Comparison of metals in eelgrass (Zostera marina L.) and the environment across the North Pacific Ocean: adaptability to a range of stressors

Abstract

Seagrass beds provide critical ecosystem services, including maintaining biodiversity, providing fisheries nursery habitat, as well as carbon and sediment sequestration. These habitats are subject to dissolved and particulate material inputs such as nutrients and heavy metals which can impact the plants and ecosystem health. Here, we present heavy metals concentration data (Cr, Cu, Zn, Cd, and Pb) from eelgrass (Zostera marina L.), sediments, and water at twenty sampling sites along the temperate North Pacific coasts of Asia and North America. We evaluated the bioaccumulation of heavy metal in eelgrass tissues, identified the potential sources of metals contaminants, and identified the potential exposure stress of heavy metals on eelgrass. Our results show that leaf tissues had higher Cd, Zn, and Cu concentrations than the belowground tissues while belowground organs had higher concentrations of Cr and Pb than were found in leaf tissues. There was a significant positive correlation between Cd concentrations in aboveground tissues and sediment, suggesting that leaf tissue might be a proxy for sediment Cd. Human activities are likely a large contributor of heavy metals to the marine environment in Asian systems, while oceanographic processes (e. g. parent rock weathering and upwelling) are important sources of metals in North America Pacific estuaries. High sediment Cr concentrations in North America Pacific estuaries maybe a cause for future concern about aquatic organisms. This study provides baseline data on heavy metal concentrations in seagrass beds in the North Pacific Ocean and provides critical data to support development of a formal ecological risk assessment of heavy metals in seagrass beds for the protection and management of seagrass beds in Asia and the USA.

1. Introduction

Heavy metals are a group of elements with density greater than 5 g cm⁻³ (Li et al., 2022). At high concentrations, some can be toxic and threaten human and ecosystem health throughout the aquatic food chain (Järup, 2003; Nagajyoti et al., 2010; Hou et al., 2020). These elements can be concentrated in the estuarine environment by natural and anthropogenic processes. In highly urbanized areas, heavy metal contamination may be a result of a variety of industrial, domestic, and agricultural processes (Hou et al., 2020). For example, over the last fifty years global production of Cr and Pb increased by 514% and 232%, reaching 37.5 Mt and 11.3 Mt per year, respectively (British Geological Survey, 2019). Depending on their role in cellular biochemistry, heavy metals can be classified as nonessential or essential (Antonovics et al., 1971; Schneider et al., 2013). For example, Cu and Zn are essential metals because of their role in cellular biochemistry but can impair plant growth and photosynthesis at excess concentrations (Sánchez-Quiles et al., 2017; Qiao et al., 2022). Nonessential heavy metals such as Cr, Cd, and Pb can impair plant functions even at very low levels (Bonanno et al., 2017; Hu et al., 2018). Consequently, understanding the accumulation, biological amplification, potential toxic impacts of heavy metals, and their environmental sources is an important step toward meeting marine management goals of seagrass conservation and restoration.

Coastal environments accumulate heavy metals from natural and anthropogenic processes. Natural sources include weathering or erosion of rocks and soils and via oceanographic processes (Batley, 1995; Ralph et al., 2006; Cassis et al., 2011; Shiel et al., 2012) and anthropogenic sources include industrial activities, sewage discharge dredging, and aquaculture, which also contribute mobilized metals to the environment (Haynes and Johnson, 2000; Govindasamy et al., 2011). Coastal and estuarine sediments are the primary reservoir of heavy metals in coastal environments (Wang et al., 2002; Liu et al., 2018). In areas with aquaculture, rapid industrial development, and intensive terrestrial farming, wastewater discharge with high levels of heavy metals to the environment is a concern to mariculture operations (Wang et al., 2015). In Asia, the aquaculture industry tends to utilize mariculture ponds, while in North America (NA) aquaculture is primarily focused on shellfish (e.g., oyster, clam) grown in natural waters. In upwelling favorable regions, the advection of nutrient- and metal-rich deep water into coastal estuaries can result in high concentrations of some heavy metals such as Cd and Zn (Cassis et al., 2011; Shiel et al., 2012). This natural accumulation of metals can be exacerbated by local anthropogenic activities. In many coastal environments, hot spots for heavy metal accumulation include habitats that slow water motion and promote particle settling such as seagrass beds, salt marshes, and mangrove forests. Metals in seagrass beds are much less studied than salt marsh and mangrove habitat. Furthermore, most studies of metals in seagrass beds are focused on a limited geographic location such as a specific embayment (Li et al., 2022). Seagrass beds are an important estuarine habitat and they are among the most productive plant communities and provide critical ecosystem services, such as biodiversity, nursery habitat for fisheries species, and carbon sequestration (Costanza et al., 1997; Orth et al., 2006; Cullen-Unsworth and Unsworth, 2013; Nordlund et al., 2016). Seagrasses can take up both essential and nonessential metals from the sediment through their roots or from the water via their leaves (Richir et al., 2013; Lin et al., 2016) and have been used as bioindicators of environmental heavy metal contamination (Brix et al., 1983; Lafabrie et al., 2007; Luy et al., 2012; Richir and Gobert, 2014; Conti et al., 2015).

When heavy metals accumulate in seagrass tissues, the plants generate free radicals, which causes oxidative stress and inhibits metabolic activity, including interference with photosynthetic pathways eventually inhibiting plant growth and development (Ralph and Burchett, 1998; Prange and Dennison, 2000; Nagajyoti et al., 2010; Signa et al., 2017). Toxic metal thresholds have been developed for several seagrass species and a few metals including Cu and Cd in Zostera marina (Qiao et al., 2022) and Ruppia sinensis (Gu et al., 2021), and Pb and Zn in Halophila ovalis (Ralph and Burchett, 1998). However, toxic thresholds for most seagrass species and heavy metals have not been developed.

Two tolerance strategies have been hypothesized for seagrasses exposed to heavy metals in the environment. Bonanno and Di Martino (2016) hypothesize that toxic effects can be prevented by sequestering metals in vacuoles in roots and rhizomes essentially keeping them from interfering with photosynthetic pathways or by concentrating metals in tissues that turn over rapidly such as leaves. Leaf abscission removes metals from the plants but makes the bioaccumulated metals available for export or to other trophic levels (e.g., microbial loop).

Zostera marina L., commonly known as eelgrass, is widely distributed along the Pacific and Atlantic coasts in the temperate northern hemisphere (Green and Short, 2003). Eelgrass is a dominant nearshore seagrass species in the North Pacific Ocean, found in the coastal waters of northern China (Zheng et al., 2013; Xu et al., 2021a), the Korean Peninsula (Lee et al., 2002), and North America (Green and Short, 2003). To date, there have been few studies of heavy metal in Z. marina beds in the North Pacific Ocean, and none comparing metals concentrations in eelgrass across the Pacific basin (Hu et al., 2019). Therefore, there is limited data to inform an ecological risk assessment of toxic metals in seagrass beds at the basin scale (Richir and Gobert, 2014; Lin et al., 2016).

