Revealing the Protective Dynamics of an Ecologically Engineered Wetland against Acid Mine Drainage: A Case Study in South Africa

Abstract

This study investigated the Zaalklapspruit valley bottom wetland in South Africa, an ecologically engineered site influenced by acid mine drainage (AMD) from a defunct coal mine upstream. Conducted in 2022, the research aimed to elucidate the dynamics of contaminant dispersal within this wetland, focusing on the sources, pathways, and receptors of metals and sulfur compounds. The analysis revealed that the wetland’s bottom sediment is rich in organic material, with pH values ranging from 6.05 to 6.59 and low oxidation-reduction potentials reaching −219.67 mV at Site S3. The significant findings included the highest adsorption rates of manganese, contrasted with iron, which was primarily absorbed by the roots of Typha capensis and the algae Klebsormidium acidophilum. The macrophyte rhizospheres were found to host diverse microbiota, including families such as Helicobacteraceae and Hydrogenophilaceae, pivotal in metal and sulfur processing. This study highlighted the complex biogeochemical interactions involving sediment, macrophyte root systems, periphyton, and microbial populations. These interactions demonstrate the efficacy of ecologically engineered wetlands in mitigating the impacts of acid mine drainage, underscoring their potential for environmental remediation. Importantly, the sustainability of such interventions highlights the need for community involvement and acceptance, acknowledging that local support is essential for the long-term success of ecological engineering solutions that address environmental challenges like AMD.

1. Introduction

Since the 1970s, natural and constructed wetlands have been effectively employed as passive treatment systems for acid mine drainage (AMD) water, capitalizing on their natural capacity to buffer, filter, and cleanse polluted water [1,2]. The complex interaction among the biotic and abiotic components in wetlands—including soil, water dynamics, and organisms—creates ecosystems that are both biodiverse and nutrient-rich [3,4]. These components act as reservoirs for particles and pollutants, which accumulate in the bottom sediment as water flows through the system [5]. Wetlands facilitate several processes, such as flocculation, sedimentation, sorption, ion exchange, precipitation, oxidation, reduction, plant absorption, and microbial metabolism. These mechanisms play a crucial role in the sequestration and transformation of metals from AMD, rendering these ecosystems highly effective in managing and remediating emerging pollutants [6,7]. Nonetheless, it is essential to thoroughly assess the resilience of natural wetlands against the severe conditions induced by AMD to maintain their ecological integrity [8].

The remediation of AMD poses significant challenges for the mining sector and environmental professionals worldwide, attributed to its severe impacts and variable site characteristics [9,10]. AMD originates from the oxidation of pyrite (FeS₂) when water interacts with metals and coal in mining operations, which results in the production of sulfuric acid, hydrogen ions, and mobilized metals [9,11].Water affected by AMD typically exhibits low pH levels and elevated concentrations of harmful elements and metals such as aluminum (Al), iron (Fe), manganese (Mn), and sulfate (SO₄²⁻) [11]. Such contamination may spread to nearby aquatic systems, consequently degrading soil and plant life and presenting substantial risks to public health and safety when the contaminated water is used for consumption [12]. The impact of AMD on water quality extends to both proximal and distant ecosystems, with the potential for water resource contamination lingering for decades post-mining operations [13,14].

The treatment of AMD water typically focuses on elevating the pH through neutralization and is divided into two fundamental approaches: active and passive treatment systems [9,15]. Active treatment strategies involve the deliberate addition of chemical alkaline agents to promote metal precipitation, whereas passive treatments harness natural processes—including physical, chemical, biological, and geological methods—to alleviate acidity [16,17,18,19]. Among these, passive treatment systems are lauded for their environmental compatibility and their added ecological advantages, positioning them as essential elements of sustainable environmental practices [15,20,21].

Ecosystem restoration involves reverting degraded ecosystems to conditions that resemble their original natural states, incorporating strategies such as wetland restoration, regeneration and rehabilitation [2,22]. Within the realm of environmental management, ecological engineering is pivotal to these restoration and even regeneration efforts [23]. Adopting a holistic and non-linear perspective, ecological engineering applies engineering solutions to environmental challenges in a sustainable manner [24,25]. A key concept in this field is ecosystem self-design, which allows ecosystems to self-regulate and adapt through strategic interventions to optimize their functionality [24,25]. This method promotes sustainable coexistence between human activities, such as mining, and natural settings, including wetlands [23]. While ecological engineering principles, especially those applied to wetlands treating AMD, have been explored in previous research, their full potential and mechanisms remain incompletely understood [24,26,27].

Moreover, there is a pressing need for comprehensive studies on ecologically engineered wetlands that consider the entire ecosystem in their approach [27]. The Zaalklapspruit wetland in South Africa, a valley bottom wetland severely impacted by AMD from an abandoned coal mine upstream, has demonstrated improved surface water quality following the implementation of ecologically engineered structures [8,28]. The concrete structures were designed to decrease the surface water flow rates, increasing the residence time of contaminated water within the wetland. Although influent–effluent comparisons indicated noticeable improvements in water quality parameters, a deeper understanding of wetland dynamics necessitates investigation into the sources, pathways, and receptors of metals and sulfur compounds. This approach is essential for developing appropriate remediation designs and management strategies, ensuring the effectiveness of ecologically engineered wetland treatment, as shown in Figure 1 [29]. An integrated source–pathway–receptor approach to wetland-based treatment systems is crucial for effectively designing, implementing, and managing ecologically engineered ecosystems.

Considering the pivotal role of mining in South Africa’s economy, environmental degradation is an almost inevitable consequence [8]. The imperative to restore equilibrium between resource utilization and its exploitation is critical. While ecological engineering has been under study since the 1960s, there remains a significant gap in the research concerning the efficacy and sustainability of wetland ecosystems engineered to mitigate the negative impacts of AMD. This research aims to delve into the dynamics within an ecologically engineered wetland affected by AMD, with several key objectives: (1) to evaluate the water quality improvement capabilities of the wetland; (2) to investigate the physical and chemical properties of the wetland’s bottom sediment; (3) to examine the metal adsorption and absorption capacities of macrophyte roots and periphyton; (4) to explore the microbiological diversity within the sediment; and (5) to identify the sources, pathways, and receptors of metals and sulfur compounds in the wetland.

2. Materials and Methods

2.1. Study Area

This research focuses on the Zaalklapspruit wetland in Mpumalanga province, South Africa, situated within the transboundary Olifants River catchment (Quaternary catchment B2OG), as depicted in Figure 2. This catchment is vital, supplying water to a wide array of ecosystems, communities, and stakeholders. Nonetheless, it faces significant challenges due to contamination from domestic, agricultural, industrial, and mining sources, marking it as one of the most polluted catchments in Southern Africa ([12,30,31]). Despite growing demands on the Olifants River for freshwater, its quality is compromised by pollution from these varied anthropogenic activities ([8,32,33,34]).

