Partition of metals in the Vistula River and in effluents from sewage treatment plants in the region of Cracow (Poland)
Email: [email protected]
Abstract
The Vistula River suffers from heavy pollution with multiple origins. In the upper reaches, metallic and chlorine pollution originates from the mining and industrial region of Upper Silesia. Downstream from Upper Silesia, urban and industrial sewage adds more metallic and organic contaminants from the large urban agglomeration of Cracow. Although the river status is monitored routinely, little is known about the partition of metals between particulate and dissolved forms. This study focuses on metal partitioning and on the impact of the two main wastewater treatment plants at Cracow on metal concentrations in the Vistula River. The Cd, Co, Cu, Mn, Pb and Zn content was measured in both dissolved and particulate fractions. High metal concentrations in the Vistula River persist, although current levels seem to be lower than those in the past. Metal concentrations in the Vistula River and effluents from the sewage treatment plants at Cracow are similar, indicating a relatively minor contribution from the treated sewage. However, untreated sewage may be a significant source of contaminants. Despite high anthropogenic metal concentrations, the metal partitioning coefficients (Kd) in the Vistula are similar to these found in unpolluted rivers. Within a narrow pH range, Kd values depend on the metal affinity to particles, but there is no evidence of dependence on particle or chloride concentrations. An important fraction of the toxic metals Pb and Cd is associated with particles, which may decrease their immediate availability to the biota of the river.
INTRODUCTION
The Vistula River is the largest river in Poland (watershed area 194 700 km2), and second-largest river flowing into the Baltic Sea in terms of water discharge (842 m3 s–1) after the Neva River in Russia ( Niemirycz 1997). The Vistula River, with a total length of 1047 km, crosses Poland from the Carpathian Mountains in the south to the Baltic Sea in the north ( Fig. 1), receiving waste waters from large cities (Cracow, Warsaw) and industrial centres such as the Upper Silesia industrial complex (coal mines, foundries, chemical plants), Pulawy, Wloclawek (chemical industries), Plock (petrochemical plants) and many others. A regular water quality monitoring system was established at a the national level in 1987, with a baseline network of 20 sampling stations ( Mierzwinski et al. 1997 ). Three of the stations are located on the main course of the Vistula River. There are also regional monitoring programmes carried out by the Voivodeships. The monitoring demonstrated that the Vistula River is polluted heavily with nutrients, salts, metals and bacteria. According to Polish classifications, in 1992, water from the entire course of the river was unsuitable for drinking water uptake, for recreation or fish breeding. Water from 25% of the river’s length was suitable for irrigation and industrial use and the remainder of the river was heavily polluted (unclassified; Dojlido 1997). The main sources of pollution are located in the upper reaches of the Vistula where waste water from the Silesian industrial region is discharged ( Helios-Rybicka 1996). The sediments of the Przemsza River, an important tributary draining Upper Silesia, are heavily contaminated. For example, in 1995 at a station close to the confluence with the Vistula, the concentrations of Zn, Cd and Pb were 6522, 116 and 418 μg per g sediment, respectively ( Bojakowska & Sokolowska 1996). Of the three national monitoring stations on the Vistula, that at Tyniec, just upstream from Cracow ( Fig. 1), invariably has the poorest water quality (the lowest dissolved oxygen level and the highest levels of biological oxygen demand (BOD), dissolved salts, SO4–2, Cl–, PO4– and heavy metals such as Cr, Cd, Cu and Pb) ( Dojlido 1997). The mean (± SD) concentrations of these metals in 1995 were 5 ± 1, 7 ± 3, 13 ± 4, 39 ± 2 μg L–1, respectively ( Cydzik et al. 1996 ).
Location of sampling sites (V1–3) in the Cracow area. The thin line on the large-scale map indicates the extent of Cracow city.
The input of waste water from the Cracow urban agglomeration and industrial complex contributes additional pollution loads to the upper Vistula River. Some 100 × 106 m3 of municipal sewage was discharged to the Vistula in 1995. Almost half of this amount was not treated at all and less then 1% passed through a full ternary treatment that included a biological reactor. Almost 30 × 106 m3 of industrial waste water, originating from the large Sendzimir (formerly Lenin) steelworks at Nowa Huta, was mechanically treated ( Turzanski 1996).
