Distribution of Phenol Derivatives by River Waters to the Marine Environment (Gulf of Gdansk, Baltic Sea)

Water samples were collected during one year in three different seasons (spring – March, summer – June and autumn – October), near the sources and at the outlets of the small rivers flowing through Pomerania, and from the estuary of the Vistula River (Fig. 1). A total of 24 samples were taken and each of them was divided into 3 subsamples. Surface water was taken from the middle part of the stream. The samples were kept in a refrigerator (+4°C) until analysis and were analysed within 72 hours. Three samples were taken at a time from each collection point, which were then analysed in the laboratory for the presence of phenol derivatives. The characteristics of individual stations are presented in Table 1.

Figure 1

To assay the concentrations of phenol derivatives in the water, the method previously described by Staniszewska et al. (2016a) was used. For this purpose, 100 cm³ of water was passed through columns with Oasis HLB (Waters) (200 mg, 5 ml). The columns were conditioned with 6.0 cm³ of a 50/50 DCM/methanol mixture (v/v) and activated with 3.0 cm³ of methanol and 3.0 cm³ of water. The analysed compounds were eluted with 6.0 cm³ (2 × 3 cm³) of the DCM/methanol mixture. The resulting extracts were evaporated until dry and topped up with acetonitrile to 0.2 cm³.

The final assays of phenol derivative concentrations in both of the analysed matrices were carried out using high-performance liquid chromatography with a fluorescence detector and a Thermo Scientific HYPERSIL GOLD C18 PAH chromatographic column (250 mm / 4.6 µm) with a generated excitation wavelength of λ = 275 nm and emission at the wavelength of λ = 300 nm. The chromatographic separation process was performed using a mobile phase (water: acetonitrile) under gradient conditions (Table 2).

The background was monitored by checking the centrifuge every 20 times the SPE column was used. The ‘background’ values for BPA, 4-NP and 4-tOP recorded in this procedure were all lower than the limit of quantification (LOQ).

All tools and vessels used at each stage, i.e. collection, transport, storage and concentration assays, were made of glass, metal or Teflon in order to avoid contact between the samples and plastics, thus avoiding potential contamination.

The linear correlation coefficient (r) of the analytical curves was > 0.999. The average amounts of BPA, 4-NP and 4-t-OP recovered, as determined through a quintuple analysis of samples containing an added known amount of the standard, were 101.0% (BPA), 80.1% (4-t-OP) and 83.5% (4-NP). The accuracy (variation coefficient) was below 2% for each of the compounds. The LOQ was determined as a tenfold signal-to-noise ratio for a sample with a very low analyte content; it amounted to 5.0 for BPA, 1.0 for 4-t-OP and 4.0 ng dm⁻³ for 4-NP.

The loads of BPA, 4-t-OP and 4-NP carried via rivers to the sea were calculated from the concentrations of phenol derivatives measured in the samples and the flow of water in the estuaries of individual rivers provided in literature. The following formula was used:

where F is the riverine flux of EDCs (ng s⁻¹), C_w is the concentration of EDCs in riverine water (ng dm⁻³), P is the water discharge in the estuary (m³ s⁻¹) and A = 31,536,000 is the number of seconds in a year.

Each of the small rivers considered in this study is no longer than 60 km, but differences in BPA, 4-t-OP and 4-NP concentrations can be observed between the water collected at the source and at the outlet. At the sources, the riverine water had concentrations of phenol derivatives that were 1.5 to 7 times higher than at the outlets (Table 3). The highest ratio of the mean BPA concentration measured at the source to the concentration at the estuary (38.62) was found in summer for water from the Kacza River (Table 4). This finding indicates that water is more heavily contaminated with bisphenol A at the source than at the outlet. This may be related to the increased amount of silty sediments in the downstream section of a river. The load of phenol derivatives is adsorbed by the suspended particles, causing the particles to sink to the bottom. Thus, phenol derivatives undergo sedimentation and the concentrations in surface water are lower (Li et al. 2004). The development of the riverside is not without significance. There are residential areas with detached houses and small farms in the vicinity of the Kacza source. The lowest source/outlet ratios were determined for samples collected in autumn from the Zagórska Struga and the Gizdepka (0.16 – 0.41). In autumn there is usually more precipitation, which results in increased leaching of pollutants which have accumulated over the summer, causing them to travel from the soil to the surface water. Near the outlets of the rivers there are beaches and recreational areas, where people discard plastic rubbish. In addition, agricultural land covers a significant proportion of the catchment area. Therefore, xenobiotics can be delivered by surface runoff from fields or farms, along with food remains or animal excrement (Stewart 2005). Leaching from the soil is also favoured by more intense weather phenomena, as strong winds cause more turbulence of bottom sediments in shallow rivers and re-suspension of the pollutants accumulated there. This may lead to higher water pollution towards the outlet.