The Zaalklapspruit wetland, significantly impacted by AMD from nearby mines, is recognized as a highly sensitive ecosystem. It is designated as a “National Freshwater Ecosystem Priority Area” and is listed among the “critically endangered wetland types” [35,36,37]. Furthermore, it is categorized as a vulnerable “Mesic Highveld Grassland group 4 wetland vegetation” as per the South African National Biodiversity Institute (SANBI) [35,36,37]. Historically, large volumes of AMD from upstream coal mines have altered Zaalklapspruit from an unchanneled to a channeled valley bottom wetland, reducing its water retention capabilities and diminishing its capacity to improve the water quality [3]. Nonetheless, adopting ecological engineering principles has facilitated gradually restoring the wetland’s functional integrity.

The Council for Scientific and Industrial Research (CSIR), in partnership with the SANBI and the Working for Wetlands (WfWET) program of the Department of Environmental Affairs (DEA), implemented several interventions in 2014 to address the challenges the wetland system encountered. Previously, the wetland had ceased functioning as a sponge, allowing storm waters to erode gullies and dry the floodplain. The rehabilitation efforts aimed to raise the streambed and employ structures to slow and diffuse the water flow. This included the construction of concrete structures, earth berms, and weirs, with funding provided by Coaltech [38]. The concrete and earth berm structures were strategically placed throughout the wetland to decelerate the water flow and enhance water retention for effective AMD treatment, as illustrated in Figure 3. These ecologically engineered interventions transformed the wetland back to an unchanneled valley bottom configuration and successfully decreased the water velocity to below 0.2 m/s, prolonging the system’s water retention time [8]. Furthermore, these modifications expanded the wetland’s surface area by 9.4 ha, bringing the total to 135.3 ha. Subsequent to these ecological engineering efforts, notable enhancements in surface water quality were observed. The improvements included an increase in the water pH and alkalinity to within the South African Water Quality Guidelines for aquatic ecosystems (Department of Water Affairs and Forestry [DWAF]) [39], a 50% reduction in the total dissolved solids (TDSs), and a 65% reduction in the sulfate levels. Additionally, the concentrations of Al, Fe, and Mn in the surface water decreased significantly, recording reductions of approximately 99%, 76%, and 96%, respectively, after the rehabilitation efforts were evaluated [8,38].

This study employed an influent–effluent comparison method for examining wetland water remediation, which typically regards the ecosystem as a “black box”. This approach provides a constrained view of the biogeochemical processes occurring within the wetland [40]. Previous research has shown that slowing the water flow in natural wetlands enhances the water residence time, thereby improving sediment infiltration and the interaction between AMD contaminants and various wetland components, such as bottom sediment [3,8,41]. This study focused on exploring the ability of the wetland’s bottom sediment and rhizosphere to accumulate metals and sulfate. Sampling took place in the dry season of May 2022 to reduce the influence of seasonal rainfall, which could potentially modify the surface water flows, groundwater levels, and AMD concentrations [28,42].

A systematic sampling strategy was employed, adhering to the guidelines set by the [43], involving the careful selection of three study sites within the wetland. These sites were strategically chosen based on their distance from the wetland inflow and were designated as follows: inflow = Site 1 (S1), middle = Site 2 (S2), and far middle = Site 3 (S3). Site S1, closest to the inflow, received surface water most directly impacted by AMD effluent; Site S2 experienced intermediate exposure to the wetland’s remedial processes; and Site S3, the farthest away, was influenced by these processes for the longest duration. The sampling regimen included the collection of both abiotic (surface water, wetland sediment, and surface soil precipitate) and biotic (macrophyte root systems and periphyton species) components. Samples were collected in triplicate (n = 3) using a simple randomized sampling method as recommended by the [43] to ensure a comprehensive and unbiased evaluation of the sites. All the collected samples were promptly sealed, stored in cool, dry containers, and transported to accredited laboratories within 24 h for analysis. These samples were analyzed for a variety of characteristics using the specific methods detailed in

Table 1. A summary of the sampled substrates and the corresponding analytical methods used for each assessment.

Wetland Substrate	Assessment	Analytical Method
Surface water	Physical and chemical characteristics	Electrometric, conductimetric, inductively coupled plasma (ICP) spectrometry [44,45]
Bottom sediment	Elemental composition	X-ray fluorescence (XRF) [46]
	Sediment texture analysis	Pipette method for sediment texture analysis [47]
	Microbiological diversity	16s RNA sequencing [48]
Macrophyte roots and periphyton	Metal adsorption and absorption	Inductively coupled plasma optical emission spectroscopy (ICP-OES) [49]
Unflooded soil precipitate	Mineralogy	X-ray fluorescence (XRF) [50]

.

Table 1 summarizes the substrates sampled and the corresponding analytical methods used for each assessment. Surface water was analyzed for physical and chemical characteristics using electrometric, conductimetric, and ICP spectrometry methods. The bottom sediment underwent elemental composition analysis via X-ray fluorescence (XRF), texture analysis using the pipette method, and microbiological diversity assessment through 16s RNA sequencing. Macrophyte roots and periphyton were examined for metal adsorption and absorption using ICP-OES, while unflooded soil precipitate was analyzed for mineralogy also using XRF. These methods collectively provided a detailed understanding of the wetland’s physical, chemical, and biological attributes, supporting a thorough evaluation of its condition and the effectiveness of AMD remediation efforts.

2.2. Substrate Sampling

2.2.1. Surface Water

Surface water samples were collected from each site at a depth of approximately 10 cm, following the techniques outlined by [51]. These samples were transferred into one-liter sterilized bottles, securely sealed, and kept chilled at 4 °C during transport. They were delivered within 48 h to Test It LAB in Bloemfontein, Free State, a SANAS-accredited facility. The analysis of the water’s physical and chemical characteristics was conducted using electrometric and conductimetric methods, alongside inductively coupled plasma (ICP) spectrometry. Electrometric methods were employed to measure parameters such as the pH and electrical conductivity, while conductimetric techniques were used to evaluate the water’s conductivity. ICP spectrometry provided detailed information on the concentrations of various metal ions. These techniques, as outlined in the Standard Methods for the Examination of Water and Wastewater (APHA, 2006) and supported by recent research [45], are crucial for assessing the water quality and understanding the impact of the wetland’s remediation processes.

2.2.2. Wetland Bottom Sediment

In situ pH and oxidation-reduction potential (ORP) measurements were conducted using a calibrated Vio70 Series handheld meter (Giorgio Bormac, Carpi, Italy) to assess the redox conditions within the wetland bottom sediment. The ORP is a crucial parameter for determining the sediment’s oxidative or reductive state, which significantly influences the mobility and bioavailability of contaminants. Including ORP measurements is essential for understanding how redox conditions affect biogeochemical processes within the wetland and their subsequent impact on pollutant dynamics and overall ecosystem health. pH assessments were also performed to provide a comprehensive view of the sediment’s chemical environment.