Although the results of monitoring indicate a high level of metal contamination in the Vistula, little is known about metal partitioning and bioavailability. This paper presents the measurements of metal concentrations (Cd, Co, Cu, Mn, Pb and Zn) in particulate and ‘dissolved’ fractions in samples from the Vistula River and in effluents from the sewage treatment plants (STP) at Cracow. This work is part of the programme entitled Biotests for the promotion of improved methods of waste-water treatment as a contribution to sustainable development in Central and Eastern European Countries. The results of ecotoxicological studies in the Vistula River are also presented in the present volume ( Pardos et al. 2000a,b ). The data shown here allow the assessment of the different geochemical behaviours of metals in a heavily polluted environment, to compare the partition coefficients (Kd) of the river and sewage effluents and to discuss the potential bioavailability of metals.
METHODS
Sampling was carried out in May 1997 on the effluent of the sewage treatment plant (STP) at Plaszow (16 May 1997), and that of an industrial treatment plant at Nowa Huta (17 May 1997). Two stations on the Vistula River (V1 and V3) were sampled (22 May 1997), upstream and downstream from Cracow, respectively, and the third (V2; 21 May 1997) between outlets from STP in Plaszow and Nowa Huta ( Fig. 1). Filtered water and suspended matter was collected. The latter was isolated from 270–720 L of water by on-line continuous-flow centrifugation (Westfalia, Bern, Switzerland). The centrifuge cut-off at a flow rate of 6 L min–1 is approximately 1 μm ( Steinmann et al. 1999 ). The suspended matter (SM) was frozen and subsequently freeze-dried in the laboratory. The SM concentration was measured from 1 L of raw water filtered using a 1.2 μm glass filter (GFF/C Whatman, Maidstone, England). The SM concentration errors were evaluated with the propagation of errors on dry SM weight and filtered volume. The filtered water for metal measurements was isolated from particles by filtration through a 1.2 μm fluoropore capsule (Millipore, Bedford, MA, USA) with a syringe, immediately after sample collection. At each sampling site, a field blank (ultrapure water) was made. In comparison to the metal concentrations in samples, the metal concentrations of blanks were negligible and constant whatever the environment of the sampling point. Each water sample was collected in a polypropylene acid-washed container, then acidified (pH < 2) with ultrapure nitric acid (Merck, Geneva, Switzerland), and stored in darkness at 4°C until the analysis. The filter-passing fractions of the metals were termed the ‘dissolved’ fractions, although they clearly contained both dissolved and colloid-bound metals.
Filtered water was digested by using a microwave digestion system (ETHOS, Milestone, Sorisole, Italy). Fifty millilitres of sample was mineralised using 3 mL ultrapure nitric acid and 2 mL ultrapure hydrogen peroxide. A multiple-stage heating programme was used: 2 min at 200 W, 10 min at 800 W, and 10 min at 250 W, each at 20 × 105 Pa pressure. Appropriate batch blanks were prepared with each set of samples. Suspended matter (500 mg) was digested during 12 h at 100°C in 2 M nitric acid. After centrifugation, the supernatant was analysed. The recoveries obtained with the HNO3 digestion procedure were tested on two lake-sediment Certified Reference Materials (LKSD1-LKSD4). The percentage of the total content extracted by using this method was > 95% for Cd, Co, Cu, Mn, Pb and Zn. Thus, the digestion method assures the quasi-complete extraction from the matrix and the measured metal concentrations can be considered representative of the total content ( Favarger et al. 1998 ).
Trace elements in filtered water and in suspended matter were analysed by ICP-MS/OES (inductively coupled plasma mass spectrometry/optical emission spectrometry; POEMS 1 -Thermo Jarrel Ash, USA) equipped with a Mistral® desolvating system (Fisons Instruments, USA). The Mistral® desolvating system consists of a heated spray chamber, where the solvent is evaporated, and then condensed by a cooling system. Typical operating conditions and detection limits are summarised in Table 1. The errors of metal measurements are given at the 95% confidence level.
