Physicochemical characteristics and microbial community analysis of a wetland system treating acid mine drainage Authors: Jing Guo, Lei Cheng, Miao Yang, Zhenli Wang Published: 2025 ABSTRACT The elevated heavy metal concentrations and acidic pH conditions characteristic of Acid Mine Drainage (AMD) pose significant environmental challenges, necessitating the development of effective remediation strategies. This study systematically evaluated the performance of a surface-flow constructed wetland system in Ma’anshan, China, for AMD treatment through comprehensive physicochemical characterization and microbial community analysis. The multi-stage wetland system, comprising four interconnected units vegetated with Phragmites australis, Typha angustifolia, and Arundo donax, was monitored for water quality parameters, sediment characteristics, and microbial diversity. Results demonstrated significant improvements in water quality, with pH increasing from 5.18 to 7.41 and substantial removal efficiencies observed for heavy metals: complete (100.00%) removal of Fe, 92.00% removal of Zn, 61.80% removal of Mn, and 63.60% removal of chemical oxygen demand (COD). Sequential extraction analysis revealed that residual fractions constituted the predominant form of sediment-bound heavy metals (65.70%-70.20% for Fe). Proteobacteria make up 58% of the microbial community. Sulfuricurvum combines denitrification and sulfide oxidation, while SRB like Desulfuromonas convert sulfate to sulfide. Nitrospinae oxidizes ammonia to nitrate, Thaumarchaeota oxidizes ammonia to nitrite, and Parkl vularcula facilitates denitrification. In order to maintain cycles and support wetland function and AMD treatment, hydrolytic bacteria break down organic materials. These findings establish the effectiveness of constructed wetlands (CWS) in AMD remediation through synergistic phytoremediation and microbial metabolic processes, offering an ecologically sustainable approach for the restoration of mining-impacted areas. Introduction The oxidation of sulfide minerals in the presence of oxygen and water results in Acid Mine Drainage (AMD)1–4. The strong acidity and elevated concentrations of sulfate and heavy metals in AMD can result in significant damage to the adjacent soil and water environment if not effectively treated5–13. It contains numerous heavy metals, including zinc (Zn), lead (Pb), cadmium (Cd), and others14. As a chronic aquatic pollutant, it disrupts aquatic ecosystems and reduces biodiversity. Characterized by strong acidity, high sulfate levels, and elevated heavy metal concentrations, this wastewater reacts with sulfides in air and water to form metal sulfides, which can further oxidize into metal oxides15. Through infiltration, runoff, and reverse osmosis, it penetrates soil, causing acidification, degradation of physical and chemical properties, and inhibition of microbial growth, particularly affecting nitrifying and nitrogen-fixing bacteria16. Additionally, it impairs plant growth and lowers crop yields. AMD also facilitates bioaccumulation and biomagnification, with heavy metals concentrating up to 100 ~ 1000 times within the food chain17, thereby posing a threat to human health. AMD can be obtained from both active and abandoned mines, and the majority of them can endure for decades or even hundreds of years18. Preventing AMD at the source is significantly more effective2. The long-term effective treatment of AMD is crucial for the maintenance of the eco-environmental health surrounding the mining area once it has been generated. In general, the current treatment technologies for AMD are typically classified as active treatment (which employs compounds) and passive treatment (which utilizes naturally occurring geochemical and biological processes)19. In comparison to passive treatment systems, active treatment necessitates continuous operation and maintenance, which incurs higher costs20,21. Traditional AMD treatment methods, such as chemical neutralization, can rapidly increase the pH value and precipitate some heavy metals. However, they typically necessitate a substantial quantity of chemical agents, have high operating costs, and generate a substantial amount of sludge, which poses a potential secondary pollution risk22. Consequently, it is of immense practical importance to create a treatment technology for AMD that is both environmentally benign and cost-effective. Construction of wetlands has become a research focus in wastewater remediation in recent years as a result of their nature-based approach23,24. This technology effectively purifies pollutants from wastewater while providing environmental benefits, low operational costs, and simple maintenance by replicating the physical, chemical, and biological processes of natural wetlands25. Constructed wetlands (CWs) have the potential to regulate water pH26 and remove heavy metals25 in the context of AMD remediation. This is achieved through synergistic mechanisms such as plant uptake, microbial degradation, substrate adsorption, and precipitation reactions. Current research is focused on four critical areas in the global effort to address AMD using CWs: the refinement of wetland configurations26, the identification of pollution-resistant vegetation27, the enhancement of substrate compositions28, and the unraveling of how microbial communities modulate pollutant removal29. Field studies have shown that the system’s ability to sequester heavy metals is significantly enhanced by strategic planting, such as the integration of reeds or cattails23,24. Simultaneously, microbial