Introduction

Large quantities of elements and compounds are discharged into coastal seas as contaminants each year (Gao et al. 2014). Heavy metal pollution poses a serious threat to coastal and marine ecosystems as well as to the inhabitants due to its persistence, bioaccumulation and ecological damage (Gao and Chen 2012; Huang et al. 2013; Pekey 2006; Wang and Wang 2007; Xu et al. 2014). Heavy metals in the sea originate from both natural processes and anthropogenic activities. Geological and geochemical processes like tectonic activities and rock weathering processes set the geochemical background values for heavy metals (Zhang et al. 2003a). In coastal zones, metals enter the sea through atmospheric deposition of metal-contained dust from burning of fuels (Meng et al. 2008; Zhan et al. 2010; Zheng et al. 2008), surface runoff, and industrial and domestic sewage discharges (Meng et al. 2008), with the last being the most important (He et al. 2009).

Sediments can act both as source and sink for heavy metals (He et al. 2009; Meng et al. 2008), providing a record of possible environmental changes caused by anthropogenic activities (Jiang et al. 2014; Qiao et al. 2013). The status of marine sediments is a significant criterion in the evaluation of the condition of aquatic systems (Meng et al. 2008; Xu et al. 2014; Zhan et al. 2010). Heavy metal concentrations in sediments are influenced by many factors including hydrodynamic condition (Yuan et al. 2012), sediment type (according to percentages of different-sized particles), sedimentation rate due to terrestrial input and hydrodynamic condition (Gao and Chen 2012; He et al. 2009; Teng et al. 2002), total organic carbon (TOC) (Gao and Chen 2012; He et al. 2009; Yuan et al. 2012; Zhan et al. 2010), the existing forms of heavy metals (He et al. 2009), and biological factors (Caçador et al. 1993).

The Bohai Sea is a nearly enclosed inland sea located in northeast China (Fig. 1a) and consists of Liaodong Bay, Bohai Bay, Laizhou Bay, and the central Bohai Sea (CBS) (Fig. 1b) (Gao et al. 2014; Wang et al. 2010). It has a total area of approximately 7.7 × 104 km2 and a water volume of 1.7 × 103 km3. The mean depth is 18 m, with the deepest region of about 70 m located near the northern coast of the Bohai Strait (Gao et al. 2014). There are more than 40 rivers flowing into the Bohai Sea, among which the Yellow River, Haihe River, Luanhe River, Shuangtaizihe River and Liaohe River are the five major ones (Gao et al. 2014; Wang et al. 2010). The amount of particulate matter inputted from the rivers into the Bohai Sea is about 1.3 × 109 t·year−1, and the sedimentation rate of the Bohai Sea is about 1.0–44.2 mm·year−1 (Li et al. 1995). Due to intensive urbanization and industrialization, the Bohai Sea with its nearby coastal areas and estuaries has been facing severe metal pollution problems (Xu et al. 2013, 2014). The bulletins released by the State Oceanic Administration, P.R. China (SOA 2012-2014) indicated that the polluted areas of the Bohai Sea in 2012 (13,080 km2) and 2013 (8490 km2) were significantly larger than that in 2011 (4210 km2). The main polluted areas were Liaodong Bay, Bohai Bay, and Laizhou Bay.

Fig. 1
figure 1

Map showing a the location of the Bohai Sea, b the sampling sites during summer in 2011, and c the sampling sites during winter in 2011 and the locations of the Luanhe Estuary and the Yellow River Estuary

Full size image

Many studies have been conducted on heavy metals in surface sediments in the Bohai Sea, mainly in the three bays, but there have been few studies in the CBS. These bays were more polluted than the CBS (Gao et al. 2014; SOA 2012-2014) due to more intense port transportation, more sewage discharge, more industrial activities, and slower water-exchange rate inside the bays. Therefore, it may overestimate the contamination status of the whole Bohai Sea with ignorance of the CBS. The concentrations of Mn, Fe, and Co and their seasonal variation in the CBS have also been rarely studied. The objectives of this paper are: (1) to measure the concentrations of Pb, Cd, Cu, Fe, Ni, Mn, Zn, and Co in sediments collected from the CBS, China, (2) to analyze the relationships between environmental variables and heavy metals, and (3) to assess pollution status of heavy metals by three methods.

