Next Article in Journal
Simulation and Prediction of Snowmelt Runoff in the Tangwang River Basin Based on the NEX-GDDP-CMIP6 Climate Model
Previous Article in Journal
Leveraging Participatory Science for Tackling Water Supply Challenges in Water-Scarce Developing Regions
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Water
Volume 16
Issue 15
10.3390/w16152081
water-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Spatio-Temporal Variability of Water Quality in the Middle Danube—The Influence of Air Temperature and Discharge

by
Antoni Grzywna
1,
Jasna Grabić
2,
Monika Różańska-Boczula
3,* and
Milica Vranešević
2
1
Department of Environmental Engineering and Geodesy, Faculty of Production Engineering, University of Life Sciences in Lublin, 20-950 Lublin, Poland
2
Department of Water Management, Faculty of Agriculture, University of Novi Sad, 21000 Novi Sad, Serbia
3
Department of Applied Mathematics and Computer Science, Faculty of Production Engineering, University of Life Sciences in Lublin, 20-950 Lublin, Poland
*
Author to whom correspondence should be addressed.
Water 2024, 16(15), 2081; https://doi.org/10.3390/w16152081
Submission received: 1 July 2024 / Revised: 19 July 2024 / Accepted: 22 July 2024 / Published: 24 July 2024
(This article belongs to the Section Water Quality and Contamination)
Browse Figures
Versions Notes

Abstract

:
Large watercourses are characterised by significant spatial and temporal changes in water quality due to both natural and anthropogenic impacts. The paper analyses changes in the Danube’s spatial and seasonal water quality in its middle part at five monitoring stations for the period 2018–2022. Examined water quality parameters include oxygen saturation (Os), ammonium (N-NH4), pH, 5-day biochemical oxygen demand (BOD), nitrate nitrogen (N-NO3), orthophosphates (P-PO4), suspended solids (SS), electrical conductivity (EC), and water temperature (WT). Furthermore, the analysis took into account the influence of two additional factors: air temperature (AT) and discharge (D). Throughout the entire period under study, all parameters were characterised by low concentration values, which met the environmental objective of good ecological status. The use of multivariate statistical methods allowed for the identification of EC, N-NH4, WT, Os, SS, and N-NO3 as determining the greatest spatio-seasonal variability of water quality in a selected section of the Danube. Regression models determined WT, EC and nitrogen nitrate changes as depending on AT, flow, and exposure time. Knowing models not only gives a better understanding of the dynamics of changes in water quality in the stretch of the Danube under study but potentially allows the prediction of these parameters based on easily measurable environmental variables.

1. Introduction

Water is the most common chemical compound on the Earth’s surface. Despite this, over 2 billion people live in areas where there is a risk of limited access to drinking water. Limited access to water is most often due to limited resources or environmental pollution [1].
Rivers are dynamic by nature, eroding their banks, moving sediments, and occasionally flooding surrounding land [2]. However, human activities causing hydromorphological changes in reshaping riverbeds, constructing dams, and forming accumulations contribute to significant changes in these natural patterns and successions. Consequently, these actions lead to changes in erosion rates, sediment deposition, and loss of habitats and biodiversity. Besides hydromorphological changes, the major concern related to rivers concerns decreasing pollution and thus improving water quality [3,4]. For the last two decades of implementation of the Water Framework Directive [5], about 20% of the water bodies in the territory of EU countries have improved their ecological status. Interestingly, this improvement was not even for all types of water bodies, e.g., water quality in rivers and transitional waters achieved less improvement in comparison to lakes and coastal waters [6]. In 2023, sustainable development goals are reducing pollution, increasing water efficiency, and implementing integrated water resources management, including transboundary cooperation [1].
Pollution entering surface waters may be of a point nature, as in the case of the discharge of untreated sewage or industrial wastewater. In addition, pollution can also be emitted over larger areas, as in the case of agricultural pollutants, along highways, or due to aerial deposition. Identifying the sources of water pollution is one of the basic elements of water management in the basin [7,8]. For example, research conducted in Poland, Nigeria, and the USA confirmed that the most common sources of water pollution are leaking septic tanks [9,10,11]. In turn, research conducted in England found that the largest source of pollution is the discharge of untreated sewage [12]. Many researchers also emphasize the impact of land use, especially agriculture, on river water quality [13]. Polluted waters often contribute to environmental degradation through eutrophication. Furthermore, water with degraded water quality is unfit for human consumption and irrigation [14]. Water quality research started almost a century ago and nowadays is conducted all over the world. Research attempts to use a variety of approaches to assess water quality in rivers [15]. Nevertheless, identifying sources of water pollution still represents a challenging task. Thus, numerous statistical tools and mathematical models have been applied [16,17].
In general, large rivers located near settlements are most frequently prone to pollution. For example, four capitals are located on the banks of the Danube River—Vienna, Bratislava, Budapest, and Belgrade. The Danube is the second longest river in Europe and its length is 2880 km. It covers a catchment area of 795,000 km2 and discharges over 6700 m3/s of water into the Black Sea. Moreover, it flows through as many as 10 countries, which makes it difficult to run a proper economy. The complicated geopolitical situation became the basis for establishing the International Commission for the Protection of the Danube River [18]. To assess water quality, ICPDR oversees the Transnational Monitoring Network (TNMN). Through the TNMN, the parties committed to the agreement monitor water quality pollution and long-term trends in water quality and pollutant loads in the main rivers of the Danube basin. Due to its course, the river is a strategic passenger and freight transport route. Since the mid-20th century, there has been a period of intensive construction of artificial reservoirs and hydroelectric power plants. In particular, the construction of two large hydroelectric power plants of the Iron Gate, on the border between Serbia and Romania, in the 1970s, led to hydromorphological changes in the lower section of the river [19,20] and also in the upper section of the river. Anthropogenic activities have led to the intensification of erosion processes [21], and in general, dams on rivers most often contribute to reducing the occurrence of floods and droughts but induce accumulation of sediments [22].
The Danube River, as one of the major river in Europe [23], stands as a focal point for extensive and enduring scientific inquiry into its water quality. This comprehensive research spans a spectrum of disciplines, encompassing rigorous examinations of its physical dynamics, chemical composition, and biological integrity. Through long-term monitoring initiatives, these studies yield vital insights into trends in nutrient levels, contaminants, and overall ecosystem health [7,24,25,26]. The findings from these scientific endeavors play a critical role in guiding effective management strategies [27,28] and policy frameworks aimed at safeguarding and enhancing the Danube’s water quality, ensuring its sustainability for both environmental resilience and societal needs [18]. Research on hydrological processes in the aspect of climate change was also carried out in the Danube River catchment [29,30].
Multivariate statistical methods are used to assess water quality in rivers around the world. Achieving accuracy in analysis processes is possible by selecting appropriate statistical methods and using appropriate tools [31]. The most commonly used methods for spatiotemporal changes in water chemical parameters include principal component analysis (PCA) and hierarchical cluster analysis (HCA). These analyses are used to assess differences between sampling sites or study periods. Multivariate statistical methods can be used to identify factors that control water quality at a given location or time. The use of statistical analysis allows for synthetic analysis of measurement data, which helps in better management of the quality of water resources. Multivariate statistical methods were used to analyze water quality for the Shailmari River (Bangladesh) [32], Shiroud River (Iran) [33], Danube (Romania, Serbia) [8,34], Litani River (Lebanon) [35], and Bug River (Poland) [9].
With regard to the mentioned, the aim of the paper was: (1) to determine the spatiotemporal variability of the analysed quality parameters included in the Serbian Water Quality Index (SWQI) for the Middle Danube for the period 2018–2022, (2) to examine whether the water quality parameters meet the criterion of good chemical status, (3) to identify the location and sources of pollution, and (4) to identify the impact of air temperature and discharge on the water quality. It should be emphasised that multivariate statistical methods were used to determine spatial and temporal changes, which allowed for a comprehensive assessment of the variability of the studied phenomenon and thus filled the existing research gap for a researched section of the Danube.