research has demonstrated that specialized microorganisms in wetland ecosystems serve a dual function: they catalyze chemical reactions that precipitate heavy metal ions and degrade organic contaminants30. Microbial metabolic activities, especially those related to sulfur (S), iron (Fe), and carbon (C) biogeochemical cycling, are one of the most important pathways for AMD restoration in CWs. Acidithiobacillus ferrooxidans-mediated production of Fe/Mn oxides and metal coprecipitates31,32 and sulfate-reducing bacteria (SRB)-catalyzed precipitation of insoluble metal sulfides33,34 are two important methods. Therefore, sustaining these microbial communities’ high metabolic activity is essential to guaranteeing the long-term stability and therapeutic effectiveness of CWs systems. Therefore, current research on CWs for AMD remediation has made progress in optimizing configurations, screening pollution-resistant vegetation, improving substrates, and exploring microbial functions, but it still has limitations: most studies lack targeted designs for AMD characteristics in specific regions, and there is insufficient in-depth exploration of the ecological processes and intrinsic mechanisms of plant-microbe synergistic remediation. Additionally, few studies integrate physicochemical monitoring, heavy metal speciation analysis, and microbial community profiling to systematically clarify the pollutant removal mechanisms of CWs in AMD treatment. In this paper, there were two specific objectives: (1) targets the local AMD pollution in Ma’anshan and constructs a multi-stage surface flow constructed wetland system using regionally adaptable plants, filling the gap of localized AMD remediation technology research in this area; (2) adopts a multidimensional analysis approach to systematically reveal the plant-microbe synergistic remediation mechanism of the wetland system, which addresses the lack of in-depth exploration of synergistic mechanisms in existing studies. Hence, this study emphasizes the significance of regional adaptation and multi-dimensional mechanism analysis for improving the efficiency of CWs in AMD remediation.It is desirable for this data to provide a reference value for AMD remediation in many similar sites. Materials and methods Study site and sampling The investigation was undertaken in Ma’anshan City, Anhui Province, China, within the metallogenic belt35 (31°46’42’’N;118°21’38’’E). The mining area is situated on the Yangtze River Delta, where a hydrological gradient flows from east to west. From a geographical standpoint, it is circumscribed by Nanjing’s Pukou and Jiangning Districts to the east, Quanjiao County (Chuzhou City) to the north, Chaohu County (Hefei City) to the west, and Jiujiang District (Wuhu City) and Wuwei County to the south. The climate of the region is monsoonal temperate, characterized by moderate temperatures and distinct seasonal fluctuations. Meteorological data suggests that the average annual temperature is 15.7 °C. The average annual precipitation is 1060 mm, and the annual evaporation is 1502.8 mm. In a Ma’anshan mining area, an acid reservoir36 is surrounded by a surface flow wetland. A continuous flow-through mode was adopted. AMD was pumped from the acid reservoir into the inlet of CWs at a hydraulic loading rate of 0.1 m³/(m²·d). The water layer is between 20 and 30 centimeters thick37. Gravel, coarse sand, limestone, iron tailings, steel slag, and blast furnace slag constitute the substrate of this CWs. The hydraulic retention time (HTR) was configured at 48 h38,39. In this wetland system, water plants like bulrushes, cattails, and reeds are planted. After being neutralized and pre-treated, the acidic mine effluent from the acid reservoir enters the wetland via pipes. The water flows downstream after being treated by the wetland system. The four constructed wetlands that make up the wetland treatment system are located in a straight line 1.0 km from the acid reservoir. This engineered wetland system is comprised of four interconnected treatment units that are arranged in sequence, as illustrated in Fig. 1. Three macrophyte species—Phragmites australis, Typha angustifolia, and Arundo donax—are strategically planted within the system. The initial plant density in the wetland is 30 plants per square meter. Through a conduit, the AMD from the reservoir is taken to the treatment system, which is located 1.0 km downstream. The AMD is subsequently discharged into the recipient waters following phytoremediation. The following is the arrangement of wetland subdivisions: CW-a (13,700 m²): Phragmites-Typha polyculture; CW-b (9,400 m²): Phragmites-Arundo polyculture; CW-c (8,900 m²): Typha-Arundo polyculture; CW-d (16,900 m²): Tri-species cultivation (Phragmites-Typha-Arundo). Plant domestication is considered complete when the pollutant absorption efficiency achieves the target level and the constructed wetland system demonstrates continuous and stable operational performance40. Each treatment unit was sampled in triplicate for both the aqueous and sediment phases in order to perform a systematic analysis. This resulted in the production of 12 composite samples, each consisting of four units and three replicates. The efficacy of spatial treatment was assessed by conducting all sampling procedures in accordance with standardized protocols for constructed wetland monitoring. Collection points were located in a strategic manner along the hydraulic flow path. Fig. 1The distribution of sampling points in the wetland. The distribution of sampling