Impact of heated waters on water quality and macroinvertebrate community in the Narew River (Poland)

1 Introduction

Man-made disturbances of water bodies may affect the species composition of aquatic communities. Riverbed engineering, hydrotechnical constructions and sewage discharge may affect the taxonomic composition of aquatic organisms [1-4]. Heated waters delivered via canals from power station cooling systems are a specific type of waste water. The constant input of cooling waters exerts an influence on the physical properties and chemical composition of the recipient waters and affects the living conditions of the aquatic flora and fauna. The permanently elevated temperature accelerates biological processes and changes the habitat of aquatic organisms [5,6].

Macroinvertebrates are a group of organisms sensitive to changes in habitat conditions. They are used as a measure of surface water quality in many biotic indices in the European Union (Directive 2000/60/EC) [7]. Macroinvertebrates are sensitive to chemical pollution of waters and to oxygen deficits. Cooling waters delivered from a power station have higher temperature than those to which they are discharged, which decreases oxygen solubility [8,9] and taxonomic diversity [10]. Although the impact of heated waters on the fresh water biocenosis is a global issue and has been studied for many years, research may still provide relevant information due to the individual character of each river [8,11] depending on e.g. the speed of water flow, the type of bottom, the occurrence of aquatic plants.

According to the Regulation of the Polish Minister of the Environment [12], the maximum allowable temperature of industrial sewage discharged into water should not exceed 35°C. The effect of cooling waters from coal-burning power stations in Poland on benthic macroinvertebrates has rarely been investigated with the exception of data on the impact of Konin and Pątnów power stations on nearby lakes [5,13]. The research by Krepski et al. [8] was also conducted to assess the influence of the Dolna Oder Power Station on changes in the qualitative and quantitative composition of the macrozoobenthos in the lower Oder River.

Studies on the effect of heated waters discharged from the Ostrołęka Power Station on physical and chemical properties of waters in the Narew River were carried out in the 1970s by Dojlido [9], while the macrobenthos of this river was analysed by Dusoge and Wiśniewski [14]. The authors showed that heated water discharge slightly increased the water temperature in the river. The increase in water temperature resulted in a slight decrease in the number of certain macrofauna species. The varying spatial and temporal density of macroinvertebrates was interpreted as a manifestation of the decreased stability of the benthic community at the sites downstream of the canal compared to the sites upstream.

In spite of the long-term operation of the Ostrołęka Power Station, the literature lacks recent information about the effects of heated waters discharged from the power station on the taxonomic composition of macroinvertebrates in the Narew River. The objective of the present study was to assess the present effect on the physical and chemical water properties and the macroinvertebrate community at the sites downstream of the heated-water inflow compared with those located upstream, about 40 years after the station was launched.

The following hypotheses were put forward in the study:

1. what effect the discharge of heated water from the Ostrołęka Power Station has on the major physical and chemical parameters of the Narew water,

2. how these alterations affect the characteristics of the macroinvertebrate community (composition and diversity).

2 Material and methods

Studies were carried out in the Narew River near Ostrołęka (21°35’E, 53°05’N), where the power station (one of the ten largest in Poland) is situated. The station with a maximum energy output of 2.2 TWh burns coal and produces 2% of the electric energy in the country [15]. Water taken from the Narew River is used in open turbine cooling systems. Water intake is situated on the left bank at 116^th km of the river course. The maximum water uptake at water temperature above 17°C and the maximum water output is 25 m³s^-1 [16].

Data were collected during four consecutive years, from 2013 to 2016, along two river stretches upstream and downstream of the canal that delivered heated waters from the power station. Three sampling sites were selected within each stretch and numbered from 1 to 6 as follows: site 1 – 3 km upstream of the canal, site 2 – 2 km upstream, site 3 – 1 km upstream, site 4 – 1 km downstream, site 5 – 2 km downstream and site 6 – 3 km downstream of the canal (Fig. 1). The sampling sites were selected in a manner to provide a similar hydromorphology. The bottom was sandy-gravel at all the sites and the water velocity was about 0.5 m s^-1 . Macrophytes (mainly Sagittaria sagittifolia L.) were scarce at the study sites. Samples were collected at a depth of about 1 m, at a distance of 2-3 m from the bank.