2. Materials and Methods

2.1. Study Area

The study area included the reach of the Danube in its middle part flowing through Serbia, 1427–1116 river km. Continental climate is characteristic of the whole Middle Danube watershed, and there are two main features concerning relief.
The first is the Pannonian Plain’s south edges, whereas the Carpathian Mountains’ southern parts characterize the second. In the Pannonian part of the Danube, the flow is smooth and uniform, a typical pattern for a lowland river. In the second part, entering the mountain region, the flow becomes turbulent, passing through a gorge it has shaped—the Iron Gate [36]. However, building dams for hydropower production disturbed the natural flow pattern in the second half of the 20th century [37]. The water flow rate in the Danube at the entrance to Serbia is 2400 m3/s. However, 300 km downstream, the flow is already 5400 m3/s. In the analysed section, the following tributaries flow into the Danube: the Drava River (670 m3/s), Sava (1720 m3/s), and Tisa (780 m3/s). Concerning the relief and vegetation, the southern rim of the Pannonian plain is relatively uniform and plain mostly converted into arable land (up to 85%). This part borders at the east southern parts of the Carpathian Mountains overgrown by forests. On the west side of the area, along the Croatia-Serbia border, are protected areas. Namely, along the left Danube bank in Serbia stretches the Biosphere Reserve Bačko Podunavlje [38], whereas on the right side in Croatia is the Kopački Rit National Natural Park [39]. Therefore, this stretch is characterised by a rich and preserved biodiversity of wetland habitats and a low impact of anthropogenic influence, both of which are reflected in the good river water quality. Data from five state monitoring stations were used for the analysis (Figure 1).

2.2. Climate

The climate of Serbia can be described as moderate-continental with more or less pronounced local characteristics. Most parts of Serbia have moderately warm and wet climates, while at higher altitudes a moderately cold and wet climate is represented, and only at the highest mountains the climate is cold and wet. The average annual air temperature for 1961–2010 is 11 °C. The exception is the urban agglomeration Belgrade, where the average annual air temperature exceeds 13 °C due to the urban heat island effect. The range of absolute minimum and maximum air temperatures for Serbia is between −39.5 and 44.9 °C. The annual precipitation ranges from 540 to 820 mm on the Pannonian Plain and an average of 700–1000 mm in the Carpathians. The normal annual amount of precipitation for the whole country is 896 mm. The occurrence of snow cover is characteristic of the colder part of the year from November to March. The annual sums of the duration of sunshine range from 1500 to 2200 h per year. The most significant wind is the košava, a southeast wind blowing from the Carpathians, which can reach a speed of up to 48 m/s [40,41].

2.3. Data Sets

The paper analyses water quality parameters of the Middle Danube River in Serbia for the period 2018–22 [42]. All analysed water quality parameters are obtained from the Agency for Environmental Protection of the Republic of Serbia and are embedded in the Serbian Water Quality Index. These parameters include oxygen saturation (Os), ammonium (N-NH4), pH, 5-day biochemical oxygen demand (BOD), nitrate nitrogen (N-NO3), orthophosphates (P-PO4), suspended solids (SS), electrical conductivity (EC), and water temperature (WT) [42]. Individual water quality parameters were subject to classification assessment of good ecological status [5]. The paper aimed to determine the temporal and spatial variability of parameters and the possible impact of environmental factors on the level of pollution. The analysis considered the influence of two factors: air temperature (AT) and discharge (D) [40]. The discharge was assessed using water level and rating curves specific to each monitoring site. The tests for all mentioned parameters were carried out following the international methodology (Table 1).
Concerning the winter period, the key indicator is ice [27] and formation of a solid ice cover over the river could substantially affect water quality and related processes in the aquatic ecosystems. However, according to the literature, it seems that due to climate changes, within the investigated reach of the Danube River, ice did not greatly affect water quality. The reason is that mild winters and higher temperatures prevented the formation of solid ice over the river for extended periods [29].

2.4. Statistical Analyses

The data defining the parameters selected to describe water quality were verified to be consistent with a normal distribution. Since the normality of the distribution (Shapiro–Wilk test) and the homogeneity of the variance (Levene’s test) have not been confirmed in most cases, a non-parametric Kruskal–Wallis test and a post-hoc test of multiple comparisons were used to verify the significant variation in parameters between measuring stations [52]. Pearson correlation was used to determine the relationship between indicators describing water quality in the Danube River [53]. The significance of the correlation was the basis for the development of cause-and-effect regression models indicating changes in WT, EC, and N-NO3 depending on AT, D. The significance of the models was tested using the F test, and the corrected coefficient of determination (R2) was used to determine the degree of fit of the models to the experimental data. Regression relationship models were developed after applying a logarithmic transformation to the dependent variable. To compare 5 selected research stations (S1–S5), from the perspective of all indicators describing the water quality in the Danube River, a cluster analysis (HCA) was performed [54]. It was based on Ward’s algorithm, which minimizes the intra-group variance, and the Euclidean distance, which is a measure of the distance between objects. The results were presented in the form of a dendrogram. The well-known principal component analysis (PCA) was used to identify factors, locations, and seasons with high water pollution levels in the Danube River in Serbia. It is a multivariate statistical technique that reduces the number of factors analysed and identifies those responsible for the variability in the initial data set [55]. PCA allows easy identification of the principal components responsible for the greatest variance. This study used PCA analysis to identify parameter variability between stations and seasons (winter, spring, summer, and autumn). The validity of using PCA analysis was confirmed via the Bartlett sphericity test and the Kaiser–Meyer–Olkin coefficient [56]. Statistical analyses were performed in STATISTICA 13.3, assuming a significance level of α = 0.05.