points in the wetland. Before sediment sampling, litter, living vegetation, and plant detritus were meticulously removed with a shovel and knife41. Three replicates of sediment samples were collected within a 1 × 1 m2 area of each sampling area. Each sample is approximately one kilogram in weight. The samples were promptly transported to the laboratory on ice in two portions and either chemically analyzed at 4 ℃ or stored at −20 ℃ until the microbiological analysis was conducted. Sediment properties The sediment samples’ physical and chemical properties were evaluated by positioning them in Petri dishes. These dishes were subsequently positioned in a dry, enclosed environment with optimal light transmission, free from dust interference, and allowed to dry naturally. The samples were subsequently passed through a 200-mesh (0.074 mm) standard sieve after air-drying at ambient temperature for a week. These samples were kept in air-sealed polyethylene containers. Unless otherwise stated, all reagents utilized were of analytical grade. In a suspension of 10 g sediment with 50 ml Milli-Q water (Yamato Millipore-filter, WT 101 UV), the pH of the sediment was measured by pH meter (Thermo Scientific, Waltham, MA, USA)42. The method described by Pansu and Gautheyrou was employed to determine the sediment organic carbon (SOC) and the total nitrogen (TN) were measured using an elemental analyzer (Vario EL cube, Elementar, Langenselbold, Germany)43. Ion chromatography, gas chromatography, and liquid chromatography were used to detect sulfate, sulfide, and sulfur in the sediment, respectively. These determinations followed the methods outlined in Water and Wastewater Analyzing Methods (State Environmental Protection Administration [SEPA], 2002). HF (1.49 g/ml), HClO4 (1.68 g/ml), and HNO3 (1.42 g/ml) were used to decompose the sediment samples for the analysis of Mn, Fe, Cu, Zn, and Cd (Guaranteed Reagent, GR). The concentrations of dissolved metals (Fe, Zn, Mn, Cu, and Cd) were measured by atomic absorption spectroscopy (WYS-2200, China). Tessier’s methodology was employed to ascertain the residual fraction of heavy metals, water-soluble fraction, combination fraction, and ion-fraction in the sediment samples44. Molecular techniques for microbial community analysis The soil DNA kit (OMEGA, USA) was employed to extract the total genomic DNA, which was subsequently amplified using specific primers that targeted the hypervariable V3 regions of the bacterial 16 S rRNA gene45. The bacterial primers were F338-GC: CGCCCGCCGCGCCCCGCGCCCGGCCCGCCGCCCCCGCCCCA-CTCCTACGGGAGGCAGCAG and 518R: ATTACCGCGGCTGCTGG. The archaeal primers were F787-GC: CGCCCG-CCGCGCGCGGCGGGCGGGGCGGG-GGCACGGGGGGCGATTAGATACCCSBGTAGTCC and 1059R: GCCATGCACC-WCCTCT. Barcodes were modified in the 3’terminus of the forward primer for sample multiplexing. For sequencing, libraries were normalized and pooled for MiSeq (Illumina, USA). Polymerase chain reaction (PCR) amplifications were performed in a 30 µL reaction mixture, which contained 1 µL of each primer (10 µmol/L), 20 ng of genomic DNA, 15 µL of 2×Taq master Mix and was added with nuclease free water to 30 µL. PCR cycling was conducted according to the following thermal conditions: an initial denaturing step at 95 ℃ for 3 min and 5 cycles of denaturing at 94 ℃ for 20 s; annealing at 55 ℃ for 20 s and extension at 72 ℃ for 30 s. For the last cycling program, the final extension step was held at 10 ℃. Using the Wizard SV Gel and PCR Clean-up System (Promega Corporation, Madison, WI, USA), the applicants were purified in accordance with the manufacturer’s instructions. Qubit (Life Technologies, Carlsbad, CA, USA) was employed to quantify the samples. The amplification inhibition of all DNA samples was verified through sample dilution. The D-code Universal Mutation System DGGE electrophoresis System from BIO-RAD was employed to analyze the PCR products. The amplified product was purified using an Easy Pure Quick Gel Extraction Kit (Beijing Quanshi Gold, China) and eluted with 30 µL of sterile water. The sequencing results of the amplified products with dissimilar results were broadcast in the NCBI database after being sent to Sangon Biotech (Shanghai, China) Co., Ltd (submissions grp:5210823; ID:1854608). The dataset was used to identify operational taxonomic units (OTUs)46 by clustering sequences at 90% sequence identity and selecting a representative sequence for each cluster. Results and discussion Effect of CWs on AMD treatment The dynamic variation characteristics of fundamental physicochemical parameters and heavy metal concentrations in AMD after treatment in the constructed wetland system are illustrated in Table 1.”Raw” refers to the untreated influent liquid of the constructed wetland, used as a baseline for comparing the effects of CW treatment on liquid-phase parameters. Results indicated that the wetland system had a substantial regulatory impact on AMD acidity. The influent pH was 5.18 (acidic), whereas the effluent pH at Outlet ⓓ increased to 7.41 (neutral), resulting in a 2.33 unit elevation. The sulfur purification efficiency was noteworthy: sulfate decreased from 4890 mg/L to 1992 mg/L (59.3% removal), sulfide decreased from 0.97 mg/L to 0.39 mg/L (59.79% reduction), and no elemental sulfur was detected. The purification efficiency of the wetland system stems from the synergy of multiple mechanisms47. This result was inconsistent with that reductive bacteria in the wetland generate alkalinity during salt reduction, which not only increases effluent pH but also reduces sulfate