Fig. 1

The location of the sampling points in the Narew River in the region of the Ostrołęka Power Station

Samples of the water and macroinvertebrates were collected on 11 October 2013, 15 September 2014, 9 June 2015, 10 June 2016 from the left bank of the river. Samples of macroinvertebrates were collected from the bottom with a 20 cm wide drag by hauling it 5 times along 1 m of the bottom surface. The content of a net attached to the drag was transferred onto a metal sieve with a mesh size of 0.5 mm. The collected animals were carefully washed, picked up with the help of magnifying lens and tweezers and placed in marked containers. This biological material was preserved in 70% ethyl alcohol. In the laboratory, the organisms were identified to the family level and sometimes (Gammaridae, Bivalvia) to the species level with the help of identification keys for freshwater fauna by Kołodziejczyk and Koperski [17]. Quantitative analysis was also performed. Macroinvertebrates identified to the family level were used to calculate the Shannon-Wiener biodiversity index (H’).

The Shannon-Wiener biodiversity index (H’) was calculated as follows:

where p_i is the relative abundance (n_i/N) of family i, n_i=the number of individuals in family i

and N = the total number of individuals in all families.

Water and macroinvertebrate samples analysed in this study were collected from the same sampling sites in the following order: water followed by macrofauna. Temperature, pH, electrolytic conductivity, concentrations of oxygen, nitrates, phosphates, chlorides and calcium were determined in the water samples. Water temperature, oxygen concentration, pH and conductivity were determined in the field with the CX-742 Elmetron multifunction probe. Other chemical parameters were analysed in the laboratory. The concentration of nitrates was determined with the use of phenyl disulfonic acid. Phosphates were determined using the molybdenum blue method. Nitrates and phosphates were determined spectrophotometrically at wave lengths of 410 and 700 nm, respectively [18]. All spectrophotometric analyses were performed using the Shimadzu UV-VIS 1800 spectrophotometer. The concentrations of chloride ions were determined using the argentometric method and Ca – complexometric titration with standard EDTA solution [19]. All analyses were performed in three repetitions. The arithmetic mean of three measurements was used for further interpretation of the data.

The Mann-Whitney test was applied to calculate the differences in values of the studied physical and chemical parameters of water upstream and downstream of the canal.

Differences in the abundance of macroinvertebrates were checked using the G-test and differences between the number of families and the number of individuals and temperatures at separate sampling sites were checked using the chi-square test. The impact of the study year, season and sampling location was checked with the use of the ANOVA and t-test. The Hutcheson test was used to compare the Shannon diversity indices. Spearman’s correlation was also conducted. Canonical Correspondence Analysis (CCA) was conducted to find relationships between environmental variables and the abundance of macrozoobenthos orders. Log+1 transformation of data (abundance of macrobenthos) was applied in the calculations. Due to the small percentages of families, orders were used to perform the Canonical Correspondence Analysis. All statistical analyses were carried out using Statistica ver. 12.

Ethical approval: The conducted research is not related to either human or animals use.

3 Results

Most of the physical and chemical parameters analysed, except for water temperature and oxygen concentration, remained at similar levels at the sampling sites upstream and downstream of the canal (Table 1). Water temperature at the sites downstream was statistically higher than at those upstream of the canal (Z=2.19, p=0.028, N=24). Significant differences were noted in the oxygen concentrations but the mean value of this parameter was higher at the sites upstream of the canal (Z=2.25, p=0.024, N=24).

The collected biological material included a total number of 33 families of macroinvertebrates, represented by Insecta, Crustacea, Mollusca and Annelidae (Fig. 2). Differences were observed in the numbers of the above-mentioned taxa at the upstream and downstream sites (G=10.15, df=3, p=0.017). Molluscs were the dominant group with a higher percentage upstream of the canal, represented by families of snails and bivalves (Table 2). Families of Viviparidae and Sphaeriidae were most frequent in this group. Insecta were the second numerous group represented mainly by the family of Chironomidae. The percentage of Crustacea was higher downstream of the canal and represented by the family of Gammaridae. The least frequently occurring were Annelida represented by Oligochaeta and Hirudinea. The last taxa were noted only in samples collected from site 3 located upstream of the canal (Table 2). The industrial water discharge was found to have an effect on two families. The number of Gammaridae increased at the locations downstream (Z=2.71, p=0.007, N=24) and the Chironomidae densities were lower there (Z=2.02, p=0.043, N=24). No significant differences between the upstream and downstream sites were noted for the remaining families (p=0.193, for 33 comparisons). The abundance of four taxa (Ephemeroptera, Heteroptera, Diptera and Oligochaeta) varied between the study years (Table 3). However, the season only affected Ephemeroptera and the location – Diptera and Crustacea.

