Effects of Water Pollution on Diatom Communities of Roșia Montană Mining Area, Romania

Abstract

We investigated the diatom communities and physicochemical water variables in the Abrud River catchment area (the Roșia Montană mining area, Romania) at 16 sampling sites, some of them impacted by acid mine drainage (AMD) and heavy metals. Diatoms serve as effective indicators of water characteristics owing to their ubiquity and sensitivity to environmental variables. This study aimed to enhance the understanding of the key environmental factors influencing the diatom flora of polluted rivers across various spatial and temporal scales, thereby informing the optimization of ecosystem management strategies. This work contributes to the knowledge of Romanian diatom flora through the identification of 274 taxa belonging to 63 genera, including 35 taxa recorded for the first time in the country. The spatial and temporal variations in the species richness patterns highlighted the effects of water pollution resulting from past mining activities, revealing distinctions between the main Abrud River stream and its tributaries, some of which exhibited extremely low species richness with few or no identified taxa. This contrasted sharply with the cleaner upstream waters of the Roșia Valley, where a notably rich diatom community (85 taxa) persisted, highlighting the severe localized impact of mining discharges on biodiversity.

1. Introduction

Diatoms are unicellular algae inhabiting world aquatic and terrestrial ecosystems, playing a key role in the biogeochemical cycles of the planet [1,2,3,4]. The health and integrity of these diatom communities are vital for maintaining aquatic biodiversity and ecosystem services, which are fundamental pillars of environmental sustainability. Benthic diatoms, in particular, are widely recognized for their utility in biomonitoring programs due to their sessile nature, rapid response to environmental changes, and diverse ecological preferences [5,6]. Understanding their response to stressors such as pollution is, therefore, crucial for developing sustainable water resource management strategies and achieving broader sustainability goals outlined in frameworks like the UN’s Sustainable Development Goals (SDGs), particularly SDG 6 [7]. Their siliceous cell walls (frustules) preserve well in sediments, providing valuable paleoecological records [8]. Assessments relying solely on a limited number of indicator species may not accurately reflect the true status of aquatic ecosystems; thus, monitoring the entire community is strongly recommended [9]. Modern, rapid techniques are currently being developed for water quality assessment. For instance, diatom metabarcoding enables faster, more sensitive water quality assessment with a higher sample throughput compared with traditional microscopy-based methods [10]. Consequently, diatoms are integral components of water quality assessments globally, including within the European Union’s Water Framework Directive (WFD; 2000/60/EC), where they are recognized as key organisms integrated into water quality assessment protocols. The WFD mandates the use of such biological indicators for assessing the ecological status of surface waters [9,11,12].

Algal communities reflect the physical, chemical, and biological characteristics of aquatic ecosystems through the presence/absence of species and the growth or decline in populations, among other responses to environmental change.

The Abrud River catchment area in Romania, significantly impacted by historical mining activities, provides a relevant case study for investigating the impacts of acid mine drainage (AMD) on diatom communities. Mining in Roșia Montană has a history spanning over two millennia, with significant industrial activity starting in the latter half of the twentieth century. Specific quantitative data on water quality indices and microalgae communities before these major phases of mining are limited. AMD originates from the oxidation of metal sulfides exposed during mining is typically characterized by a low pH and elevated concentrations of heavy metals (e.g., iron, aluminum, and manganese), known to be toxic to aquatic biota. Previous studies have demonstrated that AMD can significantly alter the composition of aquatic communities, often leading to a dominance of acidophilic species and a reduction in overall diversity [13,14].

Although the features of the Abrud River catchment area provide interesting research material for biologists, geographers, and geologists, there are few studies focusing on diatoms, and this particular group is treated rather tangentially compared with other bioindicators, like macroinvertebrates. In this context, the present study aimed to be a contribution to the knowledge of diatom communities in the Abrud River and four of its major right-bank tributaries (Vârtop Valley, Roșia Valley, Săliște Valley, and Corna Valley).

