Marine oil pollution refers to the pollution caused by oil and its refined products entering the marine environment in the process of exploitation, refining, storage, transportation and use. In particular, some major marine accidents have resulted in a large number of oil spills [1]. For example, the explosion of the Deepwater Horizon drilling platform in 2010 caused about 4.9 million barrels of crude oil to enter the Gulf of Mexico, making it the largest oil spill in the history of the United States [2]. At the beginning of 2018, the collision accident of "Sangji" tanker in the East China Sea resulted in 136, 000 tons of condensate oil entering the sea, forming a strip oil pollution distribution area of about 69 km² [3]. The spilled oil into the marine environment has significantly changed the conditions of marine biogeochemistry, poses a great threat to the marine ecosystem and human health, and has become a key issue of marine chemistry and biology research in recent years [2, 4].

Typically, the crude oil contains approximately 81 vol% alkanes and 19 vol% aromatics [5]. Therefore, it is of great significance to evaluate the transport and fate of petroleum hydrocarbons and especially alkanes in the oil spill event. The petroleum hydrocarbons in marine environment also have natural sources such as seeps at sea bottom [6], and input from terrestrial plants [7]. The source tracing of petroleum hydrocarbons is an important work to reveal their composition characteristics and evaluate the extent of oil pollution.

The South China Sea is the western part of the Pacific Ocean, located at the southern end of the Chinese mainland. Its total area is about 2.1 million km². The oil drilling and shipping have resulted in oil pollution in the South China Sea [8], and especially northeastern part with relatively heavy anthropogenic activities [9]. However, the information on the marine oil pollution characteristics in local area is lacking. Therefore, this paper has studied the distribution characteristics of petroleum hydrocarbons in the northeastern South China Sea, and identified their sources on the basis of composition profile.

A total of 50 water samples were collected from the northeastern South China Sea in July-September 2018 (Fig. 1). The sampling area is located in the south of the coastline of Zhanjiang City, Shenzhen City, Guangdong Province and Quanzhou City, Fujian Province, in the middle of Hainan Island, Taiwan Island, Xisha Islands and Luzon Island, Philippines, between 21°17′-16°21′ N and 111°02′-117°41′ E, covering the Dongsha Islands. This area is in the north of the tropics, with a tropical monsoon climate. The surface and profile water samples were collected with aid of sea anchors, and stored in 2 L high-density polyethylene (HDPE) bottles. The collected samples were stored on ice, and then transported back to the laboratory for further analysis. Table S1 (Supporting information) lists the detailed location and depth of each sampling point.

The water chemistry parameters were determined by standard methods including pH (on site), dissolved oxygen (DO, on site), chemical oxygen demand (COD, GB11892-89), total organic carbon (TOC, GB13193-91), total phosphorus (TP, GB11893-89), total nitrogen (TN, GB1189 4-8 9) and ammonia nitrogen (NH₃-N, GB7479-87). Total petroleum hydrocarbons (TPHs) including both alkanes and aromatics in water were measured from volatile (TPH_C6-C9, HJ893-2017) and extractable (TPH_C10-C40, HJ894-2017) fractions using gas chromatography-flame ionization detector (GCFID) technology. The concentration of each n-alkane (C10-C40) was also determined. The details for the analysis of petroleum hydrocarbons are provided in Text S1 (Supporting information). The gas chromatograms (C10-C40) of all the samples are provided in Supporting information.

For the water chemistry parameters (Table 1), the pH of seawater in the northeastern South China Sea was neutral or alkaline, with a range of 7.77 to 9.16. TN values ranged from 0.087 mg/L to 0.93 mg/L, with an average of 0.43 ± 0.02 mg/L. COD ranged from 0.32 mg/L to 8.16 mg/L, with an average of 2.78 ± 0.28 mg/L. The COD values at different points fluctuated greatly. However, from a macroscopic view, it showed that COD of near shore was higher than that of the far shore. The COD peaks were obtained at samples 40, 42, 47, and 49 (7.20–8.16 mg/L), and these points were located in the offshore waters. The NH3-N values were between 0.041 mg/L and 1.35 mg/L, and the average value was 0.29 ± 0.06 mg/L. The TP values ranged from 0.023 mg/L to 0.46 mg/L with the average of 0.085 ± 0.005 mg/L. There seemed to be no significant trend of the changes of TP values. The TOC content was between 0.92 mg/L and 3.98 mg/L, with an average of 1.80 ± 0.09 mg/L. The TOC got peak values at samples 15 (3.40 mg/L) and 24 (3.98 mg/L). There were 23 consecutive samples from sample 15 to sample 27, which had higher concentrations of TOC than that of other samples. The concentration of TOC of near shore was higher than that of the distant sea, and the western area had generally higher TOC than the eastern sea. For the TPHs, there were 2 samples exceeding 500 μg/L, 13 samples ranging from 300 μg/L to 500 μg/L, and the remaining 35 samples had 121.3–276.5 μg/L.