Wetland bottom sediment core samples were collected using pre-cleaned, open-ended plastic cylinders (8 cm × 8 cm), adhering to [43]. Triplicate samples (n = 3) were collected for each analysis to ensure accuracy and reliability. The samples were securely sealed with plastic wrap and wooden blocks and transported in a cool, dry container. For analysis, the samples were sent to three SANAS-accredited laboratories:

• Elemental Composition: Analyzed using X-ray fluorescence (XRF), which identifies and quantifies the elemental composition by measuring the fluorescent X-rays emitted from the sample [46].
• Sediment Texture Analysis: Performed using the pipette method, which separates and classifies sediment particles by size to determine the texture [47].
• Microbiological Diversity: Assessed through 16s RNA sequencing, a technique used to identify and characterize bacterial communities by sequencing the ribosomal RNA genes [48].

2.2.3. Precipitated Mineral Salts

Mineral salts that had crystallized on the surface of the dry, non-flooded wetland soil near the bottom sediment sampling locations were carefully identified. A sterilized plastic laboratory spatula was employed to collect these mineral salts to prevent contamination from the upper soil layer. The collected samples were then securely enclosed in watertight plastic containers, maintained at a cool temperature, and transported to Waterlab (Pty) Ltd. in Pretoria, Gauteng, for detailed analysis.

The elemental composition of the mineral salts was assessed using X-ray fluorescence (XRF) spectroscopy, a technique known for its precision in identifying and quantifying the elemental constituents within samples [46]. XRF provides a comprehensive analysis of the mineral content by measuring the characteristic X-rays emitted from the sample when excited by an X-ray beam. This method is particularly effective for detecting and analyzing metal and sulfur compounds present in the mineral salts.

2.2.4. Macrophyte Roots and Periphyton Sampling

The Zaalklapspruit wetland is characterized by diverse vegetation, with Typha capensis being a prominent macrophyte species. At each sampling site adjacent to where sediment and surface water were collected, three samples (n = 3) of T. capensis were carefully harvested to preserve their root systems. These samples were wrapped in plastic to maintain a cool temperature during transportation [1,52]. Concurrently, periphyton mats floating on the surface water near the macrophyte and sediment sampling locations were collected using methods outlined by [52]. Although periphyton species are typically abundant during the summer rainfall season, their presence is notably reduced in the dry season [53]. Some filamentous algae, including K. acidophilum, were observed in limited quantities. At Site S3, the periphyton biomass was too low for analysis and was thus excluded. The macrophyte roots and periphyton samples were transported in a cool (4 °C), dry, and dark container to the Department of Chemistry at the University of the Free State for analysis of their metal sorption properties using inductively coupled plasma optical emission spectroscopy (ICP-OES) [49].

2.3. Analytical Methods

2.3.1. X-ray Fluorescence Spectrometer

Samples of sediment and precipitate gathered from each site underwent a drying process at 50 °C for 12 h. Following this, a PANalytical X-ray fluorescence spectrometer (Malvern Panalytical, Eindhoven, The Netherlands) was employed to perform the elemental analysis. The XRF operated using Cu K radiation at settings of 30 kV and 200 mA. This method is noted for its ability to bypass organic compounds, focusing on identifying elements with atomic numbers greater than sodium (Na) [54].

2.3.2. Pipette Method for Soil Texture Analysis

To assess the distribution of the primary soil particles—clay (0–2 µm), silt (2–50 µm), and sand (50–2000 µm)—within the sediment collected from each site, the pipette method was utilized, following the protocol outlined by [55]. This method includes multiple steps: chemical and mechanical dispersion, fractionation, and quantification. Initially, chemical dispersion was carried out to strip away the colloid coatings and break down the aggregates into individual particles using dispersing agents like sodium hexametaphosphate (NaPO₃)₆, commonly referred to as Calgon. This was complemented by mechanical dispersion using a shaker to enhance the separation process. After dispersion, the particles were classified into size categories based on their settling velocity. The final quantification of these fractions was performed using the standard calculations provided by the International Soil Reference and Information Centre [56].

2.3.3. 16S rRNA Sequencing

The microbial diversity associated with the AMD treatment within the wetland was evaluated using 16S rRNA next-generation sequencing, adhering to the methodologies recommended by [57,58,59]. Sediment samples, measuring 8 cm × 8 cm, were taken to explore the diversity and abundance of the microbial communities across the designated sites. The DNA was extracted from these soil samples using the DNeasy 96 PowerSoil Pro Kit (Qiagen, Thermo Fisher Scientific, Waltham, MA, USA). This extracted DNA was then analyzed through 16S rRNA next-generation sequencing, employing the Ion 16S Metagenomics Kit (Thermo Fisher Scientific). The resulting data were processed using Ion Reporter software(Version 5.18.0), which grouped the sequences into operational taxonomic units (OTUs) based on the sequence similarity, allowing for taxonomic classification at multiple levels, such as the order, family, and genera [58].

2.3.4. Inductively Coupled Plasma Optical Emission Spectroscopy

The macrophyte root and periphyton samples underwent oven-drying at 80 °C until they reached a stable mass. These samples were then divided into two categories for analysis: adsorption, representing the metal concentration on the surface of the sample, and absorption, indicating the internal metal concentration. For the absorption analysis, the samples were milled through a 1 mm sieve and then ashed in a furnace at 550 °C overnight. The ashed material was dissolved in 5 mL of 6 M HCl, evaporated until dry, and then treated with 5 mL of 6 M HNO₃, which was heated to boiling. The resulting solution was filtered into a 100 mL volumetric flask, with the filter paper rinsed thoroughly with double-distilled water. This solution was then prepared for ICP-OES analysis by ensuring thorough mixing [60]. For the adsorption analysis, the samples were placed in 50 mL centrifuge vials filled with double-distilled water, agitated for eight hours, and subsequently filtered into a 100 mL volumetric flask. The volume was adjusted to 100 mL with 5% HNO₃ in preparation for the ICP-OES analysis. The analyses were performed using a Teledyne Leeman Labs Prodigy spec single-phase ICP-OES (SN: 8012) (Leeman Labs, Hudson, NY, USA), which operates at 230 AC voltage and a frequency of 50/60 Hz, using Salsa software (Version 4.0 SP 3). The instrument was carefully calibrated against a standard and stabilized for 30 min prior to the sample analysis to ensure its accuracy [61].