Therefore, the primary objective of this study was to determine the distribution patterns of benthic diatom communities in the Abrud River and its main tributaries affected by mining pollution. While identifying the specific factors causing observed declines necessitates further investigation, the response of the algal community serves as an indicator of water quality impairment. Ultimately, this research sought to provide data that can inform sustainable environmental management and restoration efforts in aquatic ecosystems impacted by anthropogenic activities, contributing to the long-term ecological health and resilience required for sustainability. A secondary objective was to determine the ecological profiles of the most abundant or characteristic species found in the study area concerning key environmental variables (pH, conductivity, temperature, and nitrates), contributing to the autoecological knowledge base for regional biomonitoring.

2. Materials and Methods

2.1. Study Area

This study was conducted in the Abrud River catchment area, situated within the Metaliferi Mountains, a sub-region of the Southern Apuseni Mountains, encompassing the Abrud River drainage basin. The region is characterized by its significant polymetallic ore deposits, historically exploited for gold and silver, leading to substantial environmental impacts from mining waste. The primary watercourses draining this catchment area are the Corna, Săliște, and Roșia streams, all tributaries of the Abrud River.

The geology is complex, dominated by Cretaceous sedimentary rocks and Neogene volcanic formations. The area is characterized by a moderately steep mountainous topography, with elevations ranging from 700 to 1000 m above sea level (a.s.l.), where rainwater serves as the primary source of groundwater recharge. Acid mine drainage (AMD) generated from waste rock accumulated in mine tailing ponds contaminates all environmental compartments within the Roșia Montană complex [15]. The rivers of this basin ultimately discharge into tributaries of the Danube River [16,17], presenting a significant challenge for environmental management. The river discharge increases due to low rock permeability and flow convergence. The average discharge rates during the 2001–2003 period were 0.16 m³/s in the Roșia Valley, 0.07 m³/s in the Corna Valley, and 0.16 m³/s in the Săliște Valley. The climate is continental temperate, influenced by altitude, with mean annual temperatures ranging from −4.7 °C to 16.9 °C. The annual precipitation ranges from 700 to 800 mm, consisting mainly of rain (~75%) and snow (~25%) [18].

Sampling locations were selected to assess the impact of mining operations on surface watercourses, establishing a total of 16 sampling points located at Cărpiniș, Roșia Montană, Abrud, Bucium Șasa, and Bucium-Sat (Alba County) (Figure 1). The sampling sites were strategically selected along the Abrud River and its main tributaries impacted by the Roșia Montană mining activities. The site selection aimed to represent (i) upstream conditions presumed to be less impacted (e.g., V.R.1), located above major known discharges, (ii) locations directly downstream of significant historical mining pollution sources, such as acid mine drainage (AMD) seeps or waste rock dumps (e.g., V.R.2 and V.R.3) in the highly impacted Roșia Valley, (iii) points along the main Abrud River course to assess the cumulative impact and potential downstream recovery gradients (Ab.1–Ab.6), and (iv) major tributaries known to receive different types or loads of contaminants (Vârtop, Săliște, and Corna Valleys), thus contributing differentially to the main river’s water quality.

Sampling campaigns were conducted seasonally over a two-year period, from spring 2013 to autumn 2014. This resulted in approximately 6 samples per site and a total of 96 samples analyzed for diatom communities. The sampling data were adjusted as much as possible to the beginning of each season. Winter samplings were excluded due to issues arising from adverse weather conditions and considerations regarding the physiology of the microalgae. Sampling across different seasons over two years aimed to capture variations related to hydrological cycles and temperature changes, known factors influencing both pollutant dynamics and diatom community development.

2.2. Environmental and Biological Data

Benthic diatoms were collected from stones (epilithon) according to standard protocols [19]. At each site, two samples were obtained by brushing the surface of several submerged rocks from representative microhabitats within a defined river reach. The collected algal material underwent a standardized preparation process to clean the diatom frustules. The samples were treated with strong oxidizing agents (30% H₂O₂ and concentrated HCl) to digest the organic matter, thereby isolating the silica frustules [19]. The cleaned diatom frustules were mounted onto microscope slides using Naphrax^® (Brunel Microscopes Ltd., Chippenham, UK) mounting medium (refractive index = 1.74).

