Toxicity studies on sediments near hydropower plants on the Ślęza and Bystrzyca rivers, Poland, to establish their potential for use for soil enrichment

The aim of this study is to analyze the toxicity of the sediments accumulated in the vicinity of hydropower plants (HPs) on the Ślęza and Bystrzyca rivers in Poland and the possibility of using these sediments for soil enrichment purposes. Thus far, there has been little comprehensive research related to the analysis of the impact of HPs on the properties of sediments. The analysis of the granulometric composition, physicochemical properties, heavy metals (HMs) content in sediments, and the growth of three plant species was carried out, including toxicity (HMs) and germination indices (plants). Most parameters were significant between the points upstream and downstream of the analyzed HPs. It has been shown that the most dangerous toxic factor is the high concentrations of Cu, Ni, and Zn in the sediments upstream of the HP on the Ślęza. In most cases, the HMs content was observed to decrease downstream of the HPs (e.g., Cu in Ślęza River: average of 13.44‐times), a result of changes in the particle size composition and accumulation of sediments at the site of the dam wall. Typically, the sediments tested stimulated growth in the plant species studied in comparison with the control groups (e.g., germination index for Sorghum saccharatum, Bystrzyca: 273.5% downstream of HPs). The C:N ratio increased downstream of the HPs by an average of 37.11% for the Ślęza River and 10.88% for the Bystrzyca River. The requirements for composting material were not met; however, the sediment could be used to enrich soils with an excessively wide C:N ratio.

dominate upstream, while downstream, materials with smaller particles predominate (e.g., sand and silt) (Bogen & Bønsnes, 2001). As the particles move away from the HPs, the particle size in the bottom sediments decreases, that is, stones are present in the immediate vicinity of the HPs, then gravel and sand, and further down, silt and clays (Bishkawakarma & Støle, 2010;L opez et al., 2020).
Previous studies have shown that, downstream of HPs, there is an increase in the concentration of heavy metals (especially Zn, Al, Co, Ti, and Fe) (Bai et al., 2009). This is due to changes in the saturation of bottom sediment minerals and suspended material (Klaver et al., 2007;Song et al., 2010). Moreover, because of the decomposition of organic matter present in sediments, the greenhouse gases released contribute to global warming (Agrawal & Sharma, 2012;Barros et al., 2011;Gagnon & van de Vate, 1997;Soininen et al., 2019).
The sediment composition is also modified by land use in the adjacent catchment area (Haddadchi & Hicks, 2019), especially when this is arable land (causing surface runoff of fertilizers and plant protection products rich in nutrients; Ferreira et al., 2020;Lee & Oh, 2018), as well as industrial sites (Hayzoun et al., 2014), landfills (Zinabu et al., 2019), and sewage treatment plants (sources of specific organic pollutants as well as heavy metals) (Loizeau et al., 2004;B. Wu et al., 2014). Water and sewage management in the catchment area (Wiatkowski, 2015) and potential waste discharge into rivers are also of primary concern (Jordão et al., 2002).
Various substances, including heavy metals, are present in the bottom sediments. They originate primarily from rock formations extruded from the Earth's core, but due to the by-products of industrial and agricultural activities, their quantities are constantly increasing (Amin et al., 2009;Chapman et al., 2005). At appropriate concentrations, they are necessary for the functioning of living organisms, occurring in various environments (e.g., they are a building component of cells) (Volland et al., 2014), however, in large quantities, they can be highly toxic, and affect living organisms in their vicinity (due to their long-term accumulation in tissues (Al-Reasi et al., 2007;Pokorny et al., 2015). This is because they do not biodegrade (Dra et al., 2019;Kasperek et al., 2013). For example, excess Cd disrupts nutrient absorption by plants (Jaishankar et al., 2014), while Pb causes kidney and brain damage (Z. Rahman & Singh, 2019). Because of the high toxicity of these substances, they need to be monitored in areas where their levels are high (Bhuyan et al., 2019;Deumlich et al., 2005;Manoj & Padhy, 2014), as well as finding ways to neutralize them (Cuske et al., 2017;Peng et al., 2009).
Bottom sediments can be utilized in farming as a soil substitute (Ockenden et al., 2014) or a natural fertilizer rich in nutrients necessary for plant growth (Dr ożdż et al., 2020). This nutrient density is because of the accumulation of organic and mineral substances in the sediments, for example, the decomposition of aquatic organisms (Tan et al., 2019) or the by-products of their metabolic processes (El-Radaideh et al., 2014).
The main objectives of this study are: (a) conducting toxicity studies of bottom sediments within the studied HPs on the Bystrzyca and Ślęza rivers (tributaries of the Odra River, southwestern Poland) and (b) determining the potential use of these sediments in agriculture. In order to achieve these objectives, the analysis of the grain size composition, physicochemical properties [pH, electrical conductivity (EC), C:N ratio], heavy metals content (copper, nickel, chromium, zinc, lead, and cadmium)  Studies like this are significant because of the issues of rational sediment management, which are vital on a global scale, including the very important aspect of the course of fluvial processes related to the movement of sediments in rivers and reservoirs. Moreover, it is providing part of acritical theme of international policies, and also to ensure the socioeconomic development of humanity and maintain environmental balance in water-related and water-dependent ecosystems (Arabatzis & Myronidis, 2011). Our study highlights the possibilities of sustainability in each of these aspects. Thus far, there has been little comprehensive research related to the analysis of the impact of HPs on the properties of bottom sediments; research has mostly been on the physicochemical aspects. We have used an interdisciplinary approach to also study the sedimentological and ecotoxicological aspects. A novelty in this study is the temporal and spatial study of the toxicity of bottom sediments to selected species of crops grown near HPs, together with the assessment of the possibility of their use for soil enrichment, which has not been done so far for this type of hydrotechnical structures.