2.3.5. Scanning Electron Microscopy

Scanning electron microscopy (SEM) was employed to examine the small, branched (tertiary) roots of T. capensis, offering a detailed visual analysis of the microscopic features of the root system. This technique is particularly effective for observing microbial activities and the presence of metal deposits on the roots. For the SEM analysis, a tertiary root was carefully extracted from the root system at each sampling site and prepared at the Centre for Microscopy at the University of the Free State. Each 40 mm root sample was fixed in a 0.1 M sodium phosphate-buffered glutaraldehyde solution (3%, pH 7.0) for six hours, followed by a dehydration sequence using ethanol solutions at increasing concentrations (50%, 70%, 95%, and 100%). The samples were then dried in a Tousimis Samdri-795 critical point dryer. Post-drying, the roots were mounted on SEM stubs using double-sided carbon tape and coated with iridium to enhance the conductivity, utilizing a Leica EM ACE600 coater (Leica, Vienna, Austria). The roots were finally imaged using a JEOL JSM-IT200 InTouchScope™ Scanning Electron Microscope (JEOL, Tokyo, Japan), providing critical insights into each root’s micro-environment [1,62].

2.4. Statistical Analysis

The main aim of this study was to explore the various metal and sulfur pathways within the bottom sediment that contribute to enhancing the water quality from S1 to S3. Each measurement and analysis was considered a potential pathway or receptor for metals and/or sulfur compounds to achieve this. Triplicate results from each site were averaged to obtain the mean values, utilizing Excel and XLSTAT software (Microsoft 365 MSO -Version 2407 Build 16.0.17830.20056) for the calculations. These mean values were then systematically organized into tables and graphically represented to compare the changes occurring across the wetland.

To identify the most significant differences in the substrate concentrations within the wetland, a comparison was performed between the inflow site (S1) and the far middle site (S3) using Student’s t-test. This t-test is a powerful statistical tool used to determine whether there are meaningful differences between the means of two groups. This study allowed for the detection of significant variations in the concentrations of metals and sulfur compounds between the different sites. A probability value (p-value) of less than 0.05 was considered statistically significant, following the criteria set by [53]. These significance levels were employed to evaluate the strength of the associations between specific pathways and receptors across the different sites.

In summary, the statistical analysis was crucial for interpreting the data and drawing significant conclusions about the behavior of metal and sulfur compounds within the wetland. This analysis revealed important differences between the sites and provided a thorough understanding of the interactions and patterns among the metal and sulfur pathways and receptors. As a result, it facilitated a more detailed evaluation of the wetland’s effectiveness as an ecologically engineered treatment system.

3. Results and Discussion

3.1. Surface Water Quality

Table 2. Average physicochemical properties of the surface water at each site and the percentage change from the inflow (S1) to the far middle (S3) of the wetland.

Parameter	Unit	S1	S2	S3	Change (S1:S3)	% Change (S1:S3)
pH	–	3.90	3.84	7.18	+3.28	84.10
Alkalinity	mg/L CaCO3	22.00	<5.00	143.00	+121.00	550.00
Electrical conductivity (EC)	mS/m	178.6	180.50	152.10	−26.5	14.84
Total dissolved solids (TDSs)	mg/L	1196.62	1209.35	1019.07	−177.5	14.84
Dissolved organic carbon (DOC)	mg/L	5.18	5.17	16.00	+10.82	208.88
Aluminum (Al)	mg/L	1.60	1.80	<0.01	−1.59	99.38
Iron (Fe)	mg/L	0.33	0.58	0.26	−0.07	21.21
Manganese (Mn)	mg/L	10.80	10.90	37.10	+26.30	243.52
Sulfate (SO42−)	mg/L	1310.00	579.00	774.00	−536.00	40.07

summarizes the chemical properties of the surface water in the ecologically engineered wetland, highlighting significant improvements in the water quality parameters from Site 1 (S1) to Site 3 (S3). The pH of the surface water, initially acidic at the inflow (3.9), rose to a neutral level (7.18) by the time it reached S3, indicating a more hospitable environment for aquatic organisms. The alkalinity also substantially increased, from 22 mg/L CaCO₃ at S1 to 143 mg/L CaCO₃ at S3, demonstrating a strengthened buffering capacity in the water.

The electrical conductivity (EC), which measures the water’s ability to conduct electrical current due to the presence of dissolved ions, remained fairly constant between S1 and S2. However, a slight decrease of around 15% was noted from S2 to S3. Changes in the EC are important as they reflect alterations in the concentration of dissolved salts and minerals in the water, which can impact aquatic life. According to the [39] a change greater than 15% in the TDS (total dissolved solid) concentrations, which are closely related to EC, is deemed significant in aquatic ecosystems.

The dissolved organic carbon (DOC) exhibited a substantial increase from Site S2 to Site S3, rising by 208.88%, indicating a higher organic load in the surface water at Site S3. The metal concentrations typically associated with AMD—aluminum (Al), iron (Fe), and manganese (Mn)—showed varying trends: the Al and Fe concentrations were the highest at Site S2, decreasing from S1 to S3, while the Mn concentrations were the highest at Site S3, increasing significantly by 243.52% from S1 to S3. This increase in the Mn levels at Site S3 exceeded the South African Target Water Quality Range for aquatic ecosystems [39]. Conversely, the sulfate (SO₄²⁻) concentrations decreased by approximately 40% from S1 to S3.

A previous study by [8] assessed the impact of ecological infrastructure on the Zaalklapspruit wetland, focusing on water quality improvements before and after interventions. Their findings indicated a notable enhancement in the surface water quality, with the pH increasing from 5.3 to 7.6, and significant reductions in the Al and Fe concentrations from 0.89 to less than 0.01 mg/L and from 0.29 to less than 0.01 mg/L, respectively. The current study observed similar trends in the Al and Fe concentrations but noted a discrepancy in the manganese levels. Ref. [8] reported a 96% reduction in the Mn levels, whereas this study found an increase in the Mn concentrations from the inflow to the outflow. This elevated concentration of Mn at Site S3 can be attributed to several factors. The redox conditions at Site S3, being further from the inflow and experiencing longer water retention, may facilitate the release of manganese from sediments or enhance reduction processes that increase its concentration in the water.

Additionally, the sediments at Site S3 may have accumulated higher levels of manganese over time from upstream sources, and prolonged water-sediment interactions could mobilize this manganese into the water column. The biogeochemical processes at Site S3, including microbial activity, may also contribute to manganese dissolution and its release into the water. This high manganese concentration at Site S3 underscores the need for further research into sediment-water interactions and biogeochemical processes to fully understand the sources and dynamics of manganese in the wetland. Despite these discrepancies, both studies underscore the positive impact of the ecological infrastructure on improving wetland water quality.

Similarly, the current study’s comparison of the surface water quality between the S1 and S3 sites indicates an overall enhancement in water quality, as evidenced by improvements in the pH and EC, and reductions in the Al, Fe, and SO₄ levels. The most notable changes were observed between S2 and S3. These findings underscore the effective treatment capabilities of natural wetlands and highlight the beneficial impact of ecologically engineered infrastructure on enhancing the health and function of wetland ecosystems.

3.2. Bottom Sediment Characterization

The in situ pH and ORP measurements in the bottom sediment exhibited a consistent pattern across the sampling sites. The pH values were 6.09 at S1, 6.59 at S2, and 6.52 at S3. The corresponding ORP values were −219.67 mV at S1, −206.07 mV at S2, and −217.30 mV at S3.

The analysis also highlighted the significant presence of organic matter (OM) in the bottom sediment, with notably high concentrations at S2 (25.74%) and S3 (23.43%). Organic matter refers to the sediment component derived from the decay of plant material, microorganisms, and other biological residues. OM is crucial in wetland environments as it influences various biogeochemical processes, including nutrient cycling, metal binding, and pollutant degradation. High levels of OM indicate a rich organic environment, which supports microbial activity and contributes to the wetland’s ability to remediate AMD by enhancing microbial processes that can neutralize pollutants and improve water quality.

The elemental composition of the bottom sediment samples from each site was determined using XRF analysis and was found to be relatively consistent across the wetland. The weight percentages (% w/w) of key elements such as Al, Fe, Mn, and S in their oxidized forms are detailed in

Table 3. Bottom sediment characteristics of each site and Student’s t-test results to compare the inflow (S1) to the far middle (S3) of the wetland.

Sediment Characteristic	Unit	Average Concentration			Student’s t-Test (S1; S3)
		S1	S2	S3	Mean Difference	t	p Value
pH	–	6.09	6.59	6.52	−0.43	−1.02	0.37
Oxidation reduction potential (ORP)	mV	−219.67	−206.07	−217.30	−2.37	−0.02	0.99
Organic carbon (OC)	%	9.61	14.93	13.59	−3.98	−0.83	0.49
Organic matter (OM)	%	16.57	25.74	23.43	−6.86	−0.83	0.49
Clay	%	46.29	45.25	27.19	19.10	7.00	0.00
Silt	%	13.65	16.77	8.35	5.30	3.62	0.02
Sand	%	40.06	37.97	64.47	−24.41	−8.20	0.00
Aluminum (Al2O3)	wt %	5.19	5.87	4.55	0.64	1.29	0.29
Iron (Fe2O3)	wt %	4.36	4.79	4.41	−0.05	−0.08	0.94
Manganese (MnO)	wt %	0.06	0.03	0.07	−0.02	−2.24	0.11
Sulfur (SO3)	wt %	0.64	0.56	0.63	0.00	0.21	0.85

Note: wt % = weight percentage.

. A uniform pattern was observed at all three sites, following the sequence: Al₂O₃ > Fe₂O₃ > SO₃ > MnO. This stability in terms of the elemental composition highlights the uniform distribution of these elements within the wetland’s sediment, which is essential for understanding the sediment’s role in the overall ecosystem function and its capacity to treat AMD.

Student’s t-test was used to compare the means of the bottom sediment characteristics between S1 and S3 of the wetland, as detailed in Table 3. Given the substantial distance between S1 and S3, ample time and space are provided for biogeochemical reactions to occur, potentially leading to significant changes in the sediment composition. However, the analysis showed no statistically significant differences (p > 0.05) between S1 and S3 in terms of the pH, ORP, organic matter percentage (OM%), and elemental composition. In contrast, the sediment texture analysis revealed a significant difference (p < 0.05) between these sites.

As particles enter the wetland, they undergo aggregation, flocculation, and sedimentation, settling into the bottom soil [3]. This sedimentation process ensures continuous interaction between the surface water particles and the wetland bottom sediment [6]. Ultimately, water and soil particles combine and settle to form the bottom sediment layer [7]. Characterizing the bottom sediment is crucial for understanding the functioning of natural wetlands and the dynamics of pollutants, which is fundamental for effective wetland management and restoration [63,64].

In previous studies on constructed wetlands designed to treat mine and industrial effluent, the sediment pH and ORP measurements ranged from 6.4 to 7.1 and 384 to −261 mV, respectively. These findings are consistent with the current study’s results [65,66]. Wetland bottom sediment is often characterized by high organic matter content, as observed at S2 and S3 in this study [3]. This high organic matter content is due to the balance between carbon fixation and loss, which occurs at different rates under anaerobic conditions in saturated soils. Carbon fixation typically exceeds decomposition, leading to the accumulation of carbon and organic matter [3,67]

In a study by [68], the effects of plant litter decomposition on wetland sediment were examined. Their research demonstrated that litter decomposition leads to changes in the OM, pH, EC, ORP, and the bioavailability of metals in both the sediment and the surrounding water. Figure 4 in the current study illustrates the relationship between the OM%, pH, and ORP. The highest percentages of OM were observed at S2 and S3, which also showed the highest ORP and pH values. This correlation suggests the significant impact of organic matter on the chemical properties of the wetland sediment.

Another critical indicator of soil’s physical, chemical, and biological properties, and consequently of the biogeochemical reactions within wetland bottom sediment, is the soil particle size distribution [69]. Ref. [70] found a predominance of clay in the bottom sediment, especially concentrated downstream in a wetland affected by AMD. Contrarily, the present study observed an increase in the sand content at downstream sites (Figure 5). Ref. [8] documented sediment textures in an ecologically engineered wetland, primarily composed of sand particles. However, in the current study, clay particles constituted 46.29%, 45.25%, and 27.19% of the sediment textures at S1, S2, and S3, respectively (Table 3). The differences in the particle size distribution among these studies could be attributed to variations in the surface water velocity at different points within the wetland [3].

Clay minerals are essential for cation-anion exchange and pH buffering, contributing to the relatively stable pH values observed in the bottom sediments at each site, regardless of the surface water pH [70]. Additionally, clay particles’ large surface area and cation-anion exchange capacity enhance the soil reactivity and facilitate interactions with OM [3]. However, in this study, the relationship between the OM% and clay content showed an inverse trend. S1 had the highest clay content but the lowest OM%. This discrepancy could be due to the higher water flow rate at the inflow site (S1), resulting in a shorter residence time and consequently transporting OM farther into the wetland, toward the middle (S2 and S3).

Water pollutants have a strong affinity for organic materials, leading to the accumulation of metals and sulfur compounds within the bottom sediment layers of wetlands affected by AMD [5,71]. In this study, oxidized forms of Al, Fe, Mn and S were detected in the bottom sediment, with aluminum oxide (Al₂O₃) being the most prevalent, followed by iron oxide (Fe₂O₃). Research on sedimentary metals and sulfur in natural wetlands exposed to AMD has shown a predominance of oxides of Al, Fe, and Mn, especially in oxic sediments, with a decrease in these oxides in the anoxic zones [70,72,73].

Ref. [73] noted that the accumulation, distribution, and oxidative status of metals and sulfur in wetland sediments are influenced by the availability of oxygen (redox profile) in various sedimentary layers (e.g., 0–5 cm, 2.5–7.5 cm, and >7.5 cm). They also found a positive correlation between OM and these elements. Ref. [74] observed a similar pattern in their inlet–outlet study of a natural wetland receiving AMD, where the OM% increased from the inlet to the outlet (10%–25.4%), and the metal accumulation increased for Fe (212–255 µg/g) and Mn (1.6–4.2 µg/g) but not for Al (1.21–1.05 µg/g). In the current study, the highest concentrations of Fe and Al in bottom sediments were recorded at S2, where OM was the most abundant, consistent with the findings from other studies.

