mathrm{86}}^H-SST in Table 1) were calculated according to the following formula (Kim et al., 2010),

(1)\begin{array}{l}{\mathrm{TEX}}_{\mathrm{86}}^H=\mathrm{LOG}\left({\displaystyle \frac{[\mathrm{GDGT-}\mathrm{2}]+[\mathrm{GDGT-}\mathrm{3}]+[\mathrm{Cre}]}{[\mathrm{GDGT-}\mathrm{1}]+[\mathrm{GDGT-}\mathrm{2}]+[\mathrm{GDGT-}\mathrm{3}]+[\mathrm{Cre}]}}\right),\\ =\mathrm{LOG}({\mathrm{TEX}}_{\mathrm{86}})\end{array}

(2)

BIT index proposed by Hopmans et al. (2004) was estimated by the following equation

(3)\mathrm{BIT=}{\displaystyle \frac{[I]+[\mathrm{II}]+[\mathrm{III}]}{[I]+[\mathrm{II}]+[\mathrm{III}]+[\mathrm{Cren}]}}.

Results

The distribution of GDGTs

Total concentration of the iGDGTs in the SCS sediments (15.03‒534.18 ng/g) is much higher than that of the bGDGTs (2.34‒128.62 ng/g). Among all iGDGT compounds, crenarchaeol has the highest concentration (9.72‒344.18 ng/g), followed by the GDGT-0 (2.80‒110.04 ng/g). The iGDGTs with 1‒3 rings and the crenarchaeol regio-isomer have the lowest concentrations (Fig. 3). The pattern of lipid distribution in these samples is generally characterized with 65% crenarchaeol, 12% cyclic iGDGTs, and 23% GDGT-0 (Table 2).

Spatial distribution of each individual iGDGT compound shows a similar tendency (Fig. 3). Concentration of each iGDGT compound is lower near the coast and increases offshore. Peak values of GDGT-0 and GDGT-5 appear to be shown at the region centered at (112°E, 20.75°N) (Fig. 3). However, peak values of GDGT-1‒GDGT-3 are shown at the region centered at (111.8°E, 19°N).

Concentration of the major bGDGTs (bGDGT-I+bGDGT-II+bGDGT-III) is higher than that of the total cyclic bGDGTs (bGDGT-Ib+bGDGT-Ic+bGDGT-IIb+bGDGT-IIc+bGDGT-IIIb+bGDGT-IIIc) (Lü et al., 2014). The distribution of bGDGTs is different from that of iGDGTs. Concentrations of the two types of bGDGTs decrease toward the offshore and the highest concentrations are shown at the northwest part of the study region (Fig. 4). The estimated BIT values are low, ranging from 0.02 to 0.26 at the study region(Fig. 5).

Multivariate analysis

The redundancy analysis (RDA) show that the iGDGTs distribution is mostly influenced by nutrients, water depth, spring SST, and salinity (p<0.05). The four environmental parameters contributed approximately 17%, 12%, 12%, and 9% to the variance, respectively (Fig.6(a) and Table 3).

To distinguish the controlling factors to influence the {\mathrm{TEX}}_{\mathrm{86}}^H proxy, we further performed RDA using four iGDGT compounds for the {\mathrm{TEX}}_{\mathrm{86}}^Hcalculation (Fig. 6(b)). Water depth and nitrate explain 52% and 8% of the variance, respectively. Salinity and spring-SST show an insignificant influence on it (Fig. 6(b) and Table 4).

Correlation between the fractional abundance of individual iGDGTs and water depth was analyzed. The fractional abundance of the Crenarchaeol regio-isomer and iGDGT-2/iGDGT-3 ratio has a significant correlation with water depth, showing a strong increasing trend with water depth (Fig. 7).

SST derived from core-top sediments

The performance of the TEX₈₆ calibration function could be evaluated from the residuals between the remote sensing SST and TEX₈₆-derived SST (Fig. 8). The residuals for {\mathrm{TEX}}_{\mathrm{86}}^H(Eq. (1)) ranged from – 2°C to 2.8°C in the global calibrations. The differences (ΔT) between {\mathrm{TEX}}_{\mathrm{86}}^Hderived temperature and remote-sensing temperature at depth of 0 m, 20 m, and 50 m are shown in Fig. 8. The ΔT shows a decreasing trend with depth. Moreover, the ΔT between {\mathrm{TEX}}_{\mathrm{86}}^H derived temperature and remote-sensing temperature from 0‒20 m is smaller than 1°C in the sediments with water depth>200 m (Fig. 8). The {\mathrm{TEX}}_{\mathrm{86}}^Hderived temperature in the shallow water sediment (<200 m) is closer to the remote-sensing temperature at the depth of 50m; ΔT(50 m)<ΔT (20 m)<ΔT (0 m). An inverse trend is shown in the deep water sediment (>200 m) where the {\mathrm{TEX}}_{\mathrm{86}}^H derived temperature is close to the surface temperature (0‒20 m) (Fig. 8).