It should be remembered that the concentration value alone does not reflect the extent to which riverine waters pollute coastal waters with phenol derivatives. Therefore, the average load of phenol derivatives carried by rivers to the Gulf of Gdańsk was calculated, taking into account the average water flow at the mouth of each individual river (Table 5). The authors tried to find the latest available data on the flows of individual rivers. The average annual water flow was used in these calculations, which means that the results are estimated loads. However, this does not diminish the importance of these results, because the values are significant and worthy of interest, in the opinion of the authors.

The calculated loads clearly show that it is the River Vistula that brings in the majority (99.6%), of phenol derivatives as opposed to smaller rivers. However, this does not rule out the role of smaller rivers in this supply budget. Despite the fact that the mean xenobiotic loads of these rivers are small, high concentrations of phenol derivatives incidentally occur in river water. The research did not involve frequent sampling, which perhaps would have indicated more cases of high concentrations of phenol derivatives. The load values for individual rivers in different seasons of the year indicate that the highest supply of endocrine-active phenol derivatives occurs in the summer season. These indicators are several times to several hundred times higher than those calculated for the other seasons. The load of phenol derivatives is highest in summer because of the increase in tourist traffic in the coastal region. The city’s beaches are visited by about 3 million beach-goers, who engage in water sports and general recreation from June to September. The ‘Tri-City’, which is located directly by the sea and whose population numbers in the millions, is also an urban agglomeration with a well-developed coastal industry (shipyards, refineries and pharmaceutical, cosmetics and paint producers). There may also be uncontrolled discharges of sewage from buildings into rivers. In most of the small rivers studied here, the catchment area is covered by forests and farmland. This is of great importance due to the fact that in the literature on the subject phenol derivatives have been reported at measurable levels in bird and pig faeces (Xu et al. 2018, Zhang et al. 2014, Tao et al. 2021). Importantly, 70% to 85% of bisphenol A and alkylphenols in mammals is removed with urine (Tao et al. 2021). In the catchment areas of the small rivers, there are recreational areas where so-called ‘wild garbage dumps’ are found, often with discarded plastic items or electronics. Such rubbish, decomposed under the influence of weather conditions, is a source of BPA or alkylphenols in the environment (Staniszewska et al. 2015). The intensity of runoff may increase under the influence of heavier rainfall and more frequent flooding (Saniewska et al. 2014).

It should be noted that the authors of the aforementioned studies focussed only on surface water, and that phenol derivatives have a high affinity for organic carbon (Koniecko et al. 2014, Staniszewska et al. 2016b), on which they can sorb and thus accumulate in bottom sediments. Nonetheless, sediments are not the ultimate place of storage for these xenobiotics, especially in shallow water. Due to the size of the rivers, their bottom sediments are not stable, but are sensitive to weather conditions and waves, which may facilitate the re-suspension of these pollutants back into the water column. However, it should be noted that in the remaining seasons, BPA concentrations are also elevated. In autumn, greater rainfall and more dynamic weather conditions increase the inflow of this compound even from further distances. This is related to the ability of phenol derivatives to adsorb on elemental carbon particles (Staniszewska et al. 2016). On the other hand, in spring, when the ice covers thaw, pollutants – including phenol derivatives – are released into the groundwater.

One parameter which indicates the risk of contamination for aquatic organisms is the predicted no effect concentration (PNEC), which refers to the concentration limit of individual compounds above which adverse effects on organisms may occur. These concentration limits in surface water are 150 ng dm⁻³ for BPA, 10 ng dm⁻³ for 4-t-OP and 330 ng dm⁻³ for 4-NP (EU, 2008). In the literature on the subject, recommendations can be found to tighten BPA PNECs for specific groups of organisms. A PNEC of 60 ng dm⁻³ has been proposed for species such as snails, which are extremely sensitive to the effects of this chemical (Wright-Walters et al. 2011). Currently, the lowest PNEC for BPA is 10 ng dm⁻³, for its effects on the larvae of the insect Chironomus riparious (Selvaraj et al. 2014). Accordingly, based on these acute toxicity values and an assessment factor of 100, a PNEC of 0.88 ng dm⁻³ was obtained (Ministry of the Environment, Japan, 1998).

In addition to PNEC, there are also other indicators related to concentrations of phenol derivatives that are safe for organisms. The Water Framework Directive, created by the European Union, indicates an annual average concentration (AA-EQS) and a maximum allowable concentration (MAC-EQS) for BPA, 4-t-OP and 4-NP.

In this study there were four values that exceeded the PNEC for 4-t-OP. The largest one was more than threefold higher and was found in the Kacza water collected in spring. Samples from other sites (Martwa Wisła, Gizdepka and Vistula) had instances where the limits were exceeded by a few percent each. Interestingly, one PNEC concentration exceedance of bisphenol A was also observed for the water from the Kacza. The catchment area of this river covers a wide urbanised, touristic area where the consumption of plastic products is higher; this may translate into a potentially higher load of phenol derivatives being introduced into the river. If we take into account the proposed lowered PNEC for BPA, due to its toxicity for particular groups of organisms, the number of exceedances will be higher: the concentration of 60 ng dm⁻³ was exceeded in 18.5% of the samples, while the most rigorous level (10 ng dm⁻³), which is harmful to insects, was exceeded in 89% of the samples. No dangerously high concentrations were observed for 4-nonylphenol. Neither the AA-EQS nor the MAC-EQS were exceeded in any of the environmental water samples (Table 6). Despite the fact that the concentrations measured in collected river water samples can be described as safe for aquatic organisms, it should be remembered that they were single samples. No monitoring or frequent sampling of these concentrations was carried out. Sampling could clearly indicate whether there is a risk connected with phenol derivatives being introduced to the sea via rivers. In summer, when tourist traffic increases in the area of the coastal zone of the sea, concentrations in river water may rise (Staniszewska et al. 2016a). Additionally, the mixture of these three compounds has a greater estrogenic effect on fish reproduction than each compound separately (Brian 2005). Therefore, even the low concentrations found in the small rivers in Pomerania may have chronic effects on aquatic wildlife.

During studies in 2011 and 2015 in the same area, high concentrations of phenol derivatives were observed in river outlets (in 2011, from < 4 to 55.94 ng dm⁻³ (average: 34.49 ng dm⁻³) for 4-NP, and from < 1 to 5.64 (3.04 ng dm⁻³) for 4-t-OP; in 2015, from < 5.0 to 277.9 ng dm⁻³ (56.5 ng dm⁻³) for BPA, from < 1.0 to 834.5 ng dm⁻³ (42.1 ng dm⁻³) for 4-t-OP and from < 4.0 to 228.6 ng dm⁻³ (38.8 ng dm⁻³) for 4-NP (Staniszewska, Falkowska 2011, Staniszewska et al. 2016a). It was shown that small rivers, whose course is about 100 times shorter, bring more concentrated pollutants to sea water. As in the present study, the widest range of concentrations (< 1 – 834.8 ng dm⁻³) was found in the summer for a sample taken from the Kacza River. In previous years in this region, concentrations of endocrine-active phenol derivatives were also exceeded. The highest number of exceedances was observed for 4-t-OP. In surface water, the PNEC was exceeded in 36% of water samples (on average, the values exceeded PNEC 15.7 times in river water and 3.3 times in sea water). The high exceedances of PNEC limits for 4-t-OP concentrations in riverine water point to land-based sources of phenol compounds (Staniszewska et al. 2016a). It seems that the situation has improved over the years. However, it should not be forgotten that the compared concentrations were short-term. According to the authors, it is crucial to determine the loads of endocrine-active compounds being introduced into sea surface waters. Of course, it makes sense to know the temporary concentrations because of the organisms that live in river waters, but because xenobiotics accumulate in the marine environment, they also pose a great threat to organisms on all trophic levels.

There is very little data available in the literature on the load of endocrine-disruptive phenol derivatives being introduced into marine waters by rivers. According to Yamazaki et al. (2015), 322 kg year⁻¹ are delivered to Tokyo Bay by the Arakawa River, the Edogawa River and the Tamagawa River. This suggests that the BPA load carried in spring or summer by the Vistula may be over 8 times higher, despite the fact that less bisphenol A is used in Polish industry. Research by Huang et al. (2020) indicated that the load of bisphenol analogues introduced by tributaries to the Pearl River reached 6,372 kg year⁻¹. According to Saniewska (2018), 200 kg year⁻¹ of mercury – also an endocrine-disruptive substance – enters the waters of the Gulf of Gdańsk with river runoff. Of course, in addition to mercury or phenol derivatives, a number of different pollutants are introduced into sea waters. These may act in synergy, thus increasing their bioaccumulation or biomagnification capacity. After comparing the concentrations obtained in this study with those found in the literature, it can be concluded that the concentrations were in most cases similar or lower. BPA is measured in rivers, water treatment plant waste and effluent from such plants (Santhi et al. 2003, Crain et al. 2007). The highest concentrations of this compound (reaching even hundreds of ng dm⁻³ [Yamazaki et al. 2015, Huang et al. 2020, Wu et al. 2013, Jin et al. 2004]) are found in Asian rivers, often flowing through huge metropolitan areas, where the dense population entails high production of BPA-containing waste and highly developed industry. In Europe, concentrations of these compounds are as much as two orders of magnitude lower. For example, as reported by Cladiere et al. (2013), in the Seine the concentrations of bisphenol A were 154 ng dm⁻³ and were similar to the concentrations of 4-NP (28–157 ng dm⁻³). In turn, in one of the tributaries of the Rhine in Switzerland, Voutsa et al. (2006) found endocrine-active phenol derivatives within a range of 68–326 ng dm⁻³ for 4-NP, 6–22 ng dm⁻³ for 4-t-OP and 9–76 ng dm⁻³ for BPA. In river water samples in southern Germany, BPA concentration was within a range of 500 pg dm⁻³ to 16 ng dm⁻³, while 4-nonylphenol concentrations were between 6 and 135 ng dm⁻³.