| Plasma conditions | |
| Rf power | 1250 W |
| Plasma gas flow | 17 L min–1 |
| Auxiliary gas flow | 1 L min–1 |
| Nebuliser gas flow | 0.45 L min–1 |
| Sample uptake MS | 0.9 mL min–1 |
| OES | 1.8 mL min–1 |
| Sample introduction system: Meinhard + Mistral | |
| Heater | 140°C |
| Cooler | 1°C |
Particulate organic carbon (POC) was determined by wet oxidation with dichromate–sulphuric acid oxidant. The chloride concentration was determined by photometry after reaction with mercuric thiocyanate in the presence of ferric ions. Conductivity, pH, and temperature were measured in the field by using a portable pH meter and conductimeter (WTW Instruments, Geneva, Switzerland).
RESULTS
Concentrations of Cd, Co, Cd, Mn, Pb and Zn in the dissolved and particulate fractions, and chloride, POC and SM content were measured in the samples from the Vistula River and the outlets of two sewage treatment plants at Plaszow and at Nowa Huta.
Vistula River
Trace metal concentrations in the dissolved fraction ( Table 2) generally differed between the three stations by less than a factor of three. Only the variability of Pb concentrations was much higher, with the concentration of Pb at station V1 being exceptionally high. Between V1 and the downstream stations (V2 and V3, the Pb content decreased by one order of magnitude. Other dissolved metal concentrations decreased by approximately a factor of two (Cd, Mn) or remained nearly the same (Co, Zn) between stations V1 and V3. Concentrations of Cu in the dissolved fraction were high and increased by nearly 20% from the upstream to the downstream stations. The chlorine concentration was also high and increased by 16% from the upstream to the downstream stations.
| V1 | V2 | V3 | Plaszow | Nowa Huta | |
|---|---|---|---|---|---|
| Sampling date | 22 May 1997 | 21 May 1997 | 22 May 1997 | 16 May 1997 | 17 May 1997 |
| pH | 7.4 | 8.2 | 7.6 | 7.7 | 8.8 |
| Cl (mg L–1) | 656 | 776 | 764 | 144 | 708 |
| Cond (μS cm–1) | 3530 | 3250 | 3560 | 1138 | 3150 |
| Dissolved | |||||
| Cd (μg L–1) | 0.58 ± 0.01 | 0.15 ± 0.03 | 0.23 ± 0.05 | 0.14 ± 0.01 | 0.01 ± 0.01 |
| Co (μg L–1) | 0.45 ± 0.05 | 0.50 ± 0.05 | 0.40 ± 0.05 | 2.39 ± 0.04 | 0.48 ± 0.05 |
| Cu (μg L–1) | 43.3 ± 0.6 | 20.4 ± 0.9 | 51.0 ± 0.7 | 9.4 ± 0.8 | 43.0 ± 0.6 |
| Mn (μg L–1) | 323.1 ± 0.2 | 258.1 ± 0.1 | 189.5 ± 0.1 | 149.8 ± 0.1 | 4.9 ± 0.1 |
| Pb (μg L–1) | 2.69 ± 0.01 | 0.09 ± 0.01 | 0.13 ± 0.02 | 0.30 ± 0.01 | 1.05 ± 0.01 |
| Zn (μg L–1) | 48.4 ± 0.8 | 20.90 ± 0.27 | 48.2 ± 0.4 | 31.1 ± 0.6 | 41.8 ± 0.5 |
| Particulate† | |||||
| Cd (μg L–1) | 0.1 6 ± 0.02 | 0.45 ± 0.03 | 3.30 ± 0.19 | 0.42 ± 0.03 | 0.002 ± 0.001 |
| Co (μg L–1) | 0.045 ± 0.005 | 0.18 ± 0.01 | 1.04 ± 0.06 | 0.58 ± 0.03 | 0.456 ± 0.114 |
| Cu (μg L–1) | 1.302 ± 0.122 | 1.046 ± 0.058 | 11.527 ± 0.572 | 10.604 ± 0.530 | 0.341 ± 0.085 |
| Mn (μg L–1) | 1.715 ± 0.161 | 10.02 ± 0.56 | 46.76 ± 2.32 | 22.67 ± 1.13 | 1.160 ± 0.290 |
| Pb (μg L–1) | 1.30 ± 0.12 | 2.58 ± 0.16 | 20.22 ± 1.10 | 6.08 ± 0.32 | 0.078 ± 0.020 |
| Zn (μg L–1) | 1.37 ± 0.13 | 4.76 ± 0.27 | 33.6 ± 1.7 | 40.8 ± 2.0 | 0.209 ± 0.052 |
| POC (mg L–1) | 0.435 ± 0.001 | 2.4948 ± 0.063 | 10.95 ± 0.36 | 20.45 ± 0.11 | 0.0064 ± 0.0001 |