concentration48. SRB can utilize one mole of sulphate to generate one mole of sulphide and also result in the generation of alkalinity which raises the pH of CWs with acidic conditions. Sulfur in the sediment is an intermediate product created by sulfide oxidation, primarily by a chemical reaction mechanism, rather than the direct reduction product of sulfate under sediment reducing conditions. Sulfur can build up in silt or floodwater. Yet since sulfur is quickly oxidized by air contact and only remains in the soil for a brief period of time, it is typically challenging to find in sediment. Ammonia nitrogen (NH4+) decreased from 2.42 mg/L to 0.71 mg/L (a removal of 1.71 mg/L) for nitrogen parameters, while nitrate (NO3−) increased from 3.31 mg/L to 6.87 mg/L. When NH4+ enters a CWs, it can be taken up by plants, immobilized into the organic material, or oxidized to NO3− by microorganisms49. The removal rate of 63.6% was achieved at Outlet ⓓ, as COD progressively decreased from 169.7 mg/L to 61.8 mg/L in terms of organic matter removal. Following treatment, the COD data met China’s requirements for water quality protection before being released into lakes (COD < 100 mg/L). The previous study reported that COD is reduced mainly by microbial biodegradation and substrate adsorption50. This CWs has a genuinely sensible plant arrangement, and the plants are flourishing with a wide range of microorganisms. The removal of heavy metals was exceptional: Fe concentration decreased from 8.3 mg/L to below the detection limit (< 0.03 mg/L, 100% removal efficiency); Mn decreased from 84.3 mg/L to 32.2 mg/L (61.8% removal); Zn and Co were reduced from 1.25 mg/L and 0.35 mg/L to 0.10 mg/L and 0.18 mg/L, respectively (92.0% and 48.6% removal); and Ca and Mg removal efficiencies improved to 56.4% and 60.6%, respectively. Sulfide ions produced during bacterial metabolism can form precipitates with metal ions51, thereby lowering metal concentrations. Iron removal involves plant root uptake, substrate adsorption, and oxidative precipitation induced by pH elevation52; notably, iron hydroxide precipitates form root iron plaques that can adsorb other metals, synergistically enhancing their removal. Concurrently, increased system pH enhances the negative charge density on the surfaces of clay minerals, hydrated oxides, and organic matter in wetland sediments, strengthening their adsorption capacity for heavy metals. Plant roots further intercept heavy metal ions through barrier and adsorption functions, collectively achieving efficient pollutant removal in AMD. Table 1The main physical and chemical characteristics and ionic composition of AMD and the wetland treatment system (values in parentheses represent the standard deviation (SD) of the three replicates).Liquid SamplePhysicochemical characteristicsMajor ionspHCODFeMnZnCoCaMgNH4+NO3−SO42−S²⁻(mg L− 1)(mg L− 1)Raw AMD5.18(0.05)169.7(2.1)8.3(0.2)84.3(1.5)1.25(0.02)0.35(0.01)301.5(5.8)651.3(12)2.42(0.07)3.31(0.11)4890(21)0.97(0.03)Outlet ⓐ5.98(0.05)99.9(2.2)1.17(0.1)62.3(1.4)0.48(0.03)0.24(0.01)199.3(6.2)477.2(15)1.37(0.05)4.81(0.13)2783(73)0.88(0.07)Outlet ⓑ6.68(0.07)83.2(2.5)0.78(0.08)55.3(1.9)0.3(0.02)0.22(0.01)164.4(6.1)320.3(13.2)1.09(0.08)5.62(0.12)2585(52)0.8(0.05)Outlet ⓒ7.08(0.07)66.7(2.1)0.49(0.06)50.6(1.8)0.25(0.02)0.19(0.01)146.3(5.5)301.6(11.1)1.07(0.05)6.61(0.11)2017(20)0.57(0.03)Outlet ⓓ7.41(0.09)61.8(2.1)0.03(0)32.2(1.7)0.1(0.02)0.18(0.02)131.5(4.8)256.5(9.8)0.71(0.05)6.87(0.12)1992(23)0.39(0.02)National Emission Standard6 ~ 9100 ~ 20064.02 ~ 41.0//15//1.0 Table 1The main physical and chemical characteristics and ionic composition of AMD and the wetland treatment system (values in parentheses represent the standard deviation (SD) of the three replicates).Liquid SamplePhysicochemical characteristicsMajor ionspHCODFeMnZnCoCaMgNH4+NO3−SO42−S²⁻(mg L− 1)(mg L− 1)Raw AMD5.18(0.05)169.7(2.1)8.3(0.2)84.3(1.5)1.25(0.02)0.35(0.01)301.5(5.8)651.3(12)2.42(0.07)3.31(0.11)4890(21)0.97(0.03)Outlet ⓐ5.98(0.05)99.9(2.2)1.17(0.1)62.3(1.4)0.48(0.03)0.24(0.01)199.3(6.2)477.2(15)1.37(0.05)4.81(0.13)2783(73)0.88(0.07)Outlet ⓑ6.68(0.07)83.2(2.5)0.78(0.08)55.3(1.9)0.3(0.02)0.22(0.01)164.4(6.1)320.3(13.2)1.09(0.08)5.62(0.12)2585(52)0.8(0.05)Outlet ⓒ7.08(0.07)66.7(2.1)0.49(0.06)50.6(1.8)0.25(0.02)0.19(0.01)146.3(5.5)301.6(11.1)1.07(0.05)6.61(0.11)2017(20)0.57(0.03)Outlet ⓓ7.41(0.09)61.8(2.1)0.03(0)32.2(1.7)0.1(0.02)0.18(0.02)131.5(4.8)256.5(9.8)0.71(0.05)6.87(0.12)1992(23)0.39(0.02)National Emission Standard6 ~ 9100 ~ 20064.02 ~ 41.0//15//1.0 The main physical and chemical characteristics and ionic composition of AMD and the wetland treatment system (values in parentheses represent the standard deviation (SD) of the three replicates). 5.18 (0.05) 169.7 (2.1) 8.3 (0.2) 84.3 (1.5) 1.25 (0.02) 0.35 (0.01) 651.3 (12) 2.42 (0.07) 3.31 (0.11) 4890 (21) 0.97 (0.03) 5.98 (0.05) 99.9 (2.2) 1.17 (0.1) 62.3 (1.4) 0.48 (0.03) 0.24 (0.01) 477.2 (15) 1.37 (0.05) 4.81 (0.13) 2783 (73) 0.88 (0.07) 6.68 (0.07) 83.2 (2.5) 0.78 (0.08) 55.3 (1.9) 0.3 (0.02) 0.22 (0.01) 320.3 (13.2) 1.09 (0.08) 5.62 (0.12) 2585 (52) 0.8 (0.05) 7.08 (0.07) 66.7 (2.1) 0.49 (0.06) 50.6 (1.8) 0.25 (0.02) 0.19 (0.01) 301.6 (11.1) 1.07 (0.05) 6.61 (0.11) 2017 (20) 0.57 (0.03) 7.41 (0.09) 61.8 (2.1) 0.03 (0) 32.2 (1.7) 0.1 (0.02) 0.18 (0.02) 256.5 (9.8) 0.71 (0.05) 6.87 (0.12) 1992 (23) 0.39 (0.02) Wetland sediment physicochemical indicator analysis As the distance from the acid reservoir increased, the pH values gradually increased from 6.57 at Outlet ⓐ to 7.09 at Outlet ⓓ, as shown in Fig. 2. Every recorded pH