Materials and Methods

Study area and sample collection

Two cruises were conducted onboard the vessel “Dongfanghong II” in the CBS, between 37°43′–39°38′N and 118°58′–120°55′E, from the 13th to 30th June, 2011 (summer cruise) and from the 20th November to 7th December, 2011 (winter cruise). Surface sediments were collected from 16 sites in summer (Fig. 1b) and 13 sites in winter (Fig. 1c) using a 0.1 m2 stainless steel box-corer. The top 3 cm of sediment samples were sealed in polyethylene bags, and refrigerated at −20 °C.

Sample analysis

Heavy metals analysis

After transportation to the laboratory, samples were de-frosted and air-dried at room temperature (20 °C) to constant weight, then ground and homogenized with a glass mortar and pestle. The dried samples were then passed through a nylon sieve of 160 meshes (96 μm) and stored in polyethylene bags. The pretreatment of sediment were according to Jiang et al. (2014). Eight heavy metals (Pb, Cd, Cu, Fe, Ni, Mn, Zn, and Co) in sediments were measured using an inductively coupled plasma-mass spectrometer (ICP-MS, Optima 2100 DV ICP System, Perkin Elmer Inc., USA). The instrument detection limits of metals are as follows: Pb, 0.0176; Cd, 0.0029; Cu, 0.0033; Fe, 0.0036; Ni, 0.0054; Mn, 0.0048; Zn, 0.0042; Co, 0.0048 mg L−1, respectively. All samples were analyzed in duplicate, and the data shown herein represent the average of the duplicates.

Quality assurance and quality control (QA/QC) was assured by the analysis of a marine sediment reference material, GBW-07314 (the Second Institute of Oceanography, SOA). The materials used during analytical determinations were kept in metal-free containers. The plastic and glassware were pre-cleaned by soaking in HNO3 (v:v = 1:3) for at least 24 h, followed by soaking and rinsing with de-ionized water. All reagents used in the analysis were of analytical grade.

TOC analysis

The water content of sediment was measured as a percentage of weight loss by drying the sediment at 60 °C for 24 h. The TOC content was determined by the (K2Cr2O7–H2SO4) oxidization method (SOA 2007; Mudroch et al. 1997).

Sediment type and grain size analysis

Sediment granulometry was analyzed using a Mastersizer 2000 laser diffractometer (Malvern Instrument Ltd., UK) capable of analyzing particle sizes between 0.02 and 2000 μm. The percentages of the following three groups of grain sizes were determined: <4 μm (clay, Y), 4–63 μm (silt, T), and >63 μm (sand, S). The naming of the sediment types was based on Shepard (1954).

Pollution assessment of heavy metals

Marine sediment quality (GB 18668-2002)

China national standard–marine sediment quality, GB 18668-2002 (SOA 2002), is widely used to judge the potential risks of metals in marine sediments. This standard classifies the marine sediments into three classes based on the area’s function and protection target. The first class quality area is suitable for mariculture, nature reserves, leisure activities, and industries that relate to human consumption directly; the second class quality area is used for general industry and tourism; the third class quality area is applied to sea ports and special ocean exploration.