3. Results and Discussion

Many researchers confirm that the chemical parameters of water quality vary in time and space [9,57]. Based on research conducted in 1992–2006, a satisfactory trend of improvement in the chemical condition of the Danube River in Serbia was found [58]. Individual parameters corresponded to category A1 of surface water quality intended for drinking water abstraction [59,60]. In other studies conducted in 2004–18, water quality was found to be stable and very good. It was noted that the detected sections of rivers where pollution occurs help improve regulations leading to improved water quality [61]. According to the study by Vranešević et al., it is confirmed that the water quality of the Danube in Serbia is stable and good for irrigation [62]. Slightly different research results were obtained using the ecosystem approach in 2011–17. The ecosystem approach showed that concentrations of chemical parameters exceeded the target values of the Water Framework Directive. In the study by Takić et al. [4], increased P-PO4 concentrations were found, and in the study by Radu et al. [63], an increase in diffuse nitrogen and phosphorus pollution was found. In the case of pollution coming from the city, the problem can be solved by building modern sewage treatment plants [26]. However, in the case of agricultural pollution, systemic solutions are required to reduce nitrate pollution [64]. Nutrient levels in agricultural areas are strongly influenced by local geological and hydrological conditions and high seasonal variability [25].
Hierarchical cluster analysis was used to compare five monitoring stations, taking into account a comprehensive set of water quality indicators in the Danube (Os, BOD, N-NH4, pH, N-NO3, P-PO4, SS, EC, WT, D, AT). The dendrogram (Figure 2) shows that cluster 1 was formed by three stations: Bezdan (S1), Bogojewo (S2), and Novi Sad (S3), with Bezdan and Bogojewo showing the greatest similarity from the point of view of the studied parameters. Cluster 2 was formed by the other two stations, i.e., Zemun (S4) and Smederewo (S5), showing similarity with each other. Similar observations were made based on post-hoc tests (Table 2, Figure 2). Table 2 presents the mean values and standard deviations of the analysed water quality parameters and based on the results of post-hoc tests, homogeneous groups were marked for the studied monitoring stations S1-S5. The first group usually included stations S1, S2, and S3, while the second group included S4 and S5. The main parameters differentiating in these groups were Os, pH, N-NO3, P-PO4, and SS, and, except P-PO4, these parameters had higher values for stations from group 1. However, EC, WT, and AT showed no spatial differences (Table 2).
The indicators characterizing oxygen conditions include oxygen saturation (Os) and biochemical oxygen demand (BOD). In the analyses stations on the Danube River, the average values of Os ranged from 87.0 to 104.1%, for stations S5 and S1, respectively (Table 2). The most desirable value (ideal value) is 100%. Surface waters meet the standards of good ecological status if this value does not exceed the limit below 70% or above 130% [5]. An excessive increase or decrease in oxygenation concentration is dangerous for aquatic life and contributes to phytoplankton blooms [65]. There is a very strong inverse statistical relationship between Os and BOD. The average BOD values ranged from 1.88 to 2.44 mg/L for stations S2 and S5, respectively. In the case of supplying people with drinking water, the maximum permissible value for water category A1 (water only requires filtration and disinfection) is 3 mg/L. As Table 3 shows, only 8% of water samples did not meet this standard and required physical and chemical treatment [66].
The indicators characterizing thermal conditions include water temperature. Water temperature is very strongly correlated with air temperature (Table 4) and for this reason, as we move towards the equator, the air and water temperatures increase. The average values of water temperature ranged from 13.3 to 14.6 °C, respectively for S1 and S5 (Table 2). The water quality did not meet the standards for good ecological status in 3% of the samples and for water supply of people in 1% of the samples (Table 3). An excessive increase in water temperature may contribute to the blooming of cyanobacteria [67].
The indicators characterizing salinity and acidification include electrolytic conductivity (EC), dissolved substances (SS), and pH. Mean EC values ranged from 391 to 417 μS/cm (Table 2). These values are relatively low because only exceeding the value of 850 μS/cm constitutes a threat to the functioning of aquatic life. When identifying a potential irrigation problem related to crop water availability, there is no restriction on the use of crop irrigation water [68]. The average SS values were characterized by high variability and amounted to 9 to 16 mg/L, for stations S4 and S1, respectively. In the case of supplying drinking water to people, only 10% of samples did not meet category A1 standards and required treatment. An excessive increase in SS concentration may also cause problems for the life of salmonid fish [69]. Average pH values were characterized by low variability and ranged from 7.98 to 8.19 for S5 and S1, respectively. Only a drop in the pH value below 6.5 is dangerous to the health and life of organisms [70].
The indicators characterizing biogenic conditions include ammonium (N-NH4), nitrate nitrogen (N-NO3), and orthophosphates (P-PO4). The average values of ammonium ranged from 0.04 to 0.17, respectively, for S1 and S4 (Table 2). The observed N-NH4 concentrations do not exceed the standards of good ecological status and for human consumption [5,59]. With high ammonia concentrations, aquatic life cannot develop properly and, on drastic occasions, lead to fish and invertebrate being killed [71,72]. Inland waters that are the habitat of fish in natural conditions should not contain more than 0.78 mg N-NH4/L [73]. The average values of nitrate nitrogen ranged from 0.82 to 1.44 mg N-NO3/L, respectively, for S5 and S1 (Table 2). The observed N-NO3 concentrations meet the limits for good ecological status and human water supply [5,59]. Pollution of surface and ground water by nitrates occurs mainly as a result of the use of mineral and organic fertilization. High concentrations of nitrates pose a health risk [74]. The average values of orthophosphates ranged from 0.030 to 0.052 mg P-PO4/L, respectively, for S1 and S5 (Table 2). The observed P-PO4 concentrations meet the limits for good ecological status and human water supply [5,59]. The presence of phosphates in nutrient-enriched waters has led to their recognition as a key to controlling the eutrophication process [75].
All analysed parameters were characterised by low concentration, which contributed to meeting the environmental goal of good ecological status. Only in summer were the permissible water temperatures exceeded (Table 3). The situation is slightly different regarding water suitability for supplying people. For the A1 category, the permissible standards for parameters such as BOD, SS, and WT were sometimes exceeded. However, only 10% of the water samples did not meet the standards of water for public supply, which resulted in their inclusion in category A2. Water that requires typical physical-chemical treatment is category A2 [59].
Among the analysed climatic factors, a very strong correlation was found between AT and WT (Table 4). According to Edinger’s concept of equilibrium temperature, the increase in WT with AT is a consequence of heat conduction. The relationship is linear at moderate air temperatures in the range of 0–20 °C [76]. Our research results confirm that the Danube’s water temperature is linearly related to the air temperature (Table 5). The obtained regression form indicates that with an increase in air temperature by 1 °C, the water temperature in the river increases on average by 0.86 °C. The obtained model is statistically significant (F test, p-value < 0.001), and due to the high value of the determination coefficient (0.89), it allows us to accurately predict the WT in the Danube based on known AT values.
Strong correlations were found between AT, EC, and N-NO3 (Table 4). Increased water evaporation occurs due to a lack of rainfall and high temperatures during drought. The occurrence of small water flows in the river may cause an increase in salt concentration and an increase in EC. The results of our analysis are completely different because we found a decrease in EC with an increase in AT. This was also reflected in the negative regression coefficient value in the prediction model for EC depending on AT and D1 (Table 5). The situation can be explained by the significant contribution of anthropogenic pollution during certain periods and with relatively high discharge. Excessive increase in salinity most often occurs in streams, which causes problems for human health [77]. EC is a very frequently used parameter for assessing water quality under various land uses [78]. In the case of N-NO3, the correlation coefficient indicates that there is a significant correlation with AT (Table 4). In the case of the Danube, the results suggest a high negative correlation between these two parameters, which is also reflected in the negative value of the regression coefficient for AT in the model developed for N-NO3 (Table 6). Based on this, we can conclude that N-NO3 values decrease with increasing air temperature. This is probably because high values of N-NO3 occur during spring since there is a leaching of surplus nitrogen compounds from surrounding arable land [79,80,81,82]. Later on, at the beginning of summer with increasing temperatures and light intensity increase, phytoplankton flourish, taking nutrients from the water column, thus leading to a decrease N-NO3 in the water, despite an AT increase. Our observations confirm previous research results [83,84,85].
Both air temperature and discharge are characterised by high seasonal variability and much smaller annual variability. The lowest air temperature occurred in the winter season and averaged 6.0 °C. The highest air temperature was recorded in the summer season and averaged 26.0 °C (Figure 3). Air temperatures in 2018–22 were about 3.0 °C higher than the long-term averages [40,42]. Higher air temperatures result from daily data from the moment water quality tests were used in the analyses. Meanwhile, the reports provide monthly average values. In the discharge case, the highest values were recorded in the spring season, when they ranged from 2950 to 3550 m3·s−1. High flows in the Danube were most often associated with spring snow melting increasing runoff within the Danube watershed. The lowest discharges were recorded in the autumn due to low rainfall and high evaporation during the summer. In the discharge case, there is a clear difference between individual years. The lowest discharges occurred in 2022 and the highest in 2018. The Danube basin is characterised by complex topography, high diversity of land use, and very high rainfall variability [86], thus certainly affecting flow rates monitored at certain monitoring stations and during different seasons.
Analysis PCA was used to determine the variability of parameters between stations and seasons (winter, spring, summer, and autumn). The result of Bartlett’s test of sphericity (p < 0.0001) performed for the correlation matrix (Table 5) and the Kaiser–Meyer–Olkin coefficient (0.64) provided a strong basis for the use of this multivariate technique [87]. It was performed for standardised values of indicators describing water quality in 5 research stations (S1, S2, S3, S4, S5) and 4 seasons: winter (W), spring (S), summer (SU), and autumn (A). Based on the correlation analysis, AT was removed due to its very strong correlation with WT and P-PO4 due to its very strong correlation with Os and pH. Significant PCA axes were selected with the Guttman–Kaiser criterion [88]. The first three components (PC1, PC2, PC3) had eigenvalues greater than 1 (Table 6), which allowed us to explain almost 89% of the total variability. The bolded loading values were considered the most important for the interpretation of the first three principal components PC1, PC2, and PC3, which explain 41.35%, 32.61%, and 14.69% of the total variability, respectively. The following parameters had the greatest impact on the PC1 component: N-NH4, Os, SS, and pH. The influence of N-NH4 and D was positive, while for the indicators Os, SS, and pH, it was negative. PC2 was most influenced by EC, WT, and N-NO3, and PC3 was influenced by BOD (negative correlation for WT and BOD).
Taking into account the eigenvectors obtained as a result of the PCA analysis allowed us to specify the explicit form of three new, mutually uncorrelated variables, which can be used in further research analysis on water quality:
PC1 = 0.22 · Os + 0.23 · (N − NH4) + 0.17 · pH + 0.19 · SS + 0.12 · D
PC2 = 0.23 · (N – NO3) + 0.32 · EC + 0.28 · WT
PC3 = 0.48 · BOD
Tracking their changes over time can help identify trends or seasonal variations in water quality. Dependencies (1) and (2) were visualised in the form of a biplot (Figure 4). Observations performed at measurement stations (S1–S5) about the seasons (W, S, SU, A) were also placed on the PC1-PC2 plane, which revealed several important connections. Regardless of the season, increased N-NH4 concentrations and significantly lower pH and SS concentrations were found at stations S4 and S5. However, the observed N-NH4 concentrations do not exceed the standards of good ecological status [5,59]. At high temperatures and high ammonia concentrations, aquatic life cannot develop properly, and on drastic occasions, lead to fish and invertebrate being killed [71,72]. Inland waters that are the habitat of fish in natural conditions should not contain more than 0.78 mg N-NH4/L [73].
In winter, at stations S1, S2, and S3, increased concentrations of EC and N-NO3 were found, and at the same time, lower WT values were found. In summer, the situation was the opposite for the same stations. Similar results were obtained in the Yellow River catchment, with a high correlation between EC and nitrogen ion pollution [78].