| SM (mg L–1) | 3.2 ± 0.3 | 12.6 ± 0.7 | 72.5 ± 3.6 | 54.1 ± 2.7 | 0.8 ± 0.2 |
| Concentrations in suspended matter‡ | |||||
| Cd (μg L–1) | 51.3 ± 1.6 | 36.1 ± 1.5 | 45.5 ± 1.4 | 7.8 ± 0.3 | 2.3 ± 0.5 |
| Co (μg L–1) | 14.1 ± 0.6 | 14.2 ± 0.5 | 14.3 ± 0.5 | 10.7 ± 0.3 | 570 ± 4.4 |
| Cu (μg L–1) | 407.0 ± 0.6 | 83.0 ± 0.5 | 159.0 ± 0.1 | 196.0 ± 0.4 | 426 ± 0.5 |
| Mn (μg L–1) | 536.0 ± 1.3 | 795.0 ± 1.1 | 645.0 ± 1.5 | 419.0 ± 0.9 | 1450 ± 1 |
| Pb (μg L–1) | 407.9 ± 5.9 | 204.5 ± 6.2 | 278.9 ± 6.1 | 112.4 ± 1.8 | 98.7 ± 1.5 |
| Zn (μg L–1) | 429.0 ± 4.3 | 378.0 ± 3.5 | 463.0 ± 4.1 | 754.0 ± 3.4 | 261.0 ± 3.4 |
| POC (%) | 13.6 ± 0.3 | 19.8 ± 0.5 | 15.1 ± 0.5 | 37.8 ± 0.2 | 8.0 ± 0.1 |
- *The dissolved values correspond to the fraction passing a membrane with 1.2 μm pore diameter. †The particulate values are the concentrations of suspended matter that are retained on the membrane multiplied by the concentration of metals measured in the centrifuged suspension (weight per volume of water). ‡Concentration in suspended matter is measured in the centrifuged suspension (weight per weight of suspended matter).
Metal concentrations in the SM of the Vistula River ( Table 2) had relatively low levels of variability, usually within a factor of two, except Cu, which had comparatively high concentrations at the V1 station. The homogeneity of the metal concentrations in the SM is remarkable, despite an order of magnitude increase in SM concentration between the stations V1 and V3. The comparison of metal concentrations in the SM between the upstream (V1) and a downstream station (V3) showed an important decrease in Cu (2.6-fold decrease) and moderate decrease for Pb (1.4-fold decrease), as well as a slight increase for Mn and Zn. The organic carbon concentration in the SM increased by only 10% from the upstream to the downstream stations.
As mentioned above, the intermediate station (V2) was sampled one day earlier than the other two stations. However, the metal concentrations in the SM are comparable with those at stations V1 and V3.
With relatively little variation in the SM metal concentrations between the stations, combined with a large increase in SM concentrations (> 20-fold), the particulate metal concentrations ( Table 2) increase correspondingly from station V1 to V3 (15–27-fold increase). The increase of particulate Cu is markedly less pronounced (eightfold increase).
Sewage treatment plants
Dissolved metal concentrations in effluents from the STP ( Table 2) are generally lower or equal to those measured in the Vistula, with the exception of Co in the STP effluent from Plaszow, where the Co concentration was fivefold that of the Vistula or in the Nowa Huta STP effluent.
A major difference between the two STP is observed in the SM concentration (0.8 mg L–1 at Nowa Huta and 56 mg L–1 at Plaszow). The latter is comparable to the highest value measured in the Vistula (V3). Although only mechanical treatment is used at the Nowa Huta STP, the removal of organic poor suspension is very efficient. The POC concentration at the Plaszow STP is 100-fold that from the Nowa Huta STP.