Diatom identification and counting were performed using an Olympus BX60 (Olympus Corporation, Tokyo, Japan) light microscope (LM) equipped with phase contrast and differential interference contrast (DIC) optics at 1000x magnification. At least 400 valves were counted per slide to ensure representative data. The taxonomic identification followed standard European floras and the recent taxonomic literature, e.g., [11,20]. For critical taxa or confirmation, subsamples were prepared for scanning electron microscopy (SEM). The subsamples were mounted on stubs, sputter-coated with gold or carbon, and examined using a Carl Zeiss Merlin SEM (Carl Zeiss AG, Oberkochen, Germany) equipped with energy-dispersive X-ray spectroscopy (EDX) capabilities.

Water Physicochemical and Heavy Metal Analyses

Physicochemical water quality parameters were assessed both in situ and in the laboratory. Field measurements for pH, temperature (T), electrical conductivity (EC), and dissolved oxygen (DO) were conducted using a WTW Multi 340i or WTW Multi 3630 IDS portable multiparameter probe (WTW GmbH, Weilheim, Germany). Laboratory analyses were performed using a Hach Lange DR3900 spectrophotometer (Hach Company, Loveland, CO, USA) to determine the turbidity and nutrient concentrations. Specifically, nitrate (NO₃⁻-N), sulfate (SO₄²⁻), and silicate (SiO₃) were quantified using dedicated Hach Lange kits LCK339 (Hach Company, Loveland, CO, USA) (measurement range: 0.23–13.5 mg/L), LCK349 (measurement range: 0.05–1.5 mg/L), LCK153 (measurement range: 40–150 mg/L), and LCK351 (measurement range: 0.01–1.0 mg/L), respectively, following the manufacturer’s protocols. For the heavy metal analysis (Fe, Mn, Zn, Cu, Cd, Pb, and As), water samples were filtered through 0.45 µm pore size membranes and subsequently acidified to pH < 2 using nitric acid (HNO₃). The metal concentrations were determined by flame atomic absorption spectrometry (FAAS) utilizing a Shimadzu AA-7000 instrument. The calibration of the AAS was performed using standard curves generated from serial dilutions of 1000 mg/L Merck mono-element standard solutions for each analyzed metal.

To facilitate the comparison between the diatom community distribution and environmental conditions, a water quality index (WQI), adapted from the method proposed by Nasirian [21] for mining effluents, was calculated. The WQI incorporates multiple physicochemical parameters relevant to AMD impacts (e.g., pH, sulfate, and heavy metals). The WQI was calculated using the following formula:

WQI = 0.18 (SI pH) + 0.13 (SI NO₃⁻) + 0.11 (SI DO) + 0.11 (SI Heavy Metals) + 0.08 (SI Electrical Conductivity) + 0.08 (SI SO₄²⁻) + 0.08 (SI Solid Suspended)

The phosphate parameter and radionuclides with their associated weighting factors, typically part of the index, were omitted from our calculation due to a lack of data. However, these two parameters contribute very low scores to the general formula. The water categories (e.g., very bad, bad, moderate, and good) were assigned based on the NSFWQI [22].

2.3. Ecological Profile

To develop ecological profiles potentially applicable to a regional biotic index, the following steps were undertaken:

(1)

The selection of environmental variables: Based on their relevance in AMD contexts [23] and preliminary analyses, pH, electrical conductivity, temperature, and NO₃⁻ concentration were selected as the key environmental variables (corresponding to the X-axis in the conceptual response curves; cf. Figure 2).

(2)

Field survey: The diatom communities (bioindicators) were sampled concurrently with the in situ measurement of the selected environmental variables at each site. Two replicates were obtained from each sampling point.

