Contents and spatial distribution of heavy metals in sediments
The contents of As, Cd, Cr, Cu, Pb, and Zn, specified in the China Marine Sediment Quality Standard (GB18668-2002), are listed in Table 1. In the sediments, their contents ranged from 4.02 to 41.34 mg kg−1, 0.05 to 0.81 mg kg−1, 5.06 to 81.87 mg kg−1, 16.41 to 81.48 mg kg−1, 5.41 to 67.21 mg kg−1, and 83.22 to 259.57 mg kg−1 for As, Cd, Cr, Cu, Pb, and Zn, respectively, in the order of Zn > Cu > Cr > Pb > As > Cd. The metals with the highest contents, Zn and Cu, had average contents of 175.80 mg kg−1 and 50.74 mg kg−1, respectively (Fig. 2 and Table 1). The heavy metal contents at all sampling sites met the Class-2 standard of marine sediments. The contents of As and Pb at S2, S4, S5, S8, S9, and S10 were higher than the values of the Class-1 standard, but lower than the values of the Class-2 standard. The Cd content at S1 and S2 was higher than that in the Class-1 standard and lower than that in the Class-2 standard. Only the Cr values at S9 were slightly higher than that in the Class-1 quality standard, which was 81.87 mg kg−1. The contents of Cu at S11, S14, and S15 were lower than that in the Class-1 standard. The Zn contents at S1, S2, S3, S4, S5, S6, S8, S9, and S10 were higher than the Class-1 standard value but lower than the Class-2 standard value.
Dai et al.  continuously monitored the SZB sediments from 2000 to 2007 and their results showed average contents of As, Cd, Cr, Cu, Pb, and Zn in the middle and at the mouth of the bay that were slightly higher than the results of this study. This is believed to be mainly due to the accelerated transfer and shutdown of polluting enterprises in the surrounding areas of the SZB in recent years. The high-tech industry is continuously developing, and the level of heavy metal pollution is continuously decreasing. Analysing the heavy metal content in sediments from the inside to the outside of the bay, it was found that the contents of As, Cd, Cr, Pb, and Zn gradually decreased from the inside to the outside of the bay, while Cu remained at the same level and showed no obvious downward trend. Differences in the spatial distribution of heavy metals are largely related to their different sources or the sediment origin (texture, TOC, etc.). Different pollutants often contain different heavy metals. The possible sources of As, Cd, Cr, Cu, Pb, and Zn in the SZB sediments are the discharge of domestic sewage and industrial wastewater from Shenzhen and Hong Kong into the bay.
The heavy metal contents in the sediments of the SZB were also compared with those of other bay areas and coastal regions. In Table 1, which lists the heavy metal contents obtained in the study area along with those of other regions, it can be seen that the heavy metal contents in the SZB are slightly higher than those reported for the Yellow River, Yangtze River, and Pearl River Estuary. Furthermore, the levels obtained for Zhelin Bay and Jiaozhou Bay sediments were equal or lower. In addition, the heavy metal contents obtained in this study are significantly lower than those of the world’s largest industrialised/urban ports and estuaries (Table 1), such as the Masan Bay in South Korea, Oyster Bay in Australia, Florida Bay in the United States, Izmit Bay in Turkey, and Gironde Estuary in France.
Assessment of heavy metal pollution in sediments
The Igeo index can be used to confirm the pollution level of heavy metals in sediments. As shown in Fig. 3, the Igeo index of heavy metals in the SZB sediments decreased in the order Zn > Cu > As > Pb > Cd > Cr. The average Igeo indices of Zn and Cu were 0.451 and 0.062, respectively, which indicate light pollution levels. The average Igeo indices of the other heavy metals were less than 0, indicating no pollution. For Cr, the Igeo index was less than 0 in the sediments at all sampling points, indicating no pollution. As, Cd, Cu, and Pb showed Igeo indices greater than 0 at some sampling points. For example, the values for As at S10, S9, S8, and S2 indicated medium pollution, and those at S5, S13, and S4 light pollution; Cd exhibited only slight pollution at sampling point S1, with an Igeo index of 0.12. The Igeo of Cu at all sampling points indicated slight pollution, with a maximum value at S9 (0.59), and Pb also showed slight pollution at sampling points S3 (0.54), S5 (0.71), S10 (0.75), S9 (0.84), S8 (0.80), S4 (0.76), and S2 (0.54). Zn showed pollution conditions at all sampling points except S11, S12, S14, and S15, but all were below the moderate pollution level (1 ≤ Igeo < 2). It is worth noting that the sampling points S10, S9, and S8 were polluted by all analysed heavy metals, except for Cd and Cr. Based on the Igeo index, the SZB sediments were mostly polluted by Zn and Cu. Dai et al.  studied the Igeo indices of Pb, Cu, and Zn in the same sediments from 2000 to 2007, and found that the values all ranged between 0 and 1, indicating slight pollution. Moreover, the level of pollution and heavy metal contents were lower at the bay mouth than inner bay, and both decreased gradually from the inside to the outside of the bay.
