An Assessment of Marine Ecosystem Damage from the Penglai 19-3 Oil Spill Accident

: Oil spills have immediate adverse effects on marine ecological functions. Accurate assessment of the damage caused by the oil spill is of great signiﬁcance for the protection of marine ecosystems. In this study the observation data of Chaetoceros and shellﬁsh before and after the Penglai 19-3 oil spill in the Bohai Sea were analyzed by the least-squares ﬁtting method and radial basis function (RBF) interpolation. Besides, an oil transport model is provided which considers both the hydrodynamic mechanism and monitoring data to accurately simulate the spatial and temporal distribution of total petroleum hydrocarbons (TPH) in the Bohai Sea. It was found that the abundance of Chaetoceros and shellﬁsh exposed to the oil spill decreased rapidly. The biomass loss of Chaetoceros and shellﬁsh are 7.25 × 10 14 ∼ 7.28 × 10 14 ind and 2.30 × 10 12 ∼ 2.51 × 10 12 ind in the area with TPH over 50 mg/m 3 during the observation period, respectively. This study highlights the evaluation of ecological resource loss caused by the oil spill, which is useful for the protection and restoration of the biological resources following the oil spill.


Introduction
Marine environments are frequently exposed to oil spills as a result of transportation, oil drilling and fuel usage. An estimated 5000 tons of oil per year was spilled during the period 2010-2014 due to accidents, cleaning operations or other causes [1]. Crude oil constitutes a large reservoir of the highly toxic polycyclic aromatic hydrocarbons (PAHs), which are rapidly released into the water column after the spill. As we all known, oil slicks over the sea surface not only limit gas exchange through the air-sea interface, but also reduce light penetration into the water column, and resultant affecting the photosynthesis of phytoplankton [2]. The negative impact threatens the survival of marine life that feeds on phytoplankton. In addition, phytoplankton are the primary food source for the zooplankton, and are therefore the starting points for the entry of petroleum hydrocarbons into the marine food web [3,4]. Then the petroleum hydrocarbons are further enriched by marine organisms and amplified along the food chain, and ultimately harming human health due to their mutagenic or carcinogenic activity [5,6]. In general, oil spills not only affect the growth of marine organisms, disrupt the ecological mechanism and reproduction of the ocean [7,8], but also result in environmental and economic damage to fisheries and human health [9,10]. Therefore, accurate assessment of the damage caused by the oil spill is of great significance.
The impacts of oil spills on marine organisms have been studied over the past several decades, but the conclusions are often inconsistent. Batten et al. found phytoplankton community structure did not exhibit any significant differences before and after the spill [11].
Some studies have reported that phytoplankton are affected by the oil spill [12][13][14][15]. For example, Ohwada et al. found that an oil spill could result in a reduction in phytoplankton growth [16]. On the contrary, Guo et al. found that the oil spill caused abnormal chlorophyll concentration distributions which led to an outbreak of red tide [17]. The impact of oil spill on zooplankton is generally considered detrimental [18][19][20]. Gesteira and Dauvin noted that the benthic community structure could be altered by concentrations of petroleum hydrocarbons in sediment <50 ppm, and some species may be excluded at concentrations <10 ppm [21]. Therefore, the impact of oil pollution on marine ecosystems is a pendent and hot topic in the research field of ecological environment evaluation.
On 17 June 2011, an oil slick was detected in the offshore water near platform C of the Penglai 19-3 drilling field. The Penglai 19-3 oil spill was a typical Chinese oil spill, with some 115 m 3 of heavy crude oil and 416 m 3 of mineral oil-based drilling mud seeping into the Bohai Sea. To the best of our knowledge, studies examining its potential environmental impact are rare. In fact, the Bohai Sea, covering approximately 77,000 km 2 and surrounded by land on three sides, is the only semi-closed marginal sea in China. It connects to the Yellow Sea through the Bohai Strait. The limited water exchange capacity causes a poor self-purification ability of the Bohai Sea, making it difficult to restore in a short time if the marine ecosystem is severely damaged. Therefore, oil spill pollution is a serious threat to the Bohai Sea.
In this paper, an adjoint assimilation model of oil transport is established to simulate the spatial and temporal distribution of total petroleum hydrocarbons (TPH) in the Bohai Sea. Because observations in the oceans are often sparse and the oil spill movement still cannot be predicted by using only observation data. The oil transport model which provides scientific basis for oil spill prediction and emergency decision analysis [22][23][24] is an effective tool to simulate and forecast accurate spill trajectories under actual environmental conditions. In previous study, oil transport model has been successfully implemented for simulating oil slick trajectories and the temporal and spatial distribution of pollution [25][26][27]. The adjoint assimilation model can further simulate the real ocean scenario accurately through optimizing control parameters including initial conditions, boundary conditions and empirical parameters, which has been widely applied in oceanography for decades [28][29][30][31][32][33][34][35][36]. For example, Li et al. and Zheng et al. used the transport model to simulate the distribution of pollutants in the Bohai Sea [37,38]. Huang et al. used a model of transport and degradation to estimate the degradation coefficient of petroleum hydrocarbon pollutants [39].
In order to assess the effect of crude oil spill on marine biological community, the abundance of Chaetoceros, fish larvae and shellfish were monitored during June 15-July 13. The objectives of this paper are: (1) clarifying the changes in the abundance of Chaetoceros, fish larvae and shellfish after the oil spill; (2) exploring the relationship between the mortality and TPH; (3) evaluating the biomass loss caused by the oil spill based on radial basis function (RBF) interpolation and oil spill adjoint model.