The relationship between OM and ORP is critical in wetlands. Ref. [75] explained that OM acts as the terminal electron acceptor under highly reduced redox potentials. The pH, ORP, texture, and OM% in bottom sediments may also be influenced by emergent vegetation and microbiological activities, and vice versa. This interconnected and complex nature of the biogeochemical reactions within the bottom sediment cannot be fully understood in isolation [76].

3.3. Macrophyte Sorption Abilities

Typha capensis, a macrophyte species, was collected from each site to analyze its ability to adsorb and absorb Al, Fe and Mn. The results of the inductively coupled plasma optical emission spectroscopy (ICP-OES) analysis are summarized in

Table 4. Average concentration (mg/kg) of Al, Fe and Mn on the root surface (adsorption) and in the root tissue (absorption) of T. capensis, between sites—Student’s t-test results to compare the inflow site (S1) to the far middle (S3) of the wetland.

Element	Unit	Average Concentration			Student’s t-Test (S1; S3)
		S1	S2	S3	Mean Difference	t	p Value
Adsorption
Aluminum (Al)	(mg/kg)	1.77	0.56	0.22	1.55	14.98	0.00
Iron (Fe)	(mg/kg)	0.54	0.72	0.57	−0.03	−0.55	0.62
Manganese (Mn)	(mg/kg)	2.24	14.09	7.66	−5.42	−257.72	0.00
Absorption
Aluminum (Al)	(mg/kg)	192.87	154.82	178.74	14.13	8.92	0.00
Iron (Fe)	(mg/kg)	174.11	487.33	364.01	−189.90	−135.94	0.00
Manganese (Mn)	(mg/kg)	6.62	19.68	74.35	−67.73	−117.96	0.00

, highlighting distinct patterns of metal sorption. For adsorption, the concentration order was Al (S1 > S2 > S3), Fe (S2 > S3 > S1), and Mn (S2 > S3 > S1). In terms of absorption, the concentration order was Al (S1 > S3 > S2), Fe (S2 > S3 > S1), and Mn (S3 > S2 > S1). These findings demonstrate the varying capabilities of T. capensis to uptake different metals across the sampling sites.

Student’s t-test was utilized to compare the metal adsorption and absorption capabilities of T. capensis between the inflow (S1) and the far middle (S3) of the wetland. The results revealed a significant difference (p < 0.05) in the adsorption of Al and Mn between S1 and S3. Specifically, Al was adsorbed in higher quantities at S1, whereas Mn showed greater adsorption at S3. However, the adsorption of Fe remained relatively consistent between S1 and S3 (p > 0.05). In terms of absorption, significant differences were observed for all the metals analyzed. Al was absorbed more at S1, while Fe and Mn were predominantly absorbed at S3. This indicates that the adsorption and absorption of Al by T. capensis primarily occurred at S1, whereas Fe and Mn were mainly adsorbed and absorbed at S3.

Wetland bottom sediment provides an optimal growth environment for T. capensis, with its roots playing a crucial role in rhizospheric processes that alter metals and sulfur compounds in AMD-impacted wetlands [1]. The rhizosphere, the soil–root interface, involves complex physical, chemical, and biological interactions [77].

Sorption, which includes both adsorption and absorption, is vital for metal removal in wetlands [7,78]. Adsorption involves the physical adhesion of substances to a solid surface, while absorption entails the incorporation of substances from one state to another [78]. In macrophyte roots, adsorption involves ions and molecules binding to the root epidermis surface, precipitating metals [6,78].

Metal precipitation appears as coatings in various colors, such as orange to brown for Fe oxides or dark brown for Mn oxides, indicating adsorption on the root surface [1]. This process, known as metal plaque formation, depends on the solubility of Fe and Mn and the oxidizing conditions facilitated by oxygen translocation from roots to the rhizosphere. Oxidized environments promote Mn²⁺ and Fe³⁺ oxyhydroxides’ precipitation on the root surface [6,7].

Mn adsorption was most pronounced at S2, where the ORP measurements and OM% were the highest, suggesting oxygen release from emerging roots promoted Mn precipitation. This highlights the relationship between OM, ORP, and Mn oxidation. Mn adsorption exceeded that of Fe and Al, indicating a higher oxidizing capacity for Mn. In an experimental study, Ref. [79] found Mn plaque concentrations of 983.4 µg/g and internal Mn concentrations of 421.9 µg/g at pH 6, with the plaque concentrations surpassing the internal concentrations at higher pH levels.

Sediments utilize electron acceptors in order of the decreasing free energy [73], with dissolved O₂ consumed in the following order: O₂ > Mn-oxides > Fe-oxides > SO₄. Although Mn adsorption exceeded Fe adsorption, the typical orange to brown coloration associated with iron plaque formation was observed on T. capensis roots both visually and microscopically (Figure 6).

3.4. Periphyton Sorption Abilities

The dominant periphyton species, Klebsormidium acidophilum, was collected from S1 and S2 to analyze its metal sorption characteristics using ICP-OES. The results are detailed in

Table 5. Average concentration (mg/kg) of Al, Fe and Mn on the periphyton surface (adsorption) and in the periphyton tissue (absorption) of K. acidophilum between the sites—Student’s t-test results to compare the inflow site (S1) to the far middle (S3) of the wetland.

Element	Unit	Average Concentration		Student’s t-Test (S1; S2)
		S1	S2	Mean Difference	T	p Value
Adsorption
Aluminum (Al)	(mg/kg)	0.48	0.44	0.04	0.38	0.73
Iron (Fe)	(mg/kg)	0.79	0.78	0.01	0.22	0.84
Manganese (Mn)	(mg/kg)	1.13	0.37	0.76	56.24	0.00
Absorption
Aluminum (Al)	(mg/kg)	170.3	38.61	131.68	572.71	0.00
Iron (Fe)	(mg/kg)	284.78	241.43	43.34	45.63	0.00
Manganese (Mn)	(mg/kg)	8.91	4.41	4.50	199.84	0.00

. Student’s t-test was applied to compare the mean metal sorption abilities of K. acidophilum between S1 and S2. The analysis showed no statistically significant difference in the adsorption of Al and Fe between the two sites (p > 0.05). However, there was a significant difference in Mn adsorption, with higher Mn adsorption at S1 compared to S2 (p < 0.05). Regarding metal absorption, a significant difference was observed between the sites (p < 0.05), with Al, Fe, and Mn predominantly absorbed at S1. These findings highlight the varying sorption capacities of K. acidophilum at different locations within the wetland.