The correlation analysis of water chemistry parameters indicated that TN and NH₃-N were negatively correlated with most of the parameters, while TP was significantly positively correlated with DO and NH₃-N, and was negatively correlated with other parameters. TPHs was significantly positively correlated with the values of COD, total n-alkanes, TPH_C6-C9, and negatively correlated with other parameters (Table S2 in Supporting information).

Table S3 (Supporting information) summaries the petroleum hydrocarbon concentrations in the waters of the northeastern South China Sea. The TPHs values ranged from 121.30 μg/L to 603.02 μg/L, with an average value of 259.06 ± 15.59 μg/L. Total petroleum hydrocarbons contained both volatile fraction (TPH_C6-C9) and extractable fraction (TPH_C10-C40), while the concentration of TPH_C6-C9 was extremely low even close to the detection limit. The TPH_C10-C40 accounted for more than 99.99% of the TPHs concentration. The lower concentration of TPH_C6-C9 might be ascribed to leaking during the samples transportation and easily degradation of short-chain hydrocarbons.

Dados et al. found that short-chain n-alkanes decreased rapidly in the first half process of the degradation of petroleum hydrocarbons, during the bioremediation of oil-contaminated soil using the composting method of brewing fermentation materials, while the long-chain petroleum hydrocarbons were more stable [10]. Joo et al. observed that under natural conditions on the sea surface, low-molecular-weight hydrocarbons in the crude oil declined rapidly through weathering processes such as evaporation, dissolution, and dispersion, while the content of long-chain hydrocarbons kept stable [11]. Therefore, the extremely low content of TPH_C6-C9 measured in this paper might indicate that the petroleum hydrocarbons in the surface seawater of the northeastern South China Sea have been exposed to strong physical processes of weathering. Wang determined and analyzed the n-alkanes and fatty acids in the surface sediments of the northern South China Sea and the sea area near Hainan Island, and found the distribution range of n-alkanes was C14-C33 [12], which is similar to the distribution range of n-alkanes in this study (C10-C37).

Fig. 2 presents the distribution map of TPHs concentrations in the northeastern South China Sea. The concentration of TPHs measured in five water samples exceeded 400 μg/L, which were samples 3 (515.21 μg/L), 18 (477.12 μg/L), 29 (603.02 μg/L), 39 (499.72 μg/L) and 46 (414.11 μg/L). The highest and lowest TPHs in surface water were obtained at sample 29 and sample 9 respectively. The samples of high TPHs concentrations were all located in the surface layer of the northeastern South China Sea, and mainly dominated by the near shore bay. This distribution characteristic might be related to the input of discharged industrial wastewater and domestic sewage by the northern rivers along the South China Sea.

The concentration range of total n-alkanes was 11.18-287.87 μg/L, and the average value was 80.35 ± 8.48 μg/L. The high value of n-alkanes appeared at samples 26, 29, 30, 32, 34, 39 and 40 (Table S3 in Supporting information), of which the distribution trend is slightly different from that of TPHs, indicating that the source of oil pollution was complicated and weathering effect was uneven on the surface of sea. Zhang et al. found that the concentration of n-alkanes in surface waters of the Jiaozhou Bay was between 1.76 and 39.09 μg/L [13]. The n-alkanes measured in this study are nearly ten times higher than that of Jiaozhou Bay, which might be related to the input of n-alkanes brought by oilfield exploitation in the northern South China Sea and shipping routes near harbors and docks.

