Influencing factors and health risk assessment of microcystins in the Yongjiang river (China) by Monte Carlo simulation

Introduction

Eutrophication of freshwater bodies can result in algal blooms, especially those caused by cyanobacteria. The algal toxins secreted from cyanobacteria are possibly harmful to plants, animals, and humans (Holland & Kinnear, 2013; Cao et al., 2017). So far, most of as-known 90 microcystins (MCs) have been isolated from species and strains of Microcystis (Pham & Utsumi, 2018). Among them, the most widely distributed are microcystin-LR (MC-LR), microcystin-RR (MC-RR), and microcystin-YR (MC-YR) (Zegura, 2016). These toxins are synthesized in the cells and released after cell rupture, finally appeared as MCs in the water source.

Cyanobacteria blooms exist in eutrophicated waters worldwide, so that MCs can be bioaccumulated by aquatic animals and reach human bodies. These would severely harm human health and cause illness or deaths. In 1975, the drinking water source in the small town of Pennsylvania was contaminated by Microcystis, which resulted in acute gastroenteritis for over half of the local population (Keleti et al., 1979; Keleti & Sykora, 1982; Lippy & Erb, 1976). In 1996, due to contamination by MCs occurring in a hemodialysis center in Brazil, 116 of 130 patients developed symptoms of blurred vision and nausea, and >50 individuals succumbed to mortality (Pouria et al., 1998). Studies on drinking water showed that with drinking ditch pond water containing MCs, the mortality rate of local people in Haimen and Fu Sui caused by hepatocellular carcinoma reached about 100/100,000, which was significantly higher than that of shallow wells or deep wells (20/100,000) (Ueno et al., 1996). In 2010, the International Agency for Research on Cancer (IARC) listed MCs as a “possible human carcinogen” (Group 2B) based on its potential carcinogenicity (International Agency for Research on Cancer (IARC), 2010).

Although it has been confirmed that MCs can cause acute and chronic damage to human bodies, the risk assessment of MC-LR, MC-RR, and MC-YR in the Yongjiang river of China remains lack of reports. The environmental conditions of water source are crucial in the concentration levels of toxins. However, the factors affecting the concentration levels of MCs (nutrient levels and climatic conditions) in Yongjiang have not yet been elucidated. Therefore, it is urgent to investigate the concentration and distribution of MCs in Yongjiang river as influenced by seasonal changes in water quality and the related parameters.

Owing to the limitations of conducting toxicological health risk assessments in a population, the Monte Carlo simulation (mathematical and logical model) has been widely used in recent years. It was used to understand the behavior of water systems by assuming different problems and systems, showing an advantage over experiments (Clausen et al., 2017; Sasaki et al., 2017). Moreover, the Monte Carlo simulation of uncertainties was applied in the risk assessment model by collecting limited samples to predict the overall situation. As a result, the risk uncertainty was expressed intuitively, in agreement with the order of the nature, which favors for a decision-making for risk managers and policymakers (Paladino, Moranda & Seyedsalehi, 2017; Sasi, Yozukmaz & Yabanli, 2017). The US Environmental Protection Agency (USEPA) has set the Monte Carlo simulation as a basic method in the risk analysis policy (Moolenaar, 1996). Currently, American Palisade has developed the @Risk Monte Carlo software loading into Excel simulation technology for risk assessments. It is mainly based on the analysis of stochastic simulation method of Monte Carlo, which provides various predicted results by using a variety of probabilistic simulations, including the occurrence probability of events with a risk, forecasting the risk of uncertainty quantitatively, and summarizing the characterization results. For instance, Li et al. (2017) used the Monte Carlo model to assess the quantitative risk of aluminum in Youtiao, which did not exceed the provisional tolerable weekly intake set by the Joint Expert Committee on Food Additives for the public. Jia et al. (2018) used the Monte Carlo simulation to evaluate the trace elements in the four freshwater fishes from a mine-impacted river, and found that the consumption did not exert any appreciable adverse impact on human health due to the exposure to trace elements in fish muscle.

This study investigated the current status of drinking water sources in the Yongjiang river in China with respect to the contamination of MCs. The professional health risk assessment software @Risk7.5 was used to evaluate the risks of dietary intake of MCs, including the carcinogenic and non-carcinogenic risks. These findings provide a basis to develop an effective control of water pollution and quality in order to protect the human health in the specific area.