| Sample preparation and analysis
Before analysis, the samples were stored at a temperature of approximately 20 C. After the samples were placed in a fume hood and air-dried, they were marked according to particle size distribution using the Bouyoucos areometric method as modified by Casagrande and Pr oszy nski, and dividing the samples into appropriate divisions using sieves of varying mesh size (Błażejczak et al., 2020;Waroszewski et al., 2019). They were also divided according to heavy metals (HMs) content (mineralization in aqua regia and determination by atomic absorption spectrophotometry) (He et al., 2016;Sarmani, 1989), EC (with a conductometer), and pH (potentiometric method in water).
Grain size classification was established based on the classification according to the Polish Soil Science Society of 2008 (PSSS, 2008). The following divisions with the assigned particle diameters were discovered: rock fragments >2 mm were boulders, stones, or pebbles; fine earth particles <2 mm were sand (fraction of 2-0.05 mm), silt (fraction of 0.05-0.002 mm), or clay (fraction of <0.002 mm). Depending on the percentage of non-organic matter present, specific groups and granulometric subgroups can be distinguished (Bieganowski et al., 2013;Kruczkowska et al., 2020).
Toxicity tests were performed on the seeds of three species of crops, that is, white mustard (SA), garden cress (LS), and bicolor sorghum (SS), using the Phytotoxkit™ acute toxicity microbiotest (compliant with ISO 18763). To observe the determinations, 10 seeds of each species were planted in the previously prepared sediment (with 100% humidity). Each sediment sample was placed in a transparent container, which made it possible to observe seed germination, as well as measuring the length of the roots and shoots after 3 days of growth (the samples were kept in complete darkness at room temperature). In addition, the reference soils were analyzed to determine the potential toxicity of the sediment samples within HPs. All experiments were performed in triplicate. A limitation of the test may be that the plant roots are separated from the sediment by filter paper, which may have some influence on the results; such factor may also be slightly different conditions of seed incubation in particular test dates (e.g., air temperature, humidity, and light intensity).