Biological Observation Data
Based on monitoring result, the dominant species of the phytoplankton is Chaetoceros in the oil spill area. Chaetoceros played a major contribution of the total phytoplankton abundance in the short period after the oil spill [40]. Early life stages of fish are particularly vulnerable to oil spills [41]. In addition, the Bohai Sea is of great commercial importance as one of the major fishing areas in China [42]. Marine shellfish, such as mussels, have been widely used to monitor persistent organic pollutants (POPs) in coastal environments [43]. Therefore, Chaetoceros, fish larvae and shellfish serve as indicators of risk.
The Chaetoceros, fish larvae and shellfish distribution data near the Penglai 19-3 drilling field (120 • 06 E, 38 • 23 N) during 15 June-13 July was provided by the North China Sea Environmental Monitoring Center. The sampling and counting methods of Chaetoceros are vertical trawling sampling and the concentration counting method (The specification for marine monitoring-Part 7: Ecological survey for offshore pollution and biological monitoring (GB 17378.7-2007)). The sampling and counting method of fish larvae are trawling sampling and direct counting method (Specifications for oceanographic survey Part 6: Marine biological survey (GB/T 12763. ). Shellfish was sampled by a bottom sampler and counted by direct counting method (The specification for marine monitoring-Part 7: Ecological survey for offshore pollution and biological monitoring (GB 17378.7-2007)). The sampling stations located in the area with the depth more than 20 m, and the biomass of three organisms were little affected by the tidal movements. The distribution of sampling stations is shown in Figure 1 and the locations of observations is given in Table 1. The oil spill was detected on 17 June 2011, while Chaetoceros and shellfish were observed before and after the oil spill.  ). Shellfish was sampled by a bottom sampler and counted by direct counting method (The specification for marine monitoring-Part 7: Ecological survey for offshore pollution and biological monitoring (GB 17378. ). The sampling stations located in the area with the depth more than 20 m, and the biomass of three organisms were little affected by the tidal movements. The distribution of sampling stations is shown in Figure 1 and the locations of observations is given in Table 1. The oil spill was detected on 17 June 2011, while Chaetoceros and shellfish were observed before and after the oil spill.

Radial Basis Function Interpolation
In order to analyze the effects of the oil spill on the distribution of organisms, the scatted observation data of Chaetoceros, shellfish and TPH is interpolated, with the the grid spacing of 10" × 10". Some interpolation methods, such as the Kriging, Cressman and RBF interpolations, have been proposed to alleviate this issue. Based on various interpolation methods, comparatively accurate and sufficient data are obtained in the field of marine science [44]. The interpolation method in this study is RBF interpolation, which is a well-established tool with high accuracy and numerical stability [45][46][47].
Assuming that there are n i given points, i = 1,2, . . . , m, the unknown point Z sd is calculated by RBF interpolation: where r ij is the Euclidean distance between (x i , y i ) and x j , y j , U r ij is the basis function: Note that, linear radial basis functions are used in this paper: Table 1. Observation information of Chaetoceros, fish larvae and shellfish: location and date.