(3)

The calculation of ecological optima and tolerances: The ecological optimum (weighted average abundance) and tolerance for each species with respect to each selected environmental variable were calculated using weighted averaging (WA) methods [24]. The optimum [Zn] was calculated using the Zelinka–Marvan formula:

$$\left[Zn\right]={\sum }_{j=1}^n{A}_j.S_j.V_j/{\sum }_{j=1}^n{A}_j.V_j$$

where A_j, S_j, and V_j are, respectively, the abundance, optimum, and tolerance, of the j^th taxon in the sample.

Figure 2. Model of tolerance adapted from Shelford’s law of tolerance [25].

Figure 2. Model of tolerance adapted from Shelford’s law of tolerance [25].

Statistical analyses, including correlation analysis and diversity calculations, were performed using software packages including OMNIDIA 8.1 [26] and PAST 2.17. The correlations between biological metrics (richness and abundance) and environmental variables (WQI and heavy metals) were tested using appropriate methods (e.g., Pearson or Spearman correlation, depending on the data distribution). The geometric mean of the heavy metal concentrations (Cu, Cr, Cd, Zn, Ni, and Pb) was calculated for each sample to provide a single integrated metric representing the overall metal load, a common approach when dealing with environmental contaminant data that can span several orders of magnitude [27].

3. Results

3.1. Water Quality

Table 1. Average water quality index (WQI) range and corresponding water quality classification for each sampling site.

Sites	WQI Range	Quality
V.A.1	30–40	Bad
V.A.2	40–50	Moderate
V.C.1	30–40	Bad
V.C.2	50–60	Moderate
V.S.1	50–60	Moderate
V.S.2	40–50	Moderate
V.R.1	50–60	Moderate
V.R.2	20–30	Very bad
V.R.3	20–30	Very bad
V.V.	50–60	Moderate
Ab.1	40–50	Moderate
Ab.2	50–60	Moderate
Ab.3	50–60	Moderate
Ab.4	40–50	Moderate
Ab.5	50–60	Moderate
Ab.6	40–50	Moderate

presents the calculated water quality index (WQI) values for each sampling point, averaged across the two sampling campaigns. The water quality ranged from ‘very poor’ to ‘moderate’ according to the WQI classification. Four sampling sites (V.R.2, V.R.3, V.S.1, and V.C.1) exhibited a ‘very poor’ quality, corresponding to locations in close proximity to AMD sources or downstream from significant mining impacts. Notably, sampling site V.R.1, situated upstream from known discharge points in the Roșia Valley, was classified as having a ‘moderate’ water quality. Other sites along the main Abrud River generally showed a ‘poor’-to-‘moderate’ quality, reflecting dilution effects but persistent contamination. The average physicochemical data per sampling site are included in Supplementary Table S1.

3.2. Diatom Flora Composition and Richness

A total of 274 diatom taxa were identified during the study period across all sampling sites. These 274 taxa belonged to 63 genera, comprising 264 species, 8 varieties, 1 subspecies, and 1 form. The most speciose genera were Nitzschia and Gomphonema, with 31 and 27 species identified, respectively, followed by Navicula, Pinnularia, and Eunotia. Conversely, 24 genera were represented by only a single species. Figure 3 illustrates the species distribution among the dominant genera.

The species richness fluctuated considerably across sites and seasons (Figure 4). Extremely low richness (0–4 taxa per sample) was observed consistently at sites V.R.2 and V.R.3 (downstream of the Roșia Valley) throughout the two-year study period. In contrast, sampling site V.R.1 (upstream of the Roșia Valley) exhibited high richness, peaking at over 70 taxa in summer 2014. On average, the sampling sites along the main Abrud River channel (Ab.1–Ab.6) exhibited the highest species richness, except for site Ab.4, situated immediately downstream of the Săliște Valley confluence. Conversely, tributaries generally supported poorer diatom communities, with the notable exception of V.R.1, located upstream of the primary contamination zone.

