Assessment of Lake Water Quality in Central Serbia—Using Serbian and Canadian Water Quality Indices on the Example of the Garaši Reservoir

Abstract

The water quality in lakes and reservoirs is crucial for maintaining ecological balance and ensuring public health. This research focuses on the water quality evaluation of Garaši Reservoir in Serbia, a vital source of drinking water for surrounding communities. We systematically analyzed three profiles (A1, B1, and C1) at various depths ranging from 50 cm to 1500 cm between 2021 and 2023. The study employed the Serbian Water Quality Index (SWQI) and the Canadian Water Quality Index (CWQI) to evaluate the water quality. The findings revealed significant spatial and depth-dependent differences. Higher concentrations of Aluminum (Al), Mercury (Hg) and Manganese (Mn), influenced by the inflow from the Velika Bukulja River, resulted in reduced overall water quality and suitability for drinking water. Dissolved Oxygen levels decreased with depth, indicating thermal stratification and nearly anoxic conditions, which are harmful to aquatic life. Some shallow areas exhibited poor water quality for recreational use due to high pH and metal concentrations. The study underscores the necessity of continuous and comprehensive monitoring to identify pollution sources and implement mitigation measures. Such efforts are essential to protect biodiversity and ensure the sustainable management of water resources in lakes and reservoirs.

1. Introduction

Ensuring high water quality in lakes and reservoirs is essential for maintaining ecological balance, protecting biodiversity, and sustaining human activities. Water reservoirs play a crucial role in supplying drinking water, supporting aquatic life, and enabling recreational and economic activities. Quantitative assessments of the economic impact of water quality improvements underscore the importance of sustainable water management. For example, a World Bank analysis highlights that investments in water and sanitation can yield economic returns of up to 4:1, emphasizing the significant benefits of addressing water-related challenges [1]. Additionally, reports by UNESCO and other international agencies indicate that water pollution and inadequate water management contribute to substantial economic losses globally, amounting to billions of dollars annually. These findings reinforce the ecological and economic rationale for implementing effective water quality improvement measures [2].

However, water bodies are increasingly vulnerable to pollution due to anthropogenic influences, such as industrial and agricultural pollution, wastewater discharge, and natural processes, such as torrential floods and intensive soil erosion [3]. Monitoring water quality allows for early detection of contamination sources and the implementation of measures to mitigate their effects, ensuring the long-term sustainability of these vital resources [4,5,6,7,8,9,10,11,12,13]. Advanced techniques, such as single-molecule tools, offer significant potential for detecting contaminants with ultra-high sensitivity, particularly across diverse aquatic layers. Recent studies [14,15] indicated that these innovative methods can complement existing water quality techniques by enhancing detection precision and capturing subtle changes in water quality. Incorporating such technologies into monitoring practices could provide a more robust and comprehensive approach to assessing aquatic ecosystems, ensuring early detection of potential threats and informed decision-making.

Recent studies emphasize the significance of comprehensive water quality assessment in artificial lakes. Bhateria and Jain [16] reviewed various assessment methods, highlighting the necessity of regular monitoring to maintain ecological health, while Abd Aziz et al. [17] analyzed pollution sources and their impacts on artificial lakes. The use of the Water Quality Index (WQI) for monitoring lakes and reservoirs has been widely explored: King Talal Dam in Jordan [18]; lakes and reservoirs in Korea [19]; Siling Reservoir [20], reservoirs in the lower reaches of the Yellow River [21], Lake Poyang in China [22], La Vega Dam [23] and Leon L. Dam in Mexico [24]; Loktak Lake [25], Hebbal Lake [26], Chandlodia Lake in India [27]; dam reservoirs in Poland [28]. Kadeem et al. [29] demonstrated its effectiveness in assessing water quality and protecting aquatic life. Advancements in predictive modeling, combined with machine learning techniques for water quality evaluation, as discussed by He et al. [30], underscore the growing technological progress in this field. Research on Serbian reservoirs, such as the Đerdap Reservoir [31] and Lake Vlasina [4], highlights the importance of ongoing monitoring and intervention strategies to preserve water quality.

Serbia has approximately 150 reservoirs serving various purposes, including drinking water supply, energy production, and irrigation [32]. The Garaši Reservoir is a key source of drinking water for nearby communities, which highlights the need for systematic water quality monitoring to protect public health and prevent waterborne diseases. Given its ecological and socio-economic importance, assessing the water quality of the Garaši Reservoir is necessary to identify potential contamination risks and develop effective water management strategies.

Several studies have previously investigated the environmental status and water quality of the Garaši Reservoir, providing a foundation for this research. Sunjog et al. [33] assessed contamination risks by analyzing tissue metal and metalloid concentrations in aquatic organisms, emphasizing the need for continuous monitoring. Milošković and Simić [34] examined eutrophication and phytoplankton development, identifying potential threats to biodiversity and water clarity. Marković et al. [35] explored the spatial distribution of major and trace elements in artificial lakes, assessing the suitability of water for drinking and irrigation purposes. These prior investigations highlight the necessity of systematic and long-term monitoring programs to ensure the sustainability of the Garaši Reservoir and other artificial water bodies.

This study aims to evaluate the water quality of the Garaši Reservoir by identifying sources of contamination and assessing trends over time using standardized water quality indices: the Canadian Water Quality Index (CWQI) and the Serbian Water Quality Index (SWQI). This research evaluates water quality at various depths to determine its suitability for multiple uses, thereby contributing to the sustainable management and protection of the reservoir, ensuring its continued usability for drinking water, recreation, agriculture and ecological balance. Effective water resource management relies on continuous monitoring, which facilitates informed decision-making and early intervention in response to environmental threats.

The increasing environmental pressures on artificial lakes necessitate comprehensive water quality assessments that encompass all relevant physical, chemical, and biological parameters. By integrating previous findings with contemporary assessment methods, this study aims to provide valuable insights into the current state of the Garaši Reservoir and support the development of effective water management policies.

2. Materials and Methods

2.1. Study Area and Sampling

In 1963, the upper course of the Velika Bukulja River on the Bukulja Mountain (the Sava River basin) was dammed to provide water supply for the town of Aranđelovac and the surrounding villages in Central Serbia. The reservoir is located at an altitude of 377 m a.s.l. and has a water volume of approximately 1,400,000 m³. Covering an area of 45 ha, the lake is 1500 m long, up to 300 m wide, and has a maximum depth of up to 26 m [32]. The lake is situated on the border between the Pannonian Plain and the mountains of Central Serbia. Its shores are indented with three large inlets and numerous small bays and coves, surrounded by lush greenery. None of the tributaries of the Velika Bukulja River and Garaši Reservoir are permanent watercourses. The highest spot along the watershed line reaches 696 m a.s.l. The reservoir’s drainage basin features steep slopes, which contribute to rapid runoff during intense rainfall events. This increases the likelihood of extreme discharge events, such as the torrential flood of the upper Velika Bukulja on 10 July 1999 [36]. Such events can lead to flash flooding, significant sediment transport, and reservoir siltation, impacting both water quality and reservoir capacity.

A systematic analysis of water quality was conducted at three profiles of the Garaši Reservoir (Figure 1), with the following coordinates: Profile A1 (UTM coordinates 4,904,922 m N and 7,458,491 m E, zone 34N), Profile B1 (UTM coordinates 4,905,262 m N and 7,459,101 m E, zone 34N), and Profile C1 (UTM coordinates 4,904,375 m N and 7,458,589 m E, zone 34N) [37]. Water quality was analyzed at various depths: for Profile A1 at 50 cm, 200 cm, 350 cm, 500 cm, 600 cm, 800 cm, 1000 cm, 1200 cm, and 1500 cm; for Profile B1 at 50 cm, 200 cm, and 500 cm; and for Profile C1 at 50 cm, 200 cm, 500 cm, and 600 cm. Water sampling was conducted by the Serbian Environmental Protection Agency three times (once per year) in the spring (May 2021 and March 2023) and summer (July 2022). The water quality of the Velika Bukulja River in March 2023 was used to assess its impact on the water quality of the Garaši Reservoir. It is important to note that there are no available water quality data for the Velika Bukulja River for other years.

2.2. Water Quality Indices (WQI) Analysis

This study conducted a systematic assessment of reservoir water quality based on data from the Serbian Environmental Protection Agency (2021–2023). To achieve a comprehensive assessment, two widely recognized indices were applied: the Serbian Water Quality Index and the Canadian Water Quality Index, both of which provide a standardized method for evaluating water quality across various parameters.

Water Quality Indices (WQIs) serve as an effective tool for summarizing complex water quality data into a single numerical score, allowing for easier interpretation by both experts and the general public. These indices help identify potential threats to water use, including its suitability for drinking, irrigation, aquatic ecosystems and recreational activities [39].

The Serbian Water Quality Index (SWQI) is suitable for the assessment of organic and nutrient pollution. However, a disadvantage of this index is that it does not provide information about metal pollution. This was the reason for using the Canadian Water Quality Index (CWQI). Besides metal pollution, this index includes many more parameters than SWQI. On the other hand, CWQI is not suitable for evaluating nutrient pollution. A combined approach using these two indices provides more complete and relevant results and reduces the disadvantages of each single index.

Following the collection of water samples, the Serbian Environmental Protection Agency (SEPA), operating under the Ministry of Environmental Protection of the Republic of Serbia, implemented the following methods for analysis: SRPS H.Z1.111:1987 (for pH); US EPA 120.1:1982 (for Electrical Conductivity); SEV:1977 (for Oxygen Saturation); SRPS ISO 5815:1994 (for Biochemical Oxygen Demand); HACH Method 8155 (for Ammonium Ion); UP 1.27/PC 12, Chemiluminescence detector CLD (for Total Nitrogen); SRPS H.Z1.160:1987 (for Suspended Solids); HACH Method 8048 (for Orthophosphate); SRPS EN ISO 9308-1:2010 (for Total Coliforms); UP 1.88/PC 12 (for Turbidity); SRPS ISO 5813:1994 (for Dissolved Oxygen); ISO 6058:1984 (for Calcium); HACH Method 8051 (for Sulfate); SRPS ISO 9297: 1997 (for Chloride, Nitrate + Nitrite); EPA 6020A:2014 (for Arsenic, Cadmium, Chromium, Copper, Iron, Nickel, Zinc, Manganese); EPA Method 245.7 (for Mercury); EPA 6020 A: 2007 (for Lead) [37].

2.2.1. Serbian Water Quality Index (SWQI)

The Serbian Water Quality Index (SWQI) is developed by the Serbian Environmental Protection Agency (SEPA), including following parameters: Temperature (T), pH, Electrical Conductivity (EC), Oxygen Saturation (OS), Biochemical Oxygen Demand (BOD), Ammonium (NH₄⁺), Total Nitrogen (TN), Suspended Solids (SS), Orthophosphate (PO₄³⁻) and Total Coliforms (TC) [40]. SWQI is calculated using the following equation:

$$\mathrm{SWQI}=\sum q_i\times w_i,$$

(1)

where q_i is water quality of the ith parameter, and w_i is weight unit of the ith parameter [40] (

Table 1. Assessment of water quality by SWQI parameters.

T (°C)	pH		EC (µS/cm)	OS (%)		BOD (mg/L)	NH4+ (mg/L)	TN (mg/L)	SS (mg/L)	PO43− (mg/L)	TC Coli/100 mg	${\mathit{q}}_{\mathit{i}}\times {\mathit{w}}_{\mathit{i}}$
				93–109								18
				88–92	110–119							17
				85–87	120–129							16
				81–84	130–134	0–0.9						15
				78–80	135–139	1–1.9						14
				75–77	140–144	2–2.4						13
				72–74	145–154	2.5–2.9	0–0.09				0–249	12
				69–71	155–164	3–3.4	0.1–0.14				250–999	11
				66–68	165–179	3.5–3.9	0.15–0.19				1000–3999	10
	6.5–7.9			63–65	180+	4–4.4	0.2–0.24				4000–7.999	9
	6–6.4	8–8.4		59–62		4.5–4.9	0.25–0.29	0–0.49		0–0.029	8000–14,999	8
	5.8–5.9	8.5–8.7		55–58		5–5.4	0.3–0.39	0.5–1.49	0–9	0.03–0.059	15,000–24,999	7
	5.6–5.7	8.8–8.9	0–188	50–54		5.5–6.1	0.4–0.49	1.5–2.49	10–14	0.06–0.099	25,000–44,999	6
0–17.4	5.4–5.5	9–9.1	189–239	45–49		6.2–6.9	0.5–0.59	2.5–3.49	15–19	0.1–0.129	45,000–79,999	5
17.5–19.4	5.2–5.3	9.2–9.4	240–289	40–44		7–7.9	0.6–0.99	3.5–4.49	20–29	0.13–0.179	80,000–139,999	4
19.5–21.4	5–5.1	9.5–9.9	290–379	35–39		8–8.9	1–1.99	4.5–5.49	30–44	0.18–0.219	140,000–249,999	3
21.5–22.9	4.5–4.9	10–10.4	380–539	25–34		9–9.9	2–3.99	5.5–6.99	45–64	0.22–0.279	250,000–429,999	2
23–24.9	3.5–4.4	10.5–11.4	540–839	10–24		10–14.9	4–9.99	7–9.99	65–119	0.28–0.369	430,000–749,999	1
25+	0–3.4	11.5–14	840+	0–9		15+	10+	10+	120+	0.37	750,000+	0

Source of data: Scottish Development Department [41].

).

Based on the SWQI, water quality is categorized into five distinct classes: excellent (90–100), good (84–89), medium (72–83), bad (39–71), and very bad (0–38) [42]. These descriptive quality indicators provide a clear understanding of the water quality in various bodies of water.

SWQI is a valuable instrument for long-term monitoring of water quality trends, facilitating informed decisions on water resource management and protection. By offering a standardized and consistent method for assessing water quality, the SWQI aids in ensuring the sustainability and safety of water resources in Serbia.

SWQI is mainly applied in water quality assessment in Serbia: Danube River [43,44,45]; Sava River [43]; Timok River [46]; Tisa River [43,47]; Ibar River [48]; Gradac River [49]; Drina River, Velika Morava River, Zapadna Morava River, Južna Morava River [43];, Nišava [50], Great Bačka Canal [51], Lake Vlasina [4]. SWQI is also used in research on water quality in Montenegro: Morača River basin [52] and Lake Skadar [53].

2.2.2. Canadian Water Quality Index (CWQI)

The Canadian Water Quality Index (CWQI) was developed by the Canadian Council of Ministers and Environment. In this study following parameters were used: Turbidity, Dissolved Oxygen (DO), pH, Calcium (Ca), Sulfate (SO₄²⁻), Chloride (Cl⁻), Nitrate + Nitrite (NO₃⁻ + NO₂⁻), Aluminum (Al), Arsenic (As), Cadmium (Cd), Chromium (Cr), Copper (Cu), Iron (Fe), Mercury (Hg), Manganese (Mn), Nickel (Ni), Lead (Pb) and Zinc (Zn).

Table 2. CWQI parameters.

		Overall		Drinking		Aquatic		Recreation		Irrigation	Livestock
Variables	Units	Lower	Upper	Lower	Upper	Lower	Upper	Lower	Upper	Upper	Upper
Turbidity	NTU		1		1
DO	mg/L	9.5		9.5
pH		6.5	8.5	6.5	8.5	6.5	9	5	9
Ca	mg/L		1000								1000
SO42−	mg/L		500		500						1000
Cl−	mg/L		110		250					110
NO3−NO2−	mg/L		100								100
Al	mg/L		0.005				0.005			5	5
As	mg/L		0.005		0.025		0.005			0.1	0.025
Cd	mg/L		0.005		0.005					0.0051	0.08
Cr	mg/L		0.001		0.05		0.001			0.0049	0.05
Cu	mg/L		0.002		1		0.002			0.2	0.5
Fe	mg/L		0.3		0.3		0.3			5
Hg	µ/L		0.003		1		0.1				0.003
Mn	mg/L		0.05		0.05					0.2
Ni	mg/L		0.025		0.025					0.2	1
Pb	mg/L		0.001		0.01		0.001			0.02	0.05
Zn	mg/L		0.03		5		0.03			1	50

Source of data: Canadian Councils of Minsters and Environment [54].

presents limit values (objectives) for the parameters. Besides overall water quality, CWQI is used to assess water quality for specific purposes: drinking, aquatic life, recreation, irrigation and livestock [54].

CWQI is a dimensionless value ranging from 0 to 100 and classified into the following categories: excellent (95–100), good (80–94), fair (65–79), marginal (45–64) and poor (0–44). The Canadian Water Quality Index 1.0 Calculator is applied for CWQI calculation. CWQI is calculated by the following equation:

$$CWQI=100-\left({\displaystyle \frac{\sqrt{F_1^2+F_2^2+F_3^2}}{1.732}}\right),$$

(2)

where

• F₁ (Scope) is a ratio between the number of failed variables (variables which do not meet the objective) and the total number of variables;
• F₂ (Frequency) is a ratio between failed tests (measurements which do not meet the objective) and the total number of tests;
• F₃ (Amplitude) is an asymptotic function which scales the normalized sum of excursion (nse) from the objective; nse is a ratio between excursions (variables which are greater or lower than the objective) of individual tests and the total number of tests [54].

CWQI is widely applied around the world: the Mackenzie-Great bear sub-basin in Canada [55]; Zarqa River in Jordan [56]; Khoser and Tigris Rivers in Iraq [57]; Damodar River [58]; and Lake Sukhna in India [59]; Kelani River basin in Sri Lanka [60]; Ras El-Ain Ponds in Lebanon [61]; Lake Hawassa in Ethiopia [62]. CWQI is also used in various studies of water quality in Serbia: Sava River and Drina River [63]; Zapadna Morava River and Ibar River [64]; Lake Vlasina [4].

Finally, a statistical method—regression analysis—was conducted to identify the possible relationship between SWQI and CWQI values at the shared measurement depths across all three profiles, followed by a comparative analysis of the two indices.

3. Results

3.1. Serbian Water Quality Index (SWQI)

Analysis of the data for the Garaši Reservoir from 2021 to 2023 reveals notable differences in water quality parameters across various profiles and depths. Temperature consistently decreases with depth across all profiles and years, reflecting typical thermal stratification in lake systems. Surface Temperatures reached up to 25.5 °C in 2022 at Profile C1, while at greater depths, Temperatures dropped below 6 °C, notably at a depth of 1500 cm in Profile A1 (

Table 3. Depth-Based Temperature Analysis (°C).

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard Deviation
A1	50	23.2	24.8	8.8	18.93	7.195
A1	200	20.8	24.6	8.8	18.07	6.73
A1	350	16.6	24	7.9	16.17	6.58
A1	500	10.6	20.9	7.4	12.97	5.76
A1	600	6.2	16	7	9.73	4.44
A1	800	6.8	10	6.2	7.67	1.67
A1	1000	6.5	7.8	6	6.77	0.76
A1	1200	6.3	6.5	5.7	6.17	0.34
A1	1500	5.3	6	5.6	5.63	0.29
B1	50	21.5	25.3	9.9	18.9	6.55
B1	200	21.5	25	9	18.5	6.87
B1	500	12	19.4	7.7	13.03	4.83
C1	50	23.4	25.5	8.9	19.27	7.38
C1	200	21.3	25	8.8	18.37	6.93
C1	500	10.5	21.3	8.1	13.3	5.74
C1	600	10.5	16.4	6.7	11.2	3.99

). Temperatures at deeper parts (from 800 cm to 1500 cm) were much more stable than those at lower parts and less dependent on seasonal conditions.

The pH values exhibit notable fluctuations, especially in surface waters. In 2021, high pH readings were observed at shallower depths in Profile A1, reaching up to 9.27, suggesting intensive photosynthetic activity and potential eutrophication processes. At greater depths, pH values decline towards neutrality, with values around 7.35 at 1500 cm depth in Profile A1, reflecting reduced biological activity (

Table 4. Depth-Based pH Analysis.

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard Deviation
A1	50	9.1	8.48	8.38	8.65	0.32
A1	200	9.23	8.65	8.42	8.77	0.34
A1	350	9.27	8.67	8.3	8.75	0.4
A1	500	7.95	9	7.92	8.29	0.5
A1	600	7.71	9.5	7.89	8.37	0.8
A1	800	7.43	7.39	7.8	7.54	0.18
A1	1000	7.37	7.17	7.64	7.39	0.19
A1	1200	7.36	7.26	7.64	7.42	0.16
A1	1500	7.35	7.06	7.62	7.34	0.23
B1	50	8.79	8.38	8.52	8.56	0.17
B1	200	8.82	8.38	8.46	8.55	0.19
B1	500	7.7	8.5	8.2	8.13	0.33
C1	50	9.03	8.61	8.45	8.7	0.24
C1	200	9.12	8.64	8.16	8.64	0.4
C1	500	7.62	5.8	8.05	7.16	0.975
C1	600	7.39	8.69	7.95	8.01	0.53

). The pH values showed stability at greater depths at Profile A1 (from 800 cm to 1500 cm) and at Profile B1 at depths of 50 cm and 200 cm.

Electrical Conductivity remains relatively stable across all profiles and depths, ranging between 200 and 242 µS/cm, indicating consistent mineralization levels in the reservoir. However, slight increases in Electrical Conductivity were observed in 2023, particularly in Profiles A1 and C1, which could be attributed to changes in ionic composition due to external inputs or internal processes (

Table 5. Depth-Based Electrical Conductivity Analysis (µS/cm).

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard Deviation
A1	50	204	223	235	220.67	12.76
A1	200	201	227	235	221	14.51
A1	350	200	227	236	221	15.3
A1	500	213	218	239	223.33	11.26
A1	600	215	197	238	216.67	16.78
A1	800	215	200	237	217.33	15.195
A1	1000	214	199	237	216.67	15.63
A1	1200	216	197	240	217.67	17.59
A1	1500	220	211	242	224.33	13.02
B1	50	213	237	229	226.33	9.98
B1	200	213	237	229	226.33	9.98
B1	500	214	240	236	230	11.43
C1	50	202	241	233	225.33	16.82
C1	200	201	240	238	226.33	17.93
C1	500	217	231	237	228.33	8.38
C1	600	219	221	236	225.33	7.59

). The same values were measured at depths of 50 cm and 200 cm at Profile B1. The lowest deviations across the years were recorded at depths of 500 cm and 600 cm at Profile C1.

Oxygen Saturation levels demonstrate significant fluctuations with depth and over the years. In 2021 and 2022, Profiles A1 and C1 exhibited drastic declines in oxygen saturation at greater depths. For instance, in 2022, Oxygen Saturation in Profile A1 dropped to as low as 2% at depths of 1200 cm and 1500 cm, indicating near-anoxic conditions. Such low Oxygen levels at depth may result from limited mixing, decomposition of organic matter consuming Oxygen, or other biochemical processes. In contrast, 2023 showed substantial improvement in Oxygen Saturation across all profiles and depths. In Profile A1, even at 1500 cm depth, Oxygen Saturation improved to 75%, suggesting enhanced aeration and mixing of water layers or decreased Biological Oxygen Demand (

Table 6. Depth-Based Oxygen Saturation Analysis (%).

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard Deviation
A1	50	120	110	103	111	8.54
A1	200	133	114	101	116	16.09
A1	350	134	106	101	113.67	17.79
A1	500	79	158	94	110.33	41.96
A1	600	59	210	95	121.33	78.87
A1	800	41	41	88	56.67	27.135
A1	1000	32	6	80	39.33	37.54
A1	1200	24	2	79	35	39.66
A1	1500	13	2	75	30	39.36
B1	50	114	115	109	112.67	3.21
B1	200	114	114	105	111	5.2
B1	500	77	103	98	92.67	13.8
C1	50	123	103	105	110.33	11.015
C1	200	128	107	103	112.67	13.43
C1	500	65	99	101	88.33	20.23
C1	600	39	168	98	101.67	64.58

). This improvement could be indicative of effective water management practices or natural ecological recovery.

Ammonium concentrations display significant variations and are particularly elevated in the deeper waters of Profile A1 during 2022. For example, at a depth of 1500 cm, Ammonium levels reached 1.15 mg/L, considerably higher than surface values of 0.13 mg/L. Ammonium values were most stable at Profile B1 across the years. Elevated Ammonium levels at greater depths are often associated with the decomposition of organic matter under low Oxygen conditions, leading to the release of Ammonium from sediments [65]. In 2023, Ammonium concentrations decreased across all depths, indicating an improvement in water quality and possible restoration of aerobic conditions in deeper layers (

Table 7. Depth-Based Ammonium Analysis (mg/L).

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard deviation
A1	50	0.02	0.13	0.04	0.06	0.06
A1	200	0.02	0.13	0.06	0.07	0.06
A1	350	0.08	0.12	0.07	0.09	0.03
A1	500	0.1	0.2	0.05	0.12	0.08
A1	600	0.12	0.35	0.04	0.17	0.16
A1	800	0.15	0.7	0.1	0.32	0.33
A1	1000	0.2	1	0.16	0.45	0.47
A1	1200	0.34	0.95	0.14	0.48	0.42
A1	1500	0.53	1.15	0.18	0.62	0.49
B1	50	0.02	0.05	0.08	0.05	0.03
B1	200	0.08	0.03	0.09	0.07	0.03
B1	500	0.1	0.03	0.07	0.07	0.035
C1	50	0.2	0.2	0.08	0.16	0.07
C1	200	0.23	0.28	0.06	0.19	0.115
C1	500	0.22	0.35	0.08	0.22	0.135
C1	600	0.42	0.2	0.1	0.24	0.16

).

Orthophosphate concentrations remained relatively low throughout the profiles and years, generally ranging from 0.01 to 0.064 mg/L. However, slight increases were observed at certain depths and years, such as in 2021 at Profile C1 at a depth of 600 cm with a concentration of 0.064 mg/L (

Table 8. Depth-Based Orthophosphate Analysis (mg/L).

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard Deviation
A1	50	0.038	0.029	0.01	0.026	0.014
A1	200	0.045	0.03	0.011	0.029	0.017
A1	350	0.05	0.016	0.01	0.025	0.022
A1	500	0.038	0.035	0.01	0.028	0.015
A1	600	0.042	0.019	0.01	0.024	0.0165
A1	800	0.045	0.051	0.01	0.035	0.022
A1	1000	0.048	0.054	0.01	0.037	0.024
A1	1200	0.051	0.034	0.012	0.032	0.02
A1	1500	0.051	0.02	0.016	0.029	0.019
B1	50	0.045	0.054	0.016	0.038	0.02
B1	200	0.045	0.036	0.022	0.034	0.012
B1	500	0.029	0.042	0.01	0.027	0.016
C1	50	0.038	0.054	0.01	0.034	0.022
C1	200	0.04	0.035	0.013	0.029	0.014
C1	500	0.048	0.032	0.013	0.031	0.0175
C1	600	0.064	0.038	0.016	0.039	0.024

). Elevated Orthophosphate levels can contribute to eutrophication, promoting excessive algal growth, and may indicate localized inputs or internal loading of Phosphorus [66].

Total Nitrogen exhibited noticeable increases at greater depths, particularly in 2022 in Profile A1, where concentrations reached up to 2.4 mg/L at 1000 cm depth (

Table 9. Depth-Based Total Nitrogen Analysis (mg/L).

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard Deviation
A1	50	0.49	0.44	0.9	0.61	0.25
A1	200	0.47	0.45	1	0.64	0.32
A1	350	0.49	0.45	0.9	0.61	0.25
A1	500	0.97	0.6	1	0.86	0.22
A1	600	1.01	0.8	1	0.94	0.12
A1	800	0.95	2.2	1.1	1.42	0.68
A1	1000	0.95	2.4	1.1	1.48	0.8
A1	1200	1.04	2.1	1	1.38	0.62
A1	1500	1.17	2.1	1	1.42	0.59
B1	50	0.45	0.6	1.4	0.81	0.51
B1	200	0.44	0.5	1.6	0.85	0.65
B1	500	0.97	0.4	1.4	0.92	0.5
C1	50	0.42	0.6	1	0.67	0.3
C1	200	0.42	0.7	1	0.71	0.29
C1	500	0.94	0.7	0.9	0.85	0.13
C1	600	1.19	0.5	1	0.9	0.36

). Increased Nitrogen levels can be attributed to factors such as runoff from agricultural areas, decomposition of organic matter, or atmospheric deposition. High concentrations of Nitrogen compounds can contribute to nutrient enrichment and eutrophication of aquatic systems [67].

It is important to note that data for certain key parameters like Biochemical Oxygen Demand (BOD), Suspended Solids (SS), and Total Coliforms (TC) are missing for several depths and years. The absence of these data points hinders the ability to perform a fully comprehensive assessment of the water quality and ecological status of the reservoir. Incomplete monitoring makes it challenging to identify potential sources of pollution, assess the extent of organic loading, and evaluate the sanitary quality of the water.

Based on the data for the Serbian Water Quality Index (SWQI) in the Garaši Reservoir from 2021 to 2023, a comprehensive analysis reveals significant variations in water quality across different profiles, depths, and over the specified years. The SWQI values, which classify water quality into categories ranging from very bad (0–38) to excellent (90–100), serve as an essential tool for assessing the overall health of the reservoir’s ecosystem.

Table 10. SWQI values in the Garaši Reservoir (2021–2023).

Profile	Depth (cm)	SWQI 2021	SWQI 2022	SWQI 2023
A1	50	85	87	94
A1	200	82	86	94
A1	350	85	88	93
A1	500	88	71	95
A1	600	79	67	97
A1	800	71	61	94
A1	1000	67	53	88
A1	1200	62	56	88
A1	1500	59	55	85
B1	50	89	86	93
B1	200	86	85	92
B1	500	88	91	94
C1	50	82	83	92
C1	200	79	77	95
C1	500	77	83	95
C1	600	62	76	94

displays the SWQI values for various profiles and depths in the Garaši Reservoir, recorded over the period from 2021 to 2023.

In Profile A1, the SWQI values demonstrate a notable fluctuation with depth and over the years. In 2021, the SWQI at shallower depths such as 50 cm and 350 cm was high, with values of 85, indicating good water quality. However, as the depth increased, there was a noticeable decline in water quality; at 800 cm, the SWQI dropped to 71, and further decreased to 59 at 1500 cm, which falls into the bad class. This suggests that in 2021, the deeper regions of Profile A1 were experiencing poorer water quality, possibly due to factors like stratification, reduced Oxygen levels, or accumulation of pollutants in the deeper layers.

In 2022 a continuation of this trend was observed in Profile A1, with SWQI values starting high at shallower depths (87 at 50 cm) but decreasing more sharply with depth than in the previous year. Notably, at a depth of 1000 cm, the SWQI was 53, indicating deteriorated water quality. This decline could be attributed to increased contaminants or changes in environmental conditions affecting the reservoir during that year.

In contrast, 2023 exhibited a significant improvement in SWQI values across all depths in Profile A1. The SWQI values were remarkably high, ranging from 85 at a depth of 1500 cm to 97 at a depth of 600 cm. This represents a shift to excellent water quality throughout the profile, including the deeper sections that previously showed lower quality. Such improvement may be indicative of successful water management strategies, remediation efforts, or natural ecological recovery processes that enhance the water quality across all depths.

Figure 2 depicts the SWQI values for Profile A1 from 2021 to 2023 at different depths. The data reveal a fluctuating water quality trend in 2021 and 2022, with a sharp decline in deeper layers.

Profile B1 consistently showed high SWQI values over the three years, with minimal fluctuations between depths and years. In 2021 and 2022, the SWQI values ranged from 85 to 91, classifying the water as good to excellent. In 2023, the values slightly increased, with SWQI values between 92 and 94, firmly establishing the water quality as excellent. The stability in Profile B1 suggests a more resilient part of the reservoir, less affected by negative environmental factors, or perhaps an area benefiting from consistent positive influences such as better water circulation or lower exposure to pollutants.

Figure 3 illustrates the SWQI values for Profile B1 across the years 2021 to 2023. This profile exhibits consistently high water quality with minimal fluctuations between years and depths. The stability of SWQI values suggests a resilient and well-maintained part of the reservoir, benefiting from positive environmental factors.

Profile C1 showed a gradual improvement over the three-year period. In 2021, SWQI values ranged from 62 at a depth of 600 cm to 82 at a depth of 50 cm, classifying the water quality from bad to medium. The lower value at a depth of 600 cm indicated potential localized issues affecting deeper waters. In 2022, SWQI values improved, ranging from 76 to 83, moving the water quality predominantly into the medium class. By 2023, there was a significant enhancement, with SWQI values between 92 and 95 across all depths, categorizing the water quality as excellent. This upward trend indicates effective interventions or natural ameliorations that positively impacted water quality in Profile C1 over time.

Figure 4 presents the SWQI values for Profile C1 at various depths over the years 2021, 2022, and 2023. The chart illustrates a significant improvement in water quality, particularly evident in 2023, where all depths showed excellent SWQI values around 90.

Analyzing the data collectively, there is a clear pattern of overall improvement in water quality in the Garaši Reservoir from 2021 to 2023. The initial years showed variability and areas of concern, particularly in the deeper waters of Profiles A1 and C1. The lower SWQI values at greater depths could be attributed to factors such as thermal stratification leading to reduced Dissolved Oxygen levels, accumulation of organic matter, or limited mixing of water layers, which can affect biological and chemical processes adversely. The remarkable improvement in 2023 across all profiles and depths indicates the successful implementation of measures aimed at enhancing water quality.

Furthermore, the data underscore the importance of continuous monitoring and analysis of water quality across different depths and locations within a reservoir. The significant variation with depth, especially in earlier years, highlights the complexity of aquatic ecosystems where surface measurements alone may not provide a complete picture of water quality. Deeper regions can harbor different conditions that may affect the overall health of the reservoir, such as anoxic zones or accumulation of harmful substances.

The SWQI values from 2021 to 2023 for the Garaši Reservoir indicate a positive trajectory in water quality, progressing from areas of bad and medium quality towards predominantly excellent conditions across all profiles and depths. This improvement could reflect the effectiveness of environmental management practices and the importance of sustained efforts in monitoring and protecting water resources. However, continuous monitoring is necessary to determine whether this will be a long-term trend or merely a short-term fluctuation. Ongoing assessment using the SWQI allows for early identification of potential issues and facilitates informed decision-making to ensure the long-term sustainability and ecological integrity of the reservoir. Regular monitoring and adaptive management will be essential to maintain the high water quality achieved and to address any emerging challenges that may arise due to environmental changes or human activities.

The analysis of Velika Bukulja River in 2023 was conducted at a sampling depth of 30 cm, where the SWQI value was recorded as 92, indicating excellent water quality. The measured parameters included a Temperature of 6.3 °C, a pH of 7.93, and an Electrical Conductivity of 119 µS/cm. Oxygen Saturation was 90%, while the Biochemical Oxygen Demand (BOD) was 3 mg/L. The concentration of Suspended Solids (SS) was 4 mg/L, Total Nitrogen was measured at 1 mg/L, and Orthophosphates at 0.016 mg/L. Ammonium levels were recorded at 0.07 mg/L.

3.2. Canadian Water Quality Index (CWQI)

3.2.1. Overall Water Quality

The results of CWQI at Profile A1 ranged from marginal (53) to good (85). The lowest CWQI value was calculated for a depth of 50 cm, while the highest value was at a depth of 800 cm. Generally, overall CWQI was also marginal at the various depths (600 cm, 1000 cm, 1200 cm and 1500 cm), while fair CWQI was recorded at depths of 200 cm, 350 cm and 1000 cm (Figure 5,

Table 11. CWQI values in the Garaši Reservoir (2021–2023).

Profile	Depth (cm)	CWQI 2021	CWQI 2022	CWQI 2023
A1	50	50	57	60
A1	200	84	67	100
A1	350	84	67	59
A1	500	55	84	100
A1	600	83	67	100
A1	800	81	81	100
A1	1000	79	57	100
A1	1200	76	52	100
A1	1500	67	52	84
B1	50	49	57	58
B1	200	84	84	100
B1	500	84	84	60
C1	50	47	50	56
C1	200	84	67	100
C1	500	83	67	100
C1	600	81	66	100

).

These results are attributed to increased Hg concentrations, with the highest nse at depths of 50 cm and 350 cm, increased Al with the highest nse at a depth of 500 cm, increased pH with the highest nse for 200 cm and decreased Dissolved Oxygen with the highest nse at all depths from 600 cm to 1500 cm. The lowest Dissolved Oxygen concentrations were recorded in 2022, at depths of 1000 cm, 1200 cm and 1500 cm. Increased concentrations of the following parameters also impaired overall CWQI: Cu at a depth of 50 cm, pH at a depth of 350 cm, Mn, pH and Hg at a depth of 500 cm, and turbidity at a depth of 600 cm, as well as the decreased concentration of Dissolved Oxygen at a depth of 500 cm (

Table 12. Depth-Based Dissolved Oxygen Analyses (mg/L).

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard Deviation
A1	50	10.18	9.02	12.02	10.41	1.235
A1	200	11.79	9.37	11.78	10.98	1.14
A1	350	12.94	8.83	11.6	11.12	1.71
A1	500	8.8	13.98	11.5	11.34	2.11
A1	600	7.27	20.6	11.5	13.12	5.56
A1	800	5	4.6	10.93	6.84	2.89
A1	1000	3.95	0.81	9.95	4.9	3.79
A1	1200	3	0.5	9.9	4.47	3.975
A1	1500	1.59	0.5	9.34	3.81	3.935
B1	50	9.96	9.32	11.69	10.32	0.99
B1	200	9.98	9.3	11.33	10.2	0.84
B1	500	8.24	9.36	11.11	9.57	1.18
C1	50	10.36	8.36	11.48	10.07	1.29
C1	200	11.22	8.76	11.83	10.6	1.33
C1	500	7.26	8.72	11.63	9.2	1.82
C1	600	4.54	16.44	11.37	10.78	4.88

).

CWQI values at Profile B1 were marginal (52) at a depth of 50 cm and (64) at a depth of 500 cm and fair (76) at a depth of 200 cm (Figure 5). These results are attributed to the highest nse of Hg (at depths of 50 cm and 500 cm) and pH (at a depth of 200 cm). Increased Cu and Al concentrations were recorded at 50 cm, while slightly decreased Dissolved Oxygen value was at the depth of 500 cm in 2021. However, compared with other profiles, Dissolved Oxygen values were most stable at Profile B1 at depths of 50 cm and 200 cm across the years.

CWQI values at profile C1 were marginal (49) at a depth of 50 cm and (63) at a depth of 600 cm, while fair (74) CWQI was calculated at depths of 200 cm and 500 cm. Decreased Dissolved Oxygen at depths of 500 cm and 600 cm, increased pH at a depth of 200 cm and increased Al at a depth of 50 cm were the parameters with the highest nse. Increased Cu and Hg concentrations at a depth of 50 cm, and increased pH and turbidity at a depth of 600 cm also contributed to the impairment of CWQI (Figure 5).

3.2.2. Drinking Water Quality

The results of CWQI ranged from marginal (59) at a depth of 600 cm to excellent (100) at depths of 800 cm, 1000 cm, 1200 cm and 1500 cm. CWQI was fair (77) in 200 cm, good (90) in 50 cm, (89) in 500 cm and (86) in 350 cm (Figure 6). Marginal CWQI was caused by the increased turbidity (

Table 13. Depth-Based Turbidity Analysis (NTU).

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard Deviation
A1	50	0.19	0.22	0.33	0.25	0.06
A1	200	0.23	0.23	0.33	0.26	0.05
A1	350	0.33	0.24	0.47	0.35	0.09
A1	500	0.24	0.24	0.39	0.29	0.07
A1	600	0.21	1.13	0.33	0.56	0.41
A1	800	0.15	0.49	0.29	0.31	0.14
A1	1000	0.13	0.41	0.23	0.26	0.12
A1	1200	0.2	0.45	0.23	0.29	0.11
A1	1500	0.2	0.39	0.2	0.26	0.09
B1	50	0.25	0.21	0.27	0.24	0.03
B1	200	0.26	0.22	0.41	0.3	0.08
B1	500	0.41	0.31	0.32	0.35	0.045
C1	50	0.23	0.2	0.41	0.28	0.09
C1	200	0.3	0.24	0.39	0.31	0.06
C1	500	0.29	0.31	0.42	0.34	0.06
C1	600	0.24	1.89	0.4	0.84	0.74

), which had the highest nse, while increased pH with the highest nse caused fair CWQI. Increased pH also contributed to CWQI lowering at depths of 350 cm, 500 cm and 600 cm. Mn was the variable with the highest nse at depths of 50 cm, 350 cm and 500 cm (

Table 14. Depth-Based Mn Analysis (mg/L).

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard Deviation
A1	50	0.088	0.035	0.138	0.087	0.042
A1	350			0.141	0.141	0
A1	500	0.121			0.121	0
B1	50	0.243	0.019	0.14	0.034	0.0915
B1	500				0.139	0
C1	50	0.522	0.02	0.141	0.228	0.214

).

CWQI values at profile B1 were good (80) at a depth of 200 cm, (88) at a depth 50 cm and (93) at a depth of 500 cm (Figure 6). Increased Mn values had the highest nse at depths of 50 cm and 500 cm (Table 14), while increased pH was the variable with highest nse at a depth of 200 cm and also contributed to the CWQI decline at a depth of 50 cm.

CWQI values at profile C1 were marginal (59) at a depth of 600 cm, fair (77) at a depth of 200 cm, good (80) at a depth of 500 cm and (84) at a depth of 50 cm (Figure 6). Marginal CWQI was caused by the increased turbidity (Table 13) with the highest nse; increased pH with the highest nse caused fair CWQI; decreased pH with the highest nse caused CWQI decline at a depth of 500 cm, while increased Mn with the highest nse caused CWQI decline at a depth of 50 cm (Table 14). Increased pH also contributed to the CWQI decline at the depths of 50 cm and 600 cm.

3.2.3. Water Quality for Aquatic Life

The results of CWQI at profile A1 ranged from poor (39) to fair (68). The lowest CWQI values were calculated for the depths of 200 cm, 600 cm and 1500 cm, while the highest CWQI was recorded at a depth of 350 cm. Poor CWQI was also calculated at depth of 1200 cm, while the marginal CWQI was recorded at depths of 50 cm, 500 cm, 800 cm and 1000 cm (Figure 7).

These results were caused by the highest nse of the following parameters: increased Al concentrations at the depths 50 cm, 350 cm and 500 cm (

Table 15. Depth-Based Al analysis (mg/L).

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard Deviation
A1	50	0.223	0.082	0.017	0.107	0.086
A1	350			0.029	0.029	0
A1	500	0.27			0.27	0
B1	50	0.169	0.072	0.02	0.087	0.062
B1	500			0.028	0.028	0
C1	50	0.218	0.182	0.061	0.154	0.067

), increased pH values at depth of 200 cm and decreased Dissolved Oxygen at all depths from 600 cm to 1500 cm. Increased concentrations of the following parameters also contributed to CWQI decline: Cu at the depths of 50 cm and 350 cm (

Table 16. Depth-Based Cu Analysis (mg/L).

Profile	Depth (cm)	2021 May	2022 July	2023 March	Average	Standard Deviation
A1	50	0.0029	0.0027	0.0071	0.0042	0.002
A1	350			0.067	0.067	0
A1	500	0.02			0.02	0
B1	50	0.0022	0.0044	0.0036	0.0034	0.0009
B1	500			0.0037	0.0037	0
C1	50	0.0032	0.0026	0.0113	0.0057	0.004

), pH at the depths of 350 cm and 600 cm, and decreased Dissolved Oxygen at the depths of 200 cm and 500 cm.

CWQI values at profile B1 were marginal (57) at a depth of 50 cm, and fair (68) at a depth of 350 cm and (70) at a depth of 200 cm (Figure 7). These results were caused by highest nse of increased Al values for the depths of 50 cm and 500 cm (Table 15) and decreased Dissolved Oxygen at a depth of 200 m. Increased Cu values at a depth of 50 cm (Table 16) and decreased Dissolved Oxygen values also affected CWQI values.

CWQI values at profile C1 were poor (35) at a depth of 500 cm and (39) at a depth of 200 cm, marginal (47) at a depth of 50 cm and fair (68) at a depth of 600 cm (Figure 7). These results are attributed to the highest nse of decreased Dissolved Oxygen values at depths of 200 cm, 500 cm and 600 cm and increased Al values at a depth of 50 cm (Table 16). Increased Cu values at a depth of 50 cm (Table 16) and pH values at a depth of 200 cm also contributed to the CWQI decline.

3.2.4. Water Quality for Recreation

CWQI values were poor (39) and excellent (100). Poor CWQI values were recorded at depths of 50 cm and 200 cm at profiles A1 and C1, 350 cm and 600 cm at profile A1, while excellent CWQI values were recorded at all other depths (Figure 8). Increased pH with the highest nse caused poor CWQI (Figure 9).

3.2.5. Water Quality for Irrigation

CWQI values for irrigation were good (94) at profiles B1 and C1 at a depth of 50 cm and excellent (100) at profile A1 at depths of 50 cm and 350 cm as well as 500 cm at profiles A1 and B1 (Figure 10). CWQI could not be calculated for other depths due to a lack of data. The slight decline in CWQI was caused by increased Mn values.

3.2.6. Water Quality for Livestock

CWQI values were marginal (62) at a depth of 50 cm at all profiles, fair (66) at depths of 350 cm at profile A1 and 500 cm at profiles A1 and B1, and excellent (100) at all other depths. Increased Hg with the highest nse caused CWQI decline (Figure 11).

3.2.7. CWQI of Velika Bukulja River

CWQI values were marginal (53) for overall, aquatic life (54), and livestock (62), good (90) for drinking, and excellent (100) for recreation and irrigation livestock (Figure 12). Increased Al values, with the highest nse, caused CWQI decline for overall and aquatic life water quality. Increased Mn value with the highest nse decreases CWQI for drinking. Increased Hg value with the highest nse caused a CWQI decline in livestock. Increased Mn and Hg values also contributed to overall water quality decline.

4. Discussion

Overall water quality of the Garaši Reservoir is influenced by the tributary Velika Bukulja River. Increased Al and Hg values in the tributary were in line with increased values of these parameters in the reservoir, which led to overall water quality impairment. This impact is especially visible at a depth of 50 cm at all profiles, which could be explained by the stronger impact of the tributary in the shallower parts, near the surface.

Natural geological sources play a significant role in these findings. The Velika Bukulja River carries sediment rich in Al, Hg, and Mn, contributing to elevated metal concentrations in the reservoir. For example, Al concentrations in the Velika Bukulja River reached 0.27 mg/L at a depth of 50 cm, correlating with higher Al levels observed in the reservoir’s surface layers, such as in Profile A1 (up to 0.223 mg/L at 50 cm depth). Processes like erosion and sediment transport emphasize the influence of natural geological factors on the reservoir’s water quality.

Additionally, agricultural impacts significantly contribute to water quality concerns. Agricultural runoff, particularly from fertilizers and pesticides, increases Mn levels in the reservoir, as seen in Profile C1 where Mn concentrations reached 0.522 mg/L at 50 cm depth.

Internal reservoir processes also exacerbate water quality challenges. In deeper zones of the reservoir, anoxic conditions promote the release of Mn and Al from sediments during organic matter decomposition.

On the other hand, Dissolved Oxygen decreased with depth which affected the overall water quality of the deeper parts (Figure 13). According to Karadžić et al. [68] a descent of Dissolved Oxygen with depth pointed out consistent stratification. The same authors emphasized that Oxygen deficit was the consequence of eutrophication.

Increased Mn value from Velika Bukulja River was in line with increased Mn values in the reservoir, which caused drinking water quality decline (Figure 14). Garaši Reservoir is located on granite rock, which could be also a source of increased Mn values [35]. Previous studies conducted by Karadžić et al. [68] in 2005 and Sunjog et al. [33] in 2009, 2010 and 2011 also detected increased Mn concentrations. According to the CWQI, greater depths (from 800 cm to 1500 cm) are more suitable for drinking water supply than shallower parts. However, this result could be taken with caution, because Mn concentrations were not measured at depths from 800 cm to 1500 cm.

According to the CWQI, water quality for aquatic life is critically endangered. Greater depths are not suitable for life, due to a lack of Dissolved Oxygen. These results are in line with the study of Čađo et al. [69] who found that the deepest and central parts of the Garaši Reservoir belong to a poor ecological potential according to Oxygen content. On the other hand, the strong influence of the tributary Velika Bukulja River is visible near the surface, at a depth of 50 cm, where the Al values had the highest nse, as was also recorded in tributary. The highest Al concentration (270 µg/L) was in line with the study of Marković et al. [35], when the recorded Al concentration (269.07 µg/L) in 2017, was the highest among 10 reservoirs in Serbia (Figure 15). Toxic elements can penetrate into plants and animals and further into the food chain [70]. Increased Al, Mn and Cu concentrations in water, have led to increased concentrations of the same parameters in fish tissues [33].

Poor water quality for recreation in shallower parts indicated their unsuitability for bathing and swimming while deeper parts with poor CWQI were not suitable for diving. However, excellent water quality at profile B1 indicated its suitability for recreational purposes.

Water quality was suitable for agricultural purposes, especially for livestock, where CWQI values were excellent in all measurements. The lack of data for depths of 200 m at all profiles, 500 cm at profile C1, 600 cm at profiles A1 and C1, and depths from 800 to 1500 cm at profile A1 for calculation of CWQI for irrigation, suggested that monitoring should be improved in order to obtain more comparable results. Besides that, the lack of metal parameters data for livestock, at the same depths, implies that continuous and unique monitoring (with the same parameters in all depths) is necessary to obtain more comparable results.

According to Milošković and Simić [34] agriculture and wastewater plants are the main sources of heavy metal pollution (including Hg and Mn) of the Garaši Reservoir, even though the reservoir is assessed as having low anthropogenic impact [33,71]. Inorganic polymers of Al are used in the wastewater treatment process [68], which could be a source of Al in the reservoir.

Comparative Analysis of SWQI and CWQI in the Garaši Reservoir

The comparative analysis of the Serbian Water Quality Index (SWQI) and the Canadian Water Quality Index (CWQI) provides a comprehensive understanding of water quality dynamics in the Garaši Reservoir from 2021 to 2023. Both indices highlighted reduced Oxygen as a common challenge, but their methodological differences led to contrasting insights. While SWQI focuses on parameters aligned with national standards—such as Temperature, pH, Electrical Conductivity, Oxygen Saturation, BOD, Ammonium, Total Nitrogen, Suspended Solids, Orthophosphates, and Total Coliforms—CWQI incorporates additional chemical markers, including Aluminum (Al), Mercury (Hg), Manganese (Mn), and Copper (Cu). This broader scope allows CWQI to capture the impact of specific contaminants that may remain undetected by SWQI, offering a nuanced perspective on chemical pollution.

To explore the correlation between SWQI and CWQI values at joint measurement depths (50 cm, 200 cm, and 500 cm) of all three profiles, regression analysis was employed. Since the p-value in the analysis of variance exceeds 0.10, there is no statistically significant relationship between SWQI and CWQI results at the 90% or higher confidence level. The correlation coefficient equals 0.21, indicating a relatively weak relationship between the variables. A more in-depth statistical analysis was not feasible due to the limitations of the available data series.

From a spatial and depth perspective, SWQI results demonstrated a steady improvement in water quality across profiles and depths in 2023. For instance, in Profile A1, SWQI increased from 85 at 50 cm depth in 2021 to 94 in 2023, with similar upward trends observed at depths such as 500 cm and 600 cm, despite temporary declines in 2022. Conversely, CWQI results exhibited greater variability, emphasizing chemical contamination issues at specific depths. In Profile A1, while CWQI values at 500 cm and 600 cm reached optimal levels of 100 in 2023, a significant drop was noted at 350 cm, where values decreased from 84 in 2021 to 59 in 2023. These discrepancies underscore the influence of parameters such as elevated Al, Mn, and Hg concentrations on CWQI outcomes.

Profile B1 consistently showed better water quality across both indices, with SWQI and CWQI values remaining high and stable over the three-year period. This stability reflects reduced tributary impacts and more effective water mixing in this region. On the other hand, Profiles A1 and C1 revealed greater fluctuations, driven by localized pollution sources and stratification effects. For example, CWQI results for Profile C1 highlighted increased Al and Hg concentrations at depths of 50 cm and 500 cm, which impaired overall water quality, particularly in earlier years.

In terms of temporal trends, SWQI demonstrated a marked improvement in water quality by 2023, reflecting enhanced conditions for physical and chemical. CWQI values, however, continued to reveal persistent challenges related to specific chemical pollutants, such as elevated Mn and Hg levels, which remained critical for drinking water quality. This divergence highlights the complementary roles of the two indices in evaluating ecosystem health and specific contamination issues.

The integration of SWQI and CWQI results enriches the analysis of water quality in the Garaši Reservoir, emphasizing the importance of a multifaceted approach. While SWQI provides an overview of general ecosystem health, CWQI adds depth by identifying chemical contaminant hotspots that require targeted management interventions. This dual-index perspective underscores the need for continuous and detailed monitoring, particularly in regions influenced by tributary inputs. Together, these indices offer valuable insights into sustainable water resource management and ecological preservation efforts.

5. Conclusions

Research on water quality in artificial lakes, such as the Garaši Reservoir, is vital for ecological balance and public health. This study highlights significant variations in water quality parameters across profiles and depths of the reservoir between 2021 and 2023, emphasizing high concentrations of Ammonium and changes in Oxygen Saturation in deeper layers. These findings indicate organic matter decomposition and limited water mixing, alongside risks of eutrophication driven by increased levels of Orthophosphates and Nitrogen compounds.

Notably, the data demonstrate an overall improvement in water quality by 2023, reflecting the effectiveness of implemented measures. Critical parameters like Aluminum and Manganese from the Velika Bukulja tributary warrant targeted actions, such as wastewater treatment, erosion control, and agricultural runoff reduction. To ensure sustainability, future efforts should focus on expanding seasonal monitoring to at least four times per year including metal pollution parameters in each measurement. Efforts should also prioritize the implementation of aeration and mixing systems to improve oxygen saturation in deeper layers. Remote sensing technologies should be applied as a supplementary method to traditional monitoring approaches. Additionally, strengthening local awareness and enforcing water protection regulations are crucial. Promoting best agricultural practices remains an essential component of sustainable water quality management. These recommendations aim to preserve the long-term health of the Garaši Reservoir ecosystem and secure its role as a vital resource for biodiversity and public health.