4. Conclusions

The study assesses water quality changes in the Middle Danube, using data from five measuring stations from 2018 to 2022. Using multivariate statistical methods, an attempt was made to identify the main parameters influencing the water quality within the investigated river section.
On this basis, spatial and seasonal similarities or differences between sampling sites during the monitored period were also assessed, identifying the basic indicators as EC, N-NH4, WT, Os, SS, and N-NO3. Moreover, it was shown that in winter at the stations Bezdan, Bogojevo, and Novi Sad, concentrations of EC and N-NO3 were increased, with lower WT values. In contrast, in summer, the situation was the opposite. Nevertheless, all analysed parameters were characterized by low concentration, confirming the good ecological status of the Middle Danube in Serbia. Concerning the discharge and the monitored water quality parameters, we confirmed that the lack of dependence is a consequence of mixing large amounts of water due to joining tributaries. In addition, developed fitting models for AT and WT and EC and N-NO3 provide a better understanding of the dynamics of changes in water quality within the investigated section.
The strength of the work is the use of multivariate statistical methods, which provided support in detecting polluted sections of the Middle Danube River in Serbia, and the results obtained filled the existing research gap in this area. The initial data set of 11 parameters was reduced to three significant, mutually uncorrelated variables, mainly determined by N-NO3, EC, and BOD5. These results can guide managers and authorities responsible for the sustainable management of water resources to identify the causes of water pollution. Using the recommended approach to water quality assessment, water quality can be monitored in detail and systematically, regardless of its purpose. This can minimize possible negative consequences or even avoid them altogether.

Author Contributions

Conceptualization, A.G.; methodology, A.G. and J.G.; software, M.R.-B.; validation, A.G., J.G., M.R.-B. and M.V.; formal analysis, M.R.-B.; investigation, A.G., J.G., M.R.-B. and M.V.; resources, A.G., J.G. and M.V.; data curation, A.G. and J.G.; writing—original draft preparation, A.G. and M.R.-B.; writing—review and editing, J.G. and M.V.; visualization, M.R.-B.; supervision, M.V.; project administration, A.G., J.G. and M.R.-B.; funding acquisition, A.G., J.G. and M.V. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by vouchers and the Provincial Secretariat for Higher Education and Scientific Research activity, Serbia: Determination of excess water in Vojvodina within the framework of climate change and extreme hydrometeorological phenomena (contract 142-451-3385/2023-01).

Data Availability Statement

The datasets used and analysed during the current study are available from the website: http://www.sepa.gov.rs/index.php?menu=5000&id=1304&akcija=showDocuments (accessed on 1 February 2024).

