3.1 Concentration and composition of PPCPs in surface water
During the wet season, the total concentrations of the detected PPCPs ranged from 81.26 to 149.45 ng/L (Fig. 2). Nine out of 10 PPCPs were detected in all water samples, including BT, 5-TT, CBT, XT, MP, EP, PP, BP, TCS and TCC. The concentrations of PPCPs detected in all water sites were 14.90–50.30 ng/L for BT, 5.39–36.91 ng/L for 5-TT, 5.93–12.64 ng/L for CBT, 2.13–6.03 ng/L for XT, 1.12–2.51 ng/L for MP, 0.76–2.70 ng/L for PP, 0.77–1.02 ng/L for BP, 0.32–1.29 ng/L for TCC, and 33.03–54.92 ng/L for TCS. Notably, BT and TCS were detected at higher levels compared with other PPCPs, constituting 26.89% and 43.21% of the selected PPCPs in the water phase of QTR, respectively. Additionally, the ZK site exhibited the highest concentration of PPCPs, while other sites featured relatively lower concentration. This observation indicates the potential level of domestic pollution sources in ZK. The highest PPCP concentration at the ZK site can be attributed to the higher 5-TT concentration compared with other sites..
Figure 2 illustrates the concentrations of PPCPs in the dry season. The total concentrations of the detected PPCPs detected in the dry season ranged from 98.66 to 198.55 ng/L, which exceeded those in the wet season, consist with the findings of a previous study27. Particularly, the concentrations of the detected PPCPs were 17.96–47.45 ng/L for BT, 8.54–31.51 ng/L for 5-TT, 7.89–14.79 ng/L for CBT, 3.68–13.91 ng/L for XT, 2.05–2.80 ng/L for MP, 1.49–2.16 ng/L for PP, 1.78–2.67 ng/L for BP, 3.12–8.56 ng/L for TCC, and 50.12–79.21 ng/L for TCS. BT and TCS were identified as the main contributors to PPCPs during the dry season. Notably, the higher concentrations of PPCPs detected in the dry season compared with the wet season were mainly attributed to the increased levels of TCS and TCC. Moreover, among PPCPs detected in the dry season, the ZTJ site exhibited the highest concentration, which differed from the pattern observed during the wet season. Previous studies28 have indicated that the concentration of PPCPs decreased from upstream to downstream in the surface water. However, PPCPs in QTR exhibited the opposite trend owing to the presence of WWTP located between the QB and ZTJ site. Furthermore, no significant differences were observed between the left, middle, and right banks of QTR during both wet and dry seasons.
A comparison of the concentration of PPCPs in QTR with those in other surface waters revealed that BT, CBT, TCS, TCC, and MP exhibited lower concentrations in QTR, compared with the Pearl River Delta of China5. Additionally, the highest TCS concentration was lower in QTR than that in the River Ganges of India29. Furthermore, QTR exhibited significantly lower concentration of PBs than the Yellow River and Huai River30. In comparison, QTR featured lower PPCPs concentration than many other rivers in China and worldwide.
3.2 Concentration and composition of PPCPs in sediment
The concentrations of PPCPs in sediment from different sample sites were similar, ranging from 63.24 to 79.15 ng/g (Fig. 4). During the dry season, PPCP concentrations at four different sampling sites ranged from 72.54 to75.06 ng/g. However, ZK and ZJT sites exhibited PPCP concentrations of 65 ng/g, which were lower than that of about 80 ng/g in YP and QB sites. TCS was the major contributor to PPCPs in sediment constituting 77% of the total concentration. TCS was also the main contributor to PPCPs in the water phase. Moreover, no significant difference was observed in the PPCPs concentrations between the wet and dry seasons. It was noteworthy that the PPCPs concentrations in the sediment from YP and QB sites were higher than those from ZK and ZTJ sites during the wet season. Additionally, the ZK and ZTJ sites exhibited significantly lower concentrations of TCC and PBs compared with the Cau River in Vietnam, the Yellow River, and the Huai River in China. The concentrations of BTs were higher than that (0.31–3.81 ng/g) in the Songhua River in China31. Furthermore, the TCS concentration in QTR (0-7.73 ng/g) was significantly higher than that observed in the Hanjiang River of China32. Overall, the level of PPCPs in the QTR sediment was relatively higher than that reported in other studies.
3.3 Correlation and partitioning of PPCPs between surface water and sediment
Strong correlations for PPCPs were observed between sediment and water in both wet and dry seasons, with Pearson correlation coefficients of 0.778 and 0.858, respectively (Table 1). To further investigate the diffusion tendencies of PPCPs at the sediment-water interface, the correlation and fugacity (ff) between PPCPs in water and sediment were analyzed. The ff values of PPCPs exhibited similar trends in both wet and dry seasons (Fig. 5 and Fig. 6). The uncertainty of ff was calculated to confirm the ff values of 0.3, indicating dynamic equilibrium at the sediment-water interface. According to a previous study33, ff < 0.5 suggests surface water diffusion into the sediments, ff = 0.5 indicates sediment-water equilibrium, and ff > 0.5 denotes the net desorption from the sediments into the water. The ff values of BT, 5-TT and MP ranged from 0.8 to 1.0, exceeding 0.5 ± 0.3, indicating the net desorption of BT, 5-TT and MP from the sediments into water. Conversely, BP and TCC may migrate from water to the sediment because of their ff values were below 0.5. PP, XT and TCS reached a state of sediment-water equilibrium. A previous study suggested that the organic pollutants with medium molecular weights typically achieve dynamic equilibrium, migrating from water to the sediments with high molecular weights and returning to water with low molecular weights19. The diffusion tendency of PPCPs between water and sediment was closely, correlated with their molecular weights. Additionally, the equilibrium of TCS was influenced by the hydrophilic nature of its hydroxyl groups.
3.4 Source appointment of PPCPs
3.4. Relationship between physicochemical parameters and PPCPs
A previous study has indicated that water quality could influence the distribution of PPCPs in both surface water and sediments34. To investigate the distribution dynamics and potential sources of PPCPs, Pearson correlation analysis was conducted to assess the relationship between physicochemical parameters and PPCPs, as given in Fig. 7. The analyzed physicochemical parameters mainly included nutrient parameters, such as TOC, TP, TN and NH3-N.
During the wet season, significant positive correlations were observed between BTs, TOC, and TP (wth r>0.5), consistent with findings from a previous study17. However, weak correlations were observed between PBs, TCC, TCS and TOC, TP, TN, NH3-N. A significant difference in dry season is shown in Fig. 7(b). In this season, all selected PPCPs concentrations were positively correlated with TOC and minimally correlated with TN, suggesting similarity among the detected PPCPs. The highest concentration of PPCPs at the ZTJ site near the WWTP outlet14, indicated the significant contribution of WWTP effluents as a source of PPCPs during the dry season. Additionally, the ZK sampling sites, located near the Hangzhou metropolitan area with strong anthropogenic influences, exhibited higher levels of both BTs and nutrients (TOC and TP). This indicates that anthropogenic activities and stormwater runoff were the main source of PPCPs during the wet season.
3.4.2 Quantitative source apportionment of PPCPs
To investigate the quantitative source apportionment of PPCPs, the PMF model extracted four to seven factors. As the number of factors increased five to six factors, the value of Qtrue/Qexpected decreased from 7 to 0.6, while the factors increased from six to seven, the decrease of Qtrue/Qexpected was smaller (0.6 to 0.1). Therefore, six factors were the optimal solution. Moreover, the R2 values of predicted and measured PPCP concentrations ranged from 0.77 to 0.99, indicating the reliability of the modeling results.
Factor 1 was closely associated with PP (37.6%), MP (21.7%) and TCS (20.5%) which are commonly used as preservatives and disinfectants in daily chemical products. This factor mainly represented domestic sewage. Factor 2 likely represented the sewage from WWTPs, as the most detected PPCPs were related to factor 2 and proved challenging to remove in WWTPs. Factor 3 exhibited high loadings of PP (55.4%) and TCC (36.0%), suggesting a potential source from pharmaceutical company discharge. Factor 4 was mainly linked to 5-TT (42.3%), XT (29.1%) and BP (25.8%). According to the pollution source monitoring and management platform in Hangzhou, Factor 4 originated from the wastewater of food processing factories. TCC and TCS were closely associated with Factor 5 and frequently detected in hospital drainage. Thus, Factor 5 could be attributed to hospital drainage. Factor 6 likely represented the wastewater of electroplating enterprises owing to higher loads of BTs, which served as rust and corrosion inhibitors for metals.
The contribution rate calculation by the PMF model indicated that the primary source of PPCPs was the wastewater from electroplating enterprises (Factor 6), constituting 54.0% of the total concentration. Additionally, the WWTPs discharge emerged as the second most dominant source, contributing 35.2% to the total concentration (Factor 2). This observation was consistent with findings from other studies, which also identified WWTPs discharge as a significant source of PPCPs21. Furthermore, the contribution concentrations of domestic sewage related to PPCPs cannot be ignored, accounting for 10.7% of the total concentration (Factor 1). A study20 has also reported domestic sewage as a main source of antibiotics.
3.5 Risk assessment of PPCPs in surface water and sediment
The ecological effects of PPCPs have received increasing attention, and a previous study has suggested that Daphnia magna may exhibit higher sensitivity to PPCPs compared with algae and fish35. This study examined the ecological risk posed by PPCPs on zooplankton (Daphnia magna) in both surface water and sediment, as shown in Fig. 10 and Fig. 11. In an aqueous environment, the RQ values of TCS exceeded 1.0 at all sampling sites, suggesting a high risk of TCS in both wet and dry seasons. TCC and XT exhibited medium risk levels from QB to ZTJ, with RQ values ranging from 0.01 to 0.1, but displayed low-risk levels during the dry season. However, TCC and XT posed no risks at any sampling sites during the wet season. In addition, the RQ values of BP, PP, EP, MP, CBT, 5-TT and BT remained below 0.01 in both wet and dry seasons, indicating negligible risks associated with these compounds. However, these pollutants posed slightly higher ecological risks in the dry season compared with the wet season.
The ecological risk of PPCPs in the sediment on zooplankton (Daphnia magna) exhibited similar trends in both wet and dry seasons (Fig. 11). The RQ values of TCS ranged from 0.01 and 0.10, indicating a low risk to Daphnia magna. Moreover, other detected PPCPs (BT, 5-TT, CBT, XT, MP, EP, PP, BP, and TCS) posed no risks. A comparison of RQ in both wet and dry seasons indicated no significant difference in the ecological risks posed by PPCPs. Furthermore, the ecological risks decreased from water to sediment.