Fig. 2

Percentage share of studied groups of macroinvertebrates in sites upstream and downstream the canal delivering heated water from the Ostrołęka Power Station

The number of families at the study sites varied from 15 to 20 throughout the study period but no differences were found between upstream and downstream sites (χ²=0.78, df=3, p=0.853, Table 4). However, the number of individuals at the two river stretches in each study year varied (χ²=257.75, df=3, p<0.001). The Shannon-Wiener diversity index was similar for both studied river stretches and amounted to 2.42 for the sites upstream of the canal and 2.36 for the sites downstream. The compared values of the diversity index were not statistically significant (t=0.04, df=55, p=0.969). The numbers of families were positively correlated with temperature (r=0.50, p=0.013, N=24) and conductivity (r=0.41, p=0.048, N=24) and negatively correlated with oxygen (r= -0.44, p=0.032, N=24).

The Canonical Correspondence Analysis showed that temperature and conductivity correlated positively (r=0.91 and r=0.87, respectively) and O₂, PO43− negatively with the factor represented by the first axis (r= - 0.72 and r= -0.88, respectively), while the concentration of NO3− had a positive correlation (r=0.95) with the second factor (Fig. 3). The two axes accounted for 80.3% of the predictors’ variability. Some taxa, especially Bivalvia, Hirudinea, Diptera and Oligochaeta showed a positive correlation with NO3−, whereas Trichoptera and Odonata positively correlated with PO43−. We also found that the season and year of sample collection had a greater influence on the chemical parameters than the site at which a given sample was collected (Fig. 4). Samples collected in autumn 2013 and 2014 were much more similar compared to samples collected in spring in the two other years.

Fig. 3

Results of CCA in two types of samples sites in Narew River (Cond – conductivity, Temp – temperature, Bival – Bivalvia, Coleo – Coleoptera, Crust – Crustacea, Dipt – Diptera, Ephe – Ephemeroptera, Gast – Gastropoda, Heter – Heteroptera, Hirud – Hirudinea, Odon – Odonata, Oligo – Oligochaeta, Trich – Trichoptera)

Fig. 4

Results of CCA in two types of samples sites in Narew River. Codes means site and time of collected samples : B – before canal. A – after canal. a – autumn. s – spring. 13-15 – years 2013-2016

4 Discussion

The results of the performed analyses showed that the heated waters delivered to the Narew River from the Ostrołęka Power Station do not significantly modify the chemical composition of the river waters: pH, conductivity, concentration of nitrogen, phosphates, chlorides and calcium at the sites downstream of the canal. As expected, the temperature of the river water was higher downstream of the canal compared to waters upstream. The temperature decreased with the distance from the heated water discharge. The obtained results are consistent with those obtained by Dojlido [9] in the Narew River in the first years of the Ostrołęka Power Station’s functioning. His studies showed that the input of heated waters increased the river water temperature by 8°C at a distance of 200 m from the discharge but the temperature increase did not exceed 2°C at a distance of 1.5 km downstream. The observed decrease in the water temperature with the distance from the canal is a result of mixing river water with heated waters discharged from the power station. The maximum amount of water needed for cooling at the maximum water output from the power station is 25 m³ s^-1 [16], which accounts for less than 25% of the mean long-term water flow in the Narew River near Ostrołęka estimated at 109 m³s^-1 [20]. The observed increases in the water temperature downstream of the canal also significantly affect the changes in the concentration of oxygen in the water. A decrease in oxygen concentration at the sites located downstream compared with the sites upstream of the canal was a result of the temperature increase.

The analysis of the influence of selected physical and chemical water parameters on macroinvertebrate communities includes different levels of organism identification [21-24]. In the present study, as well as in many biotic indices used for water quality assessment (for example RIVPACS [25], BMWP [26], MMI_PL [27]), the macrofauna is identified with an accuracy to the family level. During the entire study period, 29 and 27 families were identified upstream and downstream of the canal, respectively. By comparison, Krepski et al. [8] identified 26 families in the stretch of the Oder River near the Dolna Odra Power Station represented, as in the present study, by Mollusca, Insecta, Crustacea and Annelida. The families Viviparidae, Sphaeriidae, Tubificidae, Chironomidae and Gammaridae accounted for the largest percentage of the taxa inhabiting the examined stretch of the Narew River near the Ostrołęka Power Station. The literature [8,28] indicates that the increased water temperature has a positive effect on the increasing abundance of Gastropoda. The results of the statistical analysis fail to confirm the influence of the water temperature in the Narew River on the abundance of families representing Gastropoda. However, a statistically significant effect of the canal water inflow from the Ostrołęka Power Station was observed on the increased abundance of Gammaridae and the lower density of Chironomidae. Similar findings pertaining to changes in the abundance of the above-mentioned taxa at heated water sites as compared to reference sites were reported for the Lower Oder River [8]. A slight decrease in the number of Chironomidae larvae species connected with the increase in water temperature downstream of the canal was also observed in the studies previously carried out in the Narew River by Dusoge and Wiśniewski [14].

It should be noted that during the study period, the concentrations of nitrates and phosphates varied along the studied stretch of the Narew River. The results of CCA showed that the variability of macrozoobenthos composition at the sampling sites (1-6) is positively correlated with nitrates. A relatively strong correlation was determined between the concentration of nitrates and the abundance of Diptera and Oligochaeta. The Diptera order (in the present study represented mainly by the Chironomidae family) and Oligochaeta (represented by the Tubificidae family) feed on organisms, including algae, whose content in the water is determined by such factors as the concentration of nitrates [22,23]. Moreover, these taxa exhibit a relatively short life cycle compared to other macroinvertebrates. The concentration of biogenic elements in the water varies depending on such factors as the season or water convergence from the river basin. However, it does not directly relate to the working of the power station. An increase in the concentration of biogenic elements may be one of the factors that determine an increase in the number of Chironomidae and Tubificidae specimens. Other authors [22,23] also report on statistically significant correlations between the rising levels of nutrients (including nitrates) in the water and the abundance of Chironomidae [22] and Tubificidae [23]. The literature [29,30] also indicate that the riverine macroinvertebrate communities are characterised by seasonal dynamics. This is connected with the insect larvae transformation into adult forms and the inflow of biogenic substances from the river basin to the river, which changes throughout the year. This is confirmed by the CCA results, which reveal similarities between the samples collected upstream and downstream of the canal in autumn 2013 and 2014 and the differences in values of the parameters in samples collected in June 2015 and 2016. The biodiversity index H’ calculated based on the taxonomic composition and macroinvertebrate abundance did not reveal significant differences between the upstream and downstream river stretches. Similar results, confirming the lack of influence of heated water discharged from the Ostrołęka Power Station on macroinvertebrate communities in the Narew River were reported in the 1970s by Dusoge and Wiśniewski [14]. Lardicci et al. [31] also emphasize that thermal pollution did not significantly influence the abundance and variability of the benthic assemblages in the Gulf of Follonica (Western Mediterranean). Similarly, the results of the research carried out by Worthington et al. [32] in the Severn River (UK) showed that samples collected at sites located 2 km downstream of the canal outfall from the power station were generally similar to samples from the reference site in terms of the metrics. Also Coutler et al. [33] emphasize that the supply of heated waters from the power plant to the Ohio River (USA) causes slight changes in aquatic organisms only within 400 meters downstream of the power plant. In general, the input of waters heated by power stations into rivers may change the aquatic communities including the invasion of the fauna in Poland by new species [5,34]. Our study revealed the presence of two alien macroinvertebrate species (a bivalve – Sinanodonta woodiana (Lea, 1934) and a gammarid– Dikerogammarus haemobaphes (Eichwald, 1841)). Sinanodonta woodiana, originating from Eastern Asia from the basin of the Amur and the Yangtze rivers, was first found in Poland in the 1980s in canals and lakes near Konin heated by cooling waters from the Konin and Pątnów power stations [5,34]. The species occurs in heated waters. Up to now, S. woodiana has been noted in Poland at 19 sites [35]. The presence of this bivalve in the Narew River downstream of the canal is evidence of its further spreading in the Polish waters. Dikerogammarus haemobaphesis a species which comes from the Ponto-Caspian region. In Poland, it has been recorded since 1997. It occurs along almost the entire course of the Vistula River, in the lower Bug River, in the Nogat River, in the Vistula Lagoon and in the lower reaches of the Oder River. Its presence was also confirmed in some lakes of the Great Masurian Lakes complex and in the Narew River [36, 37]. This species can withstand a wide temperature range [38].

The spread of the non-indigenous invasive species could constitute a threat to biodiversity. Non-indigenous species may reduce or even eliminate native species through competition [39]. Therefore, our research is important in terms of the European Union Water Framework Directive [7] requirements.

5 Conclusions

1. The inflow of heated water from the Ostrołęka Power Station causes a considerable increase in water temperature and decrease in oxygen levels, compared to the reference sites, but it does not change the values of other chemical indices of the water.

2. The inflow of heated water from the Ostrołęka Power Station does not seem to affect the diversity of macroinvertebrates.

3. The heating of the water only led to a considerable increase in the abundance of Gammaridae represented by the shrimp Dikerogammarus haemobaphes and a decrease in Chironomidae.

4. Two invasive macroinvertebrate species: the shrimp Dikerogammarus haemobaphes and the bivalve Sinanodonta woodiana were found in the studied stretch of the Narew River near Ostrołęka.