Samples 2, 4–8 and 11 were collected in the middle or deep layer of the sea. The highest value of TPHs (366.01 μg/L) were obtained in sample 2, which is located at 315 m below the water surface and 5 m above the sea bottom. Sample 2 is near the bottom of the offshore shoal. Due to the severe disturbance of the ocean current, the migration speed of the seabed sediment there is accelerated. Samples 4–8 are located at the same latitude and longitude, with different depths, and the sampled depth reaches several kilometers. The geomorphic environment at the bottom of the sampling route changes from offshore shoals into deep sea ravines. The altitude difference is large, so there is strong current disturbance. Some studies believe that the content of organic matter in sediments is usually about 10 times the concentration of organic matter in seawater [14], so it is speculated that the high TPHs value of sample 2 is due to its special geographical location. For the vertical distribution, with the incport of petroleum hydrocarbons. For erease of the sampling depth, the TPHs concentration showed an abrupt decrease in the beginning, and then increased a little bit in the middle layer of seawater, and finally reached a steady state near the bottom of seabed (Fig. 3). The concentration of TPHs in the surface water was 515.21 μg/L. When the water depth reached 600 m, the concentration of TPHs reached a minimum value of 128.91 μg/L. As the depth continuously increasing, the concentration of TPHs gradually increased in a slower speed and reached a stable state at last, and it is especially obvious in the deep sea where the variation of the depth of landform is large.

The tendency of petroleum hydrocarbons on the vertical profile of seawater might be the results of sea rainfall, hydrodynamic factors, petroleum hydrocarbon settlement and glacial geological effects. The surface seawater had undergone physical weathering processes such as evaporation. Besides, the sampling was conducted between July and October when heavy rainfall was brought by the tropical cyclone, diluting the petroleum hydrocarbons in the surface waters. Instead, the layer of 5 m below the surface was less affected. Moreover, various endogenic hydrocarbons and organic matter were produced by plankton. Therefore, the TPHs in the superficial water usually showed a higher level.

The ocean dynamics may play a role for the distribution of petroleum hydrocarbons [15]. The geostrophic current, inertia flow, Ekman current, waves and tidal current influence the transport of petroleum hydrocarbons. For example, Fig. 4 shows the sea level and geostrophic current anomalies of the northeastern South China Sea. An average of 0.3 m/s surface geostrophic currents and numerous mesoscale eddies exist in this area. Therefore, the point pollution source of oil platform may have a considerably wide affected area.

Sedimentation is another important factor influencing the distribution of petroleum hydrocarbons. In the South China Sea, the mixed deposition of modern carbonate and siliceous debris is widely distributed [16], and the petroleum hydrocarbons could be adsorbed on the surface of siliceous debris and subjected to sedimentation. The local migration of petroleum hydrocarbons would be significantly affected by the sediment transport or sediment resuspension that adsorbs the petroleum hydrocarbons on the surface. For example, after the Deepwater Horizon oil spill, a large number of marine snow containing crude oil were formed, which accelerated the vertical transmission of petroleum hydrocarbons to the seafloor [5]. Therefore, the sedimentation or resuspension provides another explanation for the slight increase in the concentration of TPHs near seabed in this paper.

The n-alkanes in total petroleum hydrocarbons were regarded as a reliable organic geochemical index due to their relatively stable properties and slow degradation rate brought by high bond energy of carbon-carbon bonds [17], which also could be used to trace the source of organic matter [18]. As is shown in Fig. 5, the n-alkanes in the seawater was mainly concentrated in C10-C38, which is similar to that in surface seawater of Jiaozhou Bay (C11-C37) [13], coastal sediments of the Northern Cyprus (C10-C34) [14], and sediments of Lake Wanghu (C14-C33) [19]. The concentration of short chain n-alkanes (C11-C21) ranged from 0 to 16.29 μg/L with an average value of 5.27 μg/L, and the concentration of long chain n-alkanes (C21-C37) ranged from 7.46 to 287.87 μg/L with an average value of 75.08 μg/L. The ratio of LMW/HMW (short chain n-alkanes/long chain n-alkanes) was 0.21, indicating the dominance of long chain hydrocarbons in the northeastern South China Sea. n-Alkanes are widely distributed in organisms such as bacteria, algae and higher plants. Generally short chain n-alkanes (< C21) are mainly derived from marine algae and bacteria, and C15, C17 or C19 appeared as the main peak carbon. The long chain n-alkanes (C22-C40) are mainly contributed by terrestrial sources. The waxy layer of the epidermis of terrestrial higher plants contains abundant C27, C29 and C31, so that HMW odd n-alkanes often make up a dominant fraction.