The geoaccumulation index

The geoaccumulation index (I geo ) is defined by Eq. (1) (Müller 1969):

$$I_{geo} = \log_{2} \left( {C_{n} /kB_{n} } \right)$$
(1)

where C n is the measured concentration of the metal (n) in the sediment, and B n is the geochemical background value of the same metal. Factor k is the background matrix correction factor due to lithospheric effects, which is usually defined as 1.5 (Ghrefat et al. 2011; Müller 1969; Rubio et al. 2000; Shafie et al. 2013; Xu et al. 2014). Samples may be classified as:

Class 0 (uncontaminated): I geo  ≤ 0;

Class 1 (uncontaminated to moderately contaminated): 0 < I geo  ≤ 1;

Class 2 (moderately contaminated): 1 < I geo  ≤ 2;

Class 3 (moderately to heavily contaminated): 2 < I geo  ≤ 3;

Class 4 (heavily contaminated): 3 < I geo  ≤ 4;

Class 5 (heavily to extremely contaminated): 4 < I geo  ≤ 5;

Class 6 (extremely contaminated): I geo  > 5.

Ideally, B n of heavy metals should be estimated from local sediments (Teng et al. 2002; Feng et al. 2011). The background values of Cu, Pb, Zn, and Cd in the CBS were 22.1, 14.0, 65.2, and 0.088 μg g−1, respectively (Li et al. 1995). However, in the absence of the background values of Fe, Ni, Mn, and Co in the CBS, the background values of the adjacent region could be effectively substituted (Feng et al. 2011). For this reason, the B n of the Bohai Bay sediments were chosen, i.e. 3.62 % for Fe, 36.1 μg g−1 for Ni, 571.0 μg g−1 for Mn, and 13.2 μg g−1 for Co (Wu and Li 1985; Feng et al. 2011; Xu et al. 2014).

Hakanson potential ecological risk index

Hakanson potential ecological risk index (RI) evaluates the combined pollution risk of an aquatic system through a toxic-response factor for a given substance (Hakanson 1980). RI is defined as follows (Eqs. (2) and (3)):

$$E_{r}^{i} = T_{\text{r}}^{i} \times C^{i} /C_{n}^{i}$$
(2)
$$RI = \sum\limits_{i = 1}^{\text{n}} {E_{r}^{i} }$$
(3)

where T ir is the toxic-response factor for a given substance, e.g., Pb = Cu = Ni = Co = 5, Mn = Zn = 1, Cd = 30 (Hakanson 1980; Xu et al. 2008, 2014), but T ir -Fe was unavailable. C i and C i n represent the concentration of metals determined in the sediment samples and the geochemical background values of metals, respectively. The value of C i n is as same as B n above. E i r and RI represent the monomial and sum potential ecological risk factor, respectively. Samples could be classified as:

Low ecological risk: E i r  < 40, RI < 150;

Moderate ecological risk: 40 ≤ E i r  < 80, 150 ≤ RI < 300;

Moderate to high ecological risk: 80 ≤ E i r  < 160;

High ecological risk: 160 ≤ E i r  < 320, 300 ≤ RI < 600;

Very high ecological risk: E i r  ≥ 320, RI ≥ 600.

Analysis software

Surfer 8.0 (Golden Software Inc., USA) was used for drawing the distribution maps of the TOC content and metal concentrations in surface sediments. Microsoft excel 2010 (Microsoft Inc., USA) was used for drawing histograms and scatterplots. Statistical analyses were conducted with SPSS 18.0 software (SPSS Inc., USA).

Results and discussion

Environmental variables

The TOC content varied between 0.02 and 1.09 % of the dry sediment weight in summer and between 0.15 and 1.08 % in winter. The spatial distributions of the TOC are shown in Fig. 2. The higher content of TOC in summer (Fig. 2a) was generally found in the west of the CBS, decreasing coastward. The content of TOC in winter (Fig. 2b) was higher in the northwest of the CBS, near the Luanhe Estuary. As shown in Fig. 3, the sediments were predominantly composed of silt. The sediment types in the CBS have been classified into four groups: S-sand, ST-sandy-silt, TS-silty-sand, and YT-clayey-silt, according to Shepard (1954). The numbers of sites of those four groups were 2, 2, 4, and 8 in summer, and 0, 4, 2, and 7 in winter, respectively. The type YT accounted for about 50 % of the sites in both seasons.

Fig. 2
figure 2

The spatial distribution of the TOC concentrations in surface sediments of the CBS in summer (a) and winter (b). The dots represent the sites (similarly hereinafter)

Full size image
Fig. 3
figure 3

The grain size compositions in surface sediments of the CBS in summer (a) and winter (b)

Full size image

Spatial distribution of heavy metals and influencing factors

Spatial distribution of heavy metals

The metal concentrations in surface sediments collected from all sites in the CBS were ranked in decreasing order, as follows: Fe > Mn > Zn > Ni > Cu > Pb > Co > Cd in summer and Fe > Mn > Zn > Cu > Ni > Pb > Co > Cd in winter (Table 1). To compare the concentrations of these eight metals between summer and winter, an independent-sample t test was conducted and the details are represented in Table 1. There were no significant differences among them between the two seasons (p > 0.05) with exception of Cd (p < 0.01). The concentration of Cd in summer was significantly higher than that in winter.

Table 1 Average concentrations (Mean ± SD) and details about the t test of the metals
Full size table

The spatial distributions of eight heavy metals are shown in Figs. 4 and 5. The distribution patterns of individual metals were as follows: (1) Fe, Ni, Zn, and Co: Higher concentrations were generally found in the northwest of the CBS, near the Luanhe Estuary. The concentrations decreased in northwest-to-southeast direction; (2) Pb and Cu: The distribution patterns were similar to (1). Another high level of Pb in both seasons and Cu in summer was found in the southwest of the CBS, near the Yellow River Estuary; (3) Cd: The distribution was similar to (1) in summer, and similar to Pb in winter; and (4) Mn: Unlike any other heavy metals, the higher concentrations of Mn were found in north of the CBS. The Site B41 in summer showed another distinct peak situated at 38.33° N and 120.20° E.

Fig. 4
figure 4

Spatial distribution of heavy metals in surface sediments from the CBS in summer, 2011

Full size image
Fig. 5
figure 5

Spatial distribution of heavy metals in surface sediments from the CBS in winter, 2011

Full size image

The distribution of Mn showed a distinct pattern with other metals in the sediments of the CBS. This phenomenon has also been found in many other studies (Jiang et al. 2014; Li et al. 2012; Yuan et al. 2012). The concentration of Mn may not be derived from industrial activities but from geological processes (e.g., Li et al. 2012). The dissolved Mn tends to be adsorbed on, or incorporated into freshly precipitated CaCO3 in seawater (e.g. Wartel et al. 1990). According to V. M. Goldschmidt (Zhang et al. 2003b), heavy metals belonging to the same group have similar chemical properties.

Many studies have been conducted on heavy metals in surface sediments throughout the world. The range and mean concentrations of metals in sediments of the Bohai Sea and other seas are summarized in Table 2. Heavy metal concentrations (except for Cd) in the CBS were comparable to others observed in the Bohai Sea or other sea areas. The contents of Cd in summer were markedly higher than the majority of the previous studies.

Table 2 Comparison of heavy metal concentrations in surface sediments of seas
Full size table

Statistical analysis

The relationships between every pair of heavy metals, TOC, sand %, silt %, and clay % in both seasons were analyzed and the Pearson correlation coefficients are shown in Table 3. In summer, Cd and Mn showed no significant correlation with other variables except Cd-Mn pair. Six other heavy metals showed highly significant positive correlations between each other. Clay percentage presented a significant or highly significant positive correlation with those six metals. TOC had significant positive correlations with Cu, Ni, and Co. The relationships between these variables in winter were similar to summer. Unlike the situation in summer, Cd had a highly significant positive correlations with Pb, Cu, Fe, and Ni and significant positive correlations with Zn and Co in winter. Clay percentage presented highly significant positive correlations with all heavy metals except Mn. Sand percentage and silt percentage seldom had significant correlations with heavy metals in either season.