Materials and Methods

Sampling location

The Yongjiang river is a major water resource with an average annual flow of 1,292 m³/s in the Nanning City area (Fig. 1). The surface area of the river is 2,676 ha with a maximum depth of 23 m. It is the main urban water source in Nanning city, China, and the tributary channel is also a vital transportation route. Except for that, it is used for recreation as well as the source of water for domestic use, agriculture, fishery, and industry.

Determination of MCs

The air-dried samples were suspended into one mL deionized distilled water for high performance liquid chromatography (HPLC) (LC-20A; Shimadzu, Kyoto, Kyoto Prefecture, Japan) analysis. The solubilized toxin samples were analyzed using HPLC with UV detector at 238 nm and symmetrical C18 column (3.9 × 150 mm) (Waters, Milford, MA, USA). The mobile phase consists of 33% acetonitrile and 67% deionized distilled water in 0.1% phosphate buffer (pH = 3.0). The flow rate was set at one mL/min. The injection volume was 20 μL and the column temperature was 45 °C. MC-LR and MC-RR (Solarbio, approximate purity, 95%) and MC-YR (Alexis, approximate purity 98%) were used as standards. Furthermore, the concentrations of MCs were determined by calibrating such area under the peak with the corresponding standard curves. MC-LR, MC-RR, and MC-YR showed a good linearity in the range of 0.025–2 μg/L (r² = 0.9987, 0.9992, 0.9997). Under the condition that the signal to noise ratio (S/N) is 3, the detection limits of MC-RR and MC-LR is 0.0125 μg/L, and the detection limit of MC-YR is 0.014 μg/L. The recoveries ranged from 91% to 110%, the relative standard deviation was 3.0–5.6%. A series of toxin peaks were identified using retention time and compared with spikes and known standards in the blank samples. Furthermore, the concentrations of MCs were determined by calibrating such area under the peak with corresponding standard curves. The order of the peaks and time of each standard substance were as follows: MC-RR (7.599 min), MC-YR (14.225 min), and MC-LR (17.601 min). The test sample was analyzed in 18 min (Fig. 2).

Results

Concentration level, distribution characteristics, and environmental impact factors of MCs in source water

Concentration distribution characteristics of MCs in source water

The concentrations of MC-LR, MC-RR, and MC-YR in water samples were detected. The total concentrations of MCs (TMCs) in water are the sum of the concentrations of extracellular MCs (EMCs) and IMCs dissolved in the water. The results were summarized in Table 1.

Table 1:

Results of analysis of MCs in water samples.

Toxin types	Number of samples	EMCs	IMCs	TMCs (EMCs + IMCs) concentration (μg/L)
		Detection Rate (%)	Detection Rate (%)	Range	Mean ± SD
MC-RR	90	74.44	77.78	0.0224–0.3783	0.0727 ± 0.0599
MC-YR	90	64.44	56.67	0.0329–0.1433	0.0424 ± 0.0376
MC-LR	90	77.78	76.67	0.0341–0.2663	0.0763 ± 0.0637

DOI: 10.7717/peerj.5955/table-1

Note:

The total concentrations of MCs (TMCs) in water is the sum of the concentrations of extracellular MCs (EMCs) and intracellular MCs (IMCs) dissolved in the water; standard deviation means SD.

Seasonal distribution characteristics of MCs in water samples

The seasonal distribution of MCs in the Yongjiang river is shown in Fig. 3. The concentration of TMC-YR (referring to the sum of intracellular and extracellular, the same to the other twos) was significantly lower than the other two MCs. The concentrations of TMC-RR and TMC-YR at the same time reached the maximum levels in October.

Pearson’s correlation analysis of environmental factors and MCs’ concentration

Pearson’s correlation analysis of environmental factors and MCs’ concentration were analyzed and the results were shown in Table 2. According to the correlation analysis, the concentration of TMC-RR was positively correlated with water temperature and TN (p < 0.01) with a significant negative correlation with pH (p < 0.01). The concentration of TMC-YR was negatively correlated with TN:TP ratio (p < 0.05). The concentration of TMC-LR was positively correlated with water temperature, TP, and PO₄³⁻ − P (p < 0.01) with a significant negative correlation with DO and TN:TP ratio (p < 0.01). The concentration of TMCs was positively correlated with water temperature and TP (p < 0.01) with a significant negative correlation with pH, DO, and TN:TP ratio (p < 0.01).

Table 2:

Correlation coefficients between MCs and the influencing factors.