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. United Nations. Transforming Our World: The 2030 Agenda for Sustainable Development (A/RES/70/1); UN General Assembly: New York, NY, USA, 2015; Available online: https://Sdgs.Un.Org/2030agenda (accessed on 5 April 2024).
  2. Chittoor Viswanathan, V.; Schirmer, M. Water Quality Deterioration as a Driver for River Restoration: A Review of Case Studies from Asia, Europe and North America. Environ. Earth Sci. 2015, 74, 3145–3158. [Google Scholar] [CrossRef]
  3. Takić, L.; Mladenović-Ranisavljević, I.; Vuković, M.; Mladenović, I. Evaluation of the Ecochemical Status of the Danube in Serbia in Terms of Water Quality Parameters. Sci. World J. 2012, 2012, 1–6. [Google Scholar] [CrossRef] [PubMed]
  4. Takić, L.; Mladenović-Ranisavljević, I.; Vasović, D.; Đorđević, L. The Assessment of the Danube River Water Pollution in Serbia. Water Air Soil Pollut. 2017, 228, 1–9. [Google Scholar] [CrossRef]
  5. EC. Directive 2000/60/EC of 22 December 2000 Establishing a Framework for Community Action in the Field of Water Policy. Off. J. Eur. Communities 2000, 327, 1–72. [Google Scholar]
  6. Zacharias, I.; Liakou, P.; Biliani, I. A Review of the Status of Surface European Waters Twenty Years after WFD Introduction. Environ. Process. 2020, 7, 1023–1039. [Google Scholar] [CrossRef]
  7. Salvai, A.; Grabic, J.; Josimov-Dundjerski, J.; Zemunac, R.; Antonic, N.; Savic, R.; Blagojevic, B. Trend Analysis of Water Quality Parameters in the Middle Part of the Danube Flow in Serbia. Ecol. Chem. Eng. S 2022, 29, 51–63. [Google Scholar] [CrossRef]
  8. Grzywna, A.; Grabić, J.; Różańska-Boczula, M. Occurrences of Water Quality Assessment Using Improvised Water Quality Index at the Danube River, Serbia. Desalination Water Treat. 2023, 285, 67–77. [Google Scholar] [CrossRef]
  9. Grzywna, A.; Bronowicka-Mielniczuk, U.; Połeć, K. Spatio-Temporal Changes of Water Pollution, and Its Sources and Consequences in the Bug River, Poland. Desalination Water Treat. 2021, 243, 18–36. [Google Scholar] [CrossRef]
  10. Ugbebor, J.N.; Ntesat, U.B. Effects of Septic Tank Proximity to Boreholes on Groundwater Contamination at Igwuruta, Rivers State, Nigeria. Int. J. Eng. Sci. 2022, 11, 10–17. [Google Scholar]
  11. Herren, L.W.; Brewton, R.A.; Wilking, L.E.; Tarnowski, M.E.; Vogel, M.A.; Lapointe, B.E. Septic Systems Drive Nutrient Enrichment of Groundwaters and Eutrophication in the Urbanized Indian River Lagoon, Florida. Mar. Pollut. Bull. 2021, 172, 112928. [Google Scholar] [CrossRef]
  12. Rangel-Buitrago, N.; Galgani, F.; Neal, W.J. Addressing the Global Challenge of Coastal Sewage Pollution. Mar. Pollut. Bull. 2024, 201, 116232. [Google Scholar] [CrossRef]
  13. Lei, C.; Wagner, P.D.; Fohrer, N. Effects of Land Cover, Topography, and Soil on Stream Water Quality at Multiple Spatial and Seasonal Scales in a German Lowland Catchment. Ecol. Indic. 2021, 120, 106940. [Google Scholar] [CrossRef]
  14. Selvam, S.; Jesuraja, K.; Roy, P.D.; Venkatramanan, S.; Chung, S.Y.; Elzain, H.E.; Muthukumar, P.; Nath, A.V.; Karthik, R. Assessment of Groundwater from an Industrial Coastal Area of South India for Human Health Risk from Consumption and Irrigation Suitability. Environ. Res. 2021, 200, 111461. [Google Scholar] [CrossRef]
  15. Vranešević, M.; Zemunac, R.; Grabić, J.; Salvai, A. Hydrochemical Characteristics and Suitability Assessment of Groundwater Quality for Irrigation. Appl. Sci. 2024, 14, 615. [Google Scholar] [CrossRef]
  16. Grzywna, A.; Bronowicka-Mielniczuk, U. Spatial and Temporal Variability of Water Quality in the Bystrzyca River Basin, Poland. Water 2020, 12, 190. [Google Scholar] [CrossRef]
  17. Marcinkowski, P.; Piniewski, M.; Kardel, I.; Gielczewski, M.; Okruszko, T. Modelling of Discharge, Nitrate and Phosphate Loads from the Reda Catchment to the Puck Lagoon Using SWAT. Ann. Wars. Univ. Life Sci. SGGW. Land Reclam. 2013, 45, 2. [Google Scholar] [CrossRef]
  18. The International Commission for the Protection of the Danube River (ICPDR 2006). River Basin. 2006. Available online: http://www.Icpdr.Org/Main/Danube-Basin/River-Basin (accessed on 5 April 2024).
  19. Habersack, H.; Hein, T.; Stanica, A.; Liska, I.; Mair, R.; Jäger, E.; Hauer, C.; Bradley, C. Challenges of River Basin Management: Current Status of, and Prospects for, the River Danube from a River Engineering Perspective. Sci. Total Environ. 2016, 543, 828–845. [Google Scholar] [CrossRef] [PubMed]
  20. Hachemi, K.; Grecu, F.; Nougrara, Z.; El Ouarradi, A.; Amrouni, Y. Monitoring the Dynamics of the Danube Islands System, Using SAR Imagery (Orșova–Vedea Sector, 1992–2022). Environ. Earth Sci. 2023, 82, 590. [Google Scholar] [CrossRef]
  21. Ioana Toroimac, G.; Salit, F. La restauration hydromorphologique d’un hydrosystème sociale. Étude de cas du Danube en Roumanie. Cinq Cont. 2016, 6, 219–234. [Google Scholar]
  22. El Aoula, R.; Mhammdi, N.; Dezileau, L.; Mahe, G.; Kolker, A.S. Fluvial Sediment Transport Degradation after Dam Construction in North Africa. J. Afr. Earth Sci. 2021, 182, 104255. [Google Scholar] [CrossRef]
  23. Tockner, K.; Uehlinger, U.; Robinson, C.T. Rivers of Europe, Chapter 3; Academic Press: Cambridge, MA, USA, 2009. [Google Scholar]
  24. Basarin, B.; Lukić, T.; Pavić, D.; Wilby, R.L. Trends and multi-annual variability of water temperatures in the river Danube, Serbia. Hydrol. Process. 2016, 30, 3315–3329. [Google Scholar] [CrossRef]
  25. Chapman, D.V.; Bradley, C.; Gettel, G.M.; Hatvani, I.G.; Hein, T.; Kovács, J.; Liska, I.; Oliver, D.M.; Tanos, P.; Trásy, B. Developments in water quality monitoring and management in large river catchments using the Danube River as an example. Environ. Sci. Policy 2016, 64, 141–154. [Google Scholar] [CrossRef]
  26. Mănoiu, V.-M.; Crăciun, A.-I. Danube river water quality trends: A qualitative review based on the open access web of science database. Ecohydrol. Hydrobiol. 2021, 21, 613–628. [Google Scholar] [CrossRef]
  27. Hock, B.; Kovács, G. A Large International River: The Danube. Summary of Hydrological Conditions and Water Management Problems in the Danube Basin; IIASA: Laxenburg, Austria, 1987. [Google Scholar]
  28. van Gils, J. The Danube water quality model and its application in the Danube River Basin. In The Danube River Basin; Springer: Berlin/Heidelberg, Germany, 2015; pp. 61–83. [Google Scholar]