Table 3 Correlation matrix for metals, TOC, sand %, silt %, and clay % in surface sediments from the CBS in summer and winter, 2011
Full size table

Positive correlation between heavy metals suggests that those metals have common sources (Facchinelli et al. 2001) and identical behavior during transportation (Bastami et al. 2014). The concentration of Mn showed very weak correlations with the concentrations of the other metals, indicating that Mn had different sources from others. This is consistent with the discussion above about the distribution of Mn.

Grain size is one of the controlling factors affecting natural concentrations of heavy metals in sediments (Gao and Chen 2012; He et al. 2009; Jiang et al. 2014; Yuan et al. 2012). Sediment of smaller grain size such as clay tends to adsorb more heavy metals due to higher surface area-to-volume ratio (Jiang et al. 2014; Yuan et al. 2012). For this reason, the heavy metal content increased with increasing percentage of clay in sediment. TOC is another factor that controls heavy metal concentrations in sediment (Gao and Chen 2012; He et al. 2009; Jiang et al. 2014; Yuan et al. 2012; Zhan et al. 2010). Natural organic matter shows a high affinity through adsorption or complexation of heavy metals in the aquatic environment (El Bilali et al. 2002). The TOC in sediment was adsorbed onto clay mostly (He et al. 2009), accounting for the positive correlations with clay percentage as shown in Table 3. For this reason, the percentage of clay increased with TOC, leading to the increased heavy metal content. It is noteworthy that the TOC seldom showed significant correlation with heavy metals in summer (Table 3). When several sites were excluded, as shown in Fig. 6, the TOC correlated well with heavy metals (except Cd and Mn) in summer. The concentrations of heavy metals at Sites B49 and B70 (adding B65 for Pb) in summer were somewhat higher than those in adjacent regions, suggesting the metal concentrations at those sites were probably influenced by anthropogenic activity.

Fig. 6
figure 6

Relationship between the TOC and heavy metal concentrations in the sediments from the CBS in summer, 2011

Full size image

Pollution assessment

Comparison of marine sediment quality

Table 2 shows the upper limits of Pb, Cd, Cu, and Zn of the first, second, and third class of marine sediment quality according to GB 18668-2002. As shown in Table 2, the Pb, Cu, and Zn concentration at all sites in both seasons and the Cd content in winter was below the upper limit for Class I sediment. That means all sampling sites in winter were clean enough to be classified as Class I marine sediment quality. In summer, however, the Cd concentration at sites 11, 4, and 1 was within the range for Class I–II, Class II–III, and above Class III threshold values, respectively. Site B50 was heavily contaminated in summer with the Cd concentration (7.28 μg g−1) in excess of Class III. According to SOA (2012–2014), with 6.25 % of sites having bad quality in summer, the CBS was moderately contaminated.

The geoaccumulation index

As shown in Table 4, the I geo values (Müller 1969) of Cu, Fe, Ni, Zn, and Co at all sites and of Pb and Mn at most sites were below zero, indicating the CBS was uncontaminated by those metals. The I geo values of Cd indicated the CBS was moderately (2 < I geo  ≤ 3) to extremely contaminated (I geo  > 5) in summer and uncontaminated (I geo  ≤ 0) to moderately contaminated (0 < I geo  ≤ 1) in winter. The pollution status of the CBS in summer was more severe compared with winter. In summer higher values of I geo -Cd were found in the northwest of the CBS, especially at Site B50 located near the Luanhe Estuary. It is suggested that anthropogenic sources from the Luanhe Estuary might be the main contributor of Cd at Site B50.

Table 4 The values of I geo , E i r and RI in surface sediments in the CBS in summer and winter, 2011
Full size table

Hakanson potential ecological risk index

The E ir and RI values (Hakanson 1980) of sediments of the CBS are shown in Table 4. In summer, 81.25 % of sites were at very high ecological risk of Cd (E i r -Cd > 320). In winter, the E i r -Cd at 4, 5, and 4 sites were low, moderate, and moderate to high ecological risk, respectively. The E i r values of the other seven metals at all sites were below 40 in both seasons. Because of the higher T ir (T ir -Cd = 30) and relatively higher concentration of Cd compared to the background value, Cd was the priority pollutant in the CBS. In summer, the RI value at Site B50 was higher than 600, indicating a high ecological risk of metals. The RI values at 14 sites were above 300, which meant considerable ecological risk from metals. In winter, 84.62 % of sites were at low ecological risk (RI < 150). According to the values of E i r and RI, the pollution status of the CBS in summer was considerably worse than that in winter.

The results of the three methods were generally similar. The main contaminated zone was near the Luanhe Estuary. Cd was the most important pollutant in comparison with other heavy metals in the CBS. The pollution status of the CBS in summer was worse than that in winter. River runoff and anthropogenic activities are suggested as the main pollution sources in China Seas (Liu and Diamond 2005). The Bohai Sea Economic Rim, the coastal region surrounding the Bohai Sea, is one of the three most densely populated and industrialized zones in China. Due to the shallow depth, narrow connection with the outer ocean and slow water-exchange rate, the self-cleaning capacity of the Bohai Sea is very limited (Dang et al. 2013). The ecosystem of the Bohai Sea is being rapidly degraded (Gao et al. 2014). The concentrations of heavy metals in sediment may affect the heavy metal content in organisms and the bioaccumulation capacity of heavy metals by organisms (Gao et al. 2014; Liang et al. 2004). Liang et al. (2004) found that the concentrations of Cd, Cu and Zn in gastropod and oyster samples in the Bohai Sea exceeded the MPLs established by WHO (1989).

Each of the three methods has its own advantages and disadvantages. It is simple to compare the contents of the heavy metals to marine sediment quality. In terms of the concentrations of the heavy metals, this standard can classify the studied area into three classes with different function and protection target. However, the sediment qualities for many metals such as Ni, Mn, Co are not available in the standard. The geochemical background values in a certain area were not considered yet. The geoaccumulation index is used widely with the simple principle and formula, so it is convenient for comparison to previous studies. The background values of heavy metals are considered in the geoaccumulation index. Different from the above two methods, Hakanson potential ecological risk index is more complex. It is recognized as useful for the risk management of trace elements in sediments because sediment adverse biological effects and potential ecological risks are considered in Hakanson potential ecological risk index. To reduce the potential for bias and to better characterize the likelihood for adverse environmental effects in the studied sea area, the application of several metrics for pollution assessment is advisable.

Conclusions

To effectively manage potential environmental and biological impacts of contaminated sediments in a semi-enclosed sea, in the present study, the concentrations of eight metals in sediments from the CBS were measured with ICP-MS and pollution status was assessed with three methods. Higher concentrations of heavy metals (except for Mn) were generally found in the northwest of the CBS, near the Luanhe Estuary. The TOC and clay percentage were the main controlling factors affecting the distribution of heavy metals. According to the pollution assessment, the CBS was moderately contaminated in summer and uncontaminated in winter, 2011. The main contaminated zone was near the Luanhe Estuary, and Cd was the main contaminant compared to others in the CBS. Anthropogenic sources from the Luanhe Estuary might be the main contributor.

Compared the three pollution assessment methods, marine sediment quality and the geoaccumulation index methods are much simpler for calculation but they ignore the biological impacts of heavy metals in the sediments. Hakanson potential ecological risk index is a better method because biological effects and potential ecological risks are both considered. The present study provides researchers and policy makers with basic data to develop a better understanding of the geographic distribution and potential risks of metals in the CBS. In future, more background values from local sediments are needed to meaningfully assess pollution status. The combination of several pollution assessment methods is suggested in the risk management of trace metals in sediments to obtain comprehensive and accurate results.