Over the two-year study period, a slightly increasing trend in overall species richness was observed. The average number of species per site increased sequentially from 12.2 in spring 2013, to 14.2 (summer 2013), 22.6 (autumn 2013), 23.6 (spring 2014), and 23.8 (autumn 2014).

The species distribution patterns provide further detail beyond the overall richness trends (Figure 5). Across all samples, the most abundant species, based on the average relative abundance, were Achnanthidium minutissimum (11.5%), Navicula lanceolata (7.9%), Achnanthidium macrocephalum (7.4%), Nitzschia linearis (6.0%), and Cocconeis euglypta (5.0%).

Sampling sites along the Abrud River (Ab.1–Ab.6) generally exhibited similar diatom community structures, with the exception of the species-poor site Ab.4. At three points (Ab.1, Ab.2, and Ab.6), the dominant species were Achnanthidium macrocephalum, Achnanthidium minutissimum, and Navicula lanceolata. At the other Abrud River sites, Achnanthidium macrocephalum and Nitzschia palea (Ab.3); Navicula lanceolata and Nitzschia linearis (Ab.4); Navicula lanceolata, Achnanthidium minutissimum and Ulnaria ulna, together with Nitzschia linearis (Ab.5) were among the most abundant taxa. The tributary communities were highly variable. V.R.1 showed high diversity, dominated by species typical of clean, circumneutral waters, while V.R.2 and V.R.3 were extremely species-poor, dominated by a few acid-tolerant taxa like Cocconeis euglypta and Navicula lanceolata, albeit at low abundances.

Pollutant impacts on phytobenthic communities manifested in both species composition and population abundance. A significant positive correlation (r² = 0.87; p < 0.05) was found between the logarithm of total diatom abundance (cell counts) and the logarithm of species richness per sample (Figure 6).

3.3. Diatoms and Environmental Conditions

The total heavy metal concentration (expressed as a geometric mean) and species richness exhibited a statistically significant but moderate negative correlation (r² = 0.28 and p < 0.05; Figure 7), suggesting that while metal pollution contributed to the decline in diversity, it explained only a portion of the observed variance.

Furthermore, the ecological profiles of the five most abundant species were determined based on their weighted average optima and tolerances for pH, electrical conductivity, temperature, and the NO₃⁻ concentration (Figure 8a,b). Their optima and tolerances relative to these physicochemical variables were calculated using the WA method [28]. Regarding the pH optima, following Hustedt’s [29] classification, Cocconeis euglypta was profiled as ‘acidobiontic’, while Achnanthidium minutissimum and Navicula lanceolata were profiled as ‘acidophilic’. Achnanthidium macrocephalum showed a broad tolerance around neutral pH, classifying it as ‘indifferent’. Nitzschia linearis was profiled as ‘alkaliphilic’.

With respect to electrical conductivity, all five species were profiled as ‘oligohalobous’ (low-salinity-tolerant) according to the classification framework of Stoermer and Smol [6]. However, Cocconeis euglypta’s tolerance range for conductivity was relatively broad, suggesting a more euryhaline character compared with the other four species. Regarding temperature, all five species exhibited eurythermic profiles (tolerant of a wide temperature range). Finally, concerning nitrate (NO₃⁻) concentrations, Nitzschia linearis and Achnanthidium minutissimum demonstrated notably wide tolerance ranges. These two species were also the most ubiquitous diatoms across the study area.