| Data analysis
The resulting data was further analyzed as follows: 1. Determination of basic statistics for the studied parameters in individual rivers (minimum, maximum, median, mean, and SD) along with plotting box plots; wherever possible, synthetic indicators were adopted to show the variability of the tested parameters  (Kilunga et al., 2017;Santos-Francés et al., 2017), contamination factor (C f ) (Solgi & Parmah, 2015), ecological risk index (E r ) (M. S. Rahman et al., 2014;Hakanson, 1980), classification for each sample in the LAWA and geochemical methods (Table A1 in Appendix A), or calculations (other parameters; Table A2); final classification in Table A3.
3. Determination of the toxicity index for the three tested plant species (SA, LS, and SS) according to the GI, calculated from the Formula (1): Where: L S is the root length increase in the tested sample (number of seeds in the sample n = 10; mm), G S is the percentage of germinated seeds in the sample, L 0 is the increase in root length in the blank sample (n = 3; mm), and G 0 is the percentage of germinated seeds in the blank sample (García-Lorenzo et al., 2014); GI < 90%, inhibitory effect; GI = 90%-110%, no effect/non-toxic; and GI > 110%, growth stimulating effect (Baran & Tarnawski, 2013;Czerniawska-Kusza & Kusza, 2011).
4. Determining the significance of the obtained results by comparing the results upstream and downstream of the HPs using the nonparametric two-sided Mann-Whitney U test (Wilcoxon rank test).
As the null hypothesis, it was assumed that the medians for the analyzed samples are identical, and the distribution of the dependent variables does not coincide with the normal distribution (for p < 0.05 or p < 0.01); in the test, U and z values are calculated using the Formulas 2 and 3: Where: n is the number of items in the samples and P ranks is the sum of ranks in the sample.
Where: σ U is the SD of U and x U is the mean of U.

Estimating the relationship between all examined parameters for
both rivers using the Spearman's rank correlation coefficient (for p < 0.05 [Leeb et al., 2015]; selection due to data distribution other than normal), principal component analysis (PCA) and hierarchical cluster analysis (HCA). As part of the PCA, the Kaiser-Meyer-Olkin (KNO) measure of sampling adequacy, the Bartlett's test of sphericity and communalities were determined, which allowed for further analysis.
The following software was used for the analyses: SPSS Statistics 26, ORIGIN Pro 2021b, STATISTICA 13, Microsoft Office 2013, and QGIS 2.8.4.
The procedure for assessing the suitability of bottom sediment for agricultural use is presented in Figure A1, whereas conditions for the accumulation of sediment in a natural river and after the construction of a HP-in Figure A2 (Appendix A).   The plants reacted differently to contaminants in the bottom sediments. The GI value of the sediments ranged from 1% to 780% for LS, 35% to 190% for SA, and 112% to 469% for SS.

| Germination index
LS was the most sensitive to pollution in the sediment, but at the same time it had the highest GI index. In this study, the lowest GI index in each of the analyzed variants (i.e., downstream, upstream of the HP, and at the sampling sites) was maintained for the SA species.
SS plants showed the lowest sensitivity to the potential phytotoxicity of the sediment samples. Bottom sediments from all components analyzed for this species showed a growth stimulating effect.
The value of the GI index varied depending on the location of the sample points on the river (downstream, upstream of the HP, and at reference sites) and between two selected objects (Ślęza, Bystrzyca).
Visible differences were also noted between the three plants selected for study, therefore the selection of appropriate species is a very important issue, as they should be able to survive the potentially toxic effects of the sediment and its variability over time (Madera-Parra et al., 2015).
In most cases, the sediment showed a growth stimulating effect  To account for seasonal changes, sediment toxicity analyses were carried out in two separate seasons, that is, spring and autumn.
It was observed that the sediment collected in spring had a more growth stimulating effect on plants than that collected in autumn. The GI index for sediment collected in spring was up to 469% for Bystrzyca and 780% for Ślęza. However, in autumn, it was lower, in the range of up to 315% for the sediments from Bystrzyca and 413% for the sediments from Ślęza. However, in two cases, a higher growth stimulating effect of the sediment collected in autumn than in spring was noted, (i.e., for LS from Bystrzyca and SS from Ślęza).
Sediment from the two rivers showed a plant growth-stimulating effect in most cases; however, there were also cases where the sediment had no effect or even an inhibitory effect on plant growth.
When placed in order, the registered number of toxic responses in plants, which is a measure of the sensitivity of the bioassays performed, was as follows: SA > SS > LS.