  29. Ionita, M.; Badaluta, C.A.; Scholz, P.; Chelcea, S. Vanishing river ice cover in the lower part of the Danube basin–signs of a changing climate. Sci. Rep. 2018, 8, 7948. [Google Scholar] [CrossRef] [PubMed]
  30. Stančíková, A. Thermal and ice regimes of the Danube River and its tributaries. In Hydrological Processes of the Danube River Basin: Perspectives from the Danubian Countries; Springer: Berlin/Heidelberg, Germany, 2010; pp. 259–291. [Google Scholar]
  31. Syeed, M.M.M.; Hossain, S.; Karim, R.; Uddin, M.F.; Hasan, M.; Khan, R.H. Surface water quality profiling using the water quality index, pollution index and statistical methods: A critical review. Environ. Sustain. Indic. 2023, 18, 100247. [Google Scholar] [CrossRef]
  32. Islam, M.M.; Lenz, O.K.; Azad, A.K.; Ara, M.H.; Rahman, M.; Hassan, N. Assessment of spatio-temporal variations in water quality of Shailmari River, Khulna (Bangladesh) using multivariate statistical techniques. J. Geosci. Environ. Prot. 2017, 5, 1–26. [Google Scholar] [CrossRef]
  33. Mohammadi, P.; Ebrahimi, K.; Vanaei, A.; Bazrafshan, J. Spatial and Temporal Assessment of Water Quality Changes in North, Northwest and Southwest Rivers of Iran, Involving Multivariate Statistical Techniques and GIS. Water Harvest. Res. 2022, 5, 102–120. [Google Scholar]
  34. Ismail, A.H.; Robescu, D. Application of multivariate statistical techniques in water quality assessment of Danube River, Romania. Environ. Eng. Manag. J 2019, 18, 719–726. [Google Scholar] [CrossRef]
  35. Hayek, A.; Andaloussi, S.A.; Tabaja, N.; Toufaily, J.; Garnie-Zarli, E.; Hamieh, T. Multivariate Spatial and Temporal Analysis to Study the Variation of Physico-Chemical Parameters in Litani River, Lebanon. Am. J. Anal. Chem. 2021, 12, 373–391. [Google Scholar] [CrossRef]
  36. Schiller, H.; Miklós, D.; Sass, J. The Danube River and Its Basin Physical Characteristics, Water Regime and Water Balance. In Hydrological Processes of the Danube River Basin; Brilly, M., Ed.; Springer: Dordrecht, The Netherlands, 2010; pp. 25–77. ISBN 978-90-481-3422-9. [Google Scholar]
  37. Stratimirovic, D.; Batas-Bjelic, I.; Djurdjevic, V.; Blesic, S. Changes in Long-Term Properties and Natural Cycles of the Danube River Level and Flow Induced by Damming. Phys. A Stat. Mech. Its Appl. 2021, 566, 125607. [Google Scholar] [CrossRef]
  38. UNESCO. Backo Podunavlje Biosphere Reserve, Serbia. 2017. Available online: https://En.Unesco.Org/Biosphere/Eu-Na/Backo-Podunavlje (accessed on 5 April 2024).
  39. FAO. Regulation on the Establishment of the Public Institution “Nature Park Kopački Rit”. 1999. Available online: https://www.Fao.Org/Faolex/Results/Details/En/c/LEX-FAOC121853/ (accessed on 25 April 2024).
  40. RHSS. Republic Hydrometeorological Service of Serbia—Climate. 2024. Available online: www.Hidmet.Gov.Rs/Ciril/Meteorologija/Klimatologija_srbije.Php (accessed on 18 April 2024).
  41. Milovanović, B.; Stanojević, G.; Radovanović, M. Climate of Serbia. In The Geography of Serbia; Manić, E., Nikitović, V., Djurović, P., Eds.; World Regional Geography Book Series; Springer International Publishing: Cham, Switzerland, 2022; pp. 57–68. ISBN 978-3-030-74700-8. [Google Scholar]
  42. Serbian Environmental Protection Agency—SEPA. Annual Reports—Results of Surface and Groundwater Quality Testing 2018–2022; Ministry of Environmental Protection: Belgrade, Serbia, 2023. Available online: www.Sepa.Gov.Rs (accessed on 25 April 2024). (In Serbian)
  43. SRPS ISO 5813:1994; Water Quality, Determination of Dissolved Oxygen, Iodometric Method. Available online: https://iss.rs/sr_Latn/project/show/iss:proj:12941 (accessed on 18 July 2024).
  44. SRPS ISO 7150-1:1992; Water Quality, Determination of Ammonium, Part 1: Manual Spectrometric Method. Available online: https://iss.rs/sr_Cyrl/project/show/iss:proj:13025 (accessed on 18 July 2024).
  45. SRPS H.Z1.111:1987; Testing of Industrial Waters, Measurement of pH, Potentiometric Method. Available online: https://iss.rs/sr_Cyrl/project/show/iss:proj:5942 (accessed on 18 July 2024).
  46. SRPS EN ISO 5815-1:2020; Water Quality—Determination of Biochemical Oxygen Demand after n Days (BODn)—Part 1: Dilution and Seeding Method with Allylthiourea Addition. Available online: https://iss.rs/en/project/show/iss:proj:64917 (accessed on 18 July 2024).
  47. Hach Method 8171. Nitrate, MR, Cadmium Reduction Method. Available online: https://images.hach.com/asset-get.download.jsa?code=55828 (accessed on 18 July 2024).
  48. US EPA Method 365.1; Revision 2.0: Determination of Phosphorus by Semi-Automated Colorimetry. Available online: https://www.epa.gov/sites/default/files/2015-08/documents/method_365-1_1993.pdf (accessed on 18 July 2024).
  49. APHA; AWWA; WEF. Standard Methods for the Examination of Water and Wastewater, 21st ed.; Total Solids Suspended, Method 2540 D, 2-55 a 2-59; American Public Health Association: New York, NY, USA, 2005. [Google Scholar]
  50. US EPA 120.1:1982; Method 120.1: Conductance (Specific Conductance, μmhos 25 °C) by Conductivity Meter. Available online: https://www.epa.gov/sites/default/files/2015-08/documents/method_120-1_1982.pdf (accessed on 18 July 2024).
  51. SRPS H.Z1.106:1970; Testing of Water, Measurement of Temperature. Available online: https://iss.rs/sr_Cyrl/project/show/iss:proj:5941 (accessed on 18 July 2024).
  52. Sprent, P.; Smeeton, N.C. Applied Nonparametric Statistical Methods; CRC Press: Boca Raton, FL, USA, 2016. [Google Scholar]
  53. Asuero, A.G.; Sayago, A.; González, A.G. The Correlation Coefficient: An Overview. Crit. Rev. Anal. Chem. 2006, 36, 41–59. [Google Scholar] [CrossRef]
  54. Rodriguez, M.Z.; Comin, C.H.; Casanova, D.; Bruno, O.M.; Amancio, D.R.; Costa, L.D.F.; Rodrigues, F.A. Clustering algorithms: A comparative approach. PLoS ONE 2019, 14, e0210236. [Google Scholar] [CrossRef] [PubMed]
  55. McDonald, R.P. The theoretical foundations of principal factor analysis, canonical factor analysis, and alpha factor analysis. Br. J. Math. Stat. Psychol. 1970, 23, 1–21. [Google Scholar] [CrossRef]
  56. Kaiser, H.F. The Application of Electronic Computers to Factor Analysis. Educ. Psychol. Meas. 1960, 20, 141–151. [Google Scholar] [CrossRef]
  57. Wang, L.; Han, X.; Zhang, Y.; Zhang, Q.; Wan, X.; Liang, T.; Song, H.; Bolan, N.; Shaheen, S.M.; White, J.R. Impacts of Land Uses on Spatio-Temporal Variations of Seasonal Water Quality in a Regulated River Basin, Huai River, China. Sci. Total Environ. 2023, 857, 159584. [Google Scholar] [CrossRef] [PubMed]
  58. Zivadinovic, I.; Ilijević, K.; Gržetić, I.; Popović, A.R. Long-Term Changes in the Eco-Chemical Status of the Danube River in the Region of Serbia. J. Serbian Chem. Soc. 2010, 75, 1125–1148. [Google Scholar] [CrossRef]
  59. EEA. Directive 2020/2184 of the European Parliament and of the Council of 16 December 2020 on the Quality of Water Intended for Human Consumption (Recast); OJ L 435; European Environment Agency: København, Denmark, 2020. [Google Scholar]
  60. EEA. Water Resources Across Europe—Confronting Water Stress: An Updated Assessment; European Environment Agency: København, Denmark, 2021; ISBN 978-92-9480-391-7. [Google Scholar]
  61. Radu, C.; Manoiu, V.M.; Kubiak-Wójcicka, K.; Avram, E.; Beteringhe, A.; Craciun, A.I. Romanian Danube River Hydrocarbon Pollution in 2011–2021. Water 2022, 14, 3156. [Google Scholar] [CrossRef]
  62. Vranešević, M.; Bezdan, A.; Blagojević, B.; Savić, R. Change of Irrigation Water Quality on the Section of the Danube through Serbia in the Period from 2010 to 2014; Serbian Society For Water Protection: Belgrade, Serbia, 2016; pp. 177–182. [Google Scholar]
  63. Radu, V.-M.; Ionescu, P.; Deak, G.; Diacu, E.; Ivanov, A.A.; Zamfir, S.; Marcus, M.-I. Overall Assessment of Surface Water Quality in the Lower Danube River. Environ. Monit. Assess. 2020, 192, 1–15. [Google Scholar] [CrossRef] [PubMed]
  64. EC. Council Directive 91/676/EEC of 12 December 1991 Concerning the Protection of Waters against Pollution Caused by Nitrates from Agricultural Sources; OJ L 375; European Union: Brussels, Belgium, 1991. [Google Scholar]
  65. Baxa, M.; Musil, M.; Kummel, M.; Hanzlík, P.; Tesařová, B.; Pechar, L. Dissolved oxygen deficits in a shallow eutrophic aquatic ecosystem (fishpond)–Sediment oxygen demand and water column respiration alternately drive the oxygen regime. Sci. Total Environ. 2021, 766, 142647. [Google Scholar] [CrossRef]
  66. Grzywna, A.; Sender, J. The Assessment of the Amount of Water Pollution and Its Suitability for Drinking of the Tyśmienica River Basin, Poland. Environ. Monit Assess 2021, 193, 315. [Google Scholar] [CrossRef]
  67. Zhang, M.; Yang, Z.; Shi, X.; Yu, Y. The synergistic effect of rising temperature and declining light boosts the dominance of bloom-forming cyanobacteria in spring. Harmful Algae 2022, 116, 102252. [Google Scholar] [CrossRef] [PubMed]
  68. Ayers, R.S.; Westcot, D.W. Water Quality for Agriculture; Food and Agriculture Organization of the United Nations: Rome, Italy, 1985; Volume 29, p. 174. [Google Scholar]
  69. Bash, J.; Berman, C.H.; Bolton, S. Effects of Turbidity and Suspended Solids on Salmonids; University of Washington Water Center: Seattle, WA, USA, 2001. [Google Scholar]
  70. Kuo, S.H.; Shen, C.J.; Shen, C.F.; Cheng, C.M. Role of pH value in clinically relevant diagnosis. Diagnostics 2020, 10, 107. [Google Scholar] [CrossRef] [PubMed]
  71. Shin, K.W.; Kim, S.-H.; Kim, J.-H.; Hwang, S.D.; Kang, J.-C. Toxic Effects of Ammonia Exposure on Growth Performance, Hematological Parameters, and Plasma Components in Rockfish, Sebastes Schlegelii, during Thermal Stress. Fish Aquat. Sci 2016, 19, 44. [Google Scholar] [CrossRef]
  72. Grabić, J.; Benka, P.; Bezdan, A.; Josimov-Dundjerski, J.; Salvai, A. Water Quality Management for Preserving Fish Populations within Hydro-System Danube-Tisa-Danube, Serbia. Carpathian J. Earth Environ. Sci. 2016, 11, 235–243. [Google Scholar]
  73. EC. Directive 2006/44/EC of the European Parliament and of the Council on the Quality Fresh Waters Needing Protection or Improvement in Order to Support Fish Life. Off. J. Eur. Union L 2006, 264, 20–31. [Google Scholar]
  74. Verma, A.; Sharma, A.; Kumar, R.; Sharma, P. Nitrate contamination in groundwater and associated health risk assessment for Indo-Gangetic Plain, India. Groundw. Sustain. Dev. 2023, 23, 100978. [Google Scholar] [CrossRef]
  75. Akinnawo, S.O. Eutrophication: Causes, consequences, physical, chemical and biological techniques for mitigation strategies. Environ. Chall. 2023, 12, 100733. [Google Scholar] [CrossRef]
  76. Mohseni, O.; Stefan, H.G. Stream Temperature/Air Temperature Relationship: A Physical Interpretation. J. Hydrol. 1999, 218, 128–141. [Google Scholar] [CrossRef]
  77. Jakhrani, S.H.; Soni, H.L.; Shar, N.Z. Analysis of Total Dissolved Solids and Electrical Conductivity in Different Water Supply Schemes of Taluka Chachro, District Tharparkar. QUEST Res. J. 2019, 17, 1–5. [Google Scholar]
  78. Zhang, S.; Xia, X.; Xin, Y.; Li, X.; Wang, J.; Yu, L.; Li, C.; McDowell, W.H.; Tan, Q.; Yang, Z. Electrical Conductivity as a Reliable Indicator for Assessing Land Use Effects on Stream N2O Concentration. J. Hydrol. 2023, 626, 130253. [Google Scholar] [CrossRef]
  79. Kirchmann, H.; Johnston, A.J.; Bergström, L.F. Possibilities for Reducing Nitrate Leaching from Agricultural Land. AMBIO A J. Hum. Environ. 2002, 31, 404–408. [Google Scholar] [CrossRef] [PubMed]
  80. Ulén, B.; Johansson, G.; Kyllmar, K.; Stjernman Forsberg, L.; Torstensson, G. Lagged Response of Nutrient Leaching to Reduced Surpluses at the Field and Catchment Scales. Hydrol. Process. 2015, 29, 3020–3037. [Google Scholar] [CrossRef]
  81. Wang, Z.-H.; Li, S.-X. Nitrate N Loss by Leaching and Surface Runoff in Agricultural Land: A Global Issue (a Review). Adv. Agron. 2019, 156, 159–217. [Google Scholar]
  82. Mladenović-Ranisavljević, I.; Vuković, M.; Stefanović, V.; Takić, L. Multicriteria Decision Analysis of Sites with Increased Nutrient Contents in Water. Water 2022, 14, 3810. [Google Scholar] [CrossRef]
  83. Paudel, S.R.; Choi, O.; Khanal, S.K.; Chandran, K.; Kim, S.; Lee, J.W. Effects of Temperature on Nitrous Oxide (N2O) Emission from Intensive Aquaculture System. Sci. Total Environ. 2015, 518, 16–23. [Google Scholar] [CrossRef] [PubMed]
  84. Pantelić, M.M.; Dolinaj, D.M.; Leščešen, I.I.; Savić, S.M.; Milošević, D.D. Water Quality of the Pannonian Basin Rivers the Danube, the Sava and the Tisa (Serbia) and Its Correlation with Air Temperature. Therm. Sci. 2015, 19, 477–485. [Google Scholar] [CrossRef]
  85. Savic, R.; Stajic, M.; Blagojević, B.; Bezdan, A.; Vranesevic, M.; Nikolić Jokanović, V.; Baumgertel, A.; Bubalo Kovačić, M.; Horvatinec, J.; Ondrasek, G. Nitrogen and Phosphorus Concentrations and Their Ratios as Indicators of Water Quality and Eutrophication of the Hydro-System Danube–Tisza–Danube. Agriculture 2022, 12, 935. [Google Scholar] [CrossRef]
  86. Santos, I.M.; Herrnegger, M.; Holzmann, H. Seasonal Discharge Forecasting for the Upper Danube. J. Hydrol. Reg. Stud. 2021, 37, 100905. [Google Scholar] [CrossRef]
  87. Stanisz, A. Przystępny Kurs Statystyki z Zastosowaniem STATISTICA PL Na Przykładach z Medycyny. T. 3. Analizy Wielowymiarowe [Accessible Statistics Course with the Use of STATISTICA PL on Medical Examples]; StatSoft: Kraków, Poland, 2007; Volume 3. [Google Scholar]
  88. Yeomans, K.A.; Golder, P.A. The Guttman-Kaiser Criterion as a Predictor of the Number of Common Factors. Statistician 1982, 31, 221. [Google Scholar] [CrossRef]
Water 16 02081 g001
Figure 1. Investigated reach of the Danube River with selected monitoring stations.
Figure 1. Investigated reach of the Danube River with selected monitoring stations.
Water 16 02081 g001
Water 16 02081 g002
Figure 2. Dendrogram of cluster analysis.
Figure 2. Dendrogram of cluster analysis.
Water 16 02081 g002
Water 16 02081 g003
Figure 3. Seasonal heatmap for average air temperatures (a) and discharge (b).
Figure 3. Seasonal heatmap for average air temperatures (a) and discharge (b).
Water 16 02081 g003
Water 16 02081 g004
Figure 4. Biplot results of the principal component analysis (blue sections indicate parameters describing water quality, and orange dots indicate research stations (S1–S5) in different seasons (winter, spring, summer, autumn)).
Figure 4. Biplot results of the principal component analysis (blue sections indicate parameters describing water quality, and orange dots indicate research stations (S1–S5) in different seasons (winter, spring, summer, autumn)).
Water 16 02081 g004
Table 1. Methods for water quality parameters examination.
Table 1. Methods for water quality parameters examination.
ParametersUnitsMethodsMeasurementLOQ *Measurement Precision
Oxygen saturation (Os)%SRPS ISO 5813:1994 and upon it UP 3.14/PC 12, SEV:1977; [43]Iodometric method, calculation21
Ammonium (N-NH4)mg/LSRPS ISO 7150-1: 1992; [44]Spectrophotometric method
(range: 0.01–1.0 mg/L)		0.020.01
pH-SRPS H.Z1.111: 1987; [45]Potentiometric method-0.01
BOD5 (BOD)mg/LSRPS EN ISO 5815-1:2020; [46]Dilution method
(5-day incubation)		0.500.01
Nitrate nitrogen (N-NO3)mg/LMethod 8171 Hach
UP 1.98/PC 12; [47]		Spectrophotometric method
(range: 0.02–4.5 mg/L)		0.020.01
Orthophosphates (P-PO4)mg/LUS EPA 365.1; [48]Semi-automated colorimetry (range: 0.02–2.50 mg/L)0.0100.001
Suspended solids (SS)mg/LAPHA AWWA & WEF, part 2540 D: 2005; [49]Gravimetric method41
Electrical conductivity (EC)μS/cmUS EPA 120.1: 1982; [50]Conductivity meter51
Water temperature (WT)°CSRPS H.Z1.106: 1970; [51]Temperature measurement in-situ-0.1
Note: * Limit of quantification.
Table 2. Mean values ± standard deviations of the tested parameters at stations S1–S5 of the Middle Danube *.
Table 2. Mean values ± standard deviations of the tested parameters at stations S1–S5 of the Middle Danube *.
ParameterUnitS1S2 S3S4S5
Os%104 ± 15 a101 ± 12 a98 ± 10 a90 ± 10.6 b87 ± 9 b
BODmg/L1.99 ± 0.90 a1.88 ± 0.82 a2.21 ± 0.59 ac 2.40 ± 0.77 bc2.44 ± 0.50 bc
N-NH4mg/L0.04 ± 0.02 a0.04 ± 0.03 a0.07 ± 0.04 c0.17 ± 0.07 b0.15 ± 0.06 b
pH-8.19 ± 0.22 a8.15 ± 0.22 a8.16 ± 0.19 a8.02 ± 0.21 b7.98 ± 0.19 b
N-NO3mg/L1.44 ± 0.59 a1.36 ± 0.58 a1.38 ± 0.56 a0.99 ± 0.37 b0.82 ± 0.24 b
P-PO4mg/L0.030 ± 0.011 a0.031 ± 0.011 a0.042 ± 0.021 a0.049 ± 0.021 b0.052 ± 0.023 b
SSmg/L16 ± 10 a14 ± 9 a11 ± 13 a9 ± 11 b10 ± 8 b
ECμS/cm417 ± 63 a408 ± 60 a407 ± 59 a393. ± 52 a391 ± 44 a
WT°C13.3 ± 7.8 a13.7 ± 7.9 a13.6 ± 7.9 a14.0 ± 7.5 a14.6 ± 7.6 a
Dm3/s2066 ± 745 a2414 ± 776 ad2623 ± 778 bd3157 ± 948.54 c4500 ± 1797 e
AT°C14.7 ± 8.5 a14.9 ± 8.5 a14.9 ± 8.4 a15.3 ± 8.3 a14.8 ± 8.6 a
Note: * Different letters in rows, next to the values, indicate statistically significant differences between stations S1–S5.
Table 3. Assessment of water quality in the Middle Danube.
Table 3. Assessment of water quality in the Middle Danube.
ParameterMinMaxGood Status [5]PercentileA1
[36]		Percentile
Os7412870–13010070–130100
BOD1.003.804.501003.0092
N-NH40.020.300.451000.50100
pH7.408.406.50–8.401006.50–8.50100
N-NO30.501.902.2010010.00100
P-PO40.0200.0900.1011000.400100
SS329301002590
EC3206408501001000100
WT5.025.224.09725.099
Table 4. Correlation coefficients between parameters.
Table 4. Correlation coefficients between parameters.
ParameterOsBODN-NH4pHN-NO3P-PO4SSECWTD
BOD−0.90 *
N-NH4−0.76 *0.32
pH0.88 *0.35−0.63 *
N-NO30.26−0.29−0.400.25
P-PO4−0.90 *−0.110.75 *−0.93 *−0.24
SS0.83 *0.00−0.74 *0.64 *0.13−0.75 *
EC−0.08−0.28−0.06−0.030.87 *0.10−0.29
WT0.130.18−0.160.07−0.75 *−0.140.34−0.93 *
D−0.400.300.67 *−0.25−0.480.23−0.45−0.270.00
AT0.120.14−0.170.05−0.74 *−0.120.32−0.91 *0.99 *−0.02
Note: * Statistical significance at 0.05.
Table 5. Cause and effect models of selected water quality parameters.
Table 5. Cause and effect models of selected water quality parameters.
ParameterModel R ¯ 2
WTWT = 0.185 + 0.863 · AT0.89
EClog(EC) = 6.281 − 0.013 · AT − 0.029 · D10.89
N-NO3log(N − NO3) = 0.745 − 0.002 · t − 0.035 · AT0.69
Notes: t—time in months; D1 = D/1000; R ¯ 2 —coefficient of determination adjusted by multivariate regression model.
Table 6. Matrix of factor loadings calculated based on parameters of water quality.
Table 6. Matrix of factor loadings calculated based on parameters of water quality.
ParametersPC1PC2PC3
Os−0.906−0.224−0.220
BOD0.114−0.438−0.793
N-NH40.9190.018−0.265
pH−0.790−0.221−0.503
N-NO3−0.4930.825−0.122
SS−0.836−0.3680.073
EC−0.1200.969−0.171
WT−0.037−0.9120.335
D0.677−0.240−0.398
Eigenvalues3.722.941.32
Variance [%]41.3532.6114.69
Cumulative variance [%]41.3573.9688.64
Note: Bold values indicate the most important parameters for the interpretation of PC1, PC2 and PC3.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Grzywna, A.; Grabić, J.; Różańska-Boczula, M.; Vranešević, M. Spatio-Temporal Variability of Water Quality in the Middle Danube—The Influence of Air Temperature and Discharge. Water 2024, 16, 2081. https://doi.org/10.3390/w16152081

AMA Style

Grzywna A, Grabić J, Różańska-Boczula M, Vranešević M. Spatio-Temporal Variability of Water Quality in the Middle Danube—The Influence of Air Temperature and Discharge. Water. 2024; 16(15):2081. https://doi.org/10.3390/w16152081

Chicago/Turabian Style

Grzywna, Antoni, Jasna Grabić, Monika Różańska-Boczula, and Milica Vranešević. 2024. "Spatio-Temporal Variability of Water Quality in the Middle Danube—The Influence of Air Temperature and Discharge" Water 16, no. 15: 2081. https://doi.org/10.3390/w16152081

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop