Distribution and risk assessment of metals in the aquatic environment following the installation of a low-lying bridge in Yeongrang Lake, Sokcho, Gangwon State, South Korea

Reagents

A certified reference material (20 µ g/mL) containing 20 metals was purchased from AccuStandard (New Haven, CT, USA) and identified as PE-CAL2-ASL-1. Concentrated nitric acid (70 %) and perchloric acid (70 %) were purchased from Daejung Chemicals & Metals (Siheung, Gyeonggi-do, Korea). Poly(tetrafluoroethylene) (PTFE) filters were acquired from GVS Filter Technology (Bologna, Italy).

Sample collection

Water, sediment, and fish samples were collected from specified upstream and downstream sites near the low-level bridge over Yeongrang Lake on the following dates: December 28–29, 2022 (winter), March 29–31, 2023 (spring), July 18, 2023 (summer), and October 11, 2023 (fall). Eight easily accessible locations (S1–S8; Figure 1), were selected for collecting water and sediment samples. Surface water samples were collected in 250 mL amber glass bottles without headspace. Sediment samples were collected from the surface layer, approximately 15 cm deep, by combining two or more individual samples taken at least 1 m apart using a shovel. The samples were subsequently placed in zipper bags, air-dried, and sieved using a 2 mm mesh sieve.

Figure 1.

The map shows the eight sampling sites where water, sediment, and fish samples were collected in Yeongrang Lake, Sokcho, Gangwon State, South Korea..

Fish samples were collected from six locations, using stationary nets and foot poles, focusing on common fish species to facilitate comparisons between locations. The collected fish species included Tridentiger bifasciatus, Tridentiger brevispinis, Acanthogobius flavimanus, Rhinogobius brunneus, Gymnogobius urotaenia, Carassius carassius, Gymnogobius breunigii, Platichthys stellatus, Tribolodon hakonensis, Embiotocidae, Mugil cephalus, Hypomesus nipponensis, Engraulis japonicus, Gasterosteus aculeatus, Takifugu niphobles, and Konosirus punctatus.

All samples were promptly stored in a cooler and transported to the laboratory. Water and sediment samples were preserved at 4 °C, while fish samples were kept at –20 °C until analysis.

Sample preparation and analysis

The collected water samples were adjusted to pH 2 using concentrated nitric acid and passed through a 0.45 µ m PTFE filter before analysis. Sediment samples were dried in shaded areas, passed through a 2 mm sieve, and homogenized. Fish samples were lyophilized and homogenized. The prepared sediment and fish samples were divided into aliquots of 2 g and transferred into 30 mL glass tubes. Subsequently, 5 mL of concentrated nitric acid was added to each tube. After the samples were cooled, 2 mL of perchloric acid was added, and the mixture was extracted at 80 °C for 1 h. The resulting extract was passed through a 0.45 µ m PTFE filter, transferred to a 50 mL volumetric flask, and diluted with ultrapure water up to the calibration mark in preparation for analysis.

All samples underwent quantitative analysis using inductively coupled plasma-optical emission spectrometry (ICP-OES; Agilent 5100, Santa Clara, CA, USA). Calibration curves were constructed for 10 metals using nine data points ranging from 0.001 to 2 mg/L. The coefficients of determination (r²) obtained for all metals were greater than 0.99. Metal concentrations in the samples were determined by compensating for dilution factors and were expressed using appropriate units. The method detection limits (MDLs) in water were as follows: 0.006 mg/L for Cr, 0.03 mg/L for Pb, 0.004 mg/L for Cd, 0.006 mg/L for Cu, 0.002 mg/L for Mn, 0.002 mg/L for Zn, 0.01 mg/L for Ni, 0.002 mg/L for V, 0.01 mg/L for Al, and 0.005 mg/L for Fe. The MDLs in sediment and fish were as follows: 0.14 mg/kg for Cr, 0.82 mg/kg for Pb, 0.09 mg/kg for Cd, 0.14 mg/kg for Cu, 0.04 mg/kg for Mn, 0.05 mg/kg for Zn, 0.36 mg/kg for Ni, 0.05 mg/kg for V, 0.36 mg/kg for Al, and 0.13 mg/kg for Fe.