4. Discussion

4.1. Water Quality and Diatom Community Patterns

Our results show that the distribution patterns of the benthic diatom communities in the Abrud River and its main tributaries have been affected by mining pollution. This disruption of foundational biological communities not only signifies a loss of biodiversity but also indicates a compromised ecosystem functionality, thereby undermining the ecological sustainability of these water bodies. Such impacts can have cascading effects on higher trophic levels and reduce the capacity of these rivers to provide essential ecosystem services, which are critical for regional sustainability. Analysis of the WQI results (Table 1) did not reveal straightforward correlations with all biological metrics. However, it is noteworthy that the four sampling sites with the lowest WQI scores (‘very poor’ quality) also exhibited a low species richness (cf. Figure 4) and, counterintuitively, displayed the lowest relative abundances of teratological (deformed) diatom frustules. This finding contrasts with some literature reports where teratologies increased under stress [30,31,32,33]. However, it is plausible that severe deformities, potentially impairing cell motility or physiological functions, may hinder the survival and proliferation of affected individuals in extremely harsh environments, leading to their underrepresentation [23,34].

The use of diatoms for assessing environmental conditions has a long history, primarily employing two approaches: (i) autoecological indices [24,35], which leverage the known ecological preferences and tolerances of individual species, and (ii) diversity metrics [36], using community richness and evenness as indicators of ecosystem health. Both approaches have been widely applied. This study investigated the relationship between diatom community characteristics (e.g., richness and abundance) and environmental stressors, particularly heavy metal concentrations.

4.2. Influence of Heavy Metals and Acid Mine Drainage

As reported (Section 3.3), a significant negative correlation was found between the geometric mean of the heavy metal concentrations and the log-transformed species abundances (Figure 7). This moderate correlation indicates that while heavy metals are a significant stressor, other factors likely play substantial roles in structuring the diatom communities in this complex environment. Using the geometric mean to integrate the concentrations of multiple metals into a single representative value is a methodological approach employed in previous contamination research [27]. However, we acknowledge that using an aggregated metric like the geometric mean simplifies the complex reality of metal mixtures in the environment. It does not account for potential synergistic or antagonistic interactions between different metal species, which could also influence diatom community responses.

Our study observed a significant negative correlation between heavy metal concentrations and diatom species richness in the sampled sites impacted by mining activities. This finding aligns with the well-documented detrimental effects of heavy metals on algal communities [37,38,39,40,41]. Several physiological and ecological mechanisms likely contribute to this pattern. Physiologically, heavy metals can induce direct toxicity, leading to cellular damage, the inhibition of key enzymes involved in photosynthesis and nutrient uptake, or oxidative stress through the generation of reactive oxygen species (ROS) [42]. For instance, metals like cadmium and copper are known to disrupt photosynthetic processes and damage cell membranes, even at low concentrations [43,44]. Ecologically, the increased metal load likely acts as a strong selective pressure, favoring a few tolerant species while excluding many sensitive taxa, thus reducing overall richness [45]. Comparing these results with findings from other mining regions is crucial for a broader context. Similar reductions in diatom diversity linked to heavy metal contamination have been reported in mining-impacted rivers and streams globally, such as by Luis et al. (2013) [13,46], although the specific metal concentrations and community responses can vary depending on the local geochemistry and species. This knowledge is directly applicable to the development and refinement of biomonitoring tools, which are indispensable for cost-effective, long-term environmental stewardship and for tracking the efficacy of remediation efforts aimed at achieving sustainable water quality standards. Further research could explore the specific tolerance mechanisms of the dominant taxa found in our most-contaminated sites. In light of this negative correlation, the observed diatom distribution patterns within the Abrud catchment area appear to be strongly influenced by AMD inputs (particularly in the Roșia Valley) and leaching from toxic materials in waste rock dumps affecting tributaries on the right bank of the Abrud River (namely, the Vârtop, Roșia, Săliște, Corna, and Abruzel Valleys).

The composition of these waste rocks is known to be heterogeneous [47], resulting in significant variability in their potential to generate AMD and release toxic elements (including heavy metals). Servida et al. [47] identified two main types of waste rock in the Roșia Montană area: one primarily dacitic type, with a high potential for generating AMD and leaching heavy metals, and another predominantly andesitic type, exhibiting a greater acid-neutralizing capacity. According to Servida et al. [47], the waste rock dumps with the highest toxicity potential are located in the Carnic and Cetate Hills, draining toward the Roșia Valley. This aligns with our findings of poorly developed diatom communities at the downstream sites V.R.2 and V.R.3 in this valley. The drainage from these waste materials constitutes the primary pollution source for the Roșia Valley, resulting in highly acidic conditions (with reported pH values of 2.7–3.5) and elevated concentrations of Zn, Cd, Ni, Cu, Cr, Pb, and SO₄²⁻, as documented by Bird et al. [48] and RMGC [49].