Research on the biosorption of microelements has demonstrated that green macroalgae can effectively accumulate minerals, facilitating the bioaccumulation of specific elements [78]. Certain benthic filamentous algae species have shown the ability to absorb metals across various pH levels, making them suitable for passive treatment of AMD-contaminated water [80]. Ref. [80] also highlighted that the pH of surface water significantly influences algal metal absorption, with Fe absorption being most efficient at pH 3, and Mn and Al absorption improving at a neutral pH (pH 7.0). The most substantial shifts in periphyton species composition and nutrient cycling typically occur within a pH range of 4.7 to 5.6 [53,81]. However, in the current study, surface water pH values were lower at S1 (pH = 3.9) and S2 (pH = 3.84) compared to S3 (pH = 7.10), where K. acidophilum was absent.

In this study, the metal adsorption sequence for S1 and S2 was Mn > Fe > Al, while metal absorption followed Fe > Al > Mn at both sites. Ref. [82] emphasized that the algal metal accumulation can vary widely depending on the species and types of metals involved. This variability was also observed by [80], where different filamentous algae species exhibited distinct patterns of metal bioaccumulation at various sites. For example, Oedogonium crissum, Klebsormidium krebsii, and Microspora tumidula showed different metal bioconcentration patterns at four different sites: O. crissum at S1 (Al > Fe > Mn), K. krebsii at S2 (Al > Fe > Mn), M. tumidula at S3 (Fe > Al > Mn), and M. tumidula at S4 (Fe > Mn > Al). In this study, K. acidophilum displayed trends similar to M. tumidula, although the latter showed higher bioconcentrations, reaching up to 401,739 mg/kg dry weight for Fe.

Ref. [80] also found that Fe bioaccumulation increased as the pH decreased, attributed to the precipitation of Fe at pH ≥ 4, making it more available to algal species. Their results suggested that Mn bioaccumulation was favored near neutral pH conditions. However, in this study, Mn adsorption was higher than that of Al and Fe at S1. An experimental study by [83] showed that Fe removal occurred rapidly within the first hour, with up to 94.4% of Fe removed after four hours of exposure to metal-enriched water, whereas Mn exhibited continuous uptake throughout the experiment. These findings by [83] offer valuable insights into the long-term efficiency of metal uptake by periphyton species.

3.5. Microbiological Diversity

The diversity and abundance of microbiological species in bottom sediment samples, collected at a depth of 8 cm, were analyzed using 16S rRNA next-generation sequencing. The sequences were grouped into operational taxonomic units (OTUs) based on their similarities. The results were categorized into dominant and other taxa, with the dominant taxa having an abundance of ≥1%, and the other taxa having an abundance of <1%. Notably, the other taxa constituted 44.99%, 38.57%, and 37.33% of the total OTUs detected at S1, S2, and S3, respectively. Although these taxa may be present in small numbers individually, their collective presence indicates a high level of microbial diversity at all three sites. Figure 7 shows that the dominant taxa included 21, 27, and 24 microbial families at S1, S2, and S3, respectively.

Among the dominant families, those within the phylum Proteobacteria (alpha, beta, and gamma) were the most prevalent. Other significant phyla included Acidobacteria, Firmicutes, and Thermodesulfobacteriota. A previous study on the Zaalklapspruit wetland demonstrated a notable shift in the relative abundance and bacterial diversity within the bacterial consortium following ecological engineering interventions. Although Proteobacteria remained the dominant phylum, its relative abundance decreased, while Firmicutes significantly increased [8]. It is important to note that this earlier study focused on the water column bacterial consortium, which might account for the differences in the abundance percentages when compared to the sediment samples analyzed in the current study.

Several studies on natural and constructed wetlands receiving AMD have highlighted the abundance of these taxa in sediment samples, underscoring the crucial role of microbially mediated activities in transforming metallic and sulfur compounds during AMD remediation [84,85,86]. This study identified thirteen comparable families as dominant taxa in the bottom sediment across S1, S2, and S3, reflecting the consistent microbial community structure throughout the wetland. The most prominent families included the following:

S1: 

Helicobacteraceae > Hydrogenophilaceae > Rhodospirillaceae > Gallionellaceae > Xanthomonadaceae.

S2: 

Acidobacteriaceae > Gallionellaceae > Xanthomonadaceae > Helicobacteraceae > Hydrogenophilaceae.

S3: 

Helicobacteraceae > Gallionellaceae > Xanthomonadaceae > Hydrogenophilaceae > Rhodospirillaceae.

Helicobacteraceae accounted for 19.98%, 6.14%, and 23.53% of the OTUs in the bottom sediment at S1, S2, and S3, respectively. This family, part of the phylum Epsilonproteobacteria, thrives at oxic–anoxic interfaces in sulfur-rich environments [87]. Studies have linked Helicobacteraceae with mine effluents high in heavy metals and Fe concentrations, indicating their dominance in metal-polluted aquatic systems [88,89,90]. In some cases, they represent up to 65% of the microbial community in such environments.

Hydrogenophilaceae, another significant family observed in this study, are strict autotrophic or facultative anaerobic sulfur oxidizers, playing a crucial role in sulfur utilization in wastewater treatment systems [91]. The family Xanthomonadaceae constituted 6.07%, 7.44%, and 6.80% of the OTUs in S1, S2, and S3, respectively. Belonging to the phylum Gammaproteobacteria, Xanthomonadaceae includes iron-reducing bacteria that oxidize Fe(II) for growth and are commonly found in AMD-polluted water bodies [84,92]. Ref. [92] found that Xanthomonadaceae represented approximately 51% of the detected Gammaproteobacteria in sediments from AMD-induced iron mounds, highlighting their preference for Fe-rich environments.

Gallionellaceae, also associated with Fe(II)-rich sediments and iron oxidation, showed a relative abundance of 10.47% at S3, where Fe₂O₃ was prevalent in the bottom sediment (Table 3). Rhodospirillaceae, detected at all three sites, is known to inhabit stagnant water bodies with a pH of around 5.5 and tends to increase in abundance as the acidity rises [93]. This family thrives in environments with rapid OM production, aligning with the high OM levels found in this study.

The other abundant families detected at all three sites included Acetobacteraceae, Clostridiaceae, Desulfobulbaceae, Hyphomicrobiaceae, and Sphingomonadaceae. Desulfobulbaceae, part of the phylum Thermodesulfobacteria, are well-known sulfate-reducing bacteria (SRB) involved in sulfate transformation in bottom sediments, playing a crucial role in passive treatment systems like wetlands [87,94]. The SRB phyla is known to mediate AMD-polluted waters, and it includes Proteobacteria, Firmicutes, Thermodesulfobacteria, and Nitrospirae.

In summary, the 16S rRNA sequencing results indicate a rich diversity of microorganisms in the bottom sediment and rhizosphere, further illustrated by SEM images of the tertiary roots of T. capensis (Figure 8).

3.6. Precipitated Mineral Salts

An XRF analysis was conducted to identify the composition of metal and sulfur compounds in the precipitated mineral salts found on the surface of dry wetland soil. Figure 9 illustrates the precipitated mineral salts on the dry, unflooded soil surface in the ecologically engineered wetland.

Table 6. Average concentration of metal and sulfur compounds in precipitated mineral salts for each site.

Element	Unit	S1	S2	S3
Aluminum (Al2O3)	(wt %)	1.42	1.53	1.78
Iron (Fe2O3)	(wt %)	1.69	1.40	1.95
Manganese (MnO)	(wt %)	0.10	0.10	0.08
Sulfur (SO3)	(wt %)	18.50	18.56	13.67

Note: wt % = weight percentage.

details the composition of these mineral salts. The analysis revealed significant amounts of metal and sulfur compounds precipitated as mineral salts on the surface soil. Notably, sulfur trioxide (SO₃) was present in the highest concentrations, indicating that precipitation is a key process in retaining sulfur compounds, particularly in areas without surface water flow during the dry season. This suggests that SO₃ is effectively trapped in the soil during dry conditions. However, SO₃ may become mobilized during the wet season, increasing the sulfur compound concentrations in areas that had not previously been exposed to surface water. This dynamic highlights the seasonal variations in the sulfur compound mobility and the importance of precipitation in managing the sulfur levels within the wetland ecosystem.

3.7. Ecological Engineering Interventions and Ecosystem Services

The Zaalklapspruit wetland has effectively integrated ecological engineering interventions to restore and enhance its ecosystem services. These interventions, such as the strategic placement of berm structures to reduce the water velocity and increase the retention time, have improved the water quality and reinstated critical ecosystem functions. By stabilizing the pH levels, reducing harmful metal concentrations, and enhancing the wetland’s buffering capacity, these measures have facilitated the development of a more resilient and self-sustaining ecosystem.

The wetland’s ability to adopt these engineered solutions demonstrates the synergy between ecological restoration and engineering practices. The reinstatement of ecosystem services, such as improved water quality, habitat provision, and increased biodiversity, highlights the success of these interventions. The reduced metal concentrations, like Al and Fe, and the improved pH levels across the sites illustrate the effectiveness of the engineering measures implemented.

Additionally, the re-establishment of complex biogeochemical processes and microbial diversity within the wetland further underscores the positive impact of these interventions. For instance, diverse microbial communities, including families like Helicobacteraceae, Xanthomonadaceae, and Gallionellaceae, play a crucial role in transforming and stabilizing metals and sulfur compounds in the sediment.

Implementing these ecological engineering interventions has contributed to mitigating AMD impacts and promoted the wetland’s overall ecological health and resilience. By fostering natural processes and improving water quality, such interventions ensure that wetlands can continue to provide essential ecological services, support biodiversity, and enhance the quality of life for surrounding communities.

Overall, the Zaalklapspruit wetland is a model for utilizing ecological engineering to restore, regenerate and enhance the functionality of natural ecosystems. The positive outcomes observed in this study highlight the importance of continuing to invest in and apply ecological engineering principles to address environmental challenges and promote the sustainable management of natural resources. Integrating these practices not only aids in the recovery of degraded wetlands but also enhances their capacity to support diverse biological communities and maintain vital ecosystem services.

3.8. Integrated Findings

The water flow rates profoundly influence wetlands’ filtration and purification capacity. A reduced water flow promotes the aggregation, flocculation, and sedimentation of particles from AMD-affected water, leading to their settling and formation of a bottom sediment layer. This sediment layer remains in continuous contact with surface water, the primary carrier of pollutants stemming from AMD [63,64].

Wetland macrophytes undergo decomposition in oxygen-depleted, flooded environments, enriching the OM content within the bottom sediment [1,68,73]. Clay minerals, renowned for their robust ion exchange capacity, interact with OM, which metals have a strong affinity for [3,70]. Moreover, sediment characteristics such as the pH and ORP are influenced by root exudates containing oxygen and other nutrients [1,68]. Elevated pH and ORP values facilitate the precipitation and sorption of metals like Mn and Fe, often observed as Fe plaques on root surfaces [85,95,96].

Microbial communities thrive in the nutrient-rich environment provided by accumulations in the rhizosphere, modifying metals and sulfur compounds through oxidation and reduction reactions during metabolic activities [77,97]. Mobilized metals and sulfur compounds are subsequently released into the bottom sediment and readily absorbed into macrophyte root tissues [98].

The success of this passive treatment system hinges on the intricate interplay among various wetland compartments, all interdependent within the anoxic and potentially toxic environment created by AMD [15,77] Figure 10 provides a simplified schematic representation of this phenomenon in the current study, illustrating the dynamic interactions within the wetland environment.

4. Conclusions

This study’s application of a source–pathway–receptor approach to the Zaalklapspruit wetland, an ecologically engineered site, has revealed nuanced insights into the dynamics of contaminant behavior and treatment mechanisms within such systems. Our findings, which show significant improvements in critical water quality parameters from the inflow to the outflow, underscore the potent remediation capabilities of engineered wetlands against AMD impacts.

The observed increases in the pH and decreases in the Al, Fe, and SO₄²⁻ concentrations along the wetland’s course are attributable to the complex interactions between the wetland’s biotic and abiotic components. The bottom sediment’s high organic matter content, stable pH, and ORP support the wetland’s biogeochemical processes. These conditions facilitate vital reactions such as precipitation, sorption, and the redox transformations of contaminants. The role of T. capensis and K. acidophilum in metal uptake provides a pathway for contaminant removal and highlights potential areas for enhancing phytoremediation strategies in similar ecological settings.

The inherent variability of ecologically engineered systems like wetlands makes establishing a traditional ‘control’ challenging. Instead, longitudinal analyses of pre- and post-treatment dynamics offer a practical approach to assess effectiveness. This study’s comparative water quality analysis across multiple points within the wetland illustrates the system’s transformative impact on AMD effluents.

Community acceptance is critical for the sustainable operation of engineered wetlands. Our engagement efforts have aimed to educate local communities about the benefits of these wetlands in terms of the water quality improvement, biodiversity conservation, and recreational opportunities. Building community trust and support is essential, as it underpins the social license to operate these vital ecological infrastructures.

Ongoing monitoring and operational maintenance are critical to sustain the functionality of engineered wetlands. Allocating resources effectively for these activities ensures that the system adapts to load and water chemistry changes, thereby maintaining its efficiency. Regular monitoring also helps in the timely identification of issues, allowing for adjustments in management practices, which is crucial for the long-term success of any ecological engineering project.

The insights from this study are globally significant, offering valuable perspectives for the design and management of ecologically engineered wetlands in various environmental contexts. As climate change and industrial impacts intensify, the lessons learned here about the versatility and effectiveness of wetlands in pollution mitigation are increasingly relevant. Future research should focus on optimizing the design parameters that enhance the resilience and adaptability of these systems, further integrating technological advancements and community-driven approaches.