Statistical analysis

Statistical analysis was conducted using Student’s t-test, one-way analysis of variance (ANOVA), and Pearson correlation analysis. IBM SPSS 26 was used to determine significant differences at p < 0.05.

Risk assessment

Ecological risk assessment

The BCF and BSAF (biota-sediment accumulation factor) were determined by dividing the metal concentration in fish by the metal concentration in water and sediment, respectively, and expressing the result in logarithmic form. The formula below was used:

(1) BCF or BSAF=CFishCwater or Csediment

The above factors are typically denoted in L/kg or kg/kg. C_Fish is the concentration of each metal (mg/kg) in each fish sample, whereas C_Water (mg/L) and C_Sediment (mg/kg) represent the average metal concentrations in water and sediment, respectively, for each season.

Geochemical pollution was assessed using the I_geo, contamination factor (CF), and pollution load index (PLI). The I_geo and CF indices were used to evaluate aquatic toxicity and to compare metal concentrations in sediments from contemporary and pre-industrial periods. The evaluation criteria and formulas are presented below.

(2) Igeo=log2Cn1.5×Bn

(3) CF=CnBn

(4) PLI=CF1×CF2×CF3×⋯×CFnn

Where Cn and Bn represent the metal concentration (mg/kg) and background concentration (mg/kg) in the sediment sample, respectively, with n indicating the number of substances to be evaluated. The factor 1.5 in the I_geo formula accounts for the lithospheric impact; the specific Bn values are detailed in Table 1.

Table 1.

Concentrations of 10 metals in water and sediment samples, reported by previous studies, and their corresponding guideline values.

The ecological risk was assessed through a potential ecological risk index (RI) as defined by Tian et al. [18], using the formula below:

(5) EI=Ti×CFi

(6) RI=∑i=1nEI

Where the individual potential risk (EI) is calculated by multiplying the CFi of each metal by the corresponding toxicity effect coefficient (T_i). The T_i values for each metal are as follows: Cr, 2; Pb, 5; Cd, 30; Cu, 5; Mn, 1; Zn, 1; Ni, 5 [19]. Each ecological risk criterion is presented in Table S1.

Distribution of metals in water

Surface water was collected from eight points in Yeongrang Lake according to the season, resulting in 32 samples. Table 2 shows the results of the analysis of metal concentrations. Eight metals, excluding Cr and Cd, had concentrations above the MDL. Zn and Al were detected in all samples, with concentrations ranging from 10.8 to 868 µ g/L and 26.2 to 360 µ g/L, respectively. Cu, Mn, and Fe showed a detection rate of over 60 %, with concentrations up to 1,922, 915, and 924 µ g/L, respectively. The remaining metals (Pb, V, and Ni) were detected in three, 12, and five samples, respectively, indicating low detection rates. The US EPA recommends the criterion continuous concentration (CCC) and criterion maximum concentration (CMC) as measures to determine the quality of water to which aquatic life is exposed [24]. The recommended values (except for those of Mn and V) are shown in Table 1. According to the Canadian Water Quality Guidelines for the Protection of Aquatic Life and the Federal Environmental Quality Guidelines provided by Environment and Climate Change Canada, the concentrations of Mn in freshwater should be 3,600 µ g/L and 430 µ g/L, respectively, for short- and long-term exposures. Furthermore, V should be maintained below 120 µ g/L [25, 26]. In the 32 water samples in this study, five metals (Pb, Cu, Mn, Zn, Ni, and Fe) exceeded the aforementioned guidelines. Cu showed the highest excess rate at 68.8% (22/32), while Pb, Mn, Zn, Ni, and Fe exhibited low excess rates of 9.38 % (3/32), 3.13 % (1/32), 3.13 % (1/32), 3.13 % (1/32), and 3.13 % (1/32), respectively. Cu has been detected in limestone, the main raw material in cement, ranging from not detected (ND) to 67 mg/kg, and Cr is the metal with the second highest concentration ranging from 0.7 mg/kg to 63.1 mg/kg [9].

Table 2.

Distribution of 10 metals in water samples collected at eight sampling sites in Yeongrang Lake across four seasons.

Figure S1 depicts the distribution of the eight metals in the sampling points, except for Al and Fe, which can reach concentrations of up to hundreds of µ g/L in water. Pb, Cu, Mn, Zn, and Ni exhibited higher concentrations in water than Cr, Cd, and V. The concentrations of the eight metals over the sampling period followed the order of Cu > Zn > Mn > Pb > Ni > Cr > Cd > V. The overall average concentrations were 84.8 µ g/L for Cu, 52.8 µ g/L for Zn, 35.9 µ g/L for Mn, 19.6 µ g/L for Pb, 11.1 µ g/L for Ni, 2.90 µ g/L for Cr, 1.86 µ g/L for Cd, and 1.66 µ g/L for V. Cu, which exhibited the highest concentration in the spring, reached its peak at 1,922 µ g/L near the bridge (S2). The downstream points (S6–S8) had concentrations of 39.3–49.2 µ g/L, higher than the upstream points (S3–S5; 2.82–40.4 µ g/L). The average concentration of Cu was 10.5 µ g/L in the winter, 263 µ g/L in the spring, 5.97 µ g/L in the summer, and 59.6 µ g/L in the fall. Previous studies have reported average concentrations of 119 µ g/L on the coast and 1.05–19.0 µ g/L in the river for Cu, which showed the highest excess rate [18,27,28]. The concentrations of Cu in the spring and fall were approximately 14 and three times higher than those in the previous study, although they were lower in the winter and summer. However, only V showed a significant seasonal difference between spring and fall (p < 0.05), while the other metals did not exhibit seasonal variations. Moreover, a Pearson correlation analysis revealed a significant positive correlation between Pb, Cu, and Zn, and Mn, Al, and Fe in the water sample, which may have been affected by similar sources from agricultural and residential areas upstream (Table S2). In previous studies conducted in Pakistan and India, the concentrations of metals such as Cd, Zn, Co, Ni, Cu, Cr, and Fe were higher after the monsoon than before. The Water Quality Index (WQI) was also higher before the monsoon. This suggests that the high metal concentrations after the monsoon were due to runoff water from rainfall [2,29]. Therefore, the observed increase in the concentrations of Pb, Cu, Zn, and Fe in the fall—the season after summer, when Korea gets approximately 58 % of the annual precipitation—is expected to be the result of runoff water from the upstream section, where residential and rural areas are located.

Yeongrang Lake is connected to the sea and has an average salinity of 25–30 ‰, which is concerning because of the potential for marinization. Hong et al. [30] reported that the proportion of saltwater fish was approximately 30.00 % higher than in the previous period (2005–2007) based on a fish community survey conducted in 2023. Although there were no seasonal differences, the installation of the low-level bridge in Yeongrang Lake might have caused ecological changes by affecting the inflow of seawater and causing water stagnation. Moreover, certain metals exceeded the CCC values recommended by the US EPA [24]; thus, the potential for contamination from cement structures associated with the low-level bridge cannot be ignored.

Distribution of metals in surface sediment

All metals examined were detected above the MRL in all 32 surface sediment samples. The concentration range of each metal is presented in Table 3. Cr was in the range of 0.853-14.4 mg/kg; Pb, 2.17–10.3 mg/kg; Cd, 0.285–10.3 mg/kg; Cu, 2.18–8.92 mg/kg; Mn, 23.5–8.92 mg/kg; Zn, 9.47–41.3 mg/kg; Ni, 0.470–13.8 mg/kg; V, 2.76–13.8 mg/kg; Al, 1,896–10,140 mg/kg; and Fe, 2,533–17,485 mg/kg. Table 1 presents the threshold effect level (TEL), probable effect level (PEL), and average sales value (ASV), calculated according to the guidelines established for metals in sediments. All metals were detected above the MDL in all sediment samples; however, only four samples exhibited Cd levels surpassing the TEL of 0.68 mg/kg. Cd concentrations ranging from ND to 2.48 mg/kg have been found in limestone (the primary constituent of cement); however, the total Cr concentration in limestone is notably higher, reaching up to 63.1 mg/kg [9]. Nevertheless, the concentration of Cr in the sediments examined in this study was below the TEL, PEL, and ASV, indicating lower levels than in previous studies (23.1–59.4 mg/kg in coastal and river areas) (Table 1).

Table 3.

Distribution of 10 metals in sediment samples collected at eight sampling sites in Yeongrang Lake across four seasons

Figure S1 presents the distribution based on the sampling points for the remaining eight metals (except for Al and Fe), found at levels in the tens of thousands of mg/kg in the sediment. Mn accounted for the highest proportion of metals in the sediment, followed by Zn and V. The concentrations of the eight metals during the sampling period were in the order of Mn > Zn > V > Pb > Cu > Cr > Ni > Cd, with average concentrations of 101 mg/kg for Mn, 24.0 mg/kg for Zn, 6.06 mg/kg for V, 5.75 mg/kg for Pb, 3.81 mg/kg for Cu, 3.04 mg/kg for Cr, 1.26 mg/kg for Ni, and 0.563 mg/kg for Cd. The concentration of Cd in two sediment samples during the winter and one sample each during the summer and fall exceeded the TEL. The average Cd concentration was 0.659 mg/kg in the winter, 0.430 mg/kg in the spring, 0.570 mg/kg in the summer, and 0.593 mg/kg in the fall. No significant seasonal variations were observed in the concentrations of Cd and other metals (p > 0.05). Pearson correlation analyses of all samples revealed that all metals, except for Mn, exhibited positive correlations. This suggests that the distribution of metals in the sediment may have been influenced by common sources such as agricultural, residential, or crustal component (Table S2). Furthermore, seven metals (Cr, Pb, Cu, Mn, Ni, V, and Al) exhibited significant correlations with the concentration of V in water. In prior investigations, the concentration of Cd ranged between 0.063 and 0.25 mg/kg in coastal areas and 3.78 mg/kg in riverine environments (Table 1). The concentration of Cd in the current study was higher than that reported in coastal areas but lower than that in the river [31,32]. Yeongrang Lake is believed to have been connected to the east coast through the river’s upstream current, being influenced by both river and seawater. The salinity levels in the lake reach up to 31.5 ‰ at a depth of 3 m [30].

Distribution of metals in fish

Forty-nine fish samples were collected during the sampling period, and the concentrations of 10 metals are presented in Table S3. Mn, Zn, and Fe were detected in all fish samples. V, Cr, Cu, and Al were detected in 48, 46, 39, and 39 samples, respectively, indicating a high detection rate exceeding 80 %. By contrast, the detection rates for the remaining metals (Pb, Cd, and Ni) were fairly low, indicating 35 % (17/49) for Pb, 29 % (14/49) for Cd, and 29 % (14/29) for Ni. The concentrations of Mn, Zn, and Fe in all samples ranged from 1.38 to 277 mg/kg, 17.1 to 193 mg/kg, and 3.45 to 434 mg/kg, respectively. Metals such as V, Cr, Cu, and Al exhibited the next highest detection rates, with concentrations of up to 26.3, 6.04, 7.35, and 552 mg/kg, respectively.

The fish collected from Yeongrang Lake were classified as benthic and floating, based on their habitat characteristics. Out of a total of 16 fish species, eight species inhabit the benthic zone, while the other eight are floating species. Rhinogobius brunneus, Gymnogobius urotaenia, Tridentiger brevispinis, Gymnogobius breunigii, Tridentiger bifasciatus, Acanthogobius flavimanus, Platichthys stellatus, and Embiotocidae are benthic fish species. Carassius, Tribolodon hakonensis, Gasterosteus aculeatus, Hypomesus nipponensis, Konosirus punctatus, Takifugu niphobles, and Engraulis japonicus are fish species that exhibit a floating behavior. Significant variances were observed between benthic and floating fish species concerning five of the metals analyzed (Cr, Pb, Cd, Cu, and Ni) (see Figure 2; Student’s t-test, p < 0.05). The average concentrations of these metals were 1.1-4.2 times higher in the benthic fish species (1.20 mg/kg for Cr, 0.0201 mg/kg for Cd, 2.74 mg/kg for Cu, and 0.823 mg/kg for Ni). However, the Pb levels were 4.30 times higher in the floating fish species than in the benthic fish species, with an average of 0.44 mg/kg. Although no significant differences between benthic and floating

Figure 2.

Concentrations and seasonal distribution of 10 metals in fish specimens collected from Yeongrang Lake. The asterisks (*) denote p < 0.05.

fishes were observed for the other metals, Mn, Zn, V, and Al, were found to be 1.25–12.6 times higher in the benthic fish, indicating 35.8 mg/kg, 44.3 mg/kg, 2.83 mg/kg, and 99.9 mg/kg, respectively. In addition, the concentration of Fe was 1.8 times higher in benthic fish (53.5 mg/kg) than in floating fish.

A Pearson correlation analysis was conducted to establish the relationship among fish, water, and sediments as sources of metal contamination (Table S4 and S5). A positive correlation was observed between the Cr, Ni, Al, and Fe concentrations in benthic fish and Mn, Ni, and Fe concentrations in water. By contrast, among floating fish species, only Pb exhibited a positive correlation with V in water. Nevertheless, a correlation between Mn and Al in benthic fish and sediment was confirmed. A previous study conducted a correlation analysis to investigate the relationship between metal concentrations in water and sediments and metal presence in fish tissues. The study found a significant positive correlation between the combined metal concentrations in water (Zn, Pb, Cd, Ni, Cr, and Cu) and in the gills and muscles of fish. Sediment analysis revealed a positive correlation with various organs and bodily fluids, including the liver, intestines, blood, brain, spleen, and bile. However, no direct correlation of bioaccumulation was observed in the present study. Therefore, an appropriate fish species should be chosen for future research to perform a correlation analysis based on the data from the fish tissues.

According to the collection time of fish samples, seasonal variations were observed for certain metals (Figure 2). The average concentrations of Cr, Cd, and Ni in winter—the primary collection period for all fish species—were 1.77, 0.0182, and 0.751 mg/kg, respectively. These values were significantly higher than those recorded in other seasons (spring, summer, and fall) (p < 0.05). The concentrations of Mn and V were 51.2 and 4.43 mg/kg in winter, respectively, indicating higher levels than those in fall (p < 0.05). The average concentration of Cu in water, which exceeded the CCC [24], was lower during the winter and spring (first and second collection periods) than during the summer and fall (third and fourth collection periods). While Cu is an essential trace element for growth and metabolism, its accumulation in the liver or kidney can affect oxidative metabolism, lipid peroxidation, and protein levels [15]. This study was constrained by limitations related to obtaining a sufficient number of representative fish species that could be consistently collected. Further research endeavors should consider various factors, such as selecting fish species that can demonstrate bioconcentration of metals in their body tissues and organs, as well as the fish’s body size.

Risk assessment

BCF and BSAF

Based on the metal concentration in water and sediments (Figure 3), Mn and Zn exhibited the highest BCF values across all fish species, whereas the lowest values corresponded to Pb and Cd. Even in sediments, the highest BSAF corresponded to Zn, and the lowest, to Al and Fe. Twenty-two of the fish species sampled were benthic and 27 were floating. Mn and Zn exhibited elevated levels in water, with Zn having the highest concentration in sediments. For floating fish, the BCFs for 10 metals in water ranged from 0.122 to 4,043 L/kg. The average BCFs were 1,111 L/kg for Mn, 1,109 L/kg for Zn, and 187 L/kg for Cu. For benthic fish, the BCFs for these metals in water ranged from 1.71 to 24,983 L/kg, with an average of 3,300 L/kg for Mn, 2,301 L/kg for Zn, and 250 L/kg for Cu. For sediment samples, the BSAFs for these metals varied from 0.000463 to 3.58 kg/kg in floating fish and from 0.00110 to 6.79 kg/kg in benthic fish. Thus, the BSAFs in sediment are substantially lower than those in water. The average BSAF of Zn was 1.54 kg/kg in floating fish and 1.67 kg/kg in benthic fish. These findings align with prior research that found that Mn, Zn, and Cu exhibited higher values than metals such as Fe and Pb in the liver of fish from the Nile River in Egypt [33]. Similarly, Mn and Zn exhibited the highest value in all fish samples.

Figure 3.

Box plots illustrating the bioconcentration factors for benthic and floating fishes in Yeongrang Lake.

Mn, Zn, and Cu are absorbed in the gills, gastrointestinal tract, and digestive system, and are essential nutrients for immunity, enzyme function, protein synthesis, and energy metabolism [5]. Zn showed the highest BCF in the present study. Zn had a BCF ranging from 4 to 24,000 L/kg for 12 aquatic organisms in steady-state conditions [34]. However, Zn does not exhibit biomagnification properties through the food chain. Therefore, the bioconcentration of Zn may not be significant. Nonetheless, Zn should be monitored because of its toxicity (e.g., hemoglobin reduction effects) [5]. Moreover, no significant differences were observed between floating and benthic fishes regarding the BCFs and BSAFs of Mn, Zn, and Cu (p > 0.05). Given the metabolic characteristics, population distribution and size, habitat timing, and mobility of fish, acquiring a sufficiently large sample for comparison poses a challenge. Hence, this study aimed to compare the concentrations based on the benthic and floating characteristics of the fish species.

Ecological risk assessment in surface sediment

Metals in sediments are quantified using various indices such as CF, I_geo, PLI, and RI, which consider the preindustrial background concentrations. The contamination level was assessed based on the criteria specified in Table S1. Figure 4 illustrates the assessment of ecological risks by season, excluding V, Al, and Fe, due to challenges in verifying the necessary values (Bn and Ti) for analyzing the surface sediments obtained from Yeongrang Lake. The average concentration of Cd shows high CF values during the winter, summer, and fall, and considerable values in the spring. Only the CF of Cd was high, whereas other metals exhibited low CF values. By contrast, the concentration of Cd fell within an I_geo range indicating a moderate level of pollution. This can be attributed to the background concentration being multiplied by a factor of 1.5, considering the lithosphere’s influence. Furthermore, the low contamination levels of metals other than Cd suggest that the sediments were unpolluted according to the PLI.

Figure 4.

Seasonal ecological risk assessment of seven metals in sediment samples collected in Yeongrang Lake.

The RI serves as an indicator for potential ecological risk assessment, showing substantial risk throughout all seasons, mainly due to the elevated EI value of Cd. Although the contamination levels assessed by the I_geo and PLI were low during the winter, the results indicate values approaching very high levels in the potential ecological risk assessment. The seasonal ecological risk in Yeongrang Lake remains constant (ANOVA, p > 0.05) and chronic exposure to aquatic organisms is a major concern. Cd is a toxic metal that disrupts Ca ion homeostasis and damages the kidneys, skeleton, and lungs without providing any beneficial effects [5,6]. The higher concentration of Cd in benthic fish than in floating fish (p < 0.05) is of particular significance and raises concerns about the potential risk to aquatic ecosystems. Therefore, efforts to reduce ecological risks from metals are necessary.

Nevertheless, a limitation persists regarding the extent to which the installation of the bridge has significantly impacted the increased ecological risk associated with cadmium (Cd). This uncertainty primarily stems from the lack of background data from the period immediately preceding the construction of the bridge. Instead, background values have been inferred from the deepest sediment layers, which reflect conditions from the 1990s, or derived through cumulative frequency curves and linear regression analyses conducted in the South and East Seas of South Korea [32,39]. Therefore, further research is necessary to clarify the sources of metals and to conduct compositional analyses of the concrete materials.

Human risk assessment

A non-carcinogenic human risk assessment was conducted for 16 fish species collected in this study. The assessment involved oral exposure to all the metals analyzed (Table 4). V exhibited the highest values at 3.12×10¹ and 1.81×10² for the average and 95th percentile concentrations, respectively. Pb showed the second highest HQ but remained below 1. V was detected at levels ranging from 0.0600 to 14.0 mg/kg across different fish species, with an HQ ranging from 2.00×10⁻¹ to 4.66×10¹. The HI values of most fish species were above 1. Given the fish consumption rates among Koreans, with 54.3% for Engraulis japonicus, 17.9% for octopus and squid, and 17.5% for fish cakes, and considering the unlikelihood of solely consuming fish from a specific region throughout one’s lifetime, the actual risk is expected to be substantially lower. The HQ for V in Engraulis japonicus exceeded 1, while that for Pb was 0.559 (i.e., close to 1). Therefore, managing the concentrations of metals in the aquatic ecosystem is crucial to prevent risks to human health.

Table 4.

Non-carcinogenic human risk assessment of 10 metals from fish consumption.