The difference (ΔT) between {\mathrm{TEX}}_{\mathrm{86}}^H–derived temperature and seasonal remote-sensing SSTs is shown in Fig. 9. The temperature difference (ΔT) between {\mathrm{TEX}}_{\mathrm{86}}^H derived temperature and remote-sensing SSTs in summer and autumn is higher (up to 9°C) than that in annual mean SST. ΔT in spring and winter is in the range of±6°C (‒5.8°C ‒0.6°C, ‒2.8°C ‒3.4°C, respectively) (Fig. 9).

Discussion

The RDA results suggest that iGDGTs distribution is mainly influenced by water depth (WD), spring SST, nitrate, and salinity.

Influence of water depth

Water depth is the main controlling factor in the distribution of the four iGDGTs used in {\mathrm{TEX}}_{\mathrm{86}}^H (l= 52%, p=0.002). The result is different from that observed in ECS where the distribution of iGDGTs is controlled by temperature (Lü et al., 2014). Such a discrepancy may be due to the different feedback of thaumarchaeota to environmental parameters in different regions (Turich et al., 2007). Recently, Hu et al. (2011) reported that the thaumarchaeota community in shallow water (<200 m) is different from that in deep water (>200 m) based on molecular biological analysis for the thaumarchaeotal sequence from the South China Sea. Water depth in eastern coastal seas of China (ECSC) is shallower than 150 m, suggesting that the thaumarchaeota in ECSC sediments may belong to similar communities.

Thaumarchaeota are primarily aerobic, ammonia-oxidizing chemoautotrophs. Because oxidation of ammonia may be a unifying metabolism for aerobic Thaumarchaeota (Pester et al. 2011), the abundance and activity of this group is commonly enumerated by counting amoA genes and mRNA transcripts as well as 16S rRNA genes. Various studies have shown that marine Thaumarchaeota fall into shallow- and deep-water phylogenetic clades (Francis et al., 2005; Hallam et al., 2006). Thaumarchaeotal sequences from the meso-bathypelagic (>200 m water depth) waters are phylogenetically different from those retrieved from the epipelagic waters (0–200 m water depth) (Schouten et al., 2012; Villanueva et al., 2014). Environmental factors dictate the vertical distribution of these Archaea in the water column. Therefore, determining the depth in the water column in which the majority of archaeal GDGTs are produced is of paramount importance for researchers to understand the sources of these compounds in marine sediments.

The iGDGT-2/iGDGT-3 ratio also shows a significant correlation with water depth (Fig. 7(a)), suggesting that the archaea living in deep water is enriched in iGDGT-2 and the high iGDGT-2/iGDGT-3 ratio (>3) may be used as a proxy for deep water archaea. This is similar to the results reported by Kim et al. (2015) in the global core top dataset that the fractional abundances of iGDGTs and water depth have strong correlations. However, in deeper waters, water depth (>1000 m) is a better discriminator of GDGT-2/GDGT-3 ratios than SST in the modern dataset (Taylor et al., 2013). Wang et al. (2015) reported that the higher relative abundance of GDGT-2 and GDGT-3 between the Pearl River estuary and the coastal SCS (10‒200 m) may be due to the non-thaumarchaeotal species, such as Marine Group II Euryarchaeota. The difference may be due to the different archaeal community in the Pearl River Estuary compared to the SCS with water depth>200 m. In addition to the iGDGT-2/iGDGT-3 ratio, the fractional abundance of Crenarchaeol regio-isomer strongly increases with water depth (Fig. 7(b)), indicating that the crenarchaeol regio-isomer may be enriched in the deep thaumarchaeota or that it may be produced by crenarchaeota during diagenesis or both (Shah et al., 2008).