Here, we present measured heavy metal concentrations and exposure stress for Z. marina beds in the North Pacific Ocean. Our sampling sites span a gradient of anthropogenic environmental impacts from pristine marine reserves to heavily impacted urbanized areas with industry. This study aimed to: (i) investigate the heavy metal concentrations in Z. marina, sediment, and surface seawater in eelgrass beds in the North Pacific Ocean; (ii) measure the bioaccumulation of heavy metals in different tissues of Z. marina; (iii) identify potential sources of metals contaminants and identify the exposure stress of heavy metals in this geographic bioregion. Our work will provide baseline data that will be useful when conducting a formal ecological risk assessment, and reference data necessary for managing and protecting seagrass beds.

2. Materials and methods

2.1. Study sites and seagrass sampling

The present study was conducted between 2017 and 2021 in the north Pacific coasts of Asia and North America. We sampled twelve sites in the Yellow and Bohai Seas of China, one site in South Korea in the Sea of Japan, and seven sites along the Pacific Coast of North America, USA (Figure 1 and Table 1). In China, the Huludao region in the Bohai Sea has a zinc smelter and heavy ship building industries, the Rongcheng region has less industry but more mariculture while Swan Lake and the Changshan Islands are less developed. Similarly, the NA study sites fall roughly into two categories, fairly pristine areas (Neta, Coos, and Padi) and systems with mariculture (Will and Humb). Seagrass, sediment, and seawater samples were collected at all sites, except that water samples were not collected at Koje Bay, Willapa Bay, Netarts Bay, Yaquina Bay, and Coos Bay (Table 1).

Figure 1.

Sampling sites. Twelve sites were selected from the Yellow and Bohai Seas in China, including Juehua Island (Jueh), Xingcheng (Xing), Xiaohaishan Island (Xiao), Caofeidian (Caof), Swan Lake (Swan), Qingyutan (Qingy), Nanwodao (Nanw), Wawushi (Wawu), Qingdao Bay (Qingd), Xidatan (Xida), Linyang Bay (Liny), and Haxian Island (Haxi); Koje Bay (Koje) was selected from the Japan Sea of South Korea; seven sites in the North American Pacific coast of America, including Samish Bay (Sami), Padilla Bay (Padi), Willapa Bay (Willa), Netarts Bay (Neta), Yaquina Bay (Yaqu), Coos Bay (Coos), and Humboldt Bay (Humb).

Table 1.

Environmental conditions of the twenty sampling sites and sampling dates.

Region	Sampling site	Location	Sampling date	Environmental conditions	Region
Northwestern Pacific	Juehua Island, China (Jueh)	40°29’58”N, 120°47’11”E	2019/6/20	The three study sites are distributed in coastal waters of Huludao City. Huludao Zinc Plant and Bohai Ship Yard about 20–45 km from the sampling sites (Gao et al., 2014).	Huludao
	Xingcheng, China (Xing)	40°32’36”N, 120°45’07”E	2019/6/21		Huludao
	Xiaohaishan Island, China (Xiao)	40°23’59”N, 120°36’34”E	2019/6/22		Huludao
	Caofeidian, China (Caof)	39°05’44”N, 118°42’26”E	2019/6/11	The study site is surrounded by Caofeidian Port and industrial areas, possibly impacted by terrestrial pollution (Wang et al., 2018).	W. Bohai Sea
	Swan Lake, China (Swan)	37°20’55”N, 122°34’24”E	2016/5/25	Protected nature reserve with all anthropogenic activities banned, and there is no heavy industry around the lagoon (Xu et al., 2018).	Rongcheng
	Qingyutan, China (Qingy)	37°09’54”N, 122°34’24”E	2021/1/18	The three study sites are distributed in the Ailian Bay, which is a coastal embayment with an area of 5.56 km2 in Rongcheng City, Shandong Peninsula, northern China (Xu et al., 2021b). Z. marina was collected from two abandoned sea cucumber aquaculture ponds in Qingyutan and Wawushi, and one pond in use of Nanwodao.	Rongcheng
	Nanwodao, China (Nanw)	37°11’07”N, 122°35’37”E	2021/1/18		Rongcheng
	Wawushi, China (Wawu)	37°11’36”N, 122°37’21”E	2021/1/18		Rongcheng
	Qingdao Bay, China (Qingd)	36°03’35”N, 120°18’57”E	2020/6/5	Qingdao Bay, a tourist attraction, is an urban area with no direct industrial inputs. There are some sewage outlets and tourism ships in Qingdao Bay (Xu et al., 2022).	Rongcheng
	Xidatan, China (Xida)	39°13’39”N, 122°41’51”E	2019/5/17	The region, surrounded by aquacultural business, is industrial underdevelopment with low terrestrial emissions, and most of the factories along the coasts are mainly related to aquaculture. The Z. marina samples were collected from the open bay. Samples collected from sea cucumber mariculture ponds.	ChangShan
	Linyang Bay, China (Liny)	39°16’41”N, 122°35’46”E	2019/5/19		ChangShan
	Haxian Island, China (Haxi)	39°14’02”N, 122°31’11”E	2019/5/18		ChangShan
	Koje Bay, South Korea (Koje)	34°49’29”N, 128°34’47”E	2021/2/19	Koje Bay is an un-polluted area, and no sewage and industrial waste inputs were observed in this area (Lee et al., 2019).	S. Korea
Northeastern Pacific (USA)	Samish Bay, WA (Sami)	48°36’23”N, 122°29’13”W	2017/5/23	Samish Bay has been used to grow non-native Pacific oysters since 1919, and it is undeveloped with no direct industrial inputs.	Washington
	Padilla Bay, WA (Padi)	48°28’50”N, 122°28’37”W	2017/5/23	The Padilla Bay National Estuarine Research Reserve protects one of the largest beds of eelgrass-nearly 8,000 acres.	Washington
	Willapa Bay, WA (Willa)	46°43’40”N, 123°57’06”W	2017/5/25	Willapa Bay is the second-largest riverine estuary on the Pacific coast of the continental US with relatively low population density, and there are ports and logging industry present.	Washington
	Netarts Bay, OR (Neta)	45°24’37”N, 123°56’07”W	2017/5/29	Netarts Bay is an ocean dominated system with little development in the watershed.	Oregon
	Yaquina Bay, OR (Yaqu)	44°34’51”N, 123°59’41”W	2017/5/26	Yaquina Bay has moderate industry and development along the shores but has generally good water quality (Kaldy, 2006).	Oregon
	Coos Bay, OR (Coos)	43°18’53”N, 124°19’09”W	2017/5/28	The South Slough National Estuarine Research Reserve in Coos Bay, is a protected area with very little development or industry.	Oregon
	Humboldt Bay, CA (Humb)	40°51’29”N, 124°09’27”W	2017/5/31	There is relatively little heavy industry in the region surrounding the bay, and there are few sources of toxic metals other than natural mining in the small watershed (Barnhart et al., 1992).	N. Califonria

One sampling event was conducted at each sampling site, and all samples of seagrass, sediment, and seawater were collected in triplicate. Each eelgrass sample replicate consisted of approximately 30 shoots, together with rhizomes and roots. All samples were randomly collected within the seagrass meadow. Sediment samples were collected from the surface sediment layer (0–10 cm). Seagrass and sediment samples were stored in plastic bags. Surface water samples were collected in plastic buckets and transfer to clean 60 ml plastic bottles. All samples were transported on wet ice to the laboratory for processing and stabilization within 12 hours.

In the lab, eelgrass tissues were rinsed with freshwater to remove salts and sediment, divided into aboveground and belowground tissues and dried to constant weight at 60 °C. Sediments were dried at 60 °C to a constant weight, then ground and sieved prior to heavy metal analysis. Seawater samples were filtered through Whatman GF/F glass fiber filters (0.45 μm) in the laboratory, and all filtered samples were stored at −20 °C prior to analyses.

2.2. Analysis for heavy metals

Seagrass samples, 0.1 g DW, were digested using a concentrated mixed acid (1 mL HClO₄ and 5 mL HNO₃ at 140 °C for 3 hours), then heated at 180 °C until evaporated to dryness, and subsequently dissolved with 1 mL 50% (v/v) HNO₃. Millipore water was used to dilute the digested solution to 25 mL total volume. Samples were then subjected to inductively coupled plasma mass spectrometry (ICP-MS; Thermo Fisher Icap-Qc) for Cr, Cu, Zn, Cd, and Pb analysis. The detection limits of the different elements were 0.000006 mg Cr kg⁻¹, 0.000011 mg Cu kg⁻¹, 0.000111 mg Zn kg⁻¹, 0.000002 mg Cd kg⁻¹, and 0.000001 mg Pb kg⁻¹. All detection limits were based on a 98% confidence level (3 standard deviations).

Sediment subsamples (0.1 g DW) were digested using a concentrated mixed acid composed of 1 mL HClO₄, 5 mL HNO₃, and 4 mL HF at 140 °C for 8 hours, then heated at 180 °C until evaporated to dryness, and subsequently dissolved with 1 mL concentrated HNO₃. Millipore water was used to dilute the digested solution to a final volume of 25 mL. Samples were then subjected to inductively coupled plasma-optical emission spectrometry (ICP-OES; Perkin-Elmer 7300 DV) for Cr, Cu, Zn, and Pb analysis, and ICP-MS (Thermo Fisher Icap-Qc) for Cd analysis. The detection limits of the different elements were 0.0002 mg Cr kg⁻¹, 0.0004 mg Cu kg⁻¹, 0.0002 mg Zn kg⁻¹, 0.000002 mg Cd kg⁻¹, and 0.001 mg Pb kg⁻¹. All detection limits are based on a 98% confidence level (3 standard deviations).

Filtered seawater (10 mL) was diluted to 100 mL using Millipore water and subjected to ICP-MS (Icap-Qc) for Cr, Cu, Zn, Cd, and Pb analysis. The detection limits of the different elements were 0.006 μg Cr L⁻¹, 0.011 μg Cu L⁻¹, 0.111 μg Zn L⁻¹, 0.002 μg Cd L⁻¹, and 0.001 μg Pb L⁻¹.

For seagrass analysis, Saccharina japonica was chosen as certified reference material (GBW 08517), and Yellow Sea sediment was selected as certified reference material for sediment analysis (GBW 07333). The recovery rates for all metals were 90.8–104.1% and the measurement precision was <3% relative standard deviations.

2.3. Calculations and data analysis

We calculated a bioconcentration factor (BCF) which is an exposure metric to evaluate the ability of seagrass tissues to take up and accumulate heavy metals present in sediments (Lin et al., 2016). The BCF is determined by Eq. (1):

$$BCF=\frac{C_{tissues}}{C_{sediments}}$$ (1)

where $C_{tissues}$ and $C_{sediments}$ represent the measured heavy metal concentrations in the seagrass tissues (aboveground and belowground tissues, respectively) and sediments.

Results are presented as the mean ± standard error (SE). Non-parametric Kruskal–Wallis tests (one-way analysis of variance) with pairwise comparisons using t tests were used to compare the differences of metal concentrations in sediment, surface seawater, and eelgrass tissues between different sites. Non-parametric Kruskal–Wallis tests (one-way analysis of variance) with post-hoc Wilcox tests were used to compare the differences in mean concentrations of metals in the sediments and eelgrass tissues between the two geographic regions, and mean concentrations of metals between the aboveground and belowground tissues. Spearman’s correlation analysis was performed to evaluate relationships between different materials (aboveground and belowground tissues, sediments, and seawater). Spearman correlation of different heavy metal concentrations in sediments and seawaters were also evaluated, and because of the limited data on Pb in seawater, no analysis is conducted on the relationship between Pb and other heavy metals.

We used principal component analysis (PCA) to examine the distribution patterns of heavy metals (Cr, Cu, Zn, Cd, and Pb) in sediments and seagrass tissues among different sampling sites. Before PCA, univariate or multivariate normality was tested using Mardia’s test, which was conducted in R with the package “MVN” (Korkmaz et al., 2014). To achieve normality, Humb and Nanw sediment data, Swan and Xiao aboveground tissue data, and Swan belowground tissue data were removed as outliers, which was detected with “boxplot.stats” function in R. After that, metal data were log-transformed, except Pb belowground tissue data and all Zn data. The aboveground Zn concentrations were conducted normal scores transformation using Blom’s method (Altman, 1990). All data analysis was conducted using R Statistical Software (v.4.3.0; R Core Team, 2023). Statistical significance was evaluated at p < 0.05.

3. Results

3.1. Heavy metal concentrations in sediments and seawater

Heavy metal concentrations in sediments and seawater samples from eelgrass beds in the North Pacific Ocean were highly variable between sites (Table S1 and Figure 2; p < 0.001). Averaged across all sites, heavy metal concentrations in sediments exhibited the following order: Cr > Zn > Pb > Cu >> Cd. However, analyzing the data by geographical region (Asia vs North America), in Asia the order is Zn > Cr >Pb > Cu >> Cd and in North America it is Cr > Zn > Cu > Pb >> Cd (Figure 3), indicating that Cr was highest in NA sites but in Asia Zn was highest, and the Pb concentration is higher than Cu in Asian sites, and lower than Cu in NA sites. The pattern across all sites was driven by significantly higher Cr concentrations from the NA sites, where average Cr concentration was more than six times the Asian sites (Figure 3; p < 0.001), and also driven by significant higher Pb concentrations from the Asian sites, where average Pb concentration was almost three times the NA sites (Figure 3; p < 0.001). There was not much difference in concentrations of Cu, Zn, and Cd between the two geographical regions (Figure 3; p > 0.05), and the concentrations in Asia were slightly higher than NA sites.

Figure 2.

Concentrations of metals in sediment (a) and surface seawater (b); data are presented as the mean ± SE (n=3); sediment units are mg kg⁻¹ DW; water units are μg L⁻¹; NS indicates not sampled; ND indicates not detected.

Figure 3.

Mean concentrations of metals in the sediments of the two geographic regions (Asia vs North America); the unit on the y-axis represents the log-transformed value of metal concentrations (mg kg⁻¹ DW); boxes represent lower and upper quartiles with medians (line) inside the boxes (n= 20); “***”, “**”, and “*” mean significant at the 0.001, 0.01 and 0.05 levels, respectively, and “ns” represent no significant differences (one-way ANOVA).

The highest Cr concentration was measured in Humboldt Bay (1244.10 mg kg⁻¹), which is nearly ten times the mean concentration at all other sampling locations (Table S1 and Figure 2a; p < 0.001), and other Cr hot spots included Sami, Neta, and Coos. The Cr concentrations of those hot spots exceeded the China’s Level II sediment quality standard of 150 mg kg⁻¹. In addition, the Cr concentrations of Nanw, Wawu, Padi, and Yaqu exceeded the Level I sediment quality standard of 80 mg kg⁻¹ (China State Bureau of Quality and Technical Supervision, 2002).

The highest concentrations of metals in Asia occurred in Nanw with 43.40 mg Cu kg⁻¹, 143.80 mg Zn kg⁻¹, and 39.20 mg Pb kg⁻¹, and the highest Cd concentrations were measured in Xing at 0.72 mg kg⁻¹, and Nanw at 0.68 mg kg⁻¹. In contrast, Caof, Xida, and Liny all had relatively lower metal concentrations (Table S1 and Figure 2a). The Cd concentrations in sediments from Jueh, Xing, Xiao, and Nanw, and Cu concentrations from Nanw exceeded China’s Level I sediment quality standard of 0.5 mg Cd kg⁻¹ and 35 mg Cu kg⁻¹, but met the Level II criteria of 1.5 mg Cd kg⁻¹ and 100 mg Cu kg⁻¹ (China State Bureau of Quality and Technical Supervision, 2002). Additionally, at all sites, Pb concentrations were below the Level I sediment quality standard of 60 mg kg⁻¹ (Table S1 and Figure 2a).

Averaged across the China sites, heavy metal concentrations in seawater exhibited the following order: Zn > Cu > Cr > Pb > Cd (Table S1 and Figure 2b) and compared to the order of the Asian sediments (Table S1 and Figure 2a), most of the variation was found to occur in the ranking of Cu, Cr, and Pb. Seawater concentrations of all metals were below China’s Level I water quality standard (State Environmental Protection Administration, 1997), except for Zn in Swan (27.24 μg L⁻¹), which met the requirements of water quality standard Level II, 50 μg L⁻¹ (Table S1 and Figure 2b).

3.2. Heavy metal concentrations in seagrasses

Heavy metal concentrations in eelgrass tissues from the North Pacific Ocean were highly variable among sites (Table S2 and Figure 4; p < 0.01). Across all sites, the average heavy metal concentrations in aboveground tissues ranked in the following order: Zn > Cu > Cd> Cr ~ Pb; and metals in belowground tissue concentrations were ranked: Zn > Cu > Pb ~ Cr > Cd. The results indicate that variation occurred in the ranking of Cd, Cr, and Pb, which was driven by significantly higher Cd in Asian leaf tissue than in NA (Figure 5a; p < 0.001), specifically, it was driven by three sampling sites (Jueh, Xing, Xiao).

Figure 4.

Concentrations of metals in aboveground tissues (a) and belowground tissues (b); data are presented as the mean ± SE (n=3); units are mg kg⁻¹ DW.

Figure 5.

Mean concentrations of metals in the aboveground (a) and belowground (b) tissues of Zostera marina of the two geographic regions (Asia vs North America); units on the y-axis represent the log-transformed value of metal concentrations (mg kg⁻¹ DW); boxes represent lower and upper quartiles with medians (line) inside the boxes (n= 20); “***”, “**”, and “*” mean significant at the 0.001, 0.01, and 0.05 levels, respectively, and “ns” represent no significant differences (one-way ANOVA).

Comparison of results between the two geographical regions demonstrate that Asian aboveground tissue samples have significantly higher Cd and lower Zn than the NA samples (Figure 4a and 5a; p < 0.01), and this significant difference in Zn and Cd between the two regions was also observed in the belowground tissues (Figure 4b and 5b; p < 0.001). Additionally, although Cr is the highest in NA sediment samples (even without Humb) the seagrass tissue metals ranking are highest for Zn.

The Cr concentration in aboveground tissues from Swan was 18.26 mg kg⁻¹, which was significantly higher than other study sites (Figure 4a; p < 0.001), and samples from Humb had the second highest concentration at 3.21 mg Cr kg⁻¹. Aboveground tissue Cu concentrations in Xiao, Humb, and Swan were significantly higher than other study sites (Figure 4a; p < 0.05), except Koje and Qingd. Leaf Zn concentrations from Neta were significantly higher than other study sites (Figure 4a; p < 0.05), except Coos, Humb, Qingd, Sami, and Willa. Aboveground tissue Cd concentrations from Huludao city which included Jueh, Xing, and Xiao were significantly higher than the other study sites (Figure 4a; p < 0.05). Leaf Pb concentrations from Swan were significantly higher than the other study sites (Table S2 and Figure 4; p < 0.001).

Belowground tissues from Swan had concentrations of Cr that were significantly higher than all other sites (Figure 4b; p < 0.001). Belowground tissue Cd from Qingy and Qingd samples were significantly higher than at all other sites (Table S2 and Figure 4; p < 0.001), except Jueh and Xing.

Across all sites, there were no significant differences between aboveground and belowground tissue Cr and Cu concentrations (Figure 6; p > 0.05). However, concentrations of Zn and Cd in aboveground tissues were significantly higher than in belowground tissues (Figure 6; p < 0.01, p < 0.05), and concentrations of Pb in belowground tissues were significantly higher than these in aboveground tissues (Figure 6; p < 0.05).

Figure 6.

Mean concentrations of metals in the aboveground and belowground tissues of Zostera marina; units on the y-axis represents the log-transformed value of metal concentrations (mg kg⁻¹ DW); boxes represent lower and upper quartiles with medians (line) inside the boxes (n= 20); “**” and “*” mean significant at the 0.01 and 0.05 levels, respectively, and “ns” represent no significant differences (one-way ANOVA).

3.3. Bioconcentration factor

Eelgrass tissues had similar calculated metals bioconcentration factors with values <1 for Cr, Cu, and Pb, and Zn in roots/rhizomes (Figure 7). The calculated BCF was highest for Cd with values of 10.25 and 3.95 in aboveground and belowground tissues, respectively, indicating a high potential to bioaccumulate Cd from the environment, although in almost all cases Cd concentrations in the plant tissues and in the environment were relatively low. Additionally, aboveground tissues had slightly higher BCFs than belowground tissues for Cu and Zn. In turn, the BCF of Cr and Pb in belowground tissues were slightly higher than those in aboveground tissues.

Figure 7.

Bioconcentration factors (BCF) for Cr, Cu, Zn, Cd, and Pb in Zostera marina.

3.4. Multivariate statistical analysis

There was a significant correlation between aboveground and belowground tissue concentrations for all heavy metals except Pb (Figure 8; p < 0.05), and a relationship between leaf and sediment Cd and Pb concentrations (Figure 8; p < 0.05). Compared to sediment, seagrass tissues seem to have a stronger correlation with seawater, although most of these water samples were collected from Asian sites (Figure 8).

Figure 8.

Spearman correlation of metal concentrations in different materials; the upper right panels show Spearman’s correlation coefficients and the lower left panels show the scatter plots of metal concentrations (units of tissue and sediment are mg kg⁻¹ DW; units of water are μg L⁻¹); “***”, “**”, and “*” indicate significance at the 0.001, 0.01, and 0.05 levels, respectively (2-tailed).

Across all sites, there are significant correlations among Cu, Zn, and Cd in sediments, indicating that these metals have homology to each other (Figure 9). No significant correlations between Cr and any other metal (Cu, Zn, Cd, and Pb) were observed in NA (p > 0.05), whereas there were significant correlations between Cr and Cu (p < 0.001), and between Cr and Zn (p < 0.05) in Asia. It should be pointed out that there was a total of five pairs with correlations among different heavy metals in Asia, while only three pairs in the NA sites, indicating a higher degree of homology exists in Asia (Figure 9).

Figure 9.

Spearman correlations of metal concentrations in sediment; the upper right panels show Spearman’s correlation coefficients and the lower left panels show the scatter plots of metal concentrations (mg kg⁻¹ DW); “***”, “**”, and “*” mean significant at the 0.001, 0.01, and 0.05 levels, respectively (2-tailed).

Across all sites, there were significant correlations between Cu and Cr in seawater, which was driven by significant correlations from the Asian sites, as most of these water samples were collected from Asian sites. There were significant correlations between Zn and Cr (p < 0.05) in Asia. No significant correlations among any other metals were observed (Figure 10; p > 0.05).

Figure 10.

Spearman correlations of metal concentrations in seawater; the upper right panels show Spearman’s correlation coefficient and the lower left panels show the scatter plots of metal concentrations (μg L⁻¹); “***”, “**”, and “*” mean significant at the 0.001, 0.01, and 0.05 levels, respectively (2-tailed).

PCA using metal concentrations in the sediments of the study sites except for Humb with very high Cr concentrations and Nanw with high concentrations of Zn, Cu, Cd and Pb high metal concentrations, indicates that Comp.1 was dominated by Cu, Zn, Cd, and Pb concentrations explaining 56.23% of the total variance, while Comp.2 was dominated by Cr and Pb concentrations explaining 30.61% of the variance in the dataset (Figure 11). There appear to be several clusters of systems with similar sediment metals concentrations (Figure 11). For example, most of the NA sites cluster together with higher positive scores on Comp.2. Likewise, samples from Caof, Liny and Xida cluster together with lower negative scores on Comp.1 and Comp.2. Finally, the other Asian sites cluster together with higher positive scores on Comp.1.

Figure 11.

Principal component analysis of metals (Cr, Cu, Zn, Cd, and Pb) in sediments. Humb and Nanw data were removed as outliers, and all other metal data were log-transformed, except Zn concentration.

PCA using the heavy metal concentrations in aboveground Z. marina tissues of the study sites except for Swan and Xiao, indicate that the first two principal components accounted for 75% of the total variance in the dataset (54.21% and 20.79%, respectively; Figure 12a). There appear to be several clusters of systems with similar aboveground metals concentrations (Figure 12a). For example, all NA sites along with Qingd and Koje cluster together with very high Zn concentrations. Likewise, samples from the other Asian sites, except for Xiao and Swan, with low Cu, Cr, and Pb concentrations cluster together.

Figure 12.

Principal component analysis of metals (Cr, Cu, Zn, Cd, and Pb) in the aboveground (a) and belowground (b) tissues of Zostera marina. Metal concentrations of aboveground tissues in Swan and Xiao, and belowground tissues in Swan, as outliers, were removed, and all metal data were log-transformed, except concentrations of Zn and Pb of belowground tissue.

PCA results for belowground Z. marina tissues of the study sites except for Swan, indicate that the variance of Comp.1 (49.42% of the total variance) was explained by the positive contribution of Pb, Cr, Cu, and Zn (Figure 12b). While Comp.2 accounted for 20.94% of the total variance and was characterized by the positive contribution of Cd and the negative loading of Zn (Figure12b). There appear to be several clusters of systems with similar belowground metals concentrations (Figure 12b). For example, all NA sites with Koje cluster together with very high Zn concentrations, which is similar to the aboveground clusters of systems. There appear to be two clusters of systems in other Asian sites. Liny, Caof, Wawu, Haxi, Xida, and Nanw cluster together with low Cu, Cr, and Pb concentrations, and Qingy, Xing, Jueh, and Qingd cluster together with high Cd concentrations. In addition, Swan appears to be a consistent outlier for both above- and belowground tissues.

4. Discussion

4.1. Sources of heavy metals in sediment

Coastal areas are subject to inputs of heavy metals from terrestrial, oceanic, and atmospheric sources. Our sampling locations spanned a gradient from pristine marine reserves to active and historical mariculture areas and areas adjacent to heavy industry (e.g., smelting and factories) and urban environments. Coastal human activities, such as agriculture, aquaculture, and sewage discharge, which are associated with the use of fuels and artificial substances (such as pesticides/herbicides) or production of heavy metal waste, are an important source of metal pollutants for coastal sediments (Pan and Wang, 2012; Oelsner and Stets, 2019). Heavy metals produced inland can be transported to coastal areas by rivers or air deposition (Presley et al., 1980; Li et al., 2007; Gu et al., 2014), and human activities, such as mining, oil spills, and transportation, produce large amounts of heavy metals, which can be deposited in coastal sediments (Idaszkin et al., 2017; Ruiz-Fernández et al., 2019). In addition, metals are abundant in the parent rock, which are released, transported, and accumulated through natural processes such as weathering and erosion, which may be the main source for specific metals in some locations (Bem et al., 2003; Adaikpoh et al., 2005; Paul and Sinha, 2013). As a result, metals concentrations of coastal areas are variable and can be influenced by natural and anthropogenic processes.

Northwestern Pacific

Human activities are an important factor controlling the metals concentrations in many of the study sites in China, and different industry types can influence the specific metals present and the concentrations. Regional differences in industry types (Table1) may help to explain observed metals distributions.

Liu et al. (2019) investigated sources of heavy metals in supratidal wetlands along the west coast of the Bohai Sea, and found that Cu, Zn, Cd, and Pb are likely correlated with anthropogenic activities, and Cr mainly originates from the weathering of rocks and their parent materials. Previous work investigated the distribution of heavy metals in the surface sediments of the Bohai Sea, and concluded that the local mining, smelting, and transportation of metals increases the concentrations of Cu, Zn, Pb, and Cd in the area (Lin et al., 2013). Sediments from Huludao are known to have high Cd concentrations up to 1463 mg Cd kg⁻¹ as a result of rich endogenetic deposits (Gao et al., 2014). Our sites, Jueh, Xing, and Xiao were in Huludao, near the wastewater outfall for the Huludao Zinc Plant and Bohai Ship Yard (Gao et al., 2014). Shipyard activities generate wastes such as paint chips, paint residue, waste paint, and metal shavings or powder that can be transported into marine waters and then deposited in the sediments (Pourabadehei and Mulligan, 2016). Concentrations of heavy metals in sediment from Caof (Western Bohai Sea) decreased with increasing offshore distance, suggesting that terrestrial pollution may be a source of heavy metals (Wang et al., 2018).

Metals concentrations in coastal areas may be also influenced by mariculture (Zhang et al., 2012). For example, the Rongcheng region has both heavy shipyard industry (Hu et al., 2019) and a long (>50 year) history of intensive mariculture including macroalgae, shellfish and sea cucumber (Ge, 2006). Intensive mariculture significantly increases the loads of heavy metals (Zhang et al., 2012). Several of our study sites, Qingy, Wawu, Nanw, and Haxi were in active or abandoned sea cucumber aquaculture ponds. Although the heavy metal concentrations in the sediment were below the Class I upper limit of Chinese sediment quality guidelines (China State Bureau of Quality and Technical Supervision, 2002), the concentrations of heavy metals in the sea cucumber aquaculture ponds (Qingy, Wawu, Nanw, and Haxi) were relatively high compared with other study sites.

Heavy metal concentrations in sediment from Xida and Liny on the Changshan Islands in the Yellow Sea were relatively low compared to other study sites. Previous work suggests that heavy metals in the sediments of Zhangzi Island, part of the Changshan Island chain, present no ecological risks (Tan et al., 2012) because of the high degree of water exchange and a lack of heavy industry in the area.

Northeastern Pacific (USA)

We also sampled along a similar gradient of environmental impacts in the Northeastern Pacific (USA). These estuaries support diverse industries and human activities including heavy industry (e.g., shipyards) and oyster aquaculture. We believe that natural processes (e.g., weathering of ultramafic rocks and upwelling) may be an important source of metals in many Northeastern Pacific (USA) estuarine study sites.

Relative to sites in Asia, our study sites along the West Coast of the United States generally had elevated levels of Cr, with Humb being a notable outlier (1244.1 mg Cr kg⁻¹), which could be explained by the distribution of the ultramafic rocks in the West Coast of the United States. The ultramafic masses were mainly distributed in western California, southern Oregon, and northern Washington (Cohee, 1962), and ultramafic coastal rocks have elevated Cr. Oze et al. (2004) reported that weathering of ultramafic rocks and serpentinites in the Franciscan Complex of California produces serpentine soils containing high concentrations of Cr. The Cr was likely in a non-available solid mineral like enstatie, chromite, chrome spinel, or chrome magnetite, which may explain why we observed high sediment Cr concentrations but low concentrations in eelgrass tissues. Consequently, we believe that weathering processes of ultramafic rocks may play a role in controlling the bulk Cr concentrations in the PNW estuaries. Future work should focus specifically on bioavailable metals within the environment.

The West Coast of the United States is an area of intense seasonal upwelling (Barnhart et al., 1992). Cassis et al. (2011) concluded that dissolved Cd delivered by upwelling is incorporated into phytoplankton (particulate matter) and then available to either oysters or the particulates settle, delivering the Cd to the sediments. Likewise, Shiel et al. (2012), using isotope tracers, concluded that Cd in oysters was primarily from upwelling of Cd rich intermediate waters. Although these studies are oyster centric, the same fundamental metal delivery mechanism (upwelling) and incorporation into particulates likely influence sediment metals concentrations in eelgrass beds. In addition, the upwelling zone is also known to exhibit extensive hypoxia/anoxia (Roegner et al., 2011; Chan et al., 2019), and most heavy metals, including Zn (Warnken et al., 2001), Cu (Riedel et al., 1997), Cd (Riedel et al., 1999), and Cr (Palmer and Puls, 1994) are incorporated into the sediment under hypoxic/anoxic conditions (Mason et al., 2006). For example, Cr(VI) is reduced to Cr(III) an insoluble hydroxide precipitate that accumulates in the sediments in a hypoxic/anoxic environment (Palmer and Puls, 1994). Cr(III) is less toxic than Cr(VI), accumulates in the sediment, and cannot be directly used by plants (Ukhurebor et al., 2021).

Undoubtedly, human activities are also a part of the source of heavy metals in the PNW estuaries. Many estuaries support diverse industries including boat works, shipping, aquaculture, and wastewater disposal, among other activities. Lands surrounding Humb have been historically used for agriculture, with approximately 27 percent of Humboldt County land in agricultural use. Agricultural chemicals, such as pesticides and herbicides, often contain heavy metals (Defarge et al., 2016) that can be delivered to estuarine sediments (Harbor, 2020). Industrial activities in Coos Bay including agriculture, urban development, and industrial activities may influence concentrations of arsenic, chromium, mercury, and nickel within the estuary (U.S. Environmental Protection Agency, 2012; Larsen and Johnson, 2015).

4.2. Distribution and accumulation of heavy metal in seagrass

In the present study, Cr concentrations in the belowground tissues of Z. marina were generally higher than those of the aboveground tissues, and BCFs of Cr were the lowest compared with the other studied heavy metals, indicating a low rate of accumulation for Cr. This overestimate may be a result of non-bioavailable Cr species detected in the sediments, since the BCF was determined relative to total Cr content, not bioavailable Cr content. The belowground tissues tended to have higher concentrations of Pb, with low values of BCF, and this may be a result of biological compartmentalization. It has been suggested that seagrass is unable to effectively detoxify high Pb concentrations (Bonanno and Raccuia, 2018), and thus Z. marina may protect itself by concentrating Pb in root tissues, isolating photosynthetic pathways from the negative impacts of Pb (Bonanno and Di Martino, 2016). Cadmium is one of the most toxic metals with high bioaccumulation in eelgrass tissues. Cadmium is highly mobile and can bioaccumulate in plants (Devi and Bhattacharyya, 2018). Because of these characteristics, Cd has a high BCF. Zostera marina takes up Cd mainly from the water column via their leaves (Lyngby and Brix, 1982), and this likely explains the high Cd concentrations in aboveground tissues and the higher BCF (10.25) observed in this study. Thus Z. marina may remove accumulated Cd by shedding leaves, reducing the Cd burden on the whole plant (Bonanno and Di Martino, 2016). This type of concentration and compartmentalization strategy also promotes metal export from the seagrass bed (Kaldy, 2006). Alternatively, it also can act as a positive feedback resulting in a build-up of metals in the surface sediments (Hosokawa et al., 2016).

Some heavy metals are essential to cellular biochemistry. For example, Zn is a cofactor for a number of enzyme systems essential for many physiological processes such as protein synthesis, lipid, nucleic acid, and cell wall metabolism (Rai et al., 2021). Cellular requirements may explain the higher Zn tissue concentrations than observed in other metals despite the low BCF. Copper and Zn, are essential metals for eelgrass, and concentrations below 15 mg kg⁻¹ and 500 mg kg⁻¹, respectively are considered non-toxic, while concentrations > 20 mg kg⁻¹ and 1500 mg kg⁻¹ respectively can be considered toxic (Chaney, 1989; Kabata-Pendias, 2000). In our study, concentrations of Cu and Zn in eelgrass tissue were below this toxic threshold indicating that Cu and Zn levels are likely not impacting eelgrass physiology.

Our data on metals, especially for Cr, from the sites along the Rongcheng Coast is somewhat inconsistent with previous studies. Our measured sediment Cr concentrations were about twice that measured by Hu et al. (2019), while the eelgrass leaf Cr concentration was half that of Hu et al. (2019) (Table 2), which could be explained by the different collection season, generally there are significantly higher levels of heavy metals in summer season (Pasumpon and Vasudevan, 2021). However, results to date have been inconsistent as to how the seasonality significantly effects metal content, although many studies have demonstrated that seagrass metal concentrations display significant seasonal variations. The first seasonal pattern is low metal concentrations in seagrass in summer. For example, Lyngby and Brix (1982) investigated seasonal variation in Cu, Zn, Cd, and Pb concentrations in Zostera marina L. in the Limfjord, Denmark and found that the metals showed similar general seasonal variation patterns with high metal content in winter-early spring and low in summer, which could be explained by that the metal content gradually decreases as the seagrass grows as a result of dilution processes (Schlacher-Hoenlinger and Schlacher, 1998; Karimi et al., 2010; Li et al., 2012). In contrast, another seasonal pattern is high metal concentrations in seagrass in summer. For example, high Cd in Posidonia oceanica was found during summer and autumn and low in winter and spring (Malea and Haritonidis, 1989; Malea et al., 1994). In summer, the warm temperature and low salinity could facilitate metal uptake (Bond et al., 1988), which could explain our research results: high metal in seagrass during summer (Hu et al., 2019) and low during winter (this study) in the Rongcheng Coast.

Table 2.

Concentrations of metals in sediments and aboveground and belowground tissues of Zostera marina in various locations from the literature and present study; units are mg kg⁻¹ DW; the symbol “~” indicates that the value is approximated, because it is estimated from a graph; ND indicated not detected; Rongcheng Coast includes sites of Swan, Qingy, Nanw, and Wawu in the present study.

Location	Sediment or seagrass tissue	Data type	Cr	Cu	Zn	Cd	Pb	Reference
Rongcheng Coast	Sediment	Mean	~40	~21	~32	~0.36	~23	Hu et al. (2019)
	Aboveground tissue	Mean±SE	12.67±3.20	18.28±6.58	28.11±13.69	5.12±1.80	4.90±2.30	Hu et al. (2019)
	Belowground tissue	Mean±SE	18.26±5.63	10.51±4.74	32.56±14.36	1.56±0.97	16.67±7.16	Hu et al. (2019)
	Sediment	Mean±SE	78.78±8.21	28.82±5.28	95.40±19.24	0.47±0.08	28.35±3.96	This study
	Aboveground tissue	Mean±SE	5.14±4.67	7.60±1.69	40.21±5.50	4.17±0.90	2.75±1.89	This study
	Belowground tissue	Mean±SE	3.52±2.55	6.71±2.12	22.48±2.65	1.41±0.36	2.35±1.32	This study
Koje Bay	Sediment	Mean±SE	-	16.83 ± 1.39	171.82 ± 10.99	0.37 ± 0.01	75.38 ± 6.01	Lee et al. (2019)
	Aboveground tissue	Mean±SE	-	17.09 ± 3.07	20.11 ± 0.68	0.51 ± 0.04	1.00 ± 0.06	Lee et al. (2019)
	Belowground tissue	Mean±SE	-	9.94 ± 1.43	12.87 ± 0.50	0.17 ± 0.01	0.88 ± 0.08	Lee et al. (2019)
	Sediment	Mean±SE	14.33±1.28	8.03±0.84	90.03±7.24	0.29±0.02	33.47±1.60	This study
	Aboveground tissue	Mean±SE	0.54±0.11	11.94±1.00	58.08±8.46	0.87±0.06	1.38±0.15	This study
	Belowground tissue	Mean±SE	1.11±0.22	9.99±0.48	40.00±16.80	0.55±0.10	2.31±0.30	This study
Yaquina Bay	Sediment		<15	<10	<50	<5	<15	Buchman (1989)
	Sediment	Mean±SD	-	34.6±2.3	113.8±7.2	<1	17±0.6	Sherman (2003)
	Sediment	Mean±SD	66.1±7.9	13.29±8.5	59.6±29.6	0.17±0.05	12±2.1	Nelson et al. (2004)
	Sediment	Means	33.5	12.69	51.4	0.24	7.02	Britton and Siipola (1991)
	Leaf	Mean±SD	7.2±4	10±3	29±6	1.7±0.5	ND	Kaldy (2006)
	Rhizome	Mean±SD	7.4±10	8.3±14	18±8	0.8±0.6	ND	Kaldy (2006)
	Root	Mean±SD	38±23	47±64	52±16	6.4±16	ND	Kaldy (2006)
	Sediment	Mean±SE	110.57±1.60	19.97±6.18	82.47±14.13	0.30±0.03	15.57±2.74	This study
	Aboveground tissue	Mean±SE	0.79±0.05	10.24±0.80	58.52±0.68	0.23±0.00	1.48±0.45	This study
	Belowground tissue	Mean±SE	2.45±0.21	6.75±0.31	54.41±0.83	0.13±0.00	2.39±0.63	This study
Humboldt Bay	Sediment	Mean	-	6.495	45.3	-	-	Miller et al. (2022)
	Sediment	Mean±SE	1244.10±161.96	20.20±1.16	86.67±2.62	0.32±0.01	9.30±0.40	This study

The significant positive correlation in Cd and Pb concentrations at all sites between aboveground tissues and sediment indicates that Z. marina leaves could potentially be used for biomonitoring for Cd and Pb in eelgrass ecosystems. Previous work has suggested that with additional research, eelgrass could be a useful bioindicator of Cd contamination in coastal areas at multiple scales (Govers et al., 2014). However, eelgrass utility as a bioindicator of other heavy metals may be limited since concentrations of Cr, Cu, and Zn in sediment were not reflected in Z. marina leaf tissues (Figure 8).

We have summarized the variation patterns of Cr and Pb concentrations in seagrass and bioconcentration factors (BCF) with metal concentrations in sediment (Figure 13), with the increase of metal sediment concentration, BCF decreased eventhough the seagrass heavy metal concentrations increased. Suggesting that when sediment concentration increases, seagrass have a protective mechanism to avoid absorbing more heavy metals. Lee et al. (2019) transplanted Zostera marina shoots from unpolluted site to two polluted bay systems on Korean coasts, and found that polluted sites have high Pb concentrations in seagrass but with low BCF, which is consistent with our results (Figure 8).

Figure 13.

Schematic diagram of variations of Cr and Pb concentrations in seagrass and bioconcentration factors (BCF) with metal concentrations in sediment.

Our data for several metals in multiple locations suggest that high sediment concentrations are not necessarily correlated with eelgrass tissue concentrations, for example, Cr in Yaquina Bay, and Zn, Cd, and Pb in Koje Bay (Table 2). Previous work in these systems concluded that Z. marina beds in Koje Bay and Yaquina Bay were considered relatively unpolluted sites (Kaldy, 2006; Lee et al., 2019). We hypothesize that when the concentration of heavy metals in the environment is low, the ability of seagrass to accumulate heavy metals from the environment may be more affected by other factors, such as growth dynamics, rather than just the bioavailable forms and concentrations of heavy metals in environment (Lyngby and Brix, 1982; Lyngby and Brix, 1984; John and Leventhal, 1995). The relationship between metals measured in plant tissues and in the environment are complex and likely related to a number of controls on plant physiological status such as light, temperature, and nutrients, in addition to the controls on metals chemistry and bioavailability discussed earlier (John and Leventhal, 1995). Overall, further research is needed on the relationship between seagrass and heavy metal concentrations in the environment.

4.3. Heavy metal exposure stress

Excessive concentrations of heavy metals in the environment can cause heavy metal exposure stress on seagrass, affecting its survival and growth. There are some studies evaluating the toxicity thresholds and impact of metal pollution on eelgrass (Lyngby and Brix, 1984; Conroy et al., 1991; Lewis and Devereux, 2009; Qiao et al., 2022), with evidence demonstrating that effect concentrations of Cu, Cd, Cr, Pb, and Zn on eelgrass growth rate were 5 μM, 5 μM, 50 μM, 50 μM, and 50 μM in a 19-day test (Lyngby and Brix, 1984). These metal toxicity thresholds are several orders of magnitude higher than those measured in estuarine environments (Table S1) as part of this study. In addition, Hoven (1998) found that eelgrass can grow in sediment with a Pb concentration of 600 mg kg⁻¹, which is approximately 15 times that of the highest concentration of Pb in the North Pacific Ocean (39.90 mg kg⁻¹ in Nanw). Consequently, we suggest that heavy metal exposure stress is likely to be minimal at the estuarine study sites we investigated in the North Pacific Ocean.

To determine whether the concentration of heavy metals in sediments or seawaters poses a threat to aquatic life, we evaluated concentrations relative to two sediment quality guidelines (SQGs) and one water quality criteria (WQC). The SQGs were: (1) threshold effect levels (TEL) and probable effect levels (PEL); and (2) effects range low (ERL) and effects range medium (ERM) (Long et al., 1995; Canadian Council of Ministers of the Environment, 2002). The TEL and ERL guidelines represent concentrations of a given element that 95% of the tested benthic species can tolerate; below the TEL and ERL, no adverse biological effects occur. In contrast, PEL and ERM represent concentrations that are harmful to 95% of the tested benthic species; above the TEL and ERL, adverse biological effects are expected. The WQC evaluated were the criterion continuous concentration (CCC) and criterion maximum concentration (CMC) (U.S. Environmental Protection Agency, 2022). The CCC and CMC represent the highest concentration of a material in ambient water to which an aquatic community can be exposed, indefinitely and briefly, respectively, without exhibiting an adverse effect.

We compared our data on metals concentration in the sediment and water to these four parameters to evaluate exposure stress. When compared to the two SQGs, Cr concentrations exceeded PEL and ERM in 20% (Sami, Neta, Coos, and Humb) and 5% (Humb) of sites, respectively, which are located in the Northeastern Pacific. Despite some differences among the two SQGs, our results suggest that the concentrations of Cr may be of concern and may cause adverse biological effects, and it also indicates that eelgrass has the ability to thrive at the few sites exceeding PELs/ERMs. However, it should be emphasized that previous studies have shown that the HF extractable content method of measuring metals, used in this study, was found to be slightly higher than the aqua regia extractable content using in the Canadian method, especially for Cr (Santoro et al., 2017). Consequently, future studies should apply methods to evaluate bioavailable metals in both seagrass and the environment.

For seawater, all of the heavy metals were below the concentrations for CCC (Figure 4), indicating that no adverse effects are expected.

5. Conclusion

We compare metals concentrations in eelgrass and environmental pools across the Pacific basin, and this work provides baseline data on heavy metal concentrations in eelgrass beds at the basin scale and highlights the need for additional research on the phytotoxic thresholds. Our data indicate that Cd, Zn, and Cu were primarily accumulated in the aboveground tissues of eelgrass, and that Cr and Pb were preferentially stored in the belowground tissues. Aboveground tissue was a good bioindicator for Cd and Pb in sediment, while other metals (Cu, Zn, and Cr) in eelgrass tissues do not seem to reflect the metal concentrations in the environment. Additional research is needed on the factors influencing metal accumulation in eelgrass and the relationship between seagrass and metal concentrations in the environment.

Table 3.

Comparison between environmental quality standards and metal concentrations (mg kg⁻¹ DW) in the present study.

	Cr	Cu	Zn	Cd	Pb
Sediment
Min	12.30	2.92	12.51	0.08	3.60
Max	1462.40	45.20	174.20	0.81	39.90
Median	71.11	11.08	59.32	0.35	19.76
Mean	133.56	15.84	64.43	0.36	18.99
SE	34.50	1.41	4.03	0.02	1.33
Sediment quality guidelines (SQGs)
TEL	52.3	18.7	124	0.7	30.2
PEL	160	108	271	4.2	112
ERL	81	34	150	1.2	46.7
ERM	370	270	410	9.6	218
Compared with TEL and PEL
<TEL (%)	40	55	95	95	85
Between TEL–PEL (%)	40	45	5	5	15
>PEL (%)	20	0	0	0	0
Compared with ERL and ERM
<ERL (%)	65	95	100	100	100
Between ELM–ERM (%)	30	5	0	0	0
>ERM (%)	5	0	0	0	0
Water
Min	0.08	0.23	1.1	0.03	0.09
Max	2.01	2.59	52.53	0.41	0.83
Mean	0.51	1.21	7.2	0.12	0.29
SE	0.08	0.1	1.12	0.01	0.06
CCC	50	3.1	81	7.9	8.1
CMC	1100	4.8	90	33	210
Compared with CCC and CMC
<CCC (%)	100	100	100	100	100
Between CCC and CMC	0	0	0	0	0
>CMC (%)	0	0	0	0	0

Data sources: TEL and PEL: Canadian Sediment Quality Guidelines for the Protection of Aquatic Life (Canadian Council of Ministers of the Environment, 2002); ERL and ERM: Long et al. (1995); NC and MPC: Environmental Quality Standards (EQS) in The Netherlands (Crommentuijn et al., 2000); CCC and CMC: National Recommended Water Quality Criteria - Aquatic Life Criteria Table (U.S. Environmental Protection Agency, 2022).