level was more than 6.5, which is higher than the cutoff point set by China’s Class II sediment quality requirements. The ionic imbalance in sediments brought on by heavy metal leaching from the acid reservoir is thought to be the cause of this pH volatility, which in turn affects the pH properties of sediments. The luxuriant wetland vegetation and associated plant residues significantly enhance sediment nutrient enrichment, resulting in elevated TN concentrations across all four sampling sites (1.67, 1.54, 1.46, and 1.38 g/kg, respectively), which markedly exceed China’s cultivated sediment plow layer average (1.05 g/kg). SOC decomposition primarily occurs through microbial metabolic processes, where heterotrophic microorganisms utilize carbon sources for biosynthetic and nutrient cycling purposes53. In the present study, wetland sediments exhibited a mean SOC concentration of 99.1 g/kg. As a crucial nutrient for biological systems, the increased nitrogen levels ensure sufficient nitrogen availability for sediment microbiota. The wetland system demonstrates a remarkable capacity for acidic wastewater purification, as evidenced by the declining TN and SOC contents in sediments. This phenomenon likely results from reduced wastewater toxicity at sites ⓒ and ⓓ post-treatment, along with vigorous downstream vegetation growth and enhanced microbial activity that collectively promote sediment nutrient assimilation. Fig. 2Variation of pH, SO42−, S, S2−, TN and Organic carbon of wetland sediments. Variation of pH, SO42−, S, S2−, TN and Organic carbon of wetland sediments. Sulfur, as a crucial interfacial element connecting the hydrosphere, biosphere, atmosphere, and lithosphere, plays a pivotal role in governing biogeochemical cycles across various strata through its sedimentary behavior. Moreover, it exerts fundamental influences on terrestrial environmental evolution, sediment fertility dynamics, and vegetation development processes. Sulfate contamination represents a predominant sediment-related environmental concern in mining regions54. While sulfide minerals constitute the dominant sulfur species in mining debris, pyrite oxidation generates substantial sulfate quantities55. Metallurgical operations annually produce significant volumes of waste materials, predominantly comprising pyrite and other sulfide minerals. The abundant heavy metal ions present in mining areas interact with S2− to form stable metal sulfides within sediments, in addition to sulfates produced through oxidative weathering of detritus under aerobic conditions. As demonstrated in Fig. 3, sulfate accumulation in sediments derived from acidic wastewater exhibited a pronounced increase from 8.20 to 9.00 g/kg, markedly surpassing the global average sulfur concentration (0.7 g/kg). Simultaneously, sedimentary sulfide levels showed an increment of 1.50 g/kg. Sulfur concentrations exhibited a range of 193 mg/kg at Outlet ⓐ to 269 mg/kg at Outlet ⓓ, with intermediate values observed at the remaining sampling sites. Rather than direct sulfate reduction products, sulfide oxidation intermediates constitute the predominant source of sedimentary sulfur under reducing conditions. While sulfur accumulation demonstrates a positive correlation with response duration56, the transient nature of sedimentary habitats and rapid oxidative transformation upon atmospheric exposure present methodological challenges for quantitative analysis. Systematic characterization of pollution dynamics and metal behavior is made possible by thorough speciation investigation of heavy metals in a variety of substrates. Because of the significant differences in metal mobility and bioavailability among various chemical forms, speciation analysis has become a crucial metric for assessing ecological risks and toxicological effects57. The most popular speciation classification method, the well-known Tessier sequential extraction process, divides sediment-associated metals into five operationally defined fractions58: exchangeable, carbonate-bound, iron-manganese oxide-bound, organic matter-bound, and residual. This study uses this known methodology to identify the predominant kinds of metal buildup in wetland substrates and evaluate the associated ecological concerns. Fe-Mn oxide-bound (2.09%–3.98%) and residual (7.13%) fractions make up the majority of the iron in the sediments, with carbonate-bound (1.51%–2.97%) forms making up smaller amounts. These results are in line with earlier studies conducted by Li et al.59. The absence of soluble iron suggests that residual and Fe-Mn oxide-bound forms are the main speciation. Regional pyrite deposits, where metals are integrated into silicate mineral lattices with low environmental mobility and bioavailability, are suggested by the high residual iron content60. Since lower pH levels promote activation and change heavy metal solubility equilibria, sediment pH is a crucial factor affecting iron speciation61. Alkaline circumstances increase iron-manganese oxide binding (from 15.9% to 22.8%) in the reducing wetland environment, where oxide surfaces function as efficient adsorbents and iron is mostly found as Fe(OH)2 and FeOH2+62. Between Outlets ⓐ and ⓓ, manganese concentrations rose from 0.41 to 1.77 g/kg. Manganese in sediment usually exists as divalent, trivalent, or tetravalent cations; speciation is contingent upon pH, redox potential, and sediment properties63. The major forms are residual (55.1%−71.8%) and Fe-Mn oxide-bound (14.0%−35.6%), as shown in Fig. 3-②. The potential for co-precipitation between Mn, Zn, Cu, and iron oxides as well as Fe, Zn, and manganese oxides has been demonstrated in earlier research64, establishing these oxides as the main substrates for metal adsorption and precipitation. Significantly, manganese speciation is pH-dependent, with leftover manganese changing into different forms in alkaline environments65. High hydroxide concentrations cause insoluble manganese hydroxide to develop, which increases Fe-Mn oxide associations and decreases residual and exchangeable fractions. Manganese presents serious ecological concerns in this artificial wetland. Plant roots absorb soluble forms of copper, a vital trace element and transition metal that promotes biological activity66. With concentrations rising from 50.8 mg/kg at Outlet ⓐ to 138.4 mg/kg at Outlet ⓓ (Fig. 323-③), the created wetland successfully processed copper while keeping it well below China’s agricultural soil standard (200 mg/kg). The most common copper was carbonate-bound (49.8%–65.3%), with Fe-Mn oxide connections coming in second (26.2%−30.75%). According to previous research, sulfide precipitation implies SRB mediated elimination, while copper’s strong electrophilic attraction promotes widespread complexation with organic matter67. Fe-Mn oxide-bound (13.6%−16.7%) and residual (56.2%−66.1%) fractions predominated, according to zinc speciation studies, which showed consistent trends across locations. About 25% of the zinc was in potentially bioavailable forms (exchangeable, carbonate-bound, and water-soluble). Zinc solubility and chelation are strongly influenced by pH, with acidic environments increasing cation mobility68. This result supports known pH-dependent solubility behavior by confirming that water-soluble zinc decreases as pH rises69. Alkaline conditions increase reducible zinc fractions by promoting the precipitation of zinc hydroxide and improving the adsorption capacity of Fe-Mn oxide. One of the most hazardous heavy metals, cadmium (Cd), showed steady concentration increases from Outlet ⓐ to Outlet ⓓ (maximum 3.30 mg/kg), however it stayed below ecological safety criteria (T/SSSC 008 − 2,024: 6.2 mg/kg). Cadmium speciation is crucial for both ecological effects and cleanup effectiveness70. Fe-Mn oxide/hydroxide co-precipitation was a major removal mechanism, and residual, organic-bound, and carbonate-bound forms predominated, as Fig. 3-⑤ illustrates. Exchangeable portions dropped from 11.6% to 3.9% downstream, while organic cadmium averaged 24.3%. Cleaner sediments usually have lower concentrations of the highly mobile exchangeable fraction, which is an important sign of pollution. The built-in wetland successfully reduces possible environmental risks associated with heavy metals. Fig. 3The content and speciation of Fe (a), Mn (b), Zn (c), Cu (d) and Cd (e) at each sampling. The content and speciation of Fe (a), Mn (b), Zn (c), Cu (d) and Cd (e) at each sampling. The purification mechanism of wetland microorganisms for AMD Microbial communities are essential for managing AMD in constructed wetlands71. To further examine the kinds and roles of the microbial communities in the CWs, look into the metabolic processes and mechanisms of functional microorganisms, and look into how AMD is treated in the constructed wetland from the viewpoint of microorganisms. PCR-DGGE was used to investigate the diversity and composition of microbial communities in the CWs. The fact that every sequencing coverage rate is more than 0.91 indicates that the volume of sequencing being carried out this time is suitable. The sequencing results can meet sample analysis standards and precisely represent the characteristics of the microorganisms in the sample72. Analysis of the structural composition of microbial community Fig. 4Microbial community composition distribution of the samples at the phylum level. Microbial community composition distribution of the samples at the phylum level. Figure 4 illustrates the horizontal percentage of bacterial phyla in sediment samples. The microbial community formation may have been influenced by wetland plant species, as distinct microbial communities were identified in a variety of wetland environments. Proteobacteria dominated the constructed wetland system, with an average abundance of 58.0%. It was the most important phylum of microbes and is frequently observed in sewage treatment systems. A diverse array of taxa associated with the nitrogen, sulfur, and carbon cycles is found in proteobacteria, which are responsible for the removal of pollutants73. Research has demonstrated that specific bacteria from the Proteobacteria phylum will execute the sulfate reduction pathway in an anaerobic environment with high sulfate levels74. In addition, the dominant bacteria were Bacteroidetes, Acidobacteria, Chloroflexi, Gemmatimonadetes, Verrucomicrobia, and Nitrospinae. The Bacteroidetes are the primary degraders of complex carbohydrate-based biomass75,76, Actinobacteria have the capacity to metabolize organic matter, and Acidobacteria possess numerous encoded cellulases and hemicellulases77. Previous research has demonstrated these findings. The genus Bacteroidetes was found to be instrumental in the remediation of AMD in sulfidogenic bioreactors by Aoyagi et al.78 as a result of the collaborative efforts of co-existing bacteria. It is capable of hydrolyzing carbohydrates into monosaccharides and subsequently converting them into volatile fatty acids such as lactic and acetic acids, which are essential for the carbon cycle. Consequently, it is an efficient degrader of complex carbohydrates79. In addition, the phylum Nitrospinae, which includes a variety of nitrite-oxidizing bacteria80, accounted for 3.33% to 6.25%. These bacteria are essential for the nitrogen cycle, as they convert toxic nitrite into less hazardous nitrate. They prosper in a variety of environments, including marine sediments, freshwater lakes, and soil, and are frequently observed in regions with high microbial activity. Nitrospinae’s presence suggests a well-established nitrification process, which is critical for the regulation of ecological balance and the promotion of plant growth by increasing the accessibility of nitrogen. Figure 5 illustrates the distribution of archaea levels in sediment samples. The investigation identified two primary archaeal groups: Thaumarchaeota and Euryarchaeota. Through the utilization of acetic acid, Methanogens such as Methanoregula, Methanosphaerula, Methanocella, Methanoculleus, and Methanolinea are critical to the production of methane within the Euryarchaeota category80. Furthermore, the Thaumarchaeota group comprises specific archaea, such as Nitrosophaera and Nitrosopumilus, which are accountable for the oxidation of ammonium (NH4+) to nitrite (NO2−)81. Nitrogen cycling within the sediment ecosystem is significantly affected by this process. Fig. 5Archaea community composition distribution of the samples at the family level. Archaea community composition distribution of the samples at the family level. Fig. 6Relative abundance of functional genera in microbial samples. Relative abundance of functional genera in microbial samples. The bacterial functional communities in the sediment samples from the four wetland sampling locations include functional groups such as hydrolysis, nitrogen metabolism, sulfur metabolism, and Fe-Mn cycling (Fe-Mn) (Fig. 6). Each community has complete functions and can degrade a variety of complex organic substances, achieving the reduction and transformation of elements such as nitrogen, sulfur, iron, and manganese. Strong protease, amylase, and lipase activity found in hydrolytic bacterial communities (such as those of the Latescibacteria, Sphaerochaeta, Iganavibacerium, Bacillus, and Optitutus) can decompose humic acid and sediment organic matter into small molecular carbohydrates82. Cellulose and plant waste can be broken down by the Acidobacteria phylum subgroups Gp3, Gp7, and Gp13, which provide substrates for the fermentation of methanogenic bacteria83. The ideal pH range for these bacterial communities is between 7.0 and 7.5. The percentage of bacteria that produce hydrolytic acid grows from 21.9% to 31.3% as the sediment’s pH approaches neutrality, indicating that pH has a regulating influence on the hydrolytic bacterial communities. Methanogens serve as crucial carbon cycle drivers84. When there are few external electron acceptors, which include O2, Fe3+, NO3−, and SO42−, complex organic matter has to combine with Methanogens to finish the anaerobic breakdown of organic matter after producing short-chain fatty acids, H2, CO2, and alcohols85. The percentage of Methanogens in the four samples is 79.7%, 84.3%, 86.5%, and 68.0%, respectively, according to cloning library analysis. Methanosarcina and Methanosarcinales have multiple metabolic pathways and can use H2, methyl compounds, and acetic acid86, demonstrating the metabolic diversity of Methanogens in supporting the carbon cycle. Hydrogen-dependent Methanogens (genus Methanoregula, Methanolinea, and Methanocella) use H2 and formate as electron donors to reduce CO2 to produce CH487. The genus Methanomassiliicoccus only uses H2 or methanol to produce methane88, while the methyl-dependent Methanosphaerula uses methanol, methylamine, etc. The nitrogen cycle is a major factor in the biological equilibrium and succession of wetlands89. The four samples showed the presence of functional microbes related to nitrogen metabolism in the following proportions: 32.2%, 33.7%, 32.2%, and 32.3%. The Geobacter and Azotobacter genera, in conjunction with wetland plants, are involved in a position to nitrogen fixation, which involves the conversion of molecular nitrogen to ammonia and the preservation of ecological productivity90. The two-step nitrification process for the reactions NH4+→NO2− and NO2−→NO3− is catalyzed by both ammonia-oxidizing (Nitrososphaera, Nitrosopumilus genera) and nitrifying (Nitrospira, Nitrospina genera) bacteria. In anaerobic conditions, genera such as Parvularcula and Thiobacillus can progressively reduce NO3− and NO2− to N2, thereby regulating nitrogen transformation and loss91. The metabolic activities of these bacteria are crucial for the preservation of ecological equilibrium, particularly in soil environments and aquatic ecosystems. SRB are a class of anaerobic bacteria in the mining area, which can use organic matter as a carbon source and H produced in its biofilm under an anaerobic or anoxic fraction92. The sulfate is gradually reduced to hydrogen sulfide. Bacteria linked to sulfur metabolism were present in 24.0%, 25.6%, 25.8%, and 25.0% of the four samples, respectively. The genus Desulfuromonas (typical SRB bacteria) had an abundance of 2.08% to 3.13% as the pH of the sediment (6.57→7.09) increased. It can also reduce Fe3+and Mn4+. The genera Desulfobacca, Desulfarculus, etc., can oxidize acetate to produce CO2 while reducing SO42− to S2− and forming sulfide precipitates with Fe3+to lessen the toxicity of S2−93. By reducing Fe3+ to Fe2+ and forming S2− precipitates, the Geothrix and others can increase the activity of SRB bacteria94 (the abundance of SRB bacteria in the 4 samples was 12.5% to 14.6%) and facilitate the removal of sulfate (the water sample’s sulfate removal rate was 28.4%). Between 7.29% and 6.25% of the samples included sulfur-oxidizing bacteria (Sulfuricurvum, Acidiferrobacter genus). While the latter is heat- and acid-tolerant and oxidizes Fe2+, the former can use sulfides and NO3− (anaerobic denitrification). Its function varies as the wetland’s redox potential alternates. The reduction and oxidation processes that comprise the iron cycle are motivated by microorganisms95. The proportion of microorganisms implicated in iron cycling in the four samples analyzed varied from 21.1% to 27.1%. In the anaerobic oxidation of short-chain fatty acids, the Geobacter employs Fe3+ as an electron acceptor. Furthermore, it is capable of utilizing Mn(IV) and NO3− when acetic acid serves as the carbon source96. Geobacter can also transfer electrons directly to the Methanosarcina genus, which collaborates to degrade acetic acid and generate methane. It secretes riboflavin, which functions as an electron transporter by transporting electrons through the extracellular membrane c-cytochrome. To decrease Fe3+ and Mn(IV), heterotrophic iron-reducing bacteria from the Geothrix and Albidiferax genera utilize organic matter as an electron source97. Under neutral conditions, Fe2+ is easily oxidized. The Sideroxydans genus can oxidize biotic iron at oxygen levels below 50µmol/L98. The Thiobacillus genus, which is nitrate-dependent, oxidizes Fe2+ by reducing it to ammonium and using NO3− as the electron acceptor99. This process demonstrates how microorganisms regulate and drive various iron cycle processes. The degradation reaction mechanism Fig. 7Speculation on the reaction mechanism of degradation of CWs. Speculation on the reaction mechanism of degradation of CWs. The degradation reaction mechanism of the constructed wetland is suggested to be built on the experimental results and the microbial community structure analysis. It hypothesizes that the reaction mechanism is depicted in Fig. 7. Macrophytes like Phragmites australis, Typha angustifolia, and Arundo donax oxygenate the rhizosphere and supply carbon sources through root exudates100, thus establishing an ideal habitat for microbial growth in the constructed wetland system. Through microbial metabolism, the system increases the pH of AMD from acidic (5.18) to neutral (7.41), which is favorable for the processes that remove pollutants later on. Methanogens (Methanoregula, Methanosarcina) use the hydrolytic bacteria (Latescibacteria, Acidobacteria) that break down complex organic materials into short-chain fatty acids as substrates to encourage carbon mineralization in microbially driven element cycling101. Utilizing organic matter as carbon sources, SRB (Desulfuromonas, Desulfobacca) convert sulfate to sulfide, which then reacts with heavy metal ions to generate precipitates of insoluble sulfide102. To improve metal adsorption on iron-manganese oxides, Sideroxydans oxidizes Fe2+ in a microaerophilic environment. Meanwhile, iron-reducing bacteria (Geobacter) and manganese-reducing bacteria (Geothrix) transform oxidized metals into precipitable forms103. Heavy metals are eliminated by precipitation, adsorption, and speciation change in the context of heavy metal immobilization. In particular, sulfide from sulfate reduction and hydroxide from pH increase create stable precipitates (such as Fe(OH)₃ and ZnS), and cationic metals (such as Mn²⁺ and Cd²⁺) are adsorbents of negatively charged substrates and root iron plaques, transforming metals from bioavailable fractions to residual fractions with minimal ecological risk. In combination, these procedures effectively remove pollutants like COD, sulfate, Fe, and Zn; the removal effectiveness of Fe reaches 100%, while that of Zn reaches 92.0%. This maintains the constructed wetland system’s purification efficiency steady. Conclusion The pH rose from 5.18 to 7.41 as a result of the system’s significant neutralizing capacity. It also showed remarkable removal efficiency for a range of contaminants. Sediments, which are mostly found in low-bioavailability residual fractions, were found to be the main reservoir for heavy metals. Increased pH levels (6.57–7.09) and improved substrate adsorption promoted the development of less bio-available metal species, such as ion-exchangeable, water-soluble, and carbonate-bound forms, thereby reducing environmental hazards. Proteobacteria was found to be the predominant phylum in microbial community analysis, with a variety of functional groups: hydrolytic bacteria (like Latescibacteria) that break down organic matter; nitrogen-cycling bacteria that mediate nitrification/denitrification processes; sulfur-metabolizing bacteria that reduce sulfate and precipitate metal sulfides; and Fe-Mn cycling bacteria that facilitate redox transformations. Rhizosphere oxygenation and carbon supply from vegetation (Phragmites, Typha, Arundo) were critical factors that supported microbial activity. System stability was preserved through consistent biomass harvesting. This system remediates contaminants via pH regulation, metal immobilization, and vegetation-microbe synergy, offering a scalable, eco-friendly solution for heavy metal pollution and ecological restoration. Future research should clarify pH-driven metal speciation via in situ micro-spectroscopy, quantify vegetation-microbe synergy with multi-omics, validate long-term stability or scalability, and strengthen sediment heavy metal risk assessment using bioassays and trophic transfer analysis.