In addition, we found that the SST differences (ΔT) between {\mathrm{TEX}}_{\mathrm{86}}^Hderived temperature and remote-sensing SST decreases with depth and the ΔT is lower than 1°C in the sediments with water depth>200 m. Furthermore, the {\mathrm{TEX}}_{\mathrm{86}}^H derived temperature is close to the surface temperature (0‒20 m) in the deep water sediment (>200 m) (Fig. 8), suggesting that the {\mathrm{TEX}}_{\mathrm{86}}^H could be applied to trace the SST in the region with water depth>200 m (Wucher and Schouten, 2005). This also implies that GDGTs from the deep water sediment may predominantly originate from surface water.

Influence of seasonal temperature

Temperature is the secondary factor to influence the distribution of iGDGTs, different from the results reported in the previous studies (Wei et al., 2011; Jia et al., 2012; Ge et al., 2013). If RDA is performed for six iGDGT compounds, spring SST plays a less important role on the iGDGTs distribution than WD (p<0.006, l=12%) (Table 3). However, if RDA is performed for only the four iGDGT compounds, the autumn SST and summer SST showed less influence than the WD (Table 4). The difference (ΔT) between {\mathrm{TEX}}_{\mathrm{86}}^Hderived temperature and remote-sensing SSTs in summer and autumn is higher than those in the annual mean SST and ΔT in spring and winter, suggesting that {\mathrm{TEX}}_{\mathrm{86}}^H-derived SSTs are skewed towards spring/winter SSTs (Fig. 9). Moreover, it is noted that temperature is overestimated in colder (spring/winter) seasons with {\mathrm{TEX}}_{\mathrm{86}}^H. This is consistent with the fact that the seasonality of water temperature seems unlikely to be a major factor controlling the distribution of iGDGTs in the sediments of coastal SCS, as reported by Wang et al. (2015).

Influence of nutrients and salinity

Nutrient concentrations show impacts on the relative abundance of thaumarchaeota and thus affect GDGT concentrations and distributions. It is found that nitrate concentrations and the abundance of crenarchaeotal cells have a significant negative correlation (Murray et al., 1999). Our study shows that nitrate has a significant correlation with the relative abundance of individual iGDGT compounds in the RDA analyses (p<0.024) (Table 3). This is similar to the result reported by Turich et al (2007) that the mesopelagic zone/upwelling GDGTs showed a significant correlation with nitrate concentration. However, the correlation coefficient of the nitrate and {\mathrm{TEX}}_{\mathrm{86}}^H is low (R=0.1, p=0.60), indicating that the nitrate concentration may have a strong impact on the iGDGT concentrations, but not {\mathrm{TEX}}_{\mathrm{86}}^H. It may be due to that fact that nutrient and the other environmental factors could jointly influence iGDGT distribution. Therefore, a partial RDA is performed to distinguish controlling factors. Obviously, nitrate can explain the largest contribution to variance (10.7%; Table 5), although the joint effect is 51%.

Some previous studies also reported that salinity may influence the abundance and community structure of Archaea associated with iGDGTs and archaeol (Turich and Freeman, 2011; Xie et al., 2014). Interestingly, the correlation between iGDGTs distribution and salinity is significant (l=9%, p=0.044) when six iGDGT compounds are considered in RDA, while the correlation is insignificant with only the four compounds. It suggests that salinity has a more significant influence on GDGT-0 and Crenarchaeol than on the other iGDGT compounds. The low correlation coefficients are found between salinity and GDGT-0 and crenarchaeol (R²=0.16, 0.02, respectively). Wuchter et al. (2004) performed mesocosm studies with marine Crenarchaeota incubated at salinities of 27‰ and 35‰ to test the validity of the TEX₈₆ proxy. Their results show that salinity does not substantially influence the GDGTs distribution in terms of TEX₈₆ values. The reason remains unknown.

Conclusions

Concentration of total iGDGTs ranges from 15.03 to 534.18 ng/g, much higher than that of total bGDGTs (2.34‒128.62 ng/g). Among the six iGDGT compounds, crenarchaeol and GDGT-0 are the predominant compounds, which account for 65% and 20%, respectively. Concentrations of each iGDGT compound increase gradually eastward. RDA shows that water depth is the most significant factor to influence the iGDGTs distribution at the study region, different from that in ECS where temperature is the primary controlling factor . This may be due to the fact that thaumarchaeota communities are dependent on water depth. Furthermore, nitrate and salinity are found to show some influence on iGDGTs distribution. This study shows that the{\mathrm{TEX}}_{\mathrm{86}}^H may be considered as a robust proxy to trace the paleo-SST in the region with water depth>200 m.