Fish Larvae Shellfish
Longitude

Oil Transport Model
In this paper, an adjoint assimilation model of oil transport is established to simulate the spatial and temporal distribution of TPH after the oil spill based on the observation data of the oil spill in the Bohai Sea. The observed data used in the model correspond to the period from June 2011 to July 2011, when petroleum hydrocarbon samples were collected on the sea surface in the Bohai Sea ( Figure 2). The model can obtain satisfactory results and reduce the error by adjusting control variables and optimizing simulation results iteratively. The monitoring data of different time and space dynamically is fitted, considering the hydrodynamic process. The model assimilates the monitoring data of different time and space to the same time through the numerical iteration process of the monitoring values.
The model assimilates the monitoring data of different time and space to the same time through the numerical iteration process of the monitoring values. The adjoint assimilation model is based on the rectangular coordinates, where the parameters settings refer that in Wang et al. (2013) [48]. The computing area spans over the domain of 117.5° E~122.5° E and 37° N~41° N, with the grid spacing of 4′ × 4′. The model is divided into five layers vertically, whose thicknesses are 10 m, 10 m, 10 m, 20 m and 25 m from top to bottom, respectively. The time step is set as 1 h.   The model assimilates the monitoring data of different time and space to the same time through the numerical iteration process of the monitoring values. The adjoint assimilation model is based on the rectangular coordinates, where the parameters settings refer that in Wang et al. (2013) [48]. The computing area spans over the domain of 117.5° E~122.5° E and 37° N~41° N, with the grid spacing of 4′ × 4′. The model is divided into five layers vertically, whose thicknesses are 10 m, 10 m, 10 m, 20 m and 25 m from top to bottom, respectively. The time step is set as 1 h.   Considering the convection and diffusion processes, the governing equation of the marine oil transport model is presented as follows: where C denotes the concentration of TPH; t is the time; x, y, z are components of the Cartesian coordinate system in the eastern, northern and vertical direction, respectively; u, v, w are velocities in the x, y, z directions, respectively; A H and K H denote the horizontal and vertical diffusion coefficients, which are set as 100 m 2 s −1 and 10 −5 m 2 s −1 , respectively. The equations of the adjoint model are presented in Appendix A.

Average Abundances of Chaetoceros, Fish Larvae, and Shellfish
The abundances of Chaetoceros and shellfish before and after the oil spill are shown in Figure 4. The average abundance is calculated as follows: where C is the average abundance. C i is the observation abundance at the ith sampling stations. n denotes the number of sampling stations.
Considering the convection and diffusion processes, the governing equation of the marine oil transport model is presented as follows: where C denotes the concentration of TPH; t is the time; x, y, z are components of the Cartesian coordinate system in the eastern, northern and vertical direction, respectively; u, v, w are velocities in the x, y, z directions, respectively; AH and KH denote the horizontal and vertical diffusion coefficients, which are set as 100 m 2 s −1 and 10 −5 m 2 s −1 , respectively. The equations of the adjoint model are presented in Appendix A.

Average Abundances of Chaetoceros, Fish Larvae, and Shellfish
The abundances of Chaetoceros and shellfish before and after the oil spill are shown in Figure 4. The average abundance is calculated as follows: where ̅ is the average abundance. is the observation abundance at the ith sampling stations. n denotes the number of sampling stations.
The average abundances of Chaetoceros and shellfish before the oil spill are 11941 ind/m 3 and 47 ind/m 2 , respectively. While that after the oil spill are 1651 ind/m 3 and 12.1 ind/m 2 , respectively. In addition, the average abundance of fish larvae in June and July is 0.13 ind/m 3 and 0.007 ind/m 3 , respectively. According to the information of observation (Table 1), Chaetoceros and shellfish were observed before and after the oil spill, fish larvae was observed after the spill in June and July. By comparing the abundance before and after the oil spill, it is found that Chaetoceros and shellfish decreased significantly in the short term after the oil spill. The oil pollution also had a negative impact on fish larvae. The effects of oil spills on marine organisms are still unclear and many findings often contradict each other ( Table 2). Most investigations have reported that oil spills had negative effects on planktonic species, while a few investigations reported that spills had no significant effects on plankton. In our study, Chaetoceros, fish larvae and shellfish decreased significantly due to oil pollution, which is consistent with most of the previous reports. Compared with the previous study, for example, Peterson found phytoplankton first decreased and then followed by blooms [9]. Our result is consistent with the "first decreased", because the observation time in this study is only a few days after the oil spill. As for the effects of zooplankton and zoobenthos, both the previous reports and our result showed that they could be affected by oil spills. The average abundances of Chaetoceros and shellfish before the oil spill are 11,941 ind/m 3 and 47 ind/m 2 , respectively. While that after the oil spill are 1651 ind/m 3 and 12.1 ind/m 2 , respectively. In addition, the average abundance of fish larvae in June and July is 0.13 ind/m 3 and 0.007 ind/m 3 , respectively. According to the information of observation (Table 1), Chaetoceros and shellfish were observed before and after the oil spill, fish larvae was observed after the spill in June and July. By comparing the abundance before and after the oil spill, it is found that Chaetoceros and shellfish decreased significantly in the short term after the oil spill. The oil pollution also had a negative impact on fish larvae.
The effects of oil spills on marine organisms are still unclear and many findings often contradict each other ( Table 2). Most investigations have reported that oil spills had negative effects on planktonic species, while a few investigations reported that spills had no significant effects on plankton. In our study, Chaetoceros, fish larvae and shellfish decreased significantly due to oil pollution, which is consistent with most of the previous reports. Compared with the previous study, for example, Peterson found phytoplankton first decreased and then followed by blooms [9]. Our result is consistent with the "first decreased", because the observation time in this study is only a few days after the oil spill.
As for the effects of zooplankton and zoobenthos, both the previous reports and our result showed that they could be affected by oil spills.

Variation in the Abundance of Chaetoceros, Fish Larvae, and Shellfish
Oil pollution results in the massive loss of Chaetoceros, fish larvae and shellfish. Mortality, which is the ratio of the total number of organisms that died in a period of time to the initial value, is used to measure the effect of oil pollution. To explore the relationship between the mortality and TPH, the mortality of Chaetoceros, fish larvae, and shellfish and TPH are fitted through the least-squares method ( Figure 5). After the oil spill, the four-day mortality of Chaetoceros, fish larvae and shellfish ranged from 89.4% to 90.0%, 91.9% to 100% and 47.5% to 95.9%, respectively, indicating that oil pollution caused a negative impact on the growth of marine organisms. The mortality of shellfish is lower than that of Chaetoceros and fish larvae, suggesting that shellfish are less vulnerable to oil pollution as they live in the seabed while Chaetoceros and fish larvae live in surface waters. Furthermore, the mortality of fish larvae is positively correlated with TPH concentration while the mortality of shellfish is negatively correlated with TPH. However, due to the limitation of data, the complicated mechanism behind the negative correlation needs to be further explored.
According to the water quality standards in China, when the concentration of TPH is above 50 mg/m 3 , TPH may cause ecological damage to different species of marine organisms. Calculations showed that the four-day mortality of Chaetoceros is over 90% and shellfish is over 78.9% by calculation when the concentration of TPH exceeds the water quality standards in China, suggesting the TPH has a significant ecological damage. Mortality is the ratio of the total number of organisms that died in a period of time to the initial value. In order to explore the daily variation of the abundance after the oil spill, the daily mortality for Chaetoceros and shellfish are calculated based on the four-day mortality (90%, 78.9%). In short, when the TPH concentration is over 50 mg/m 3 , the daily mortality of Chaetoceros is over 43.8% and shellfish is over 32.3%. by 40% to 70% [52]. This biomass decrease is different from our result, probably because the observation time of data is different.

Spatial Distribution of the Mortality and the Death Rate
The spatial distribution of Chaetoceros ( Figure 6) and TPH ( Figure 7A) are obtained by RBF interpolation, and the spatial distribution of Chaetoceros mortality ( Figure 7C) and death rate ( Figure 7B) are also obtained. According to the interpolation result, the average abundance of Chaetoceros is 12001 ind/m 3 , and the death rate is positively correlated with TPH, with a correlation coefficient of 0.70. However, the correlation between the mortality and TPH is not significant. The equations of mortality and death rate are presented in Appendix B.  Ainsworth et al. observed that the biomass of large reef fish decreased by 25% to 50% in areas most affected by a spill, and biomass of large demersal fish decreased even more, by 40% to 70% [52]. This biomass decrease is different from our result, probably because the observation time of data is different.

Spatial Distribution of the Mortality and the Death Rate
The spatial distribution of Chaetoceros ( Figure 6) and TPH ( Figure 7A) are obtained by RBF interpolation, and the spatial distribution of Chaetoceros mortality ( Figure 7C) and death rate ( Figure 7B) are also obtained. According to the interpolation result, the average abundance of Chaetoceros is 12,001 ind/m 3 , and the death rate is positively correlated with TPH, with a correlation coefficient of 0.70. However, the correlation between the mortality and TPH is not significant. The equations of mortality and death rate are presented in Appendix B. by 40% to 70% [52]. This biomass decrease is different from our result, probably because the observation time of data is different.

Spatial Distribution of the Mortality and the Death Rate
The spatial distribution of Chaetoceros ( Figure 6) and TPH ( Figure 7A) are obtained by RBF interpolation, and the spatial distribution of Chaetoceros mortality ( Figure 7C) and death rate ( Figure 7B) are also obtained. According to the interpolation result, the average abundance of Chaetoceros is 12001 ind/m 3 , and the death rate is positively correlated with TPH, with a correlation coefficient of 0.70. However, the correlation between the mortality and TPH is not significant. The equations of mortality and death rate are presented in Appendix B.   The distribution of shellfish is shown in Figure 8. The average abundance of shellfish is 43 ind/m 2 . The death rate of shellfish is positively correlated with TPH ( Figure 9), with the correlation coefficient of 0.91. While the correlation between the mortality of shellfish and TPH is not significant, which is similar to Chaetoceros.

General Evaluation of the Biomass Loss
The daily mean distribution of TPH is obtained by the adjoint assimilation model. After assimilation, the mean absolute error of concentration decreases from 57.02 mg/m 3 to 28.53 mg/m 3 , with a decreasing ratio of about 50%. According to the water quality The distribution of shellfish is shown in Figure 8. The average abundance of shellfish is 43 ind/m 2 . The death rate of shellfish is positively correlated with TPH ( Figure 9), with the correlation coefficient of 0.91. While the correlation between the mortality of shellfish and TPH is not significant, which is similar to Chaetoceros. The distribution of shellfish is shown in Figure 8. The average abundance of shellfish is 43 ind/m 2 . The death rate of shellfish is positively correlated with TPH ( Figure 9), with the correlation coefficient of 0.91. While the correlation between the mortality of shellfish and TPH is not significant, which is similar to Chaetoceros.

General Evaluation of the Biomass Loss
The daily mean distribution of TPH is obtained by the adjoint assimilation model. After assimilation, the mean absolute error of concentration decreases from 57.02 mg/m 3 to 28.53 mg/m 3 , with a decreasing ratio of about 50%. According to the water quality The distribution of shellfish is shown in Figure 8. The average abundance of shellfish is 43 ind/m 2 . The death rate of shellfish is positively correlated with TPH ( Figure 9), with the correlation coefficient of 0.91. While the correlation between the mortality of shellfish and TPH is not significant, which is similar to Chaetoceros.

General Evaluation of the Biomass Loss
The daily mean distribution of TPH is obtained by the adjoint assimilation model. After assimilation, the mean absolute error of concentration decreases from 57.02 mg/m 3 to 28.53 mg/m 3 , with a decreasing ratio of about 50%. According to the water quality

General Evaluation of the Biomass Loss
The daily mean distribution of TPH is obtained by the adjoint assimilation model. After assimilation, the mean absolute error of concentration decreases from 57.02 mg/m 3 to 28.53 mg/m 3 , with a decreasing ratio of about 50%. According to the water quality standards in China, an area with the concentration of TPH over 50 mg/m 3 is considered to be toxic. The spatial distribution of TPH from June 18 to June 21 indicates that the range of the high concentration area decreases ( Figure 10). standards in China, an area with the concentration of TPH over 50 mg/m 3 is considered to be toxic. The spatial distribution of TPH from June 18 to June 21 indicates that the range of the high concentration area decreases ( Figure 10). Chaetoceros and shellfish were observed pre-spill and post-spill, but fish larvae were only observed after the oil spill. Thus, the data of Chaetoceros and Shellfish were used to evaluate the biomass loss combined with the distribution of TPH obtained by the adjoint assimilation model. According to the water quality standards in China, the mortality of Chaetoceros and shellfish are estimated so as to comprehensively evaluate the loss of Chaetoceros and shellfish during the observation period. Based on the average value of observation data (Chaetoceros: 11,941 ind/m 3 ; shellfish: 47 ind/m 2 ), the total loss of Chaetoceros and shellfish in toxic area are 7.25 × 10 14 ind and 2.51 × 10 5 ind, respectively (Table 3). While, according to the interpolation result, the total loss of Chaetoceros and shellfish are 7.28 × 10 14 ind and 2.30 × 10 5 ind, respectively. The comparison indicates that the total death of Chaetoceros and shellfish calculated based on the interpolation are close to those of the observation data. Combining these two results, the total death of Chaetoceros is 7.25 × 10 14~7 .28 × 10 14 ind and shellfish is 2.30 × 10 12~2 .51 × 10 12 ind during the observation period. Chaetoceros and shellfish were observed pre-spill and post-spill, but fish larvae were only observed after the oil spill. Thus, the data of Chaetoceros and Shellfish were used to evaluate the biomass loss combined with the distribution of TPH obtained by the adjoint assimilation model. According to the water quality standards in China, the mortality of Chaetoceros and shellfish are estimated so as to comprehensively evaluate the loss of Chaetoceros and shellfish during the observation period. Based on the average value of observation data (Chaetoceros: 11,941 ind/m 3 ; shellfish: 47 ind/m 2 ), the total loss of Chaetoceros and shellfish in toxic area are 7.25 × 10 14 ind and 2.51 × 10 5 ind, respectively (Table 3). While, according to the interpolation result, the total loss of Chaetoceros and shellfish are 7.28 × 10 14 ind and 2.30 × 10 5 ind, respectively. The comparison indicates that the total death of Chaetoceros and shellfish calculated based on the interpolation are close to those of the observation data. Combining these two results, the total death of Chaetoceros is 7.25 × 10 14 ∼ 7.28 × 10 14 ind and shellfish is 2.30 × 10 12 ∼ 2.51 × 10 12 ind during the observation period.

Conclusions
Oil spills can cause marine environmental pollution, resulting in a decline of marine ecosystem services function. The Penglai 19-3 oil spill is one of the typical oil spills in the Bohai Sea. The assessment of ecological loss following the oil spill based on observation data is however limited at present. In this paper, based on the observation data of Chaetoceros, fish larvae and shellfish near the platform C of the Penglai 19-3 drilling field, the negative effect of oil spill on biological growth is evaluated, and the damage of biological resources caused by oil spill accident is estimated. The oil spill caused the biomass loss of Chaetoceros, fish larvae and shellfish.
According to the observation data, the average abundances of Chaetoceros and shellfish before the oil spill were 11,941 ind/m 3 and 47 ind/m 2 , and after the oil spill were 1651 ind/m 3 and 12.1 ind/m 2 , respectively. The average abundance of fish larvae in June and July were 0.13 ind/m 3 and 0.007 ind/m 3 , respectively. The abundances of all these three organisms decrease in large quantities in the short term after the oil spill. The four-day mortality of Chaetoceros and fish larvae was around 90%, and that of shellfish was 78.9%, which is lower than that of Chaetoceros and fish larvae. It can be obtained that the daily mortality of Chaetoceros and shellfish were 43.8% and 32.3%, respectively. The spatial distribution of the death rate, mortality and TPH are obtained through RBF interpolation. According to the result, the average abundances of Chaetoceros and shellfish were 12,001 ind/m 3 and 43 ind/m 2 and the death rate of Chaetoceros and shellfish are both positively correlated with TPH. However, the correlation between the mortality and TPH is not significant.
In order to evaluate the total loss of organisms, the temporal and spatial distribution of TPH is obtained by assimilating the monitoring data of TPH through the oil spill adjoint model. Based on the spatial and temporal distribution field, the area within the concentration of TPH over the water quality standards in China (50 mg/m 3 ) is obtained. According to the daily mortality and the reference value (the average abundance before oil spill based on the observation data and RBF interpolation result), the total losses of Chaetoceros and shellfish caused by the oil spill were 7.25 × 10 14 ∼ 7.28 × 10 14 ind and 2.30 × 10 12 ∼ 2.51 × 10 12 ind in the area where the TPH was over 50 mg/m 3 during the observation period.
In this paper, Chaetoceros, fish larvae and shellfish served as the indicators of risk to assess the damage of marine ecological resources caused by an oil spill, which is useful in future environmental monitoring efforts regarding pollution. The results of this study also highlight the necessity of early monitoring and sampling efforts shortly after the occurrence of oil spills. Understanding the alterations to organism community structure which can ultimately affect the higher trophic levels in these ecosystems can permit an improved assessment of the short-term influence of an oil spill on the marine ecosystem.