It is important to point out the potential influence of natural seasonal cycles and hydrological factors (rainfall and river flow) on diatom communities. It is well known that these factors generally play a significant role in structuring aquatic communities. While our study focused primarily on the impact of the strong pollution gradient originating from mining activities, we acknowledge that natural seasonality and hydrology likely interact with these impacts. Concurrently, detailed hydrological data were not available for the sampling period, limiting our ability to quantitatively assess their specific contributions. However, the observed spatial patterns, particularly the drastic differences between upstream and downstream sites within the same tributaries, strongly suggest that pollution (AMD and heavy metals) is the dominant driver shaping the diatom communities in the most affected areas, potentially masking or overriding typical seasonal succession patterns.

4.3. Spatial Heterogeneity and Responses of Key Taxa

Reflecting this polluted environment, significant spatial and temporal (seasonal and potentially inter-annual) dynamics were observed in the benthic diatom communities within the study area, affecting both species richness and relative abundances. For instance, Achnanthidium macrocephalum, one of the most abundant species in the main Abrud River channel, was nearly absent from the most heavily contaminated tributary sites. Regarding species richness, a sharp decline was noted between point Ab.3 (the richest site on the main river, with 102 taxa identified over the study period) and Ab.4 (only 38 taxa), located approximately 500 m downstream. This sharp drop in richness is explicitly linked to the potential influence of the Sălişte tributary and possible changes in water chemistry. Although classified as having a ‘moderate’ WQI, it may introduce specific toxic contaminants or alter physicochemical conditions (e.g., conductivity or specific metal ratios) in a way that excludes many species present at Ab.3. The shift in the dominant species, with Nitzschia linearis becoming prominent at Ab.4, supports this, as this species showed a broader tolerance to certain stressors like nitrates and, potentially, others not fully captured by our calculated ecological profiles.

As noted previously, the community composition also differed markedly between these two adjacent sites. The elevated species richness at Ab.3 might be partially attributed to nutrient enrichment from urban discharges originating upstream from the town of Abrud. Concurrently, the confluence with the Săliște Valley tributary just upstream of Ab.4 likely contributes to the poorer water quality and lower species richness observed at this point. The Săliște stream itself supports a relatively poor diatom community and has an average discharge of 0.16 m³/s [36].

Diatom communities in the tributaries draining the Roșia Montană mining area exhibited distinct characteristics compared with those in the main Abrud River. As illustrated by the species richness data (Figure 4), these tributary communities were generally species-poor (except V.R.1 and, to a lesser extent, V.V. and V.C.2) and displayed high variability in their species composition, even within the same tributary, such as the Roșia Valley, where community heterogeneity between upstream (V.R.1) and downstream (V.R.2 and V.R.3) sites was pronounced. At the severely impacted sites V.R.2 and V.R.3, the cumulative number of taxa recorded over the study period was low (eight and six taxa, respectively), suggesting only a few species could tolerate these extreme conditions. The extremely acidic conditions measured/reported in Roșia Valley (pH 2.7–3.5) directly explain the dominance of acidobiontic taxa like Cocconeis euglypta and the overall extremely low species richness observed at sites V.R.2 and V.R.3. Interestingly, the taxonomic compositions of the sparse communities at these two adjacent downstream sites differed considerably. Species such as Cocconeis euglypta, Navicula lanceolata, Gomphonema olivaceum, and Navicula exilis Kützing were characteristic dominants in these low-diversity communities, although their overall abundances remained low.

Two of the most species-rich genera found in this study were Nitzschia and Gomphonema, with 31 and 27 species, respectively. This observation aligns with the findings reported by Sabater [50] in the Guadiamar River (Spain). The study identified Nitzschia palea and Gomphonema parvulum as key indicator species, whose dominance signified a marked decrease in water quality, particularly in the zones immediately downstream from the pollution source, where heavy metal concentrations in the sediment were notably high.

The strong positive correlation found between log(abundance) and log(richness) (Figure 6) potentially reflects the generally disturbed nature of the aquatic ecosystems in this region, where stable, high-density communities may be uncommon. Deviations from this linear relationship might occur in less disturbed or nutrient-enriched conditions, where competitive exclusion could lead to less diverse but highly abundant communities. Some data points (e.g., samples 8, 27, and 28, corresponding to V.C.2 and upstream Abrud River sites Ab.1/Ab.2 in spring/summer 2013) may represent such conditions, plotting slightly above the general trend line (Figure 6). Overall, however, the significant linear correlation across most samples suggests that factors limiting richness also limit abundance in this system. Furthermore, the observed spatial patterns of diatom communities are inevitably linked to the hydrogeochemical transport dynamics of contaminants within the catchment. Factors such as variable flow rates, water residence times, sediment interactions, and the specific locations and release rates from heterogeneous sources like waste rock dumps control the downstream evolution of pollutant concentrations. Detailed transport modeling was beyond the scope of this study, including incorporating frameworks that describe advection–dispersion processes in heterogeneous media, such as those proposed by Ding et al. [51].

4.4. Contribution to Romanian Diatom Flora

Based on available literature surveys of Romanian diatom flora [52,53,54,55,56,57,58,59], this study identified 35 taxa not previously recorded in Romania, along with 7 varieties and 1 form, also representing new records for the country. These new records constitute approximately 13% of the total taxa identified in this study. Previous research has highlighted the diversity and heterogeneity of freshwater benthic diatom assemblages in Romania [52,53,54,55,56,57,58]. While numerous studies have examined diatom communities within the broader Arieș River catchment, few have specifically included sampling sites on the Abrud River itself. Notably, two studies by Momeu et al. [60,61], primarily focused on the Arieș River, reported finding few or no diatoms (‘no relevant taxa’) at their sampling points located on the Abrud River. Similarly, Szekely-Andorko et al. [62], studying the Arieș River catchment area, listed approximately 100 taxa from a single sampling site on the Abrud River. Finally, a broader limnological investigation by Battes et al. [63] identified 58 diatom species in samples collected within the Roșia Montană area. Regardless, the substantial number of new national records generated by this study underscores that the Romanian freshwater diatom flora remains incompletely characterized, warranting further taxonomic and ecological investigations. Our study pointed out that these new records significantly enhance the known diatom flora of Romania and underline the need for further biodiversity surveys.

5. Conclusions

The diatom species richness patterns in the studied region are primarily driven by pollution impacts, specifically acid mine drainage (AMD) and associated heavy metals originating from mining activities. The spatial distribution of the diatom communities clearly reflects the gradient of contamination, with extremely impoverished communities in heavily impacted tributaries (e.g., downstream of the Roșia Valley) and richer, more diverse assemblages in upstream reference sites (V.R.1) and along the main Abrud River where dilution occurs. While seasonal variations were observed, they did not alter the fundamental spatial pattern of the pollution impact, which remained consistent across seasons. The identification of numerous new diatom records for Romania highlights the need for continued floristic research in the country’s diverse aquatic ecosystems. The ecological profiles generated for dominant species provide valuable autoecological data that can refine diatom-based water quality assessment tools for rivers affected by mining pollution in this region and, potentially, elsewhere.

This study confirms the utility of benthic diatoms as sensitive indicators of complex mining-related pollution impacts. These findings paint a stark picture of the environmental cost of unsustainable industrial practices and highlight the urgent need for interventions that can steer these ecosystems back toward a trajectory of ecological recovery and long-term sustainability.