Heat maps have been widely used in studies of potential ecological risks of heavy metals. The colour scale ranges from blue to red and indicates increasing ecological risks . The toxicity values in the environment are 30, 10, 5, 5, 2, and 1 for Cd, As, Cu, Pb, Cr, and Zn, respectively. The ecological risk of heavy metals in the SZB sediments is shown in Fig. 4. Except for As at S10 (41.3) and Cd at S1 (48.8), the potential ecological risks of the six heavy metals were all lower than 40, indicating low potential ecological risks. The comprehensive potential ecological hazards were found in the following order: RI As = 274.05 > RI Cd = 265.07 > RI Cu = 126.87 > RI Pb = 111.29 > RI Zn = 32.96 > RI Cr = 20.31. The results showed that As and Cd in the SZB sediments pose moderate potential ecological risks, implying that their pollution cannot be ignored; however, the overall potential ecological risks of the other heavy metals were relatively low. The moderate ecological risks of As and Cd are mainly due to their high ecological toxicity values; although their contents are relatively low, the ecological risk value is high. Therefore, special attention should be paid to the monitoring and management of As and Cd. In general, although the Zn contents in the sediments of the entire study area exceeded the background value, its potential ecological risk was only 32.96, which corresponds to a slight pollution level. These results are consistent with those of Zuo et al. , who used the ecological model evaluation method to analyse the heavy metals in SZB sediments in 2006 and showed that Cd had the largest potential ecological hazard coefficient. Comparing different sampling points, the comprehensive and individual ecological risks of heavy metals in the SZB sediments corresponded to the changes in the heavy metal content, and also showed a gradual decrease from inner bay to the outside of the bay.
The distribution of the potential ecological risks showed that: (1) the risk of Shenzhen River Estuary was the lowest, and at a slight level. This is mainly attributed to the fact that the surface sediments in the estuary area are mainly silt and sandy silt with coarse particle sizes that have a small adsorption capacity for heavy metals, which resulted in a low degree of pollution. The concentrations of Cr, Pb, and Cu indicate no pollution. (2) The potential risk to the Shekou zone was at a medium level. Because this area is located in Shenzhen Bay Estuary, the hydrodynamic force was strong, and the surface sediments were easily disturbed; at the same time, this area was far away from Shenzhen city and there was no large river to flow into. Furthermore, the heavy metals that were transported in the water were deposited in the inner bay where the water exchange is weak. (3) The potential risks of the sampling points except Shenzhen River Estuary and Shekou area were all at a high level. These sampling points were all located inside the SZB where the Dasha, Xinzhou, and Fengtang rivers bring a large amount of heavy metals into the bay.
Source apportionment of heavy metals in sediments
The correlation between heavy metal contents can be used to evaluate the potential common source and migration processes of these elements. A high correlation between two elements indicates that they have a common source and similar migration process. The correlation coefficients between the heavy metals in the SZB sediments are shown in Fig. 5. Cr–Pb (0.96), Cu–Zn (0.95), As–Pb (0.85), As–Cr (0.82), Cr–Cu (0.65), and Cr–Zn (0.67) showed a strong significant correlation (p < 0.01), while Pb–Cu (0.58) and Pb–Zn (0.64) showed a significant correlation (p < 0.05). These results suggest that these pairs of metals may have the same pollution sources and migration processes.
CA is widely used to identify pollutant sources of heavy metals in sediments to distinguish natural from anthropogenic contributions (Loska and Wiechula, 2003; . The clusters of six heavy metals obtained from the cluster analysis are presented by the relationship between groups of variables, as shown in Fig. 6, with a lower value on the horizontal axis showing a more significant association. There are three statistically significant clusters, among which cluster 1 only includes the heavy metals Zn that has been identified to have a light pollution degree from the Igeo index. Cluster 2 includes Cu, Pb, and Cr, and can be further divided into two sub-clusters because the distance between Cu and Pb–Cr was relatively long. Sub-cluster 1 includes Cr and Pb, which were shown to have no pollution and a slight pollution degree according to the Igeo index, respectively, Sub-cluster 2 includes Cu, which was shown to have a slight pollution degree according to the Igeo index. Cluster 3 includes Cd and As, which posed moderate potential ecological risks based on the PERI. The CA results are consistent well with the results from the Pearson correlation analysis.
The correlation coefficients among Zn, Pb, Cr, and Cu were all higher than 0.5, indicating that these four elements may have similar sources and migration processes. The correlation coefficients between Zn and Cu, and Pb and Cr in particular were very high, 0.95 and 0.96, respectively, representing an extremely strong correlation. This may be because these elements are all sulfophilic elements; that is, these elements often form less soluble sulfides, which accumulate in water sediments. The correlations between As and Cd and other heavy metals were relatively low, indicating that there may be differences between their sources and those of the other elements in the sediment. For example, the source of As may be related to the transportation of fossil fuels, such as coal, in the SZB .
Heavy metals in sediments can originate from natural pollution sources or from human activities. Using the correlation between Fe and heavy metals, it was possible to distinguish between the natural and anthropogenic sources of the heavy metals. Naturally sourced heavy metals are often significantly correlated with Fe. Huang et al.  found that Pb, Cu, Zn, and Fe were relatively poorly correlated (Pb had a certain correlation with Fe at low concentrations), indicating that these heavy metals are related to anthropogenic pollution. The accumulation of historical heavy metals is an important cause of changes in the sedimentary environment of the SZB. During the rapid economic development of Hong Kong in the 1970s, industries in the northern New Territories developed rapidly, and a large amount of wastewater was discharged directly into the SZB or transported there via Shenzhen River. When entering the bay, the heavy metals were quickly incorporated into the sediments. Furthermore, the period from 1985 to 2000 was an era of economic rise in Shenzhen and other parts of the Pearl River Delta. A large amount of industrial wastewater, especially wastewater from electronics and electrical appliances, was polluted with heavy metals, which were directly discharged into the SZB. Heavy metals were further transported to the sea through the mouth of the SZB via the runoff and tidal currents of the Pearl River Estuary [25, 26].
Effect of land reclamation on heavy metals in sediments
Coastlines are formed through long-term evolution under various dynamic factors and are in a state of relatively dynamic equilibrium. Land reclamation activities are processes of creating new, dry land on the seabed, which changes the natural coastal pattern in a short time and on a small scale, and have a strong impact on the marine environment, causing an imbalance. Furthermore, the physical and chemical properties of the reclamation material itself may have a long-term impact on the heavy metal content in bay sediments.
Massive changes have occurred both on land and in the seawater in the SZB area since 1980. By 2005, the reclaimed land area had reached 1963 ha (Additional file 1: Figure S3). After 2005, however, no large-scale reclamation activities were carried out . Nevertheless, land reclamation disrupted the dynamic balance of the original coastline, reduced the SZB area, led to a rapid decline in the tidal volume and tidal cycle of the bay, and created favourable conditions for sedimentation. The exchange capacity of the seawater was greatly reduced, resulting in pollutants remaining in the bay, especially the inner bay, for a long time. Some studies have shown that the average residence time of pollutants in the inner bay is approximately 10–14 days in the dry season, 8–9 days in the rainy season, 7–8 days near the Shenzhen–Hong Kong Western Corridor, and only 3–4 days in the outer bay . In this case, the ability of seawater to dilute pollutants in upstream or coastal areas was greatly reduced, causing water pollution. Because the rivers entering the bay carry a large amount of sediment, the longer the pollutants remain in the water, the greater the amount of pollutants deposited, which causes the pollution of sediments to increase. From 1940 to 2010, heavy metal concentrations were not significantly affected by land reclamation activities and remained stable (Fig. 7). If land reclamation activities created favourable conditions for sedimentation still needs to be answered, however, the heavy metal content has remained stable. In fact, the amount of heavy metal pollution entering into the SZB gradually decreased since the 1980s, and if there were no land reclamation activities, the heavy metal contents in the sediments would have decreased. Thus, the proportion of heavy metals entering the sediments actually increased due to the weak hydrodynamic conditions in the bay and, eventually, the heavy metal contents become relatively stable. Reports indicate that harmful substances (heavy metals and organic toxins) in the form of pollutants accumulate in water bodies or sediments, causing environmental disasters in the mangrove ecosystem along the SZB. Wang et al. (2010) reported that the release of sediment in the inner SZB had an impact of 25% on the water quality; the release of sediment pollution in the estuary bay also accounted for 8%. The pollution load in the sediments of the inner bay exceeded the capacity of the inner bay. With the reduction of land-based pollution from Shenzhen and Hong Kong, this proportion will continue to rise.
It is generally recognised that the filling materials used in the process of land reclamation may affect the heavy metal content of sediments. Chen and Jiao  conducted a chemical analysis of the reclamation materials used in a certain reclamation project in the SZB and showed that their heavy metal contents were much lower than those in the sediments (Table 2). Accordingly, it is reasonable to assume that the heavy metal contribution of the filler material to the SZB sediments is negligible. Therefore, it can be concluded that the heavy metals contents in the SZB sediments are probably related to historical land-source pollution from both sides of the bay, i.e., Shenzhen and Hong Kong.