In this paper, a series of petroleum hydrocarbon characteristic indices including carbon predominance index (CPI), terrigenous/aquatic ratio (TAR), Pr/Ph, Pr/C17, and Ph/C18 were introduced to analyze the source of petroleum hydrocarbons in the northeastern South China Sea. Hydrocarbon molecules with abundant odd carbon atoms usually originate from terrestrial higher plants, while petroleum hydrocarbons derived from crude oil would display an evenly distribution between even and odd carbon molecules [20]. CPI stands for the relative abundance of even numbered n-alkanes and odd numbered n-alkanes, and is used to evaluate the contribution of terrestrial organic matter to the organic pollution of hydrocarbons (Eq. 1) [21]:

(1)

A higher CPI index indicated that n-alkanes are mainly composed of odd numbered carbon molecules, and the petroleum hydrocarbons mainly come from terrestrial plants [21]. Most of the samples in this paper peaked at C20, C21, C24, C28 and C29. The calculated CPI index was 0.009–24.14, and each point showed a significant difference (Table S4 in Supporting information). The CPI value of 16 samples could not be calculated accurately due to continuous absence of C25-C34. There were 18 samples showing CPI < 1 (close to 0.15), while 16 samples with CPI > 1. The results suggested the coexistence of terrestrial contributions dominated by higher plants and marine endogeny produced by algae and bacteria. In terms of the average CPI, the calculated value was 4.417 (including the 16 samples without CPI as 0) or 6.099 (excluding the 16 samples without CPI value), suggesting the dominant input from land sources.

The TAR index is the ratio of terrestrial and seawater endogenous sources. It is used to calculate the contribution of terrestrial higher plants to organic pollutants in seawater compared to endogenous seawater sources (Eq. 2) [22]:

(2)

The TAR value measured in this paper generally was mostly greater than 1, indicating that the input of terrestrial higher plant accounted for the main proportion.

Pristane (Pr) and phytane (Ph) are common isoprenoids in marine sediments. Pr and Ph are regarded as the diagenetic products of phytyl-side-chain of chlorophyll, which are more likely to produce Ph under reducing conditions, and to produce Pr under oxidizing conditions. Therefore, the Pr/Ph ratio could reflect the oxidative or reductive nature of the formation environment of alkanes [23]. From Table S4, it could see that only 22 samples were available for analysis of Pr/Ph ratio due to the undetection of Pr or Ph in other samples. The calculated Pr/Ph ranged between 0.001 and 3.49. 13 samples (5 samples were below sea level) got Pr/Ph < 1, showing a reductionstrong reduction environment, while the other 9 samples got Pr/Ph > 1, showing an oxidative-strong oxidizing environment. The sea surface presented a complex distribution with a combination of reduction and oxidizing environment.

Pr/nC17 and Ph/C18 are biodegradation parameters for petroleum hydrocarbons. Pr/C17 > 1 indicates that planktonic algae contribute more for the biodegradation of petroleum hydrocarbons, while Ph/C18 > 1 represents that the marine bacteria played a more important role [24]. There were 23 effective values of Pr/C17 measured in this paper, of which 18 samples (78.26%) had a value greater than 1; only 11 samples could be taken the effective value of Pr/C18, with the value of Ph/C18 > 1 accounting for 72.7% (Table S4).

The comprehensive analysis of various characteristic indices showed that the petroleum hydrocarbons in the northeastern South China Sea were mainly from terrestrial higher plant input, which were mixed with oil pollution caused by land-based oil pollution, shipping leaks and oil drilling. The land-based oil pollution might be related to the industrial oil pollution discharged from the Pearl River Estuary. The middle and deep layer of seawater showed obviously reducing environment, while the surface water was an obvious oxidation environment. A considerable number of samples in the offshore area showed the endogenic characteristic by plankton, algae and bacteria. However, the terrestrial sources made a greater contribution in the northeastern South China Sea.

In addition, a large number of unresolved complex mixture (UCM) were found in the samples (Table S4), which was obtained by deducting the alkanes from the TPHs [25]. UCM is widely used to describe the gas chromatographic characteristics, which indicates the presence of fossil fuel hydrocarbons in the samples [26]. The gas chromatogram showed that over time, there was a slowly rising and slightly swinging pattern track on the line. With the end of the temperature program, there were some small peaks at the top of the line, gradually returning to the column loss background of GC column. The content of UCM was much higher than that of n-alkane, indicating that the northeastern South China Sea was in a long-term chronic oil pollution state.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

This work was supported by the National Natural Science Foundation of China (Nos. 91851110, 41701541).

Appendix A. Supplementary data

Supplementary material related to this article canbe found, in the online version, at doi:https://doi.org/10.1016/j.cclet.2020.06.020.