Environmental Factors	Correlation Coefficient
	TMC-RR	TMC-YR	TMC-LR	TMCs
Water temperature	0.436**	0.085	0.480**	0.614**
PH	−0.729**	−0.029	−0.164	−0.566**
DO	−0.063	−0.076	−0.768**	−0.570**
TN	0.286**	−0.179	−0.079	0.055
TP	0.043	0.073	0.851**	0.610**
PO43− − P	−0.047	−0.071	0.399**	0.204
TN:TP ratio	0.040	−0.229*	−0.434**	−0.347**

DOI: 10.7717/peerj.5955/table-2

Notes:

* Correlation is significant at the 0.05 level (two-tailed).

** Correlation is significant at the 0.01 level (two-tailed).

Stepwise multiple linear regression analysis of MCs’ concentration and environmental factors

A stepwise multiple linear regression analysis of MCs’ concentration and environmental factors is shown in Table 3. The results indicated that TP (χ³) is the dominant factor affecting the contents of TMC-LR and TMCs. pH (χ²) and TN:TP ratio (χ⁷) are the primary factors affecting the content of TMC-RR and TMC-YR, respectively. These findings were in agreement with the results of correlation analysis.

Table 3:

Results of stepwise multiple linear regression.

MCs	Fitting equation	Correlation coefficient R	Adjusted R2	Tolerance	VIF	Model fitting		Partial regression coefficient test	D–W statistic
						F	P
TMCs	y = 116.848 + 46.449χ3	0.581	0.337	1	1	44.792	<0.001	χ3(t = 6.693, P < 0.001)	1.979
	y = 805.526 + 42.979χ3 − 92.886χ2	0.716	0.513	0.989	1.001	45.87	<0.001	χ3(t = 7.148, P < 0.001)
								χ2(t = −5.608, P < 0.001)
	y = 538.069 + 35.557χ3 − 75.968χ2 + 6.452χ1	0.741	0.549	0.693	1.443	34.938	<0.001	χ3(t = 5.495, P < 0.001)
								χ2(t = −4.396, P < 0.001)
								χ1(t = 2.622, P = 0.01)
TMC-RR	y = 799.874 − 99.263χ2	0.718	0.515	1	1	93.396	<0.001	χ2(t = −9.664, P < 0.001)	1.828
	y = 825.53 − 101.047χ2 − 35.61χ4	0.738	0.544	0.994	1.006	51.98	<0.001	χ2(t = −10.066, P < 0.001)
								χ4(t = −2.375, P = 0.02)
	y = 676.095 − 91.169χ2 − 39.668χ4 + 3.42χ1	0.759	0.576	0.846	1.181	38.908	<0.001	χ2(t = −8.682, P < 0.001)
								χ4(t = −2.709, P < 0.001)
								χ1(t = 2.522, P = 0.014)
TMC-YR	y = 1.54 − 0.013χ7	0.268	0.072	1	1	6.83	0.011	χ3(t = −2.613, P = 0.011)	2.299
TMC-LR	y = 11.216 + 41.829χ3	0.806	0.650	1	1	163.585	<0.001	χ3(t = 12.79, P < 0.001)	1.461
	y = 63.349 + 32.481χ3 − 6.653χ6	0.848	0.718	0.678	1.476	111.004	<0.001	χ3(t = 9.06, P < 0.001)
								χ6(t = −4.592 P < 0.001)

DOI: 10.7717/peerj.5955/table-3

Note:

The fitting result of TMC-YR after lg10 conversion; D–W statistic (Durbin–Watson statistic); Variance Inflation Factor (VIF).

Assessment of MCs’ exposure in source waters

Distribution fitting of the concentration of MCs in source water

The @Risk7.5 software is used to fit the processed samples. The concentration of MCs in source water were characterized as continuous data. The fitting results were followed by the optimal fitting distribution models: Gamma, Invgauss, Lognorm, Expon, and Loglogistic. Three main methods were used to test the goodness of the fittings: the Chi-Sq (Chi-squared) test, the K–S (Kolmogorov–Smirnov) test, and the A–D (Anderson–Darling) test (Lipton et al., 1995; Cummins et al., 2009). Above all, the sample fitting results (Table 4) were used to determine the fitting distribution types of the optimal probability of the pollution data: MC-RR had Gamma and Invgauss distribution, MC-YR had Lognorm, Expon, and Loglogistic distribution, and MC-LR had Gamma and Expon distribution. Furthermore, the results of the distribution parameters after fitting and the comparison with the sample data parameters are summarized in Table 5. The probability distribution of the mass concentration of MCs, MC-RR, MC-YR, and MC-LR in source water is shown in Figs. 4–6. The fitting results can be visually observed from the coincidence of the blue rectangular shape and the area under the red curve.

Table 4:

Fitting distribution and related parameters of MCs in source water (μg/L).

MCs	Fitting of Distribution	Distributed parameters		Fit test sort			50% Confidence value	90% Confidence value	95% Confidence value
		Mean	Std Dev	K-S	A-D	Chi-Sq
MC-RR	Gamma	0.073	0.061	1	1	1	0.056	0.154	0.194
	Invgauss	0.073	0.066	5	2	2	0.053	0.154	0.202
MC-YR	Lognorm	0.047	0.071	4	2	1	0.026	0.103	0.154
	Expon	0.042	0.039	2	1	4	0.030	0.093	0.120
	Loglogistic	0.081	–	1	4	6	0.026	0.124	0.216
MC-LR	Gamma	0.076	0.070	1	1	2	0.055	0.168	0.216
	Expon	0.075	0.073	2	2	1	0.053	0.170	0.221

DOI: 10.7717/peerj.5955/table-4

Table 5:

Estimated value of the quantile for overall sample in different theoretical distributions.

Projects	Pollutant	Real value (μg/L)	Types	Predicted value (μg/L)	Relative difference of real value (%)
P50	MC-RR	0.058	Gamma	0.056	3.01
			Invgauss	0.053	7.81
			Lognorm	0.026	8.82
	MC-YR	0.024	Expon	0.030	25.17
			Loglogistc	0.026	10.03
	MC-LR	0.057	Gamma	0.055	2.02
			Expon	0.053	6.10
P75	MC-RR	0.011	Gamma	0.099	13.00
			Invgauss	0.095	16.91
			Lognorm	0.053	13.88
	MC-YR	0.062	Expon	0.057	8.17
			Loglogistc	0.056	9.41
	MC-LR	0.112	Gamma	0.104	6.79
			Expon	0.104	7.46
P90	MC-RR	0.136	Gamma	0.154	12.93
			Invgauss	0.154	13.09
			Lognorm	0.103	4.31
	MC-YR	0.108	Expon	0.093	14.10
			Loglogistc	0.124	15.12
	MC-LR	0.166	Gamma	0.168	1.09
			Expon	0.170	2.46
P95	MC-RR	0.167	Gamma	0.194	16.26
			Invgauss	0.202	21.06
			Lognorm	0.154	29.08
	MC-YR	0.119	Expon	0.120	0.58
			Loglogistc	0.216	80.78
	MC-LR	0.215	Gamma	0.216	0.26
			Expon	0.221	2.56
P99	MC-RR	0.378	Gamma	0.285	24.71
			Invgauss	0.321	15.02
			Lognorm	0.325	127.11
	MC-YR	0.143	Expon	0.183	27.56
			Loglogistc	0.735	412.77
	MC-LR	0.266	Gamma	0.326	22.62
			Expon	0.338	27.03

DOI: 10.7717/peerj.5955/table-5

Daily exposure calculation

The @Risk7.5 software was utilized for the random extraction of the MCs concentration profiles from the water to calculate the daily exposure of direct drinking water by different populations to MC-RR, MC-YR, and MC-LR. Each simulation cycle was performed for 10,000 cycles, and the simulation results are shown in Table 6. A significant difference between adults and children was observed in the daily exposure. P50, P85, P90, and P95 (Table 6) represent the high exposure sites of each population. The MCs exposed to drinking water showed that the children’s daily intake was twofold higher than that of the adults, suggesting that children are more susceptible to the pollution of MCs than adults.

Table 6:

Daily exposure to MCs intake through drinking water (μg/kg/d).

Projects	MC-RR		MC-YR		MC-LR
	Adult	Child	Adult	Child	Adult	Child
Mean	0.002	0.005	0.001	0.003	0.002	0.005
P50	0.002	0.004	0.001	0.001	0.002	0.004
P85	0.004	0.008	0.003	0.006	0.004	0.009
P90	0.004	0.009	0.003	0.007	0.005	0.010
P95	0.005	0.010	0.003	0.007	0.006	0.013

DOI: 10.7717/peerj.5955/table-6

Discussion

The WHO established a guide value of one μg/L for MC-LR concentration in drinking water (World Health Organization (WHO), 2017). A comprehensive Australian report shows that the concentration of total MCs in drinking water should not exceed 1.3 μg/L, expressed as MC-LR toxicity equivalents. Furthermore, a cell density of approximately 6,500 cells/mL (biovolume of 0.6 mm³/L) would be equivalent to the guideline of 1.3 μg/L MC-LR toxicity if the toxin was fully released into the water (NHMRC & NRMMC, 2011). In the study of a Canadian group, the recommended MAL in drinking water is 0.5 μg/L of MC-LR, or in the absence of potency equivalency values for other MCs, one μg/L of total MCs (Watanabe et al., 1996).

By monitoring of the water quality in the Yongjiang river, we demonstrated that although no major algal bloom occurred, MC-RR, MC-YR, and MC-LR were present in the water column during the monitoring period. The TMCs concentrations varied from 0.0313 to 0.4585 μg/L. An earlier study of MCs in Guangxi showed that the average concentration of MCs in source water and treated water supplies were 0.277 μg/L and 0.221 μg/L, respectively (LV et al., 2005). Another survey showed that the concentration of MCs in the source water of high-incidence areas of liver cancer in Guangxi was 15.64 ± 2.08 ng/L, and the concentration in treated water supplies was 14.42 ± 2.28 ng/L (Li et al., 2016). These results suggest that MCs are detected in parts of Guangxi, but without considering the influencing factors and health risk assessment of MCs. The collected data indicated that the peak level in October followed by a sharp drop in concentration when using TMCs content as an indicator (Fig. 3). The significant decrease in TMCs content may result from the decreasing of temperature from November to December, and result in a slow growth of Microcystis. These phenomena were similar to those described by previous groups in Tai, Yang-cheng, and Xuanwu lakes in China (Xu et al., 2010; Li, Gu & He, 2014). TMC-RR concentration reached maximum level in October and then decreased to an average concentration level in November. As compared to the concentration of the above two toxins, TMC-YR concentration was the lowest of the three toxins studied; these results were identical to findings by other researchers, which suggested that the MCs are primarily dominated by TMC-RR and TMC-LR (Yang et al., 2006; Bi et al., 2017). It was clearly shown that the concentration of TMC-LR gradually increased from September to November. Such variation may be influenced by the differences in nutrients and climates, which are in favor of TMC-RR, TMC-LR and to a lesser extent TMC-YR.

Previous studies demonstrated that the algal toxins are produced by algae and consequently the concentration of toxins in water depends mainly on algal abundance (or biomass) such as chlorophyll-a concentration or algal cell counts, which in turn, is regulated by the environmental factors. The relationship of physical and chemical water parameters to the concentration levels of toxins are shown in Table 2. In this study, it was evident that temperature was positive and significantly correlated with concentration levels of TMCs, TMC-LR, TMC-RR, and weakly associated with concentration levels of TMC-YR. The highest concentrations of TMCs and TMC-RR, TMC-LR were observed in October and November with surface water temperature were around 25.6 °C and 26.2 °C, respectively. When water temperature increased, even higher concentration of TMCs and TMC-RR, TMC-LR concentration were detected. These findings are in agreement with previous reports which showed the concentrations of TMCs and TMC-RR, TMC-LR were temperature dependent, and TMC-RR which are generally detected at lower temperatures as compared to TMC-LR which favors at higher temperatures (Wang et al., 2010; Mantzouki et al., 2018). Intriguingly, the pH value was also shown to be related to the concentration level of toxins in the Yongjiang river. The maximum toxin concentration was detected at a pH below or above the medium level. As a result, TMC-RR and TMCs were negatively correlated with pH value (Table 2), which was similar to the results of other studies. Notably, the phytoplankton is known to affect the pH, and then, further affects the concentration levels of toxins. Therefore, the pH value cannot be used as an appropriate parameter to determine the concentration levels of toxins. A majority of the blue-green algae can grow adequately in the water at pH of 6.5–7.9 (Wang et al., 2002). The pH of Yongjiang river was within this range. DO concentration ranged from 2.0 to 12.5 mg/L during the study period. The reported environmental standard for river water is five mg/L (Guobiao (GB) Chinese National Standard, 2002). DO of Yongjiang river was partially lower than the reported standard during the monitoring period; these results show that the water is contaminated by organic matter, the oxygen consumption is severe, DO cannot be replenished in time, and the anaerobic bacteria in the water will multiply quickly (Wang et al., 2002). DO showed a negative correlation of TMC-LR with TMCs in Yongjiang river. However, some studies indicated that increases in oxygen saturation were correlated with algal biomass (Bi et al., 2017). Nonetheless, the algal abundance (or biomass) such as chlorophyll-a concentration or algal cell counts was not measured, and thus DO has no direct effect on the concentration levels of toxins. The correlation analysis results indicated that increasing the TP concentration could increase the concentration levels of toxins, especially that of TMC-LR. The current observations were in agreement with those from a study conducted in the large eutrophic Lake Erie in the USA (Harke et al., 2016), which demonstrated positive correlations between TP and the abundance of toxic Microcystis and MCs. Consistent with the trend, Vézie et al. (2002) also found that higher P concentrations were beneficial to the growth of toxic Microcystis. Although TP was a dominant explanatory variable, the effect of TN on the concentration levels of toxins could not be ignored. The concentration of TMC-YR was negatively correlated with the TN/TP ratio. Previous studies also demonstrated that decreasing the TN/TP ratio concentration could promote the growth and toxin concentration of Microcystis (Yu et al., 2014; Lei et al., 2015). According to stepwise multiple linear regression (Table 3), TP was found to be the dominant factor affecting the contents of TMC-LR and TMCs, and pH and TN/TP ratio as the main factors affecting the content of TMC-RR and TMC-YR. These findings were in agreement with the results of correlation analysis.

The Monte Carlo simulation model determined the risk level and putative human exposure scenarios associated with the blooms in the Yongjiang river used for drinking. The whole process of security risk assessment was always accompanied by the uncertainty. The entire process of risk assessment was conducted in two steps: exposure assessment and hazard characterization. Although the extrapolation of the experimental results does not lead to certainty, it could be carried out from experimental animals to the general population and from the general population to specific populations (sensitive populations). The variations in human individuals involved parameters such as genetics, age, sex, environment (nutritional status), and other factors. On the other hand, missing data or limitations led to uncertainties, including NOVEL, time differences, and lack of exposure data. Recent studies have gradually established superior methods, such as benchmark dose (BMD) and chemical-specific adjustment factor (CSAF), to address and reduce the uncertainty in the risk assessment (Ibelings et al., 2015). The USEPA and Health Canada have gradually started utilizing the BMD and CSAF methods to develop the health guidance values (Zeller, Duran-Pacheco & Guerard, 2017).

Several countries that regulate cyanotoxins in drinking water use a parametric value based on the WHO Guidelines for one μg/L MC-LR (World Health Organization (WHO), 2017). With respect to the drinking source waters, most countries use guideline values based on cyanobacterial biomass (cell density, chlorophyll-a, biovolume) indirectly reflecting the potential hazardous MCs concentrations (Valerio et al., 2009; Menezes, Churro & Dias, 2017). Our results indicated that the risk of carcinogenicity of MC-RR to children health under high exposure was greater than the maximum acceptable risk level recommended by USEPA (1 × 10⁻⁴); the annual risk of carcinogenic exposure in adults with MC-RR was greater than the maximum acceptable risk level recommended by the ICRP (5 × 10⁻⁵). The TMC-RR concentrations varied from 0.0224 to 0.3783 μg/L during the monitoring period. Therefore, the guideline value of TMC-RR in the Yongjiang river should be <0.3783 μg/L, so as not to pose a health risk to humans. Furthermore, we must take into account the increasing usage of the Yongjiang river, not only for the production of drinking water, but also for ludic activities, such as water sports, fishing, sailing, and swimming. Thus, the relevant departments must attach great importance to the potential risks associated with the Yongjiang river in order to protect the health of their users.

Conclusion

This study analyzed the influencing factors and the health risk assessment of MCs by Monte Carlo simulation in the Yongjiang river, China. The results showed that TP content may be related to TMC-LR and TMCs concentration, while pH and TN/TP ratio may be related to TMC-RR and TMC-YR concentration, respectively. The health risk assessment results showed that the risk of MC-RR for human health hazards is higher than that of MC-LR and MC-YR, and children are more vulnerable to MCs contamination than the adults. The risk of carcinogenicity of MC-RR to children health under high exposure was greater than the maximum acceptable risk level recommended by the USEPA. The annual risk of carcinogenic exposure in adults with MC-RR was greater than the maximum acceptable risk level recommended by the ICRP. The non-carcinogenic HI for MCs was <1. Therefore, MCs in the water bodies should be monitored with regarding to the carcinogenic risk to human health.

Supplemental Information