| Statistical significance test (Mann-Whitney U test)
The statistical analysis performed to compare the medians for groups of points upstream and downstream of the HPs in the Bystrzyca and Ślęza rivers showed statistical significance for most of the parameters tested for p < 0.01 (exceptions were: GI-SA for the Ślęza River, EC and C:N for the Bystrzyca River-significance for p < 0.05; GI-LS and GI-SS for the Ślęza River and pH, GI-SA, GI-LS and GI-SS for the Bystrzyca River-no statistical significance). This means that in most cases, HPs affect the analyzed parameters such as heavy metals, grain size composition, EC, and C:N ratio, as indicated by the calculated values of U and z. For the Ślęza River, the greatest impact was for Cu, Ni, D 50 , Zn, and EC, while for the Bystrzyca River, it was for Pb and D 50 (Table D1; Appendix D).  Tables E1-E6 and Figures E1 and E2 (Wuana & Okieimen, 2011), that is, immobilization of pollutants in the solid phase such as changing the properties of the sediments, namely their reaction and sorption capacity using various treatments such as

| Germination index
In Trojanowska's research on the use of Phytotoxkit tests to assess the risk posed by bottom sediments in a dam reservoir, in each of the analyzed cases, the highest phytotoxicity was observed for LS (Trojanowska, 2011).
The lowest GI index maintained for the SA species in this study (Figures 2 and 3) may be due to it being a species that shows a high F I G U R E 5 (a-c) Hierarchical cluster analysis, principal component analysis, and Spearman correlation matrix results for parameters tested in the sediments of the Bystrzyca River (* significant at p < 0.05) [Colour figure can be viewed at wileyonlinelibrary.com] sensitivity to a very wide range of chemicals (Kalčíková et al., 2014;Vaverková et al., 2020;. In this study, SS plants showed the lowest sensitivity to the potential phytotoxicity (Figures 2 and 3), while in other studies, SS was classified as the most sensitive species of the three studied (Baran & Tarnawski, 2013;Czerniawska-Kusza & Kusza, 2011;Mamindy-Pajany et al., 2011).
Despite the increased content of Cu and Ni observed in the sediments (Table C1), their toxic effects on the plants were not demonstrated, which is explained by the ability of trace elements to combine with organic matter. As a result, the chance of them being released from the sediment is much lower, making them less toxic (Baran et al., 2019). The higher GI index upstream of the HP for SA and LS may be due to the change in the distribution of components in the alluvia, for example, greater accumulation of nutrients necessary for plants upstream of a HP (Kabala et al., 2020;M. E. Rahman et al., 2020;Vukosav et al., 2014).
It was observed that the sediment collected in spring had a more growth stimulating effect on plants than that collected in autumn

| Statistical significance test (Mann-Whitney U test)
The reduction of heavy metal concentration downstream of HPs was demonstrated, among others, by X. Zhao et al. (2018) in the Yangtze River, China; Shim et al. (2015), in Geum, South Korea; Aradpour et al. (2020) in Salaban, Iran (this study: results- Figures B5-B10, Appendix B, statistical analysis- Table D1, Appendix D). Among other parameters, the pH was also examined, and no changes were observed after the Klingenberg Dam on the Main River in Germany (Hahn et al., 2018), while there was a pH increase for the Vaussaire Dam on the Rhue River in France (Frémion et al., 2016), which is a correlation identical to that for the Ślęza River, and opposite to the Bystrzyca River ( Figure B2, Appendix B). Regarding the granulometric composition, the results for the Klingenberg Dam in Germany (Hahn et al., 2018) and the Platanovrisi Dam on the Nestos River in Greece (Kamidis & Sylaios, 2017) coincide with the results ( Figure B1, Appendix B); more coarse-grained material (sand and clay) was recorded downstream of the dams, with silt upstream. This situation is due to the accumulation of large amounts of bottom sediments upstream of HPs, which are a vast storehouse of various substances (e.g., it has been estimated that 15% of the river phosphorus load is upstream of dams) (Maavara et al., 2015). In addition, the accumulation of fine-  Rolka and Wyszkowski (2021) showed that in the studied soils, there is a significant correlation between Zn and Cd (compliance with this study), while for Pb and Ni there is none, which is inversely proportional contrary to the reported results. The inversely proportional correlation between heavy metals and D50 is because, as the particle diameter of the sediment decreases, its sorption capacity and ability to accumulate various types of substances increases. Exemplary studies confirming this claim were conducted in Chile (Parra et al., 2014), Russia (Minkina et al., 2011), China (He et al., 2016), Australia (Strom et al., 2011), Italy (Bartoli et al., 2012), and Nigeria (Aladesanmi et al., 2016). The interaction of all parameters can be modified by various factors, such as anthropogenic pressures that modify the physicochemical properties of bottom sediments, the erosive action of rivers changing their granulometric composition, or chemical reactions occurring inside sediments, which affect the variability of the forms of various substances and their ability to migrate in the environment (Gabbud & Lane, 2016;Zhang et al., 2014).

| CONCLUSIONS
The conclusions of the analysis are as follows: 6. After passing through HPs, the granulometric composition of the sediment changes as follows-D 50 is higher, more coarse-grained formations (mainly sands) dominate, while upstream, fine-grained (clays and silts). Fine-grained materials have a greater tendency to absorb various substances, therefore, in most samples upstream of HPs, the concentration of pollutants is higher than downstream.
7. The C:N ratio increased downstream from HPs by an average of 37.11% for the Ślęza River and 10.88% for the Bystrzyca River.
The median values of C:N ranged from 13.7:1 to 18.6:1, which are typical values for soils in this climate zone (12:1). The requirements for composts were not met (the optimal ratio is from 20:1 to 30:1), however, the sedimentary material can be used to enrich soils with an excessively wide C:N ratio, for example, peat.
8. There was an anomalous relationship for the pH; in the case of the Ślęza River, there was an increase in pH (in spring by 14.0%, in autumn by 7.64%), and in the Bystrzyca River, a decrease (spring-9.66%, autumn-12.65%). In the case of sediments from the Ślęza River, the pH is appropriate for the cultivation of most plants (average: 7.31), while for sediments from the Bystrzyca River (average: The conducted research shows that the tested bottom sediments do not meet the compost conditions (due to the low C:N ratio), however, phytotoxic tests prove that they can be used for the cultivation of selected species of crop plants (with a wide range of ecological tolerance). The greatest factor limiting the agricultural use of the tested sediment, apart from the low C:N ratio, is the high content of Cu, Ni, and Zn in the sediment above the Ślęza HP, as well as the low pH of the sediments from the Bystrzyca River (Kirk et al., 2010), unsuitable for most of the cultivated plants (exception: pineapples; Taira et al., 2005). However, acidic sediment could find other uses, for example, in the reclamation of alkaline soils in industrial areas, within a cement plant. The topic of the properties of bottom sediments within HPs is still poorly researched and requires in-depth analyzes, especially from a practical and interdisciplinary point of view.

CONFLICT OF INTEREST
The author declares that there is no conflict.

DATA AVAILABILITY STATEMENT
The data that supports the findings of this study are available in the supplementary material of this article. Łukasz Gruss https://orcid.org/0000-0002-7548-9758