Surface Water Contaminants (Metals, Nutrients, Pharmaceutics, Endocrine Disruptors, Bacteria) in the Danube River and Black Sea Basins, SE Romania

Abstract

The assessment of surface water quality of the Danube River and Black Sea was performed taking into account the amounts determined for heavy metals (As, Cd, Cr, Cu, Hg, Mn, Ni, Pb, Zn), nutrients (compounds of N and P, chlorophyll a), emerging contaminants (CECs) (pharmaceutics and endocrine disruptors—19 quantified compounds, out of 30 targeted chemicals), heterotrophic bacteria and total coliforms, in thirty-two locations from the lower Danube sector (starting with km 375 up to the river mouths), the Danube Delta Biosphere Reserve (three Danube branches—Chilia, Sulina, and Sf. Gheorghe) and the Romanian coastal area of the Black Sea. The heavy metals levels were found in the following ranges: 3.0–6.5 μg/L As; 0.51–1.32 μg/L Cd; 21.6–61.2 μg/L Cr; 10.2–28.6 μg/L Cu; 196–351 μg/L Mn; 12.3–47.67 μg/L Ni; 5.2–15.5 μg/L Pb; 44–74 μg/L Zn; 0.01–0.08 μg/L Hg. The nutrient concentrations vary in the intervals: 0.04–0.45 mg/L N-NH₄; 0.01–0.06 mg/L N-NO₂; 0.07–1.9 mg/L N-NO₃; 1.0–3.2 mg/L N total; 0.01–0.05 mg/L P-PO₄; 0.02–0.27 mg/L P total, and 0.8–17.3 μg/L chlorophyll a. The concentrations of CECs from various classes (sulfamethoxazole, trimethoprim, ciprofloxacin, flumequine, amoxicillin, cefuroxime, dicloxacillin, carbamazepine, pravastatin, erythromycin, piroxicam, ketoprofen, diclofenac, naproxen, enilconazole (imazalil), clotrimazole, drospirenone, 17α-ethinylestradiol, and bisphenol A) were compared with values reported for European rivers and the Danube River water in various river sectors. The highest detection frequencies were registered for bisphenol A (100%), sulfamethoxazole (96%), carbamazepine and diclofenac (87%), trimethoprim (78%), pravastatin (46%), and imazalil (34%). Bisphenol A exhibited the largest concentrations (342 ng/L), followed by diclofenac (132 ng/L), carbamazepine (38 ng/L), and sulfamethoxazole (36 ng/L). For most of the contaminants, Black Sea coastal water showed lower concentrations than the Danube water and good ecological status for surface water. Correlations between CECs and total coliforms suggest insufficient treated wastewater effluents as a common contamination source and possible use of CECs as indirect fecal pollution indicator in aquatic systems. This is the first study carried out in the connected system Danube River–Danube Delta–Black Sea for a large palette of toxicants classes and microbial pollutants, which will serve as a baseline for future monitoring of water quality in the region.

1. Introduction

The Danube River, one of the most significant rivers in Europe, plays a vital role in the hydrological and environmental balance of the continent. Crossing through ten countries before flowing into the Black Sea, the river forms the Danube Delta, a unique and complex wetland ecosystem recognized as a UNESCO Biosphere Reserve. However, due to its vast hydrographic basin, the Danube is highly susceptible to pollution from multiple anthropogenic sources, which ultimately impact the ecological status of both the river and the adjacent Black Sea Basin [1,2,3]. An important waterway for the transport of goods, technological and cooling waters for industries, habitat for commercially valuable fish species and, last but not least, support for biodiversity, are the main services provided by the Danube River along its 2850 km to its discharge into the Black Sea. At the same time, the Danube River is the main collector and emissary to the Black Sea of all municipal and industrial wastewater from the entire river basin, affecting the quality of the Danube Delta waters, but also the coastal area of the Black Sea. Moreover, the input and diversity of pollution sources increase as the consumer market imposes new requirements to which suppliers must respond with new technologies and processing methods [4]. The end result leads to residual contaminants, which in various chemical combinations and mixtures can have harmful effects on human health, even when individual chemicals are below the “safety level” [5]. The main pollution sources in the receiving waters of the large hydrographic basin are numerous and diverse, such as urban sprawl, agricultural practices, human and animal waste, hospital sewage, industrial activities, wastewater treatment effluents, shipping, etc. The expansion of cities, the use of fertilizers, pesticides, and herbicides in agriculture, the improperly managed human waste from septic tanks or untreated sewage, the waste from livestock, the medical and pharmaceutical discharge of hospitals can lead to an increased pollution and harm of the water bodies with contaminants, heavy metals, pathogens, nutrients, endocrine disruptors, pharmaceuticals compounds and their metabolites. These pollution sources collectively contribute to the degradation and contamination of water bodies quality, posing numerous risks to biomes and ecosystems, human health, and overall environmental sustainability [6,7,8]. Among these, heavy metals and metalloids (HMs) are of great concern for the aquatic environment since these metals are highly persistent, toxic at high levels, and have the propensity to bioaccumulate in the food chain and, ultimately, harm human health [9,10,11].

Nutrients play an important role in the eutrophication process and pose a serious problem to the monitoring and estimation of their effects on water quality in a riverine environment and are expressed in phosphorus (total and dissolved forms) and nitrogen (ammonia, nitrite, nitrate and organic nitrogen). The important sources of nutrients in water bodies include point sources (municipal, industrial, and agricultural facilities) and diffuse sources (erosion and surface runoff, groundwater inflow and atmospheric deposition) throughout the catchment area, with direct or indirect impacts on aquatic life, biomass growth, oxygen concentrations, water clarity, and sedimentation rates [1,12].

Surface waters contamination with heterotrophic bacteria and coliforms is in correlation with the level of anthropogenic development, as a result of human population density, along with the growing urban, agricultural, and industrial activities. Bacteria are considered ideal markers of microbial contamination of surface waters because of their rapid adaptability of the environmental intrinsic, extrinsic, and biological factors, being predominant in aquatic environments. Generally, heterotrophic bacteria, which constitute most of the microbiota, play a crucial role as bio-decomposers, which makes them an essential bio-catalyzers in the aquatic ecosystems. They are responsible for the organic matter’s biodegradation. Fecal and non-fecal coliforms, or total coliforms, are frequently utilized as markers of the overall bacteriological indicators of surface-waters. Total coliforms offer information about the rate of organic contamination and fecal coliforms are useful indicators for assessing fecal contamination and the potential presence of pathogens [13]. Using culture methods, ref. [14] showed that a high concentration of heterotrophic bacteria in the water makes it difficult to determine the presence of fecal coliforms. In these conditions, the evaluation of the total coliforms can serve as an indicator of water pollution.

In the 1990s, the issue of pharmaceutical residues in water gained attention. Following numerous published reports, only in 2013, three pharmaceutical substances ended high up on the watch list of Directive 39/2013 within the Water Framework (diclofenac, 17-β-estradiol, 17-α-ethinylestradiol). Starting with Directive 2013/39/EU [15] to current Decision 2022/1307 [16], the Watch List mechanism established in 2013 by EU Directive was intended to improve the available information on the occurrence of substances of greatest concern. In the context of the great number of reports regarding the occurrence of the contaminants of emerging concern (CECs) in the environment and the continuous updating of the regulation in the field of water policy [15,16,17,18,19,20], there is a need for constant monitoring of these contaminants in the water environment [21,22,23,24,25,26,27]. A better knowledge of pollutants’ occurrence will help to derive prioritization strategy and development of environmental quality standards.

The Danube River is among the world’s top 10 rivers considered at risk [1]. As the second largest catchment region in Europe, the Danube River Basin was subjected to numerous monitoring campaigns along the entire river length, either within the Joint Danube Surveys (JDS), coordinated by the International Commission for the Protection of Danube River (ICPDR) [28,29,30,31,32], or individual surveys conducted on shorter sectors [10,12,23,33,34]. The Joint Danube Survey JDS4 included 51 sampling sites in 13 countries across the Danube River Basin [31], out of which in Romania only 7 sampling sites have been surveyed.

The MONITOX project, supported by the EC through the Joint Operational Programme Black Sea Basin 2014–2020 [35], has established a complex network of 32 sampling points on Romanian territory along the Danube River, Danube Delta and up to the Black Sea littoral, starting from Ostrov/Silistra at Romania-Bulgaria border (Figure 1), in order to assess the quality of surface water from a physico-chemical, chemical, and microbiological point of view. The Black Sea was also selected due to the semi-enclosed sea character, the size of the hydrographical basin, and its hydrobiological features which make the Black Sea a unique ecosystem, extremely sensitive and exposed to organic and inorganic threats [36].

Due to the limited information reported in literature, a total of 30 CECs—including pharmaceuticals, azolic pesticides, synthetic hormones, and endocrine disruptors—were selected for monitoring in the present study, performed within the MONITOX program using liquid chromatography-tandem high resolution mass spectrometry (LC-HRMS/MS). The selection of emerging organic contaminants was based on several criteria: (i) ecotoxicological potential; (ii) inclusion on the watch list of the last Decision in the Water Framework; (iii) compounds not routinely monitored; (iv) contaminants which had been previously detected in the Danube River. Besides the selected CECs, the surface water physico-chemical parameters (9 metals, 7 nutrient compounds) and two microbiological indicators were investigated in the fluvial, deltaic, and marine environments belonging to the connected system Danube River–Danube Delta–Black Sea in SE part of Romania. This work is part of a complex research performed in the period 2018–2021 on contaminants from various classes in different environmental matrices of large aquatic ecosystems from Romania, Republic of Moldova, and Greece and evaluation of ecotoxicological and health risk.

2. Material and Methods

2.1. Water Sampling and Analysis of Physico-Chemical Parameters

To assess the spatial variation of contaminants, water samples were collected in the period 22 June–1 July 2020 from a total of 32 sites located in the connected system Danube River–Danube Delta–Black Sea in SE part of Romania (Figure 1).

The freshwater sampling started from Ostrov (at the border with Silistra/Bulgaria), and continued until the Danube flows into the sea through its 3 arms (Chilia—Musura bay, Sulina, and Sf. Gheorghe); the seawater sampling was done in Sacalin and 7 representative points of the Black Sea, Romanian littoral. The geographical coordinates of the sampling points were recorded, and samples were coded sequentially as 1–32 (Figure 1). The target sites included the main localities on the lower, pre-deltaic, deltaic, and maritime sector of the system, points of confluence with the main tributaries Siret and Prut, protected areas and several border sites. The expeditions were organized by the project teams of Danube Delta National Institute for Research and Development, Tulcea (DDNI), and Dunarea de Jos University of Galati (UDJG), Romania, by boat and cars.

Surface water samples were collected according to European standards [37,38], from the water column (0–60 cm) for both monitoring points located along the Danube and in the coastal area. For the Danube, samples were taken from the navigable channel area, as this area captures the maximum degree of homogenization of pollutants, and in the case of the coastal area, sampling was carried out at depths of 1.5–2 m, at a distance of 15–25 m from the shoreline. The containers used were prepared in advance, and at the time of sample collection, they were rinsed well with surface water [12].

For dissolved nutrients (nitrates, nitrites, ammonia) and total phosphorus analysis, the samples were taken in 1 L polyethylene containers, labeled from 1 to 32, stored at a temperature of 2–5°, to minimize deterioration prior to chemical analysis. For organic nitrogen analysis, the water samples were preserved by adding 1 mL of concentrated sulfuric acid per 100 mL of sample, for acidulation at pH < 2. After collecting and preserving the samples, the bottles were sealed to ensure the integrity of the samples. During transport, the samples were protected from light and excessive heat, as the quality of the sample can change rapidly due to gas exchange, chemical reactions, and metabolism of the organisms. The samples were kept at a temperature of 2–5 °C (refrigerated box) [37]. The same principle of sample acidification was also used for the preservation of heavy metals, for which 1 mL of superpure nitric acid was added to a volume of 100 mL of the sample. The samples selected for microbiological analysis were transported to UDJG by car on the same day of collection.

For nutrient (N, P) compounds, the measurements of the total forms of nitrogen and phosphorus and dissolved forms of nitrogen (nitrite, nitrate, ammonia) and dissolved phosphorus were performed by validated procedures at DDNI, in a chemistry laboratory accredited since 2005, according to the ISO/IEC 17025:2017, General requirements for the competence of testing and calibration laboratories [39], using molecular absorption spectrometry with the aid of a UV-VIS Lambda 650 Perkin Elmer spectrophotometer and ISO standards, and their quality was controlled through the analysis of blank samples and control standards [12]. All the reagents were of analytical grade quality. For quality assurance, flow charts were made with the specific certified reference materials.

The N species determined and reported in this study as nitrogen are referred to as N-NO₃, N-NO₂, N-NH_4, and total nitrogen (N total), expressed in mg/L. The P forms discussed here are orthophosphates, expressed as P-PO₄, and total phosphorous (P total), expressed in mg/L.

Determination of chlorophyll “a” (labeled as Chloroph. a) and pH was performed in situ using the Xylem YSI EXO2 submersible multiparameter at an approximate depth of 20 cm for 5 min. The EXO2 probe is a multiparametric instrument that collects water quality data. The probe collects data with up to six sensors, each sensor measuring its parameters through a variety of electrochemical, optical, or physical detection methods [12].

2.2. Metal Analysis by ICP-MS and CVAAS

In order to evaluate the concentrations of arsenic, cadmium, chromium, copper, manganese, nickel, lead, and zinc from surface waters, it used the inductively coupled plasma mass spectrometry (ICP-MS) technique, with the aid of a PerkinElmer Elan DRC–e II ICP-MS instrument (PerkinElmer LAS (UK) Ltd., Seer Green, UK).

For mercury analysis, it employed the cold vapor atomic absorption spectrometry (CVAAS) procedure, a physical method based on the absorption of radiation at 253.7 nm by the mercury vapor. The Flow Injection Mercury System (FIMS) FIMS 400 Perkin Elmer is a stand-alone mercury analyzer that contains a light source and detector specific for mercury, where organic mercury compounds are oxidized, and the mercury is reduced to the elemental state and aerated from solution in a closed system [40].

Both instruments were calibrated using the external calibration technique, in which the concentrations for the measured sample set were extrapolated using linear regressions made from raw counts per second data (ICP-MS) and absorbances (peak height) (FIMS) [41]. The experimental conditions, including instrumental selection, were set to validate and regulate the quality of data, accuracy, and stability of calibration.

The calibration curves were drawn up using the ICP multi-element standard VI, Merck, Lot HC15474231 Certified Reference Material. The recovery values ranged from 95% (Cd) to 99% (Pb). For Hg, the calibration curve was drawn using the Mercury (Hg) Pure Plus Standard (10 µg/mL Mercury in 5% HNO₃) of Perkin Elmer Single Element Calibration Standard for AA and ICP-OES, the recovery degree being 96%. The linear regression coefficient (R²) had the following values: As (0.9985), Cd (0.9995), Cr (0.9962), Cu (0.9994), Mn (0.9982), Ni (0.9975), Pb (0.9987), Zn (0.9979), Hg (0.9969).

2.3. Analysis of CECs

2.3.1. Chemicals and Reagents

Pharmaceuticals including antibiotics, nonsteroidal anti-inflammatory drugs (NSAIDs), hormones, antiepileptic carbamazepine, endocrine disruptor bisphenol, and five widely spread azole antifungals were selected as target compounds. Analytical standards of selected compounds were supplied from Sigma–Aldrich–Merck (Darmstadt, Germany). The stock standard solutions were prepared in methanol at 1 mg/L. Working standards were made by diluting the stock solutions in ultrapure water to concentrations ranging from 2.5−50 ng/mL. Both stock and working standards were stored at 4 °C until further use. For UHPLC–HRMS analysis, LC–MS-grade methanol and water were purchased from Merck (Darmstadt, Germany) and acetic acid and formic acid LC grade from Fisher Scientific (Loughborough, UK). Solid phase extraction (SPE) Strata-X cartridges (500 mg/6 mL) from Phenomenex (Torrance, CA, USA) were used for pollutants’ extraction from water samples.

2.3.2. Extraction

Aliquoted volumes of 100 mL of the sample were filtered using a glass filter to remove solid components. After the pH adjustment to 3 using concentrated acetic acid, samples were subjected to solid phase extraction using Strata-X cartridges, pre-conditioned with 6 mL ACN, followed by 6 mL water prior to extraction. Water samples are eluted through the SPE column at a flow rate of 3–5 mL/min. Afterwards, 6 mL of 10% methanol followed by 6 mL of pure water was added to remove water-soluble interferences. The residual water in the column was removed under low pressure vacuum for 10 min. The analytes were eluted from the SPE columns using 6 mL methanol at a flow rate of 1 mL/min. The eluent was evaporated in an evaporation unit under a stream of nitrogen at 40 °C (Thermo-Scientific, Bremen, Germany). The final extract was reconstituted with 0.250 mL methanol: water (1:9, v/v), and injected into the UHPLC-HRMS/MS system.

2.3.3. Instrumentation

HRMS-MS analyses were performed at MORAS research center of UDJG, with an UltiMate 3000 UHPLC System (Thermo Fisher Scientific), coupled with a Q Exactive-Orbitrap^TM mass spectrometer equipped with Heated Electrospray Ionization (HESI) probe (Thermo Fisher Scientific). Chromatographic separation used a Syncronis C18 column (50 mm × 2.1 mm, 1.7 μm) at 30 °C. A 15 min gradient elution at a 0.3 mL/min flow rate solvent A (water with 0.01% formic acid) and solvent B (methanol with 0.01% formic acid) was used for quantitative LC-HRMS-MS analysis.

Full scan MS followed by data dependent MS2 (ddMS2) in both positive and negative mode was used for quantitative analysis of the selected compounds listed in

Table 1. The exact mass, retention time, and method parameters for the selected contaminants.

Compound	Class	Formula	Exact Mass	[M + H]+	[M − H]−	RT (min)	MS-MS Fragments	Recovery (%)	LOD ng/L	LOQ ng/L
Sulfamethoxazole	Sulfonamides	C10H11N3O3S	253.0521	254.0593	252.0448	5.20	108.4450; 156.0115; 92.0496	90.5	1	3
Trimethoprim	Diaminopyrimidinen	C14H18N4O3	290.13789	291.1451	289.1306	4.85	230.1162; 123.0665; 245.1032	100.2	1.5	3.7
Ciprofloxacin	Quinolones	C17H18FN3O3	331.1332	332.1404	330.1259	4.26	245.1086; 288.1508; 207.0653	95	1.9	5.7
Norfloxacine	Quinolones	C16H18FN3O3	319.1332	320.1404	318.1259	4.12	302.1302; 276.1511; 233.1086	95.4	2.8	6.9
Flumequine	Fluoroquinolines	C14H12FNO3	261.0801	262.0873	260.0728	6.25	244.0768; 220.0407; 202.0287	91	3.2	9.7
Oxytetracycline	Tetracycline	C22H24N2O9	460.1481	461.1554	459.1409	4.66	184.0520; 128.0621; 115.0544	50.2	8	24.3
Doxycycline	Tetracyclines	C22H24N2O8	444.1532	445.1605	443.1460	6.35	168.0571; 152.0621; 139.0542	97	8.8	26.6
Amoxicillin	Antibiotic	C16H19N3O5S	365.1045	366.1117	364.0972	3.39	160.0433; 114.0378	62.5	6.7	20.1
Cefuroxime	Penicillines	C16H16N4O8S	424.0688	425.0761	423.0616 *	8.12	318.1451; 284.2901; 207.0990	87	3.2	9
Dicloxacillin	Penicillines	C19H17Cl2N3O5S	469.0265	470.0338	468.0193	6.48	156.9607; 108.9841	96	2.4	6.8
Clindamycin	Lincosamides	C18H33ClN2O5S	424.1798	425.1871	423.1726	10.25	407.1762; 377.1842; 126.1278	95	5.2	15
Carbamazepine	Antiepileptic	C15H12N2O	236.0949	237.1022	235.0877	7.45	194.0968; 192.0809; 179.0725	108	2	6.2
Clofibric acid	Lipid regulator	C10H11ClO3	214.0396	215.0469	213.0324	8.20	126.9957; 85.0295; 169.0661	67.5	1.9	5.7
Pravastatin	Lipid-lowering	C23H36O7	424.2461	425.2533	423.2388	5.86	321.1703; 303.1601; 101.0607	98	1.2	3.7
Erythromycin	Macrolide	C37H65NO12	715.4506	716.4579	714.4434	8.12	576.3721; 558.3648; 421.3601	37.2	8.5	25.1
Piroxicam	NSAIDs	C15H13N3O4S	331.0626	332.0699	330.05543	7.45	95,0605;121.0398; 164.0820	92	3.9	12
Ketoprofen	NSAIDs	C16H14O3	254.0942	255.1015	253.0870	8.32	138,9949; 129.0102; 174.0915	99.5	4.4	12.5
Indomethacin	NSAIDs	C19H16ClNO4	357.0767	358.0840	356.0695	9.80	138.9949; 129,0102; 174.0915	78	6.4	19.4
Carprofen	NSAIDs	C15H12ClNO2	273.0556	274.0629	272.0484	9.65	230.0538; 228.0567; 193.0890	48.8	9.8	29.6
Diclofenac	NSAIDs	C14H11Cl2NO2	295.0166	296.0239	294.0094	9.70	215.0497; 250.0188; 180.0811	105.1	3.8	11.9
Meclofenamic acid	NSAIDs	C14H11Cl2NO2	295.0166	296.0239	294.0094	10.25	278.0133; 243.0445	68.8	11.7	35.5
Naproxen	NSAIDs	C14H14O3	230.0942	231.1015	229.0870	8.60	185.0963; 153.0704; 170.0726	60.9	3.7	11.5
Enilconazole	Azole antifungal	C14H14Cl2N2O	296.0483	297.0555	295.0410	6.87	255.0099; 158.9765; 109.0762	100.5	1	3
Ketoconazole	Azole antifungal	C26H28Cl2N4O4	530.1487	531.1560	529.1415	8.45	489.1459; 255.0091; 82.0526	87	0.5	1
Fluconazole	Azole antifungal	C13H12F2N6O	306.1040	307.1113	305.0968	5.65	238.0791; 220.0685; 169.0459	106.3	1.6	5
Clotrimazole	Azole antifungal	C22H17ClN2	344.1080	345.1152	343.1007	9.61	278.0835; 165.0689	100.2	4.4	12.3
Miconazole	Azole antifungal	C18H14Cl4N2O	413.9860	414.9932	412.9787	10.45	281.9769; 156.9766; 69.0449	76.4	0.5	1.7
Drospirenone	Synthetic progestin	C24H30O3	366.2194	367.2267	365.2122	10.56	349.2163; 257.1532; 171.1154	87.0	0.7	2.0
17-α Ethinylestradiol	Synthetic estrogen	C20H24O2	296.1776	297.1848	295.1703	12.04	279.1744; 214.1308; 159.1169	69.0	0.5	1.0
Bisphenol A	Endocrine disruptor	C15H16O2	228.1150	229.1222	227.1077	4.02	219.0901; 147.1170; 95.0857	100.4	0.8	2.1

* in bold—negative ion used for the identification of the compound.

.

Parameters for MS analysis were set as follows. The applied voltage was 3.6 kV, and the capillary temperature was 320 °C for positive ionization. Negative mode ionization used a spray voltage of 2.9 kV and a capillary temperature of 300 °C. The normalized collision energy (NCE) of the cell was set at 35 eV in both positive and negative mode. Nitrogen was used as collision and auxiliary gas, at flow rates of 10 and 40 arbitrary units, respectively, with a temperature of 350 °C. Full scan covered the 100–1000 m/z range; data were acquired at a resolving power of 70,000 FWHM at 200 m/z, while a resolution of 35,000 was used for MS-MS analysis. The Automatic Gain Control (AGC) target was set to 10⁶, with the maximum injection time of 200 ms. The scan rate was set at 3.7 scan/s. The precursor ions are filtered by the quadrupole which operates at an isolation window of 2 m/z.

Data were processed with the Xcalibur software version 2.1.1. The mass tolerance window was set to 5 ppm. The identification of the target compounds was carried out based on the accurate mass of the molecular ion, retention time, and the fragmentation pattern resulting from MS-MS analysis (Table 1). Typical chromatograms are presented in Figures S1 and S2 (Supplementary Materials).

Calibration solutions were prepared in the 2.5–50 ng/mL concentration range for each compound of interest by serial dilution with 10% methanol in water of the 1 mg/L standard mixture. The linear calibration curves for each compound were forced through the origin. The area of the parent compound in full MS analysis was used for quantitative analysis. The performance of the method was assessed regarding linearity, recovery, limits of detection (LOD), and quantification (LOQ) by several tests performed with spiked aliquots water samples with appropriate amounts of mix standard solution at 2.5; 5; 10 and 50 ng/L. The results of the evaluation are listed in Table 1.

The statistical analysis was performed with XLSTAT software (version Basic+, 2023.3.0.1415) [42].

2.4. Analysis of Total Coliforms and Heterotrophic Bacteria

Heterotrophic bacteria and total coliforms were the microbiological indicators analyzed at BioAliment research center of UDJG for the assay of bacterial contamination of the analyzed samples. The assay for heterotrophic bacterial count was carried out by pour plate method, cultivating on plate count agar medium for 48 h at 37 °C. The level of contamination was expressed as colony-forming units/mL of analyzed sample (CFU/mL) [43]. The total coliforms count was assayed according to the ISO 4831:2006 (E) guidelines [44] and expressed as the Most Probable Number (MPN) per 100 mL of analyzed sample (MPN/100 mL of sample). Data processing was performed by Microsoft Excel 2019, at INPOLDE research center of UDJG.

3. Results and Discussion

3.1. Physico-Chemical Parameters

Physico-chemical parameters were measured in the frame of the MONITOX monitoring network and can be considered in the Water Framework Directive as supporting elements governing the development of the biological communities [45]. Water Framework Directive [46] establishes five ecological statuses for water bodies: high, good, moderate, poor, and bad, transposed in Romanian legislation with Order no. 161 of 16 February 2006 of the Minister of Environment and Water Management for the approval of the Normative on the classification of surface water quality in order to establish the ecological status of water bodies. In this Romanian Order, limit values corresponding to five quality classes are established for all physico-chemical indicators [47]. Our results (parameter values and analytical errors) are presented in Table S1 (Supplementary Materials), and the water quality classes for each target site are presented in

Table 2. Physico-chemical parameters of water in the target region and site classification in ecological classes.

No.	Sampling Sites	N-NH4 mg N/L	N-NO2 mg N/L	N-NO3 mg N/L	N Total mg N/L	P-PO4 mg P/L	P Total mg P/L	Chloroph. a µg/L	pH pH Unit
1	Ostrov ferry	0.09	0.013	1.130	2.64	0.043	0.116	4.9	8.83
2	Ostrov, Danube old branch	0.05	0.012	1.333	1.94	0.053	0.091	2.9	7.81
3	Fetesti	0.20	0.016	0.944	1.57	0.038	0.103	5.9	7.77
4	Cernavoda bridge	0.07	0.011	1.370	2.79	0.034	0.102	9.9	8.87
5	Cernavoda Seimeni	0.13	0.015	1.148	1.50	0.038	0.095	5.7	8.78
6	Braila harbor upstream	0.10	0.019	0.870	2.94	0.027	0.096	6.7	8.73
7	Braila harbor downstream	0.07	0.016	1.000	2.40	0.038	0.079	5.7	8.81
8	Siret R. upstream	0.25	0.048	1.481	3.17	0.033	0.172	4.4	8.66
9	Siret R. downstream	0.19	0.047	1.611	1.58	0.022	0.107	7.3	8.6
10	Galati downstream	0.09	0.017	0.963	1.39	0.026	0.089	5.6	8.82
11	Galati shipyard downstream	0.07	0.016	1.278	2.19	0.033	0.266	4.4	8.71
12	Prut R. upstream Giurgiulesti	0.08	0.023	0.778	2.15	0.012	0.061	7.7	8.66
13	Prut R. downstream	0.08	0.019	1.019	2.43	0.033	0.140	6.3	8.78
14	Reni downstream	0.08	0.019	1.056	2.56	0.031	0.117	2.8	8.68
15	Isaccea downstream	0.11	0.021	1.037	2.66	0.033	0.205	5.7	8.59
16	Ceatal Chilia	0.09	0.027	1.111	1.96	0.029	0.128	2.4	8.63
17	Izmail downstream	0.08	0.035	0.998	1.56	0.021	0.118	3.2	8.44
18	Ceatal Sf.Gheorghe	0.05	0.044	1.052	1.42	0.032	0.098	3.4	8.52
19	Chilia veche upstream	0.07	0.050	1.031	1.23	0.035	0.121	3.7	8.48
20	Chilia veche downstream	0.09	0.050	1.058	1.33	0.034	0.088	3.7	8.48
21	Sf.Gheorghe upstream	0.07	0.047	1.020	1.26	0.038	0.184	3.0	8.63
22	Musura bay mouth	0.04	0.039	0.881	1.46	0.024	0.064	2.0	8.9
23	Sulina mouth	0.05	0.048	0.996	1.17	0.035	0.164	3.4	8.42
24	Sf.Gheorghe mouth	0.05	0.042	0.963	1.04	0.030	0.139	3.0	8.56
25	Sacalin	0.07	0.041	0.999	1.30	0.029	0.113	4.1	8.48
26	Gura Portitei	0.28	0.035	1.870	2.19	0.006	0.024	1.4	8.99
27	Corbu	0.21	0.029	1.871	2.11	0.006	0.022	2.3	9.01
28	Mamaia	0.20	0.029	0.950	1.18	0.005	0.019	4.4	9.24
29	Constanta	0.34	0.024	0.873	1.23	0.006	0.023	0.8	9.05
30	Costinesti	0.26	0.060	1.250	1.57	0.005	0.020	6.6	8.98
31	Mangalia	0.45	0.044	0.716	1.21	0.005	0.018	17.3	9.23
32	Vama veche	0.29	0.022	0.787	1.10	0.011	0.045	2.1	9.09
Ecological status according to Romanian Order no. 161/2006 [47]
Highest ecological status		0.4	0.01	1	1.5	0.1	0.15	25	6.5–8.5
Good ecological status		0.8	0.03	3	7	0.2	0.4	50
Moderate ecological status		1.2	0.06	5.6	12	0.4	0.75	100
Poor ecological status		3.2	0.3	11.2	16	0.9	1.2	250
Bad ecological status		>3.2	>0.3	>11.2	>16	>0.9	>1.2	>250

.

Ammonium may be found in surface waters as a result of degradation of proteins and organic matter from vegetable and animal waste contained in the sediment, industrial and domestic water discharge [48]. The highest value of 0.45 mg/L for ammonia was obtained in Site 31 (Mangalia) and the lowest of 0.04 mg/L in Site 22 (Musura bay mouth). In accordance with Romanian legislation, from the point of view of ammonium concentrations, water samples have values corresponding to the high ecological status (except Site 31 (Mangalia) with good ecological status).

Along the Danube, up to Ceatal Chilia, in terms of nitrite concentrations, the values correspond to the good ecological status, with values between 0.011 and 0.023 mg/L (except from sites 8 (Siret R. upstream) and 9 (Siret R. downstream) with moderate ecological status). In the Danube Delta, water bodies have moderate ecological status, with nitrite concentrations from 0.035 mg/L to 0.050 mg/L. Across the Black Sea coast, the ecological status varies from good (sites 27 (Corbu), 28 (Mamaia), 29 (Constanta) and 32 (Vama Veche)) to moderate (sites 30 (Costinesti) and 31 (Mangalia)) and poor (site 30 (Costinesti)).

The presence of nitrates in natural waters can be explained by water contact with the ground watershed or by water discharge from farmland [48]. For nitrate concentrations, expressed in nitrogen, all water bodies are framed in good ecological status (65.6% of stations) and high ecological status (34.4% of stations). The highest nitrate concentrations were determined at sites 27 (Corbu) (1.871 mg/L) and 26 (Gura Portitei) (1.870 mg/L) and the minimum (0.716 mg/L) corresponded to site 31 (Mangalia).

Total Nitrogen has a similar trend with nitrate, with good ecological status (53.1% of monitoring points) and high ecological status (46.9% of sites). The Total Nitrogen concentrations ranged from 1.10 mg/L (site 32 (Vama veche)) to 3.17 mg/L (site 8 (Siret R. upstream)).

Orthophosphates, expressed in dissolved phosphorus, are 100% bioavailable to plants. All dissolved phosphorus concentrations are above the maximum acceptable limit of 0.1 mg/L (high ecological class). The maximum value (0.053 mg/L) was measured in site 2 (Ostrov, Danube old branch) and the minimum (0.005 mg/L) in 3 sites on the Black Sea coast: 30 (Costinesti), 31 (Mangalia), and 28 (Mamaia).

Total phosphorus represents the sum of dissolved phosphorus and particulate phosphorus (a long-term source for algae and plants). The concentrations obtained for this parameter revealed that surface water in the monitored area corresponds to high ecological class, except the sites 8 (Siret R. upstream), 11 (Galati shipyard downstream), 15 (Isaccea downstream), 21 (Sf. Gheorghe upstream), and 23 (Sulina mouth), that correspond to good ecological class. The relative contribution of dissolved phosphorus to total phosphorus varied between 12.260% (site 11 (Galati shipyard downstream) and 58.47% (site 2 (Ostrov, Danube old branch)).

The values determined for chlorophyll “a” ranged from 1.4 µg/L (site 26, Gura Portitei) to 17.3 µg/L (site 31, Mangalia), indicating a high ecological class (quality standard of 25 µg/L) in all 32 sites.

In general, the examined surface waters were in the alkaline pH range. Only seven surface waters had pH values within the 6.5−8.5 range recommended by the Romanian legislation. The minimum value (7.77 pH units) was obtained in Site 3 (Fetesti) and maximum (9.23 pH units) in site 31 (Mangalia).

A comparison of nutrient levels in Danube water can be made with the findings published in the JDS4 report of the ICPDR for the 2019 survey [31]. The report includes three sites in the monitored sector of lower Danube and Delta in the SE part of Romania, starting with km 375—JDS4-48 (Chiciu-Silistra), JDS4-50 (Reni), and JDS4-51 (Vylkovo, RO-UA, Chilia branch)—which are similar with the sites of our monitoring network nos. 1 (Ostrov ferry), 14 (Reni downstream), and the combination of sites 20 (Chilia veche downstream) and 22 (Musura bay mouth), respectively. Our results indicate higher concentrations for nitrates and nitrites in all selected sites, orthophosphate in site 1, total nitrogen in sites 1 and 14, and comparable levels of ammonium and total phosphorous across all the sites. For orthophosphate, the concentrations at sites 14 and 20/22 are lower than the ones reported by JDS4 for this sector.

3.2. Metals

In this study, 9 of the 13 elements in the heavy metal and metalloid categories were analyzed as potential toxic elements, for which maximum permissible limits are defined in the Order no. 161 of 16 February 2006 by the Minister of Environment and Water Management for the approval of the Normative on the classification of surface water quality in order to establish the ecological status of water bodies [47]. The metal concentrations (values and analytical errors) are presented in Table S2 (Supplementary Materials), and the water quality classes for each site are shown in

Table 3. Metal concentrations in surface water in the target region and site classification in ecological classes.

No.	Sampling Sites	As µg/L	Cd µg/L	Cr µg/L	Cu µg/L	Mn µg/L	Ni µg/L	Pb µg/L	Zn µg/L	Hg µg/L
1	Ostrov ferry	5.8	0.96	31.1	14.9	320	31.2	15.5	44	0.02
2	Ostrov, Danube old branch	5.0	0.51	26.0	19.8	284	37.7	10.0	59	0.02
3	Fetesti	4.6	0.89	34.5	12.7	270	40.1	12.4	66	0.03
4	Cernavoda bridge	5.2	0.78	39.6	15.2	296	44.2	11.8	45	0.02
5	Cernavoda Seimeni	4.0	1.32	49.5	23.0	244	28.4	9.4	66	0.04
6	Braila harbor upstream	5.0	1.12	55.6	21.7	287	30.2	10.3	59	0.05
7	Braila harbor downstream	5.5	1.00	48.6	28.6	308	31.2	8.7	45	0.04
8	Siret R. upstream	5.2	1.20	43.1	27.4	295	29.3	9.3	57	0.03
9	Siret R. downstream	5.0	1.07	61.2	17.9	284	30.4	10.1	50	0.03
10	Galati downstream	5.4	0.89	55.6	19.7	306	27.2	9.2	47	0.03
11	Galati shipyard downstream	4.9	1.20	51.6	19.6	280	31.3	11.1	46	0.03
12	Prut R. upstream Giurgiulesti	4.3	0.82	46.9	27.4	256	38.0	8.3	73	0.02
13	Prut R. downstream	4.7	1.32	54.3	17.9	272	27.8	9.4	45	0.05
14	Reni downstream	5.7	1.12	52.4	19.7	318	47.7	10.3	55	0.03
15	Isaccea downstream	5.9	1.00	48.2	17.9	325	32.6	8.7	64	0.05
16	Ceatal Chilia	5.0	0.65	51.6	19.7	288	27.2	11.3	47	0.05
17	Izmail downstream	4.9	0.72	39.3	23.8	283	29.6	9.5	58	0.04
18	Ceatal Sf.Gheorghe	3.0	0.75	41.9	22.4	196	19.7	10.1	62	0.05
19	Chilia veche upstream	5.0	0.68	31.2	24.6	284	24.9	9.9	47	0.06
20	Chilia veche downstream	5.2	0.66	40.1	24.2	294	26.9	9.7	44	0.05
21	Sf.Gheorghe upstream	4.8	0.80	29.0	13.1	276	29.6	7.1	53	0.04
22	Musura bay mouth	4.6	0.74	38.7	10.2	268	26.7	9.5	46	0.04
23	Sulina mouth	4.5	0.66	27.0	18.2	266	30.0	8.1	56	0.06
24	Sf.Gheorghe mouth	4.7	0.80	32.1	12.6	274	27.2	6.0	49	0.06
25	Sacalin	5.6	0.59	21.6	19.8	312	23.4	8.5	47	0.08
26	Gura Portitei	5.8	0.87	31.1	25.9	323	12.3	8.7	74	0.05
27	Corbu	3.5	0.71	32.2	14.2	221	15.1	10.0	65	0.01
28	Mamaia	4.6	0.69	32.6	17.3	270	23.1	6.0	47	0.01
29	Constanta	6.5	0.70	45.0	21.4	351	16.6	5.2	71	0.01
30	Costinesti	6.3	0.83	28.4	16.5	344	19.9	6.3	53	0.01
31	Mangalia	6.0	0.72	35.9	19.8	282	17.6	5.8	47	0.01
32	Vama veche	6.1	0.86	40.9	21.3	284	12.6	6.1	55	0.01
Ecological status according to Romanian Order no. 161/2006 [47]
Highest ecological status		10	0.5	25	20	50	10	5	100	0.1
Good ecological status		20	1	50	30	100	25	10	200	0.3
Moderate ecological status		50	2	100	50	300	50	25	500	0.5
Poor ecological status		100	5	250	100	1000	100	50	1000	1
Bad ecological status		>100	>5	>250	>100	>1000	>100	>50	>1000	>1

.

The results obtained (Table 3) showed a good ecological status for at least 72% of the 32 monitoring points for chromium and cadmium—elements with high toxic potential. Specifically, 79% of the determinations carried out for chromium showed values lower than the quality standard of 50 µg/L, and in the case of cadmium concentrations, the percentage was 72% for values lower than the threshold of 1 µg/L. The rest of the values, comprised in a percentage of 21% (chromium) and 28% (cadmium), were located in the range corresponding to moderate ecological status for both elements. The maximum values of cadmium concentrations (1.32 µg/L) characteristic of moderate ecological status were identified in sites 5 (Cernavoda Seimeni) and 13 (Prut R. downstream), while for chromium, the highest values were registered in the sites 9 (Siret R. downstream) (61.2 µg/L), as well as 6 (Braila harbor downstream) and 10 (Galati downstream) (55.6 µg/L).

It is encouraging that two other elements with high toxicity potential, such as arsenic and mercury, had concentration values corresponding to high ecological class, under the threshold limits (10 µg/L for As and 0.1 µg/L for Hg), for the entire monitored area, both at the Danube level and for the coastal area. In addition to these two elements with a high degree of toxicity, zinc was another element with values corresponding to a very good ecological status.

Regarding copper concentrations, 66% of the monitored points correspond to a high ecological status with values below the threshold of 20 µg/L. The remaining values (34%), between the standard value of 20 µg/L and 30 µg/L, correspond to a good ecological status and were identified in points in sites 5, 6, 7, 8, 9, 12, 20 located along the Danube, and sites 26 and 29, in the coastal area of the Black Sea.

Manganese, the 7th element with toxic potential monitored in our study, is known for its presence in relatively large quantities throughout the lower Danube [49]. The values obtained following the analysis carried out by us in the 24 sampling points along the Danube are in the range of 196 μg/L in site 18 (Ceatal Sf. Gheorghe) and 320 μg/L in site 1 (Ostrov ferry), values comparable to those reported by ICPDR in 2007 (max 228 μg/L) [49]. From the ecological status point of view, 72% of surface waters are framed in moderate ecological class and 28% in poor ecological class. Similar concentrations were also identified in the Black Sea coastal area, with an increasing trend in the south, near the border with Bulgaria.

For nickel, the lowest value of 12.6 μg/L was recorded in site 32 (Vama veche) and the highest value of 47.7 μg/L in site 14 (Reni downstream). According with the Romanian legislation in force [47], 31.25% of surface waters had good ecological status (quality standard of 25 μg/L) and 67.75% had moderate ecological status (quality standard of 50 μg/L). It was observed that the entire coastal area presented values at least 50% lower compared to the lower Danube sector. Most likely, the high concentrations that were identified at the Danube level are attributed to the riverside industrial capacities, which might have major influences on the concentrations of heavy metals, including nickel or chromium and cadmium which correlated very well with the areas neighboring industrial activities.

The values of lead concentrations classify the surface waters in 71.87% of sites of the MONITOX network in good ecological status, whose quality standard is 10 μg/L (71.87% sites), and 28.13% of sites in moderate ecological status, with a quality standard of 25 μg/L. Along the Black Sea coast, all the sampling sites had good ecological status. The minimum value of 5.2 μg/L was obtained in site 29 (Constanta) and the maximum of 12.4 μg/L in site 3 (Fetesti).

Compared with the results reported by ICPDR for JDS3 in 2013 [30], the levels of heavy metals recorded in the Danube water (sites 1–24) in this work (3.0–5.9 μg/L As; 0.51–1.32 μg/L Cd; 26.0–61.2 μg/L Cr; 10.2–28.6 μg/L Cu; 19.7–47.67 μg/L Ni; 6.0–15.5 μg/L Pb; 44–73 μg/L Zn; 0.02–0.06 μg/L Hg) are higher than JDS3 values obtained for the entire river length (1.09–2.46 μg/L As; <0.01–0.145 μg/L Cd; 0.29–6.73 μg/L Cr; 1.06–9.93 μg/L Cu; 0.78–24.63 μg/L Ni; 0.2–8.08 μg/L Pb; 1.13–12.95 μg/L Zn; <0.002–0.007 μg/L Hg). For selected metals, our results are similar to the maximum values obtained for Danube tributaries (5.33 μg/L As; 1.05 μg/L Cd; 67.13 Cr; 60.73 μg/L Zn) [30], suggesting the fact that the affluents may add an important input of pollutants into the Danube River.

3.3. CECs

Nineteen out of the thirty selected analytes were identified in the surface water samples: sulfamethoxazole (SMX), trimethoprim (TMP), ciprofloxacin (CIP), flumequine (FLU), amoxicillin (AMX), cefuroxime (CFX), dicloxacillin (DCX), carbamazepine (CBZ), pravastatin (PRV), erythromycin (ERY), piroxicam (PIR), ketoprofen (KET), diclofenac (DCF), naproxen (NAP), enilconazole (imazalil) (IMZ), clotrimazole (CLO), drospirenone (DRO), 17α-Ethinylestradiol (EE2), and bisphenol A (BPA). The quantitative analysis results (in ng/L) are presented in

Table 4. Concentrations of CECs for each sampling site (ng/L).

Site No.	SMX	TMP	CIP	FLU	AMX	CFX	DCX	CBZ	PRV	ERY	PIR	KET	DCF	NAP	IMZ	CLO	DRO	EE2	BPA
1	28.9	6.4	ND	ND	3.8	1.6	3.5	15.4	6.4	4.3	ND	ND	28	6.2	8.3	ND	ND	1.6	238
2	18.6	ND	ND	ND	ND	ND	ND	11	ND	ND	ND	ND	18.7	ND	6.2	ND	ND	ND	186
3	24	8	ND	ND	ND	ND	ND	14.9	7	ND	ND	ND	41.3	ND	ND	ND	ND	ND	182
4	15.4	2.4	ND	ND	ND	ND	ND	8.4	ND	ND	ND	ND	16	ND	ND	ND	ND	ND	164
5	21	6.2	ND	ND	ND	ND	ND	10	ND	ND	ND	ND	13.2	ND	ND	ND	ND	ND	157
6	26	8.6	ND	ND	ND	ND	ND	16.6	18	ND	14.6	ND	69	ND	ND	ND	1.8	1.8	186
7	18.4	4.1	ND	ND	ND	ND	ND	35	6.4	ND	ND	ND	87	ND	ND	ND	2.3	ND	173
8	32	6.7	ND	ND	ND	ND	ND	30.4	21.4	ND	32	26	112	6.1	27.6	ND	1.8	2.15	297
9	28	10.1	ND	ND	ND	ND	ND	37	20.7	ND	28.6	20.8	80	4.9	24.6	6.4	2.05	2.1	310
10	35	11	5.2	6.4	ND	4.9	7	26	6.8	ND	8.9	7.5	87	5.2	6.8	ND	1.04	2.4	138
11	21	6.4	ND	ND	ND	ND	ND	12.2	4.3	ND	16	ND	46	ND	10.5	ND	1.75	1.3	142
12	24	12	4.1	ND	ND	ND	ND	38	24.8	ND	ND	12.6	132	8.6	31.4	8.2	3.4	3.05	342
13	25	7.5	3.4	4.6	ND	ND	9.4	27.8	24	ND	ND	8.2	114	6.2	24.6	5	2.6	1.62	317
14	36	6.2	5	ND	ND	ND	3.6	18.9	16.2	ND	5.6	5.6	95	ND	10.2	ND	ND	1.15	156
15	22	5.9	ND	ND	ND	ND	4	20	9	ND	ND	ND	32	ND	ND	ND	ND	ND	141
16	14	6.3	ND	ND	ND	ND	ND	16.3	4.2	ND	ND	ND	12.9	ND	ND	ND	ND	ND	92
17	18	6.1	ND	ND	ND	ND	ND	12	ND	ND	ND	ND	9.4	ND	ND	ND	ND	ND	85
18	11.8	2.9	3.4	ND	ND	ND	ND	17	3.1	ND	ND	ND	24.8	ND	ND	ND	ND	ND	63
19	7.4	4.2	ND	ND	ND	ND	ND	6.2	ND	ND	ND	ND	4.6	ND	ND	ND	ND	ND	49
20	12.5	3.5	ND	ND	ND	ND	ND	8	ND	ND	ND	ND	5.5	ND	ND	ND	ND	ND	54
21	14	3.9	3.1	ND	ND	ND	ND	18	ND	ND	ND	ND	25.3	ND	ND	ND	ND	ND	48
22	24	9.7	ND	ND	ND	ND	ND	10.2	ND	ND	ND	ND	12.8	ND	4.8	ND	ND	ND	183
23	21	6.3	1.8	ND	ND	ND	ND	26.7	ND	ND	ND	ND	18.4	ND	6.2	ND	ND	ND	51
24	15.7	5.3	2.6	ND	ND	ND	ND	19.5	ND	ND	ND	ND	21.7	ND	ND	ND	ND	ND	34.5
25	3.2	ND	ND	ND	ND	ND	ND	5.6	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	87
26	4	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	114
27	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	124
28	6.5	ND	ND	ND	ND	ND	ND	8.6	ND	ND	ND	ND	5.7	ND	ND	ND	ND	ND	218
29	7.9	ND	ND	ND	ND	ND	ND	26.7	ND	ND	ND	ND	14.8	ND	ND	ND	ND	ND	237
30	5.2	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	6.2	ND	ND	ND	ND	ND	107
31	11.4	ND	ND	ND	ND	ND	ND	8.9	ND	ND	ND	ND	8.4	ND	ND	ND	ND	ND	164
32	10	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	127

ND = not detected; Note: the CECs compounds’ abbreviation is given in the text (Section 3.3) and Table 5.

.

Table 5. Minimum (Min.), maximum (Max.), and average values of the CECs measured concentrations, the Predicted no-effect concentration (PNEC) values and the detection frequency of the identified substances.

Compound	Abbreviation	Min. (ng/L)	Max. (ng/L)	Average (ng/L)	Lower PNEC * ng/L Fresh/Marine Water	No. of Positive Results
Sulfamethoxazole	SMX	3.2	36	18.13	100/60	31
Trimethoprim	TMP	2.4	12	6.35	100	25
Ciprofloxacin	CIP	1.8	5.2	3.58	89/8.9	8
Flumequine	FLU	4.6	6.4	5.5	1500/150	2
Amoxicillin	AMX	3.8	-	-	78/7.8	1
Cefuroxime	CFX	1.6	4.9	3.25	1290/130	2
Dicloxacillin	DCX	3.5	9.6	5.5	5.1/0.51	5
Carbamazepine	CBZ	5.6	38	18.05	50	28
Pravastatin	PRV	3.1	24.8	12.31	4570/460	15
Erythromycin	ERY	4.3	-	-	300/30	1
Piroxicam	PIR	5.6	28.6	17.62	490/49	6
Ketoprofen	KET	5.6	26	13.45	2100/210	6
Diclofenac	DCF	4.6	132	40.75	50/5	28
Naproxen	NAP	4.9	8.6	6.2	1700/170	6
Enilconazole (Imazalil)	IMZ	4.8	31.4	14.64	870/87	11
Clotrimazole	CLO	5	8.2	6.53	30/3	3
Drospirenone	DRO	1.04	3.4	2.09	120/12	8
17α-Ethinylestradiol	EE2	1.15	3.05	1.91	0.035/0.0035	9
Bisphenol A	BPA	34.5	342	155.2	240/1600	32

* lower PNEC values according to NORMAN database (https://www.norman-network.com/nds/prioritisation/) (accessed on 13 May 2024) [50] and EU watch lists [16,17,18,19,20].

shows the minimum (Min.), maximum (Max.), and average values of the CECs measured concentrations, the Predicted no-effect concentration (PNEC) and the detection frequency of the identified substances.

The concentrations of pharmaceuticals in the analyzed surface waters varied between 1.04 ng/L (drospirenone) to 132 ng/L (diclofenac). Triazine pesticide enilconazole (imazalil) concentrations ranged from 4.8 to 31.4, while bisphenol concentrations ranged from 34.5 ng/L to 342 ng/L.

Pharmaceuticals detected in almost all samples—sulfamethoxazole, trimethoprim, carbamazepine, and diclofenac—can be considered as river basin specific CECs. The widely used pharmaceutical diclofenac was detectable at 28 sampling sites, but the PNEC was exceeded only in 8 sampling points corresponding to Siret and Prut tributaries and Danube River in the points corresponding downstream Siret and Prut rivers, Galati town, and Prut-Giurgiulesti. The measured concentration of β-lactam antibiotic dicloxacillin exceeded PNEC in two points: downstream Galati town and in the Prut River. For the remaining pharmaceuticals, the level does not exceed PNEC. Bisphenol A was detected in all the samples, the measured concentration exceeding PNEC in 4 sampling points in the tributary Siret and Prut rivers, indicating a risk to the aquatic environment. The agricultural impact on the ecosystem of the Danube River was shown by measured imazalil concentrations, with a peak of 31.4 ng/L in the Prut River confluence site.

To enable comparisons between variables with different orders of magnitude, the data were standardized. The univariate analysis resulted in a box plot that highlights the variation and presence of extreme values for each substance (Figure 2).

Several substances, such as ethinylestradiol and bisphenol A, exhibit numerous extreme values, indicating possible sporadic pollution events or high variability in their concentrations. Compounds such as carbamazepine and diclofenac show higher variability compared to others, which may be attributed to diverse pollution sources or varying environmental factors.

Regarding the spatial distribution, the highest number of pollutants was identified in the tributary rivers Siret and Prut, along with the sampling points corresponding to Ostrov village and the towns of Galati, Giurgiulesti, and Reni. In marine water, a significantly lower number of contaminants were found, including sulfamethoxazole, carbamazepine, diclofenac, and bisphenol A. Furthermore, the measured concentrations were lower than those in freshwater samples. The hierarchical clustering analysis (HCA) dendrogram confirmed the patterns among sampling locations based on contaminant concentrations (Figure 3). The analysis revealed three clusters (C1, C2, C3) as follows: cluster C1 (green): includes locations such as Mangalia, Costinesti, Gura Portitei, and Galati shipyard downstream, suggesting similarity in contamination patterns; cluster C2 (blue): includes Prut R. upstream, Siret R. upstream, and Chilia veche upstream and downstream, which may indicate distinct pollution sources; cluster C3 (red): includes Braila and Galati harbor downstream, Cernavoda bridge and Reni downstream, suggesting a unique pollution profile. The between-cluster variance is high (81.63%), indicating that the three clusters are well-separated and validate the chosen classification. Clusters 1 and 3 are more similar to each other, as indicated by their lower centroid distances.

The mobility and persistence of pharmaceuticals in aquatic environments are strongly influenced by physico-chemical parameters such as pH, temperature, dissolved oxygen, and nutrient concentrations. These parameters affect the solubility, degradation, and adsorption of contaminants, ultimately shaping their spatial distribution. Understanding these interactions is essential for assessing pollution sources and potential environmental risks.

To investigate these relationships, a Principal Component Analysis (PCA) and Random Forest regression models were conducted, examining the correlations between key physico-chemical parameters and pharmaceutical pollutants across different sampling sites. The PCA biplot provides an insight into how contaminants cluster together, which locations exhibit higher pollution levels, and which environmental factors may drive these trends (Figure 4).

The axes F1 (37.18%) and F2 (13.16%) together explain 50.34% of the total variance of the dataset’s structure. Nutrient Parameters (N-NO₃, N-NO₂, N-NH₄, P-PO₄, P total) are positioned in the negative F1 region, suggesting that they are negatively correlated with many pollutants. Sites like Mangalia and Costinești are aligned with these parameters, indicating potentially nutrient-rich but less polluted environments. The pH is in the negative quadrant, showing a reverse correlation with pharmaceutical contaminants. The PCA confirms that certain locations are CECs pollution hotspots, particularly in tributary rivers and industrialized areas. Sulfamethoxazole, trimethoprim, carbamazepine, bisphenol A, ethinylestradiol, and diclofenac form a cluster in the positive F1 direction, suggesting they share common sources or behavior in the aquatic environment. They tend to group with locations like Siret R. upstream, Prut R. downstream, and Galati downstream, which could be indicative of wastewater discharges.

The PCA analysis also highlights distinct spatial trends in pollution distribution across the sampling sites. Pollution hotspots, such as Prut R. downstream, Siret R. upstream, and Galati downstream, show a strong association with pharmaceutical pollutants, indicating high anthropogenic impact, likely driven by wastewater treatment plant (WWTP) effluents discharges. In contrast, less contaminated sites, including Sulina mouth, Sf. Gheorghe upstream, and Chilia veche downstream, exhibit lower pollution levels, suggesting they represent more natural water bodies with reduced contamination. The marine area samples appear to be less contaminated compared to those from freshwater sources, due to lower anthropogenic influence or greater dilution effects in the marine environment.

To better understand the relationships between physico-chemical parameters and pharmaceutical contamination, a Random Forest regression model was applied. This approach allows for the identification of key environmental factors influencing the distribution of contaminants while handling complex interactions and non-linear patterns in the data. The model was configured with regression-type analysis, random input selection, and a sampling method without replacement. The number of variables (predictors) considered at each tree split (Mtry) was set to 2, selecting two variables at each split, with a sample size of 29. A total of 300 trees were built out of the 500 initially required. The model’s predictive performance was evaluated using the Out-of-Bag (OOB) error, residual analysis, and variable importance measures (Table S3, Supplementary Materials).

The scatter plot of observed vs. predicted sulfamethoxazole concentrations reveals a moderate predictive power, with some deviations suggesting potential over- or underestimation (OBB 53.4) (Table S3, Supplementary Materials). The variable importance analysis identified P-PO₄ (phosphate), pH, total phosphorus (P total), and ammonium nitrogen (N-NH₄) as key predictors, while N-NO₃ (nitrate) showed negative importance, suggesting it may introduce noise rather than contribute to predictions. These findings highlight the influence of nutrient levels on Sulfamethoxazole persistence in aquatic environments and suggest potential model refinements for improved accuracy.

The Random Forest regression model applied to trimethoprim concentrations (Figure 5) showed an Out-of-Bag (OOB) error of 6.198, indicating a relatively low prediction error. The scatter plot of observed vs. predicted values demonstrates moderate predictive power, with points generally following the 1:1 diagonal. The variable importance analysis highlights P-PO₄ (phosphate), total phosphorus (P total), pH, and N-NH₄ (ammonium nitrogen) as the most influential parameters affecting trimethoprim concentrations. These findings suggest a strong link between nutrient levels and the presence of trimethoprim in aquatic environments.

The Random Forest regression model for carbamazepine resulted in a high Out-of-Bag (OOB) error of 121.876, indicating substantial prediction uncertainty (Table S3, Supplementary Materials). However, the variable importance analysis identifies pH, total phosphorus (P total), phosphate (P-PO₄), and chlorophyll-a as the influential parameters affecting carbamazepine concentrations showing that carbamazepine behavior in aquatic environments is strongly influenced by physico-chemical parameters.

An analysis of the relationship between physico-chemical properties (pKa, LogP) and Random Forest model performance (OOB error, variable importance) highlights the influence of environmental parameters on the persistence and distribution of pharmaceutical compounds in water (Table S3, Supplementary Materials).

Compounds with moderate pKa values (4–6), such as sulfamethoxazole (SMX, pKa = 5.6), ketoprofen (KET, pKa = 4.5), and diclofenac (DCF, pKa = 4.0), show significant correlation with pH (DCF-pH: 5.32; SMX-pH: 5.06), indicating that their solubility and ionization state are strongly pH-dependent. The correlation analysis highlights that compounds with higher pKa and log P values exhibit a positive correlation with OOB error (0.31 and 0.32, respectively), suggesting that more hydrophobic and weakly acidic compounds introduce greater prediction uncertainty (Table S3, Supplementary Materials).

Carbamazepine (CBZ), pravastatin (PRV), and diclofenac (DCF) correlate positively with P-PO₄ and P total; Log P is negatively correlated with N-NH₄ (−0.68), indicating that hydrophobic compounds are less influenced by ammonium presence, likely due to stronger sorption to organic matter. pKa negatively correlates with N-NH₄ (−0.47) and N-total (−0.48), showing that ionizable compounds are more influenced by nitrogen species (Table S3, Supplementary Materials).

The detection of pharmaceutical residues in the Danube River and the Black Sea within our study confirms the increasing concern regarding emerging contaminants (CECs) in aquatic environments. The presence of sulfamethoxazole, trimethoprim, carbamazepine, diclofenac, and bisphenol A in the analyzed samples (Table 4) aligns with findings from previous studies conducted in other European river basins.

Diclofenac, a widely used nonsteroidal anti-inflammatory drug (NSAID), was detected at concentrations reaching up to 132 ng/L in the present study (Table 4). This finding is consistent with reports from other major European rivers, such as the Rhine and Elbe, where diclofenac concentrations ranged from 50 to 200 ng/L [51,52]. Similarly, a study on the Danube River conducted as part of the Joint Danube Survey (JDS3, 2013) found diclofenac levels exceeding 100 ng/L in several sections, particularly near urban areas with high wastewater discharge [53].

Carbamazepine, an anticonvulsant known for its persistence in aquatic environments, was found in 28 out of 32 sampling sites in this study (Table 4), with concentrations averaging 18.05 ng/L. Research on the Elbe River, Czech Republic and its tributaries reported median carbamazepine concentrations ranging from 12 ng/L to 54.5 ng/L [54]. A study assessing human health risks associated with carbamazepine in surface waters of North America and Europe found that the 90th percentile measured environmental concentrations (MECs) ranged from 150 to 220 ng/L, while the 90th percentile predicted environmental concentrations (PECs) ranged from 333 to 658 ng/L [55]. The widespread presence of carbamazepine, despite its relatively low usage compared to NSAIDs, suggests its high resistance to conventional wastewater treatment processes and its potential to serve as a tracer of anthropogenic pollution.

Sulfamethoxazole and trimethoprim, commonly used antibiotics, were detected in 31 and 25 sites, respectively, with concentrations reaching 36 ng/L (SMX) and 12 ng/L (TMP) (Table 4). These findings are comparable to those reported in the Danube basin by [23], who identified sulfamethoxazole concentrations ranging from 10 to 50 ng/L in Romanian and Bulgarian sectors of the river. Additionally, a study on the Tiber River in Italy found sulfamethoxazole at similar concentrations (50–70 ng/L), suggesting that antibiotic contamination is a widespread issue in European riverine ecosystems [56].

Bisphenol A (BPA), a known endocrine disruptor, was found in all analyzed samples, with concentrations ranging from 34.5 to 342 ng/L (Table 4). These values are consistent with those reported in other studies, such as those from the Ebro River in Spain (50–450 ng/L) [57]. A comprehensive 19-year study analyzing 5057 samples from European and North American water bodies found that BPA was detected in 67% of the samples, indicating significant contamination. In European surface waters, reported concentrations varied between 7 ng/L and 300 ng/L, with a median of 29 ng/L in marine waters [58].

The synthetic estrogen 17α-ethinylestradiol (EE2) was also detected in the present study, in low concentrations (1.15–3.05 ng/L) (Table 4), which are consistent with the values reported for surface waters in the UK (1.5–5 ng/L), France (1.0–2.9 ng/L), and Greece (3.0 ng/L) [59]. A review of monitoring studies reports higher EE2 concentrations ranging from 0.2 ng/L in Spain to 101.9 ng/L in Portugal, with median values of 5.6 ng/L in Poland, 2.0 ng/L to 34 ng/L in Italy, 1.5 ng/L to 5 ng/L in the UK, and 0.8 ng/L to 17.2 ng/L in Germany [59]. These values exceed the proposed European environmental quality standard (EQS) of 0.035 ng/L, highlighting the potential risks for aquatic life.

The concentrations obtained in this work for sulfamethoxazole, ciprofloxacin, amoxicillin, and carbamazepine (Table 5) are lower than the values reported for JDS4 2019 survey of the entire river length (1.55–71.7 ng/L SMX; 8.53–24.87 ng/L CIP; 0.84–52.46 ng/L AMX; 0.13–57.6 ng/L CBZ) [33], while for trimethoprim and diclofenac, our values (Table 5) are higher than the JDS4 ones (<LOQ–1.78 ng/L TMP; 2.12–63.15 ng/L DCF) [33]. For dicloxacillin, the values are similar with the JDS4 findings (3.34–5.49 ng/L DCX) [33], with the average concentration presented in Table 5 (5.5 ng/L) being close to the maximum value of JDS.

While for the cumulative CECs concentrations in Danube water, an increasing spatial profile was detected in the three sites comprised in the JDS4 monitored sector in the SE part of Romania (JDS4-48 (Chiciu-Silistra), JDS4-50 (Reni), and JDS4-51 (Vylkovo, Chilia branch)) [31] (Annex Figure A, page 329), our results drawn in Figure S3a (Supplementary Materials) reveal a different trend, with a strong influence of emission sources related to large cities and Siret and Prut tributaries in Braila-Reni sector. When BPA concentrations are taken into account at calculation of the cumulative CECs concentrations (Figure S3b, Supplementary Materials), these influences are more pronounced, even for the river mouth corresponding to the Chilia RO-UA transboundary artery (site 22—Musura bay). These findings may suggest the necessity of establishing much denser regional monitoring networks in order to assess the contribution of a larger number of localities to the contaminant emissions.

3.4. Microbiological Contaminants

Heterotrophic bacteria were present in all of the samples examined (

Table 6. Bacteriological contamination of surface water in Danube River and Black Sea.

No.	Sampling Sites	Heterotrophic Bacteria, CFU/mL	Total Coliforms, MPN/100 mL
1	Ostrov ferry	1760	70,000
2	Ostrov, Danube old branch	12,000	5000
3	Fetesti	2700	7000
4	Cernavoda bridge	2240	6000
5	Cernavoda Seimeni	2300	25,000
6	Braila harbor upstream	390	1300
7	Braila harbor downstream	110	7000
8	Siret R. upstream	2150	60,000
9	Siret R. downstream	5700	70,000
10	Galati downstream	1900	2500
11	Galati shipyard downstream	250	11,000
12	Prut R. upstream Giurgiulesti	1210	250
13	Prut R. downstream	270	2500
14	Reni downstream	575	7000
15	Isaccea downstream	2050	2500
16	Ceatal Chilia	435	2000
17	Izmail downstream	525	10
18	Ceatal Sf. Gheorghe	531	1300
19	Chilia veche upstream	2750	7000
20	Chilia veche downstream	340	600
21	Sf.Gheorghe upstream	715	2000
22	Musura bay mouth	1700	250
23	Sulina mouth	125	250
24	Sf.Gheorghe mouth	950	2500
25	Sacalin	2400	6000
26	Gura Portitei	400	600
27	Corbu	65	130
28	Mamaia	72	120
29	Constanta	70	200
30	Costinesti	53	300
31	Mangalia	42	120
32	Vama veche	35	250
Microbiological pollution quality classes for bathing waters [60]		Indicator of organic pollution	Indicator of fecal pollution
Class I—Low		<500	<500
Class II—Moderate		500–10,000	500–10,000
Class III—Critical		10,000–100,000	10,000–100,000
Class IV—Strong		100,000–750,000	100,000–1,000,000
Class V—Excessive		>750,000	>1,000,000

). The samples taken from sites 2 (Ostrov, old Danube branch) and 9 (Siret downstream the confluence with Danube) showed the highest contamination level, with site 2 being classified in Class III—Critical according to the classification set up by ICPDR for bacteriological parameters in bathing water [60]. Heterotrophic bacterial counts in samples taken from sites with codes 3, 4, 5, 8, 10, 12, 15, 19, 22, and 25 show that in these areas the surface waters are also organically contaminated, being classified in the Class II (moderate pollution) of water quality (Table 6) [60]. Heterotrophic bacteria group includes common genus, such as Gram-negatives: Aeromonas, Acinetobacter, Alcaligenes, Citrobacter, Enterobacter, Flavobacterium, Klebsiella, Moraxella Pseudomonas, and Proteus, and Gram-positives: Bacillus and Micrococcus [61]. These bacteria are found in a large variety of aquatic environments, including water supplies, wastewater, seawater. Thus, the heterotrophic plate count is used as microbiological indicator of the water quality and the efficiency of water treatment procedures [14]. While heterotrophic bacteria are not generally included in risk groups, some of them are reported as opportunistic pathogens, such as Aeromonas spp. (gastroenteritis) and Pseudomonas spp. (skin and lung infections) [62].

Water containing total coliforms may be the result of natural processes rather than fecal contamination. In addition to human discharge fecal pollution, non-fecal sources like local flora and wildlife also contribute to the overall coliform population. In the water environment, coliforms, whether fecal or not, have the potential to not only survive but also to grow if the physical-chemical and biological conditions are favorable. While coliforms by themselves are generally not thought to pose safety risks, their presence in water microbiota suggests that fecal contamination may have taken place, and the presence of the pathogens should be taken into account [63]. In all analyzed samples, total coliforms were detected(Table 6). The sites with the highest levels of contamination were those coded 1,9,8,5, and 11 (Class III—Critical) [60], in which the MPN/100 mL of water was higher than 10⁴. The water collected from the site coded 17 presented the most reduced contamination level with total coliforms.

Both heterotrophic bacteria and total coliforms showed a negative correlation with pH, as shown by [64]. High concentrations of heterotrophic bacteria in water can lead to organic contamination or the existence of physical, chemical, and biological conditions that facilitate the growth of these bacteria. The concentration of total coliforms is correlated with contamination of waters by feces, and when present in a high concentration, it can increase the risks of pathogens spreading.

The amounts and complexity of water pollution with pharmaceuticals have a significant impact on the microbiota, particularly on the microorganisms that pose a risk to human and animal health. In addition, the pathogen bacteria Salmonella spp., Shigella spp., Yersinia spp., Escherichia coli, and all members of the coliform group (Enterobacteriaceae family) are important markers of contamination because they are resistant to environmental conditions. Multi-drug resistance and gene transfer pathways are also important indicators of water contamination in relation to pharmaceutical pollutants, type of compounds, and levels of contamination [65,66,67]. Biotechnology innovations based on next-generation sequencing (NGS) techniques, such as eDNA metabarcoding, have emerged as a valuable biomonitoring tool for pharmaceutical pollution, providing numerous benefits, including the ability to assess and predict pollution status, identify pollution sources, pollution dynamics, and assess the impact on the diversity and resistance of the microorganisms in water microbiota [68].

The correlation analysis between total coliforms (MPN/100 mL) and pharmaceutical concentrations (Table S4, Supplementary Materials) revealed significant associations, suggesting a common anthropogenic origin of both microbiological and chemical contamination. The highest positive correlations were observed for piroxicam (PIR, r = 0.648), ketoprofen (KET, r = 0.621), and naproxen (NAP, r = 0.546), all nonsteroidal anti-inflammatory drugs (NSAIDs) frequently excreted in unmetabolized form. These findings suggest that the occurrence of such pharmaceuticals may serve as indirect indicators of recent fecal pollution in aquatic systems. Moderate correlations were also found for sulfamethoxazole (SMX, r = 0.453), enilconazole (IMZ, r = 0.502), 17α-ethinylestradiol (EE2, r = 0.473), bisphenol A (BPA, r = 0.480), and amoxicillin (AMX, r = 0.570), highlighting the co-occurrence of antibiotics, antifungals, endocrine disruptors, and NSAIDs alongside fecal indicator bacteria. A multiple linear regression model (Table S4, Supplementary Materials) identified amoxicillin (AMX) and piroxicam (PIR) as significant predictors of total coliform concentrations, jointly explaining 80.4% of the observed variance (R² = 0.804; adjusted R² = 0.791). AMX and PIR resulted in strong predictive performance, possibly due to overlapping input sources (e.g., pharmaceutical residues from human excretion). Taken together, these results support the hypothesis that both pharmaceutical and microbial contaminations share a common source, particularly from insufficiently treated effluents.

As the sampling was carried out during the strict COVID-19 lockdown in 2020, which closed many activities, it is expected that the CECs levels recorded during this time interval were lower than in other periods [69]. This presumption, corroborated with the conclusion derived from another study related to the significant decrease in fecal pollution in all Danube sections in the period 2001–2019 [32], leads to the necessity of a more thorough monitoring of fluvial and marine chemical and microbiological pollution [34]. Moreover, pharmaceutics and heavy metals are widespread environmental contaminants considered potential markers of social and economic activities [70]. The scarcity and dispersion of the studies related to water-borne emerging and persistent pollutants and their synergies in lower Danube-Black Sea region [8,33,71] impose concerted seasonal monitoring campaigns in relevant sampling points.

4. Conclusions

Pollution of water bodies in large hydrographic basins from diverse sources, including urbanization, agriculture activities, untreated sewage, and industrialization, pose a great threat, with potentially severe ecological and human health impacts over time. Effective management and regulatory directives are crucial in order to mitigate this impact and protect water quality. Also, continued research and monitoring are essential to further understand the sources, pathways, and adverse effects of these contaminants, thereby ensuring a sustainable environment for both ecosystems and human populations.

The present study provides a comprehensive assessment of surface water contamination in the lower Danube River, the Danube Delta, and the Romanian coastal waters of the Black Sea, highlighting the presence of various pollutants, including heavy metals, nutrients, pharmaceuticals, endocrine disruptors, and microbial contaminants.

The findings indicate that certain sectors of the Danube River and its tributaries, particularly the Siret and Prut rivers, exhibit elevated levels of contamination. Industrial activities, untreated wastewater discharges, and agricultural runoffs contribute significantly to the presence of heavy metals and emerging contaminants in these areas. While some metals, such as arsenic and mercury, were found within acceptable ecological limits, others, including manganese, nickel, and lead, exceeded threshold values in specific locations, raising concerns about potential bioaccumulation and ecological impact.

The detection of pharmaceutical residues, particularly diclofenac, carbamazepine, and bisphenol A, confirms the increasing issue of emerging contaminants in surface waters. These substances, often originating from wastewater treatment plants, have been detected at levels that may pose ecological risks. Given their persistence and potential effects on aquatic organisms, continuous monitoring and improved wastewater treatment technologies are necessary to mitigate their impact.

Nutrient pollution, primarily from nitrates and phosphates, remains a major concern, as it contributes to eutrophication, leading to excessive algal growth and oxygen depletion. The presence of high nutrient concentrations in certain areas highlights the urgent need for better management of agricultural runoff and wastewater discharges to prevent further ecological degradation.

Microbiological contamination, evidenced by high levels of total coliforms in several sampling sites, suggests a significant impact from untreated sewage and wastewater effluents. The correlation between coliform concentrations and anthropogenic activities underscores the necessity of stricter wastewater management policies and improved sanitation infrastructure to safeguard both environmental and public health. The microbial adaptation of xenobiotics, their multi-drug resistance, and gene transfers are aspects that must be investigated by single strain analysis behavior in in vitro conditions, in order to establish the correlation between pollution and microbiota.

Overall, this study underscores the importance of continuous water quality monitoring and the need for effective measures to reduce pollution sources. Strengthening environmental policies, modernizing wastewater treatment facilities, and promoting sustainable agricultural practices are crucial steps toward ensuring the long-term protection of water resources in the Danube–Black Sea system. This work aims to establish a baseline for future monitoring of regional water quality, with the data obtained providing valuable insights for further research and policy development to preserve the ecological integrity of these interconnected aquatic ecosystems.

Ongoing work focuses on synthesizing pollutant data to highlight the spatiotemporal dynamics of water quality within the MONITOX network, including Danube—Black Sea sediment contamination with metals, radioelements, and other trace elements, as well as regional pollution mapping.