References

[1] Basse A, Zhu  C, Versteegh G J M,  Fischer G,  Hinrichs K,  Mollenhauer G (2014). Distribution of intact and core tetraether lipids in water column profiles of suspended particulate matter off Cape Blanc, NW Africa. Org Geochem, 72: 1–13

[2] De Rosa M, Esposito  E, Gambacorta A,  Nicolaus B,  Bu'Lock J D (1980). Effects of temperature on ether lipid composition of Caldariella acidophila. Phytochemistry, 19(5): 827–831

[3] Francis C A, Roberts  K J, Beman  J M, Santoro  A E, Oakley  B B (2005). Ubiquity and diversity of ammonia oxidizing archaea in water columns and sediments of the ocean. Proc Natl Acad Sci USA, 102(41): 14683–14688

[4] Ge H, Zhang  C, Dang H,  Zhu C, Jia  G (2013). Distribution of tetraether lipids in surface sediments of the northern South China Sea: implications for TEX86 proxies. Geoscience Frontiers, 4(2): 223–229

[5] Gliozzi A, Paoli  G , De RosaM ,  GambacortaA (1983). Effect of isoprenoid cyclization on the transition temperature of lipids in thermophilic archaebacteria. Biochimica et Biophysica Acta (BBA) - Biomembranes, 735(2): 234–242

[6] Hallam S J, Konstantinidis  K T, Putnam  N, Schleper C,  Watanabe Y,  Sugahara J,  Preston C,  de la Torre J,  Richardson P M,  DeLong E F (2006). Genomic analysis of the uncultivated marine crenarchaeote Cenarchaeum symbiosum. Proc Natl Acad Sci USA, 103(48): 18296–18301

[7] Herndl G J, Reinthaler  T, Teira E,  van Aken H,  Veth C, Pernthaler  A, Pernthaler J (2005). Contribution of Archaea to total prokaryotic production in the deep Atlantic Ocean. Appl Environ Microbiol, 71(5): 2303–2309

[8] Hopmans E C, Weijers  J W H, Schefuß  E, Herfort L,  Sinninghe Damsté J S,  Schouten S (2004). A novel proxy for terrestrial organic matter in sediments based on branched and isoprenoid tetraether lipids. Earth Planet Sci Lett, 224(1‒2): 107–116

[9] Hu A, Jiao  N, Zhang C (2011). Community structure and function of planktonic Crenarchaeota: changes with depth in the South China Sea. Microb Ecol, 62(3): 549–563

[10] Hu J, Kawamura  H, Hong H S,  Qi Y (2000). Review on the currents in the South China Sea seasonal circulation South China Sea warm current and Kuroshio Intrusion. J Oceanogr, 56(6): 607–624

[11] Huguet C, Hopmans  E C, Febo-Ayala  W, Thompson D H,  Sinninghe Damste' J S,  Schouten S (2006). An improved method to determine the absolute abundance of glycerol dibiphytanyl glycerol tetraether lipids. Org Geochem, 37(9): 1036–1041

[12] Jia G, Zhang  J, Chen J,  Peng P, Zhang  C (2012). Archaeal tetraether lipids record subsurface water temperature in the South China Sea. Org Geochem, 50: 68–77

[13] Karner M B, DeLong  E F, Karl  D M (2001). Archaeal dominance in the mesopelagic zone of the Pacific Ocean. Nature, 409(6819): 507–510

[14] Kim J H, Schouten  S, Rodrigo-Gámiz M, Rampen S,  Marino G,  Huguet C,  Helmke P,  Buscai R,  Hopmans E,  Pross J,  Sangiorgi F,  Middelburg J B M,  Sinninghe Damsté J S (2015). Influence of deep-water derived isoprenoid tetraether lipids on the paleothermometer in the Mediterranean Sea. Geochim Cosmochim Acta, 150: 125–141

[15] Kim J H, van der Meer  J, Schouten S,  Helmke P,  Willmott V,  Sangiorgi FKoç N, Hopmans  E C, Damsté  J S S (2010). New indices and calibrations derived from the distribution of crenarchaeal isoprenoid tetraether lipids: implications for past sea surface temperature reconstructions. Geochim Cosmochim Acta, 74(16): 4639–4654

[16] Li F Y (1988). Determination of recent sedimentation rates by ²¹⁰Pb method in the South China Sea. Mark Sci, 3: 64–66

[17] Li F Y, Yuan  W (1991). Profile model of ²¹⁰Pb in the South China Sea, South Huanghai Sea and Bohai Sea. Mar Geol & Quaternary Geol, 11(3): 35–43 (in Chinese)

[18] Liu K K, Chao  S Y, Shaw  P T, Gong  G C, Chen  C C, Tang  T Y (2002). Monsoon-forces chlorophyll distribution and primary production in the South China Sea: observations and a numerical study. Deep Sea Res Part I Oceanogr Res Pap, 49(8): 1387–1412

[19] Lü  X, Yang  H, Song J,  Versteegh G J M,  Li X, Yuan  H, Li N,  Yang Y, Ding  W, Xie S (2014). Sources and distribution of isoprenoid glycerol dialkyl glycerol tetraethers (GDGTs) in sediments from the east coastal sea of China: application of GDGT-based paleothermometry to a shallow marginal sea. Org Geochem, 75: 24–35

[20] Murray A E, Blakis  A, Massana R,  Strawzewski S,  Passow U,  Alldredge A,  Delong E F (1999). A time series assessment of planktonic archaealvariability in the Santa Barbara Channel. Aquat Microb Ecol, 20: 129–145

[21] Ose T, Song  Y K, Kitoh  A (1997). Sea surface temperature in the South China Sea—An index for the Asian monsoon and ENSO system. J Meteorol Soc Jpn, 75: 1091–1107

[22] Pernthaler A, Preston  C M, Pernthaler  J, DeLong E F,  Amann R (2002). Comparison of fluorescently labeled oligonucleotide and polynucleotide probes for the detection of pelagic marine bacteria and archaea. Appl Environ Microbiol, 68(2): 661–667

[23] Pester M, Schleper  C, Wagner M (2011). The Thaumarchaeota: an emerging view of their phylogeny and ecophysiology. Curr Opin Microbiol, 14(3): 300–306

[24] Peterse F, Kim  JH, Schouten S,  Kristensen DK,  Koc, N, Sinninghe Damsté  JS(2009). Constraints on the application of the MBT/CBT palaeothermometer at high latitude environments (Svalbard, Norway). Organic Geochemistry, 40: 692–699

[25] Powers L, Werne  J P, Vanderwoude  A J, Sinninghe Damsté  J S, Hopmans  E C, Schouten  S (2010). Applicability and calibration of the TEX₈₆ paleothermometer in lakes. Org Geochem, 41(4): 404–413

[26] Qu T D (2001). Role of ocean dynamics in determining the mean seasonal cycle of the South China Sea surface temperature. J Geophys Res, 106(C4): 6943–6955

[27] Schouten S, Hopmans  E C, Schefuß  E, Sinninghe Damste J S (2002). Distributional variations in marine crenarchaeotal membrane lipids: a new tool for reconstructing ancient sea water temperatures? Earth Planet Sci Lett, 204(1): 265–274

[28] Schouten S, Huguet  C, Hopmans E C,  Kienhuis M M,  Sinninghe Damste J S (2007). Analytical Methodology for TEX₈₆ Paleothermometry by High-Performance Liquid Chromatography/Atmospheric Pressure Chemical Ionization-Mass Spectrometry. Anal Chem, 79(7): 2940–2944

[29] Schouten S, Pitcher  A, Hopmans E C,  Villanueva L,  van Bleijswijk J,  Sinninghe Damsté J S (2012). Intact polar and core glycerol dibiphytanyl glycerol tetraether lipids in the Arabian Sea oxygen minimum zone: I. Selective preservation and degradation in the water column and consequences for the TEX₈₆. Geochim Cosmochim Acta, 98: 228–243

[30] Shah S R, Mollenhauer  G, Ohkouchi N,  Eglinton T I,  Pearson A (2008). Origins of archaeal tetraether lipids in sediments: Insights from radiocarbon analysis. Geochim Cosmochim Acta, 72(18): 4577–4594

[31] Shen S, Lau  M K (1995). Biennial oscillation associated with the East Asian summer monsoon and tropical sea surface temperatures. J Meteorol Soc Jpn, 73(1): 105–124.

[32] Shintani T, Yamamoto  M, Chen M T (2011). Paleoenvironmental changes in the northern South China Sea over the past 28,000 years: a study of TEX86-derived sea surface temperatures and terrestrial biomarkers. Journal of Asian Earth Science, 40(6): 1221–1229

[33] Sinninghe Damsté JS,  Ossebaar J,  Abbas B,  Schouten S,  Verschuren D (2009). Fluxes and distribution of tetraether lipids in an equatorial African lake: constraints on the application of the TEX86 palaeothermometer and BIT index in lacustrine settings. Geochimica et Cosmochimica Acta 73: 4232–4249

[34] Taylor K W R,  Huber M,  Hollis C J,  Hernandez-Sanchez M T,  Pancost R D (2013). Re-evaluating modern and Palaeogene GDGT distributions: implications for SST reconstructions. Global Planet Change, 108: 158–174

[35] Turich C, Freeman  K H (2011). Archaeal lipids record paleosalinity in hypersaline systems. Org Geochem, 42: 1147–1157

[36] Turich C, Freeman  K H, Bruns  M A, Conte  M, Jones A D,  Wakeham S G (2007). Lipids of marine Archaea: patterns and provenance in the water-column and sediments. Geochim Cosmochim Acta, 71(13): 3272–3291 

[37] Villanueva L, Schouten  S, Sinninghe Damste′  J S (2014). Depth-related distribution of a key gene of the tetraether lipid biosynthetic pathway in marine Thaumarchaeota. Environ Microbiol, 10(17): 3527–3539

[38] Walsh E M, Ingalls  A E, Keil  R G (2008). Sources and transport of terrestrial organic matter in Vancouver Island fjords and the Vancouver – Washington Margin: a multiproxy approach using d13Corg, lignin phenols, and the ether lipid BIT index. Limnol Oceanogr, 53(3): 1054–1063

[39] Wang J X, Wei  Y L, Wang  P, Hong Y H,  Zhang C L (2015). Unusually low TEX₈₆ values in the transitional zone between Pearl River estuary and coastal South China Sea: impact of changing archaeal community composition. Chem Geol, 402: 18–29

[40] Weber Y, De Jonge  C, Rijpstra W I C,  Hopmans E C,  Stadnitskaia A,  Schubert C J,  Lehmann M F,  Sinninghe Damsté J S,  Niemann H (2015). Identification and carbon isotope composition of a novel branched GDGT isomer in lake sediments: evidence for lacustrine branched GDGT production. Geochim Cosmochim Acta, 154: 118–129

[41] Wei Y, Wang  J, Liu J,  Dong L, Li  L, Wang H,  Wang P, Zhao  M, Zhang C (2011). Spatial variations in Archaeal lipids of surface water and core-top sediments in the South China Sea and their implications for Paleoclimate studies. Appl Environ Microbiol, 77(21): 7479–7489

[42] Weijers J W H,  Schouten S,  van den Donker J C,  Hopmans E C,  Sinninghe Damsté J S (2007). Environmental controls on bacterial tetraether membrane lipid distribution in soils Geochim Cosmochim Ac, 71(3): 703–713

[43] Wuchter C, Schouten  S (2005). Temporal and spatial variation in tetraether membrane lipids of marine Crenarchaeota in particulate organic matter: Implications for TEX86 paleothermometry. Paleoceanography, 20: PA3013, 

[44] Wuchter C, Schouten  S, Coolen M J L,  Sinninghe Damsté J S (2004). Temperature dependent variation in the distribution of tetraether membrane lipids of marine Crenarchaeota: Implications for TEX86 paleothermometry.  Paleoceanography, 19, PA4028, 

[45] Xie W, Zhang  C L, Zhou  X D, Wang  P (2014). Salinity-dominated change in community structure and ecological function of Archaea from the lower Pearl River to coastal South China Sea. Appl Microbiol Biotechnol, 98(18): 7971–7982

[46] Yang H, Pancost  R D, Tang  C, Ding W,  Dang X, Xie  S (2014). Distributions of isoprenoid and branched glycerol dialkanol diethers in Chinese surface soils and a loess–paleosol sequence: implications for the degradation of tetraether lipids. Org Geochem, 66: 70–79

[47] Zell C, Kim  J-H, Dorhout D,  Baas M, Sinninghe Damsté  J S (2015). Sources and distributions of branched tetraether lipids and crenarchaeol along the Portuguese continental margin: implications for the BIT index. Cont Shelf Res, 96: 34–44

[48] Zhang C L, Wang  J X, Wei  Y L, Zhu  C, Huang L Q,  Dong H L (2012). Production of branched tetraether lipids in the lower Pearl River and estuary: effects of extraction methods and impact on bGDGT proxies. Front Microbiol, 2(274): 1–18

[49] Zhou H, Hu  J, Spiro B,  Peng P, Tang  J (2014). Glycerol dialkyl glycerol tetraethers in surficial coastal and open marine sediments around China: indicators of sea surface temperature and effects of their sources. Palaeogeogr Palaeoclimatol Palaeoecol, 395: 114–121

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg