Polycyclic aromatic hydrocarbons (PAHs) in the upstream rivers of Taihu Lake Basin, China: spatial distribution, sources and environmental risk

The polycyclic aromatic hydrocarbon (PAH) pollution in Taihu Lake Basin has caused widespread concern. However, the spatial temporal distribution of PAHs in the upstream rivers of Taihu Lake Basin remains largely unknown. Thus, this study aims to investigate the level, spatial distribution, sources, and environment risk caused by PAHs in upstream rivers of Taihu Lake Basin. The concentrations of total 16 PAHs (∑16PAHs) ranged from 188.64 to 1060.39 ng/g, with an average of 472.62 ng/g. High-molecular-weight (HMW) PAHs were the predominant compounds in most sample sites. The results of source analysis demonstrated that the PAH pollution was mainly sourced from mixture of combustion and direct petroleum spillage. The ecological risk assessment showed that moderate ecological risk caused by the PAH contaminants might occur in most sample sites. The incremental lifetime cancer risks (ILCRs) ranged from 2.07 ×10−4 − 2.66 × 10−3 for children and 9.66 ×10−5 − 1.24 × 10−3 for adult, indicating moderate cancer risk of PAH-contaminated sediments.


Introduction
Polycyclic aromatic hydrocarbons (PAHs) are ubiquitous in various environmental medium and primarily produced by petroleum spillage and incomplete combustion of organic materials (Ravindra et al. 2008; Thompson et al. 2017). PAHs can enter the aquatic environments by atmospheric deposition, surface runoff, oil leakage, and waste water discharge (Hu et al. 2017). Sixteen PAHs have been identified as the priority pollutants in control by the US Environmental Protection Agency (US EPA) (Manoli et al. 2000;Wang et al. 2010), and 7 of them (BaA, Chr, BbF, BkF, BaP, DaA, and InP) were identified as carcinogenic compounds (Zheng et al. 2016), which cause both ecological and health risks (Kim et al. 2013;Sarria-Villa et al. 2016).
Due to their low water solubility, high lipid solubility, and relatively high persistence, PAHs are easily absorbed by suspended particular matters from water and accumulated in the sediment. PAH concentrations are higher in sediments than in other phases ). Thus, sediment was considered to be an important reservoir for hydrophobic organic pollutants in aquatic environment. Meanwhile, PAHs in the sediment can be re-suspended into the aquatic environment, resulting in a long-term contamination of aquatic environments. PAHs can also be bio-accumulated in organisms, which ultimately pose considerable threats to ecosystem and human health through the bio-magnify in the food chain (Geffard et al. 2003;Humans 2010;Ma et al. 2018). Therefore, the PAHs in sediments can provide valuable records of pollution and denote environmental risks.
Taihu Lake is the third largest freshwater lake in China, which is important for drinking water, tourism, recreation, shipping, aquaculture, and industry (Qin et al. 2007). With the rapid industrialization and urbanization of the surrounding region, massive industrial wastewater, municipal sewage, shipping, roadway runoff, and agricultural non-point sources have been discharged into the lake, leading to serious deterioration of water quality of the lake (Chen et al. 2018;Li et al. 2019;Niu et al. 2020;Tao et al. 2018;Xu et al. 2014;Zhao et al. 2017). Several studies have detected the PAHs in the sediment of the lake, and concluded that the PAH pollution was mainly from the upper reaches of the lake Lei et al. 2014;Lei et al. 2016;Li et al. 2019;Qiao et al. 2006;Tang et al. 2015;Zhao et al. 2017). Indeed, as there are various industrial enterprises surrounded in the upper reaches of Taihu Lake, the major upper reaches were more polluted than the lower reaches ). However, the occurrence, environmental risk, and source of PAHs in the upstream rivers remain largely unknown. Thus, this study aims to elucidate the level, spatial distribution, source, and environment risk of PAHs in upstream rivers of Taihu Lake Basin. The results will help to further understand the characteristics and risk of PAHs, and provide a valuable reference data for PAH management in the Taihu Lake Basin.

Sample collection
The sediment samples (depth 0-5cm) in 18 sample sites were collected in Taihu Lake Basin in May 2019, including 16 in the upstream rivers and 2 in the wastewater treatment plants (WTPs) (Fig. 1). A stainless steel grab sampler was used to collect the sediment samples, and then wrapped, stored, and transported to the laboratory. Before the sample analysis, all samples were stored at −20 °C.

Sample extraction and cleanup
Sediment samples were dried by freeze-drying before extraction. Accelerated solvent extraction (ASE) in static conditions was used to extract the samples. Each sample (10g) was mixed with diatomite and milled in the disc. The samples were then transported to the extraction tank and mixed with 20mL hexane and acetone (1:1, v/v) solution for 30 min (Belo et al. 2017;De Nicola et al. 2005;Liu et al. 2017). The extracts were collected into a roundbottomed flask and dehydrated by filtering through anhydrous sodium sulfate (Bortey-Sam et al. 2014), which were then concentrated, solvent-exchanged with n-hexane, and further concentrated to approximately 2 ml under a gentle N2 stream. Cleanup solid-phase cartridges filled with silica and copper powder were used to fractionate the PAH extracts. Before use, the solid-phase cartridge was successively eluted with 4 ml dichloromethane and 5 ml hexane. The extracts were eluted with 2 mL hexane for three times and 10 mL dichloromethane/hexane (1:9, v/v). The eluate was re-concentrated to 1 ml under a gentle N2 stream.

Chemical analysis
Gas chromatography/mass spectrometry (GC-MS, Agilent 9000/5977B) equipped with a fused silica capillary column was used to quantify the concentration of 16 PAHs. Helium was used as carrier gas with flow rate of 1 ml/min. Samples (1μl) were injected with an auto-sampler for analysis. The initial oven temperature programmed at 40°C for 4 min, increased to 80°C at a rate of 10°C/min, which was maintained for 2 min, and then increased at 5°C/min to 200°C, maintained for 1 min, and finally heated to 280°C at a rate of 10°C/min, maintained for 10 min. The injection port temperature was set at 280°C. Sixteen USEPA priority PAH compounds including naphthalene (Nap), acenaphthylene (Acy), acenaphthene (Ace), fluorine (Fluo), phenanthrene (Phe), anthracene (Ant), fluoranthene (Flua), pyrene (

Quality assurance and quality control
Quantitation was performed using an internal standard calibration method; the correlation coefficients (r 2 ) of the calibration curves were all higher than 0.999. Procedural blanks, method blanks, matrix spikes, and sample duplicates were routinely processed to ensure data quality (Li et al. 2015). Known quantities of PAH mixture surrogate standards (p-dichlorobenzene_d 4 , naphthalene_d 8 , phenanthrene_d 10 , chrysene_d 12 , and perylene_d 12 ) were added to the samples prior to extraction to determine the recovery efficiencies. The average recoveries were 63.0 ± 2.6%, 81.3 ±2.4%, 92.1 ± 3.2%, 97.0 ± 4.5%, and 86.7 ± 7.6%. The detection limits (LODs) were calculated as three times the signal-to-noise level of the chromatogram in blank samples spiked with surrogate standards. The LODs of PAHs were ranged from 0.3 to 2.0 ng/g dry weight. Method blank operation showed no PAH was observed from the reagents and procedures.

Ecological risk assessment
The sediment quality guidelines (SQGs) were applied to assess the ecological risk of PAHs in the sediment in this study. The concentrations of PAHs were compared with concentration limits of SQGs. Two sets of SQGs include (a) effect range low (ERL)/effect range median (ERM) and (b) threshold effect level (TEL)/probable effect level (PEL) (MacDonald et al. 2004;McCready et al. 2006). The PAH concentrations were classified into three different ranges: adverse ecological risk rarely occurred (<ERL or TEL), occasionally occurred (ERL-ERM and TEL-PEL), and frequently occurred (≥ERM or PEL) (Long et al. 1995, Macdonald et al. 1996. In addition, mean PEL quotient (mean PEL-Q) was used to estimate the probable ecological risk of multiple contaminants (Long and MacDonald 1998). The mean PEL-Q equations is as follows: In the formula, C i is the concentration of the PAH i in sediment; PEL i is the probable ecological risk of the PAH i, and n is the number of PAHs with available sediment quality guideline. The PEL-Q values were divided into three categories: (1) PEL-Q≤0.1, indicates low ecological risk.

Toxicity and carcinogenic risk assessment
Toxic equivalent factors (TEFs) are taken as the reference chemical to compute the toxic equivalent concentrations (TEQ BaP ) of PAHs in the sediment. The TEQ BaP of each PAH is the PAH concentration multiplied by its TEF value (Soltani et al. 2015;Tian et al. 2013). The total TEQ BaP of 16 PAHs (TEQ ∑16PAHs ) and 7 carcinogenic PAHs (TEQ ∑7PAHs ) were both calculated according to the following equation: where C i is the concentration of single PAH i; TEF i refers to TEF of this compound, and the TEF values of PAHs were shown in Table 1. TEQ i is the toxic equivalent of the PAH i and TEQ PAH stands for total toxic concentration of PAHs.
In general, oral ingestion and dermal absorption are the two main ways of human exposure to PAHs in sediments. To assess the potential carcinogenic risk of PAHs in the sediment, the incremental lifetime cancer risk (ILCR) was estimated based on the US EPA standard models. The ILCR through oral ingestion and dermal absorption pathways was estimated using the following equations (Soltani et al. 2015;Wang et al. 2015): where C sed is the toxic concentration of 7 carcinogenic PAHs in the sediment (ng/g) (Keshavarzifard et al. 2017). Other parameters referred to in the model for children and adults are based on the Risk Assessment Guidance of US EPA and related publications (Table 2). According to the guidelines recommended by the US EPA, the value of ILCRs less than or equal to 10 −6 , between 10 −6 and 10 −4 and exceeding 10 −4 , signifies acceptable level, potential moderate risk, and potentially high risk, respectively (Liang et al. 2019;Soltani et al. 2015;USEPA 2004).

Statistical analysis
R software 4.0.0 and IBM SPSS22.0 for Windows were used for statistical analyses. The spatial distributions of PAHs in the sediment were analyzed using ArcGIS 10.2 software. PCA was performed to analyze the source distributions of different PAHs. The data for PAHs were standardized to unit variance prior to PCA analysis (Bemanikharanagh et al. 2017).

Level of PAHs in the sediment
The PAH concentrations of 18 sediments samples are presented in Table 1. The concentrations of total 16 PAHs (Σ16PAHs) ranged from 188.64 to 1060.39 ng/g, with an average of 472.62 ng/g. The PAH concentration was similar with the studies conducted by Yuan Zhang ) and Bingli Lei (Lei et al. 2014), but showed a decrease than the level reported by Yuqiang Tao ). The concentrations of total 7 carcinogenic PAHs (Σ7PAHs) ranged from 58.13 to 504.76 ng/g, with an average of 192.04 ng/g, which account for 40.63% of the Σ16PAHs. Among 16 PAHs, Nap was the main pollutant with mean concentration of 93.65 ng/g, whereas the concentrations of Fluo, Acy, and DbA were below the detection limit in almost all samples.

Spatial distributions of PAHs
The spatial distribution of Σ16PAHs in the sampling sites is shown in Fig. 1. The highest concentration of Σ16PAHs (1060.39 ng/g) was detected at WTP2, followed by 912.15ng/g at S12, 850.68 ng/g at S13, 749.74 ng/g at S7, and 586.84 ng/g at S16. The lowest concentration of Σ16PAHs (188.64ng/g) was observed at S6 in the Yongan river, one of the upstream rivers. The WTP2 is a wastewater treatment plant of Yixing city nearby chemical industrial park, which is responsible for treating surrounding chemical wastewater, suggesting that the highest level of PAHs at site WTP2 might be associated with chemical plant emission. Sites S12, S13, and S16 are located in the lake estuary, where many chemical plants were concentrated, which may attribute to the chemical industrial activities. S7 is located at the Beijing-Hangzhou Canal of Changzhou, which is near the printing and dyeing industries, suggesting that wastewater from printing and dyeing operations might be the main source of PAH pollution (Liu et al. 2016;Wang et al. 2018). The geographic distribution of PAHs showed that Σ16 PAH concentrations in the downstream rivers were higher than those in adjacent upstream rivers, which could be associated with domestic industrial wastewater and adhesion of upstream PAH pollution in sediment .
The composition patterns of the PAHs by ring size in the 18 samples are depicted in Fig. 2. The two-ring PAHs (Nap) accounted for 0-43.28% of the total PAH content, three-ring PAHs 14.17-31.88% (Acy, Ace, Fluo, Phe, Ant), four-ring PAHs 15.87-39.08% (Flua, Pyr, BaA, Chr), five-ring PAHs 4.89-39.30% (BbF, BkF, BaP, DbA), and six-ring PAHs 0-22.73% (InP, BgP). Except for S4, S5, S6, S10, and S11, high-molecular-weight (HMW) PAHs (4-6 rings) were the predominant compounds in most sample sites (51.71~72.29%). Indeed, due to the high water solubility and benthic recycling in aquatic environment, low-molecular-weight (LMW) PAHs (2-3 rings) are more likely to dissolve and degrade, while HMW PAHs are more resistant to degradation and easier to accumulate in the sediment Montuori et al. 2016). Additionally, researches showed that PAHs with LMW are abundant in petrogenic and low-temperature pyrolytic sources (e.g., petroleum spillage and incomplete combustion), while those with HMW are abundant in compounds from pyrolytic sources (Li et al. 2015). Thus, the composition of PAHs in the sediments is dominated by high rings, indicating that the combustion at high temperature is the major source of PAH pollution in sediment.
The results of diagnostic ratio for the sampling sites are shown in Fig. 3. In this study, ratio of Ant / (Ant + Phe) > 0.1 was found at most of the sample sites except S6, which indicate combustion origins. Sample sites with Flua / (Flua + Pyr) <0.4 accounted for 44.4%, and sites with> 0.5 accounted for 55.6%, which indicated that PAH pollution were from mixed source of petroleum and combustion of grass, wood, and coal. And similarly, the BaA / (BaA+Chr) of all of the samples were >0.35 in this study, which implied combustion of grass, wood, and coal. The ratio of InP / (InP + BgP) was<0.2 at 77.8% of the sample sites, suggesting the direct petroleum spillage pollution (Bemanikharanagh et al. 2017;Bortey-Sam et al. 2014;Yunker et al. 2002). Therefore, these results showed that the PAH pollution was mainly sourced from combustion and direct petroleum spillage.

Principal component analysis
Principal component analysis (PCA) was applied to further explore the PAH sources in this study (Bemanikharanagh et al. 2017;Lin et al. 2018;Tao et al. 2010;Zheng et al. 2016). Flua, Phe, Ant, and Pyr usually imply coal combustion. Ace and Acy are the foremost product of coke burning. BaA, Chr, and BaP are regarded as typical pollutants of biomass and coal combustion. BkF and BbF are typical pollutant of diesel emissions, and InP, BgP, and DbA are typical markers of traffic emission (Kannan et al. 2005;Liu et al. 2017;Liu et al. 2016).
Barttlet's sphericity test was used to verify if the PCA was applicable in this study, and the P-value was <0.01, indicating the applicability of PCA here (Zheng et al. 2016). The factor loadings for PAH concentration by variamax rotation are shown in Table 3. Two components (PC1 and PC2) were extracted from sediment data responsible for 81.88% of the total variation of PAHs. The PC1 explained 71.87% of the total variance, which was highly loaded with BaA, Chr, Ant, Ace, and Phe and relatively highly loaded of Flua, Pyr, BbF, BaP, and BkF, which suggest that the PAH pollution was mainly from the petroleum spillage and combustion of grass, wood, and coal, while traffic emission (e.g., gasoline and diesel exhaust) is also an important factor for PAH pollution (Zheng et al. 2016). PC2 accounts for 10.01% of the total variance, and InP and BgP account for high loadings, indicating that traffic emission was predominant in PC2, such as gasoline combustion and diesel combustion (Fig. S1).
Therefore, consistent with the results of previous studies, these findings further confirmed that PAH pollution was mainly sourced from mixture of petroleum spillage; combustion, such as grass, wood, and coal; and traffic emission , which may be associated with Fig. 3 The cross-plot for the ratios of (a) Ant / (Ant + Phe) vs Flua / (Flua +Pyr), (b) BaA / (BaA+Chr) vs InP / (InP + BgP) intensive traffic (e.g., shipping), discharge of urban sewage, and industrial wastewater (Li et al. 2015).

Ecological risk assessment
The results of ecological risk assessment in sediment are shown in Table 4; the sample sites were classified into three different ranges: ecological risk rarely occurred (<ERL or TEL), occasionally occurred (ERL-ERM and TEL-PEL), and frequently occurred (≥ERM or PEL). The results of SQGs showed that the concentrations were below than their respective ERM and PEL values, except for compound DbA in sample WTP2. The concentration of Ace at all of the sample sites was between ERL/TEL and ERM/PEL, indicating that the ecological risk caused by Ace may occur occasionally at all sites. The BaA and BaP concentrations were at levels where ecological risk may occur occasionally (≥TEL and <PEL). Meanwhile, the concentrations of Acy, Ant, and Pyr at a few sites were between the TEL and PEL levels. The Nap concentration was below the ERL at most site, while TEL level was exceeded at most sites. In addition, the concentrations of Phe and Flua were lower than the ERL level at most of the sites. Therefore, these findings suggest that the pollution of Ace, BaA, BaP, Acy, Ant, and Pyr may cause potential ecological risk occasionally at some sites. The mean PEL-Q values ranged from 0.09 to 0.29. Most of the sample sites had mean PEL-Q exceeded 0.1 but lower than 1.0 (<PEL-Q≤1.0) and only two sites had mean PEL-Q <0.1(S1 and WTP1), indicating that PAH contaminants may cause moderate ecological risk in most sample sites (Fig. S2).

Toxicity and health risk assessment
The TEF of BaP was used to evaluate TEQ BaP of PAH compounds. The TEF value of PAH compounds and total their TEQ BaP concentrations is shown in Table 1. In the areas, the TEQ ∑16PAH values ranged from 19.76 to 208.75 ng/g, with mean of 44.10 ng/g. The site WTP2 has the highest TEQ value, followed by S13, S12, and S16. The TEQ ∑7PAH values ranged from 14.84 to 190.64 ng/g, with mean of 38.70 ng/g, accounting for 87% of the TEQ ∑16PAH , suggesting that the 7 PAHs were major carcinogenic contributors.
The results of carcinogenic risk of seven carcinogenic PAHs are shown in Fig. 4. In this study, the ILCRs ranged from 2.07 ×10 −4 to 2.66 × 10 −3 for children and 9.66 ×10 −5 to 1.24 × 10 −3 for adults, which is higher than the baseline of acceptable risk. The highest risk was found in site WTP2, and followed by S13, S12, and S15, which is consistent with the spatial distributions of PAHs in sediment. Therefore, the results revealed that the PAH-contaminated sediments at most of the sites posed a potential moderate cancer risk to human health via both ingestion and dermal contact pathways.

Conclusion
The levels, spatial distributions, sources, and environmental risks of PAHs in sediment from the upstream rivers of Taihu Lake Basin were analyzed. The concentrations of Σ16PAHs varied from 188.64 to 1060.39 ng/g, with mean of 472.62 ng/g. The result of spatial distribution showed that PAH concentrations in the lake estuary were higher than those in upstream rivers, including S12, S13, S16, and S15. The source identification revealed that PAH pollution were mainly sourced from mixed source of combustion and petroleum spillage. Ecological risk assessment showed that moderate ecological risk caused by PAH contaminants may occur in most sample sites, and Ace, BaA, BaP, Acy, Ant, and Pyr were likely to cause adverse biological effects occasionally in some sites. The PAH-contaminated sediments at most of the sites may cause a moderate potential cancer risk to human health via both ingestion and dermal contact pathways.

Supplementary Information
The online version contains supplementary material available at https:// doi. org/ 10. 1007/ s11356-021-17598-w. Availability of data and materials All data generated or analyzed during this study are included in this published article

Declarations
Ethics approval and consent to participate Not applicable

Consent for publication Not applicable
Competing interests The authors declare no competing interests.