Characteristic differences between municipal and industrial STP are also shown by the trace metal concentrations in the SM ( Table 2). In the Nowa Huta STP, effluent Co concentrations in the SM are one order of magnitude higher than in the suspensions from the Vistula River and the Plaszow STP effluent, while Mn concentrations are three-fold more. Concentrations of other metals in the STP suspended matter are lower or equal to these measured in the Vistula SM except for Zn in Plaszow.
Because of a very efficient removal of suspended matter at the Nowa Huta STP, the particulate metal concentrations (in μgmg L–1) are very low, even for Mn and Co, while those from the STP at Plaszow and in the Vistula River are similar.
DISCUSSION
First, concentrations of metals in the Vistula River will be compared with data obtained by using similar methods from a number of polluted European rivers. Then, discussion will focus on metal partition coefficients.
Dissolved concentrations
The concentrations of dissolved metals and Cl in the Vistula near Cracow are higher than in the ‘world average river’ ( Meybeck 1988; Meybeck & Helmer 1989), and Co, Cu, Mn, and Zn concentrations are also markedly higher than in the Po River, downstream of its most polluted tributary, the Lambro River ( Pettine et al. 1996 ; Table 3). The concentrations measured in this study are lower for Cd, Pb, and Zn, and higher for Cu and Mn than are than the yearly means reported in 1995 at Tyniec station ( Cydzik et al. 1996 ). The contribution of dissolved metals from the sewage treatment plant to the Vistula River was negligible at the time of sampling.
| Vistula (the | World | |||
|---|---|---|---|---|
| present study) | Vistula | Po | average | |
| Cl | 656–775 | 512–1266 | – | 0.6–25 |
| Cd | 0.2–0.6 | 4.0–10.0 | 0.01–0.59 | 0.001 |
| Co | 0.4–0.5 | – | 0.004–0.15 | 0.1 |
| Cu | 20.4–51.0 | 9.0–17.0 | 0.95–3.07 | 1.4 |
| Mn | 189.5–323.2 | 100.0–140.0 | 0.1–17.6 | 10 |
| Pb | 0.1–2.7 | 16.0–62.0 | 0.05–1.20 | 0.04 |
| Zn | 20.9–48.4 | 132.0–530.0 | 0.3–14.9 | 0.2 |
Concentrations in suspended matter
With the exception of Mn, metal concentrations in the SM of the Vistula River at Cracow are higher than in the SM of the ‘world average river’ ( Förstner & Wittmann 1979; Martin & Meybeck 1979). A comparison with three major European rivers ( Table 4) is particularly pertinent, as the sampling and analysis followed the same procedure and were performed by the same laboratory ( Santiago et al. 1993 ; Diserens 1997; Bostina & Bostan 1998). The most striking feature in the Vistula SM is an extremely high concentration of Cd and relatively high concentration of Pb as compared with other polluted rivers. Copper and zinc concentrations were also high, but comparable to the few most polluted sites on the Rhone (downstream from Lyon), Danube (downstream from Bucharest industrial region) and Po (downstream from Milan) rivers. It should, however, be noted that the SM metal concentrations measured in this study are clearly lower than those reported from the Vistula SM in 1983 ( Helios-Rybicka 1983), confirming a trend to decrease the metal loading from the Upper Silesia area. In terms of metal loading, the SM from STP in the region of Cracow contribute little to the receiving Vistula waters, as the concentrations in Plaszow were lower than in the Vistula suspension and the SM load at Nowa Huta STP was very low.
| Vistula (the present study) | Vistula | Rhone | Danube | Po | World average | |
|---|---|---|---|---|---|---|
| Cd | 36–51 | 130–210 | 0.3–4.2 | 1–2 | 0.3–4.1 | 0.3 |
| Co | 14 | – | – | 14–17 | – | 20 |
| Cu | 83–407 | 310–710 | 50–200 | 50–100 | 40–390 | 50 |
| Mn | 536–795 | 960–3050 | 300–1080 | – | 800–1700 | 1100 |
| Pb | 204–408 | 550–1070 | 55–122 | 40–90 | 20–135 | 40 |
| Zn | 378–429 | 3550–7710 | 120–325 | 130–350 | 100–750 | 240 |
Total metal concentration
While metal concentrations in filtered water and in SM were measured directly, the latter must be multiplied by the concentration of SM to obtain the concentration of particulate metal (μg L–1). As the SM concentrations are highly variable even on a short time scale, our SM concentration measurements are, almost certainly, not representative and must be used with caution.
The most striking observation is that the total metal concentrations in the Vistula River were equal or higher than in effluents from the STP, except for the Co concentration in the Plaszow STP. This indicates that the metal level in the upper Vistula is elevated upstream from Cracow, in agreement with the results of monitoring ( Bojakowska & Sokolowska 1996). However, at the time of sampling, the total metal concentrations increased downstream and the total metal enrichment at the downstream station (V3) relative to the upstream station (V1) was 5.1, 4.7, 2.9, 1.6 and 1.4 for Pb, Cd, Co, Zn and Cu, respectively. It should be noted, however, that the SM concentration increased 22-fold, a much greater increase than the increase in total metal concentration. This suggests an important loading of suspended matter with a lower metal concentration (in μg g–1) to the Vistula River in Cracow between sampling points V1 and V3, at least at the time of sampling. Based on 1995 data, an average daily suspended load originating from municipal sewage is 48 t day–1 ( Turzanski 1996). Because an average Vistula discharge at Tyniec (station V1) is 65.8 m3 s–1 and an average suspension concentration is 22 mg L–1, the load of SM may increase by 40% from this source alone. At the exceptionally low SM concentration of 3.2 mg L–1 that was measured on the sampling day at the V1 station, the load may increase fourfold. Even if one considers additional SM matter input from a few smaller turbid effluents and a time lag of a few hours in sampling, it is difficult to explain a 20-fold increase in SM concentration between stations V1 and V3. The river water downstream from Cracow municipality displays peachiness in turbidity and therefore the measured value at station V3 may be biased.
Partition of trace metals
Sorption processes at the particle–water interface are of importance for the transport and bioavailability of the chemical species in the natural environment. The dissolved fraction may generally be considered as more mobile and bioavailable for planktonic organisms than the particulate fraction, which is removed from water by sedimentation in a low-energy environment and may be of low or no bioavailability. Distribution coefficients (Kd) were calculated to quantify the partition between the particulate and dissolved fractions as:
where [Me]P (w/w) and [Me]d (w/v) are the concentrations in particles and in filtered water, respectively. An elevated Kd indicates a higher affinity for the particulate phase. It should be noted that in this paper the ‘dissolved’ fraction was defined as the fraction passing through the 1.2 μm filter. In addition, the concentration in the SM denotes the concentration measured in the particles retained by the continuous flow centrifuge with a cut-off of approximately 1 μm. The limits of the ‘particle-bound’ and ‘dissolved’ fractions are always somewhat arbitrary, but with the limit of 1 μm, the separation between entities subjected only to advective or turbulent transport and those also subjected to settling in low-energy conditions is more justified than the traditional limit of 0.45 μm.
Field distribution coefficients (Kd) for metals in the Vistula River and in the effluent from the treatment plants are plotted against the SM concentration in Fig. 2 and compared in Table 5 with the Kd from other rivers. The Kd of metals from the Vistula River decrease in the following order: Pb > Cd > Co > Zn ≥ Cu ≥ Mn. A similar sequence is observed in the effluent from the Plaszow STP, although the absolute Kd values for Pb, Cd and Co are much lower and for Cu much higher than in the river. The pH of the water from all sampling points is similar (7.4–8.8). For a given metal, the differences in Kd values between sampling sites are neither related to the particle concentration, as shown in Fig. 2, nor to water ionic strength (expressed as conductivity or Cl concentration). Apparently, the metal partition is mainly determined by the affinity of metals to particles and the changing particle composition may mask the so-called particle concentration effect. For example, a high Kd for Cu in the Plaszow STP effluent may be related to a high concentration of particulate organic matter. The ranking of metal Kd obtained for the effluent of the STP in Nowa Huta (Co > Mn > Cd > Pb > Cu > Zn) is very different from that in the Vistula. Also, the absolute values of Kd for Co and Mn in the Vistula are twofold those found in the Nowa Huta STP effluent. This treatment plant receives waste water from steelworks and associated industries, which are potential sources of particulate cobalt and manganese. Although the bulk of these particles settle in the STP and the effluent is poor in SM, the remaining small particles are highly enriched in Co and Mn. It is possible that such particles do not exchange metals with water over a short time scale, contrary to the possible rapid release of pollutants adsorbed on particle surfaces. It is possible that a high concentration of Co in the SM results from sorption on Mn–oxihydroxides, as this association has been previously observed in many surface waters ( Oztürk 1995; Sirinawin et al. 1998 ).
Metal distribution coefficients (Kd) plotted against particle concentrations for water from the Vistula River (V1–3) and effluent from wastewater treatment plants (Plaszow, Nowa Huta). (●), Cd; (○), Co; (▾), Cu; (▿), Mn; (▪), Pb; (□), Zn.
| Non-polluted | ||||
|---|---|---|---|---|
| Sample | Vistula | Plaszow | Nowa Huta | rivers |
| Cd | 4.9–5.5 | 4.7 | 4.1 | 3.7–5.8 |
| Co | 4.5–4.6 | 3.7 | 6.1 | 4.5 |
| Cu | 3.5–4.0 | 4.3 | 4.0 | 4.3–5.1 |
| Mn | 3.2–3.5 | 3.4 | 5.5 | 3.9 |
| Pb | 5.2–6.5 | 5.6 | 5.0 | 5.0–6.0 |
| Zn | 3.9–4.3 | 4.4 | 3.8 | 3.3–5.1 |
Metal partitioning coefficients found in the river and in the effluent from Plaszow were of the same magnitude as in the unpolluted rivers at a similar pH ( Table 5), despite a relatively high water salinity in the Vistula. A relatively constant Kd for each metal in the Vistula River and their lack of dependence on particle concentration result in a strong dependency of the metal fraction transported in particulate form on the SM concentrations ( Fig. 3). Most of the Pb and Cd is transported in the particulate form at the V2 and V3 stations, while merely 30% of the Pb and 20% of the Cd are bound to particles at the V1 stations. A previous study has shown that Cd sorbs on particles as soon as the SM concentrations rise ( Thouvenin et al. 1997 ). In general, the particulate metal fraction increases from V1 to V3 for all metals; however, it never attains more than 40% for Zn, Mn, and Cu. In the effluent from the Plaszow STP, more than 50% of Cd, Cu, Pb and Zn and in the Nowa Huta STP 80–100% of all metals are transported in particulate form.
Fraction of metals associated with particles in the Vistula River (V1–3) and effluent from wastewater treatment plants (Plaszow, Nowa Huta). (▪), Cd; (), Co; (), Cu; (), Mn; (), Pb; (), Zn; (), suspended matter (mg L–1). Values for metals are percentages, values for suspended water are in mg L–1.
Conclusions
The impact of effluents from two major wastewater treatment plants on the Vistula River in terms of heavy metal load is negligible, as the river is overwhelmed by the major contamination sources located in the Upper Silesia mining and industrial region, upstream from Cracow. With respect to metal and chloride concentrations, water quality in the Vistula River is equal to or worse than that of mechanically treated municipal sewage. Metal distribution coefficients depend, at similar pH, on the metal’s affinity to particles, as shown by similar Kd ranking for individual metals at different sampling sites. Only the effluent from the industrial wastewater treatment has different Kd sequences, which is probably related to the enrichment in Co- and Mn-rich particles from the steelworks. The absolute Kd values depend on the quality of the suspension but show no dependence on particle or chloride concentrations. A relatively high fraction of the most toxic metals, Pb and Cd, is associated with particles, which may result in a decrease of their immediate bioavailability.
Acknowledgements
The authors are grateful to the Fonds National Suisse de la Recherche Scientifique for financial support of this research project (7PLPJ048296). Jean-Daniel Steiner and André Dumont (Applied Research Laboratories, Switzerland) are acknowledged for providing the Mistral® system. The Fondation pour l’etude et la protection du patrimoine lacustre has provided financial help in purchasing the microwave oven. We are also grateful to all our colleagues from the IZWiOS (Instytut Zaopatrzenia W. Wode I. Ochrony Srodowiska) at the Technical University of Cracow for help in sampling.
