The Study of Nitrogen and Phosphorus Removal Efficiency in Urbanized River Systems Using Artificial Wetland Systems with Different Substrates

Abstract

This study evaluated the effectiveness of five commercial substrates (zeolite, volcanic rock, gravel, magic rack, and ceramic pellets) in removing nitrogen (N) and phosphorus (P) from urban river systems using constructed wetlands. By employing X-ray CT and NGS technologies, we analyzed the physical structure of the substrates and the microbial communities they harbor. The results indicated that volcanic rock and ceramic pellets, due to their high porosity and specific surface area, performed exceptionally well in nitrogen and phosphorus removal. Specifically, the microbial systems with these two substrates achieved ammonia nitrogen removal rates of 89.86% and 88.45%, total nitrogen removal rates of 78.78% and 74.97%, and total phosphorus removal rates of 92.67% and 80.82%, respectively, within a 7-day period. Furthermore, the microbial communities on volcanic rock and ceramic pellets were more diverse, which correlated with their high pollutant removal efficiency. The study further elucidated the synergistic role of substrate characteristics and microbial community structure and function in nitrogen and phosphorus removal, enhancing the understanding of the purification mechanisms in constructed wetlands. These findings provide a scientific basis for the ecological restoration of urban rivers and are significant for improving the quality of urban water environments.

1. Introduction

Urban point sources and agriculture are the main drivers of N and P pollution worldwide, triggering eutrophication processes in many freshwater lakes and streams [1,2]. Their excessive input disrupts the balance of aquatic ecosystems, leading to degraded water quality and ecological health [3]. Constructed wetlands, as an ecological engineering technology, have become one of the preferred technologies for treating polluted river water due to their large buffering capacity, easy management, and low infrastructure and operational costs [2]. In constructed wetland systems, the choice of substrate is crucial for the removal efficiency of pollutants [4]. The substrate’s irregular, porous structure, as well as its different specific surface area, porosity, and surface charge, has a significant impact on the adsorption of pollutants in water [5]. These different structural characteristics provide various carriers for microbial growth, promoting the formation of biofilms with varying characteristics, thus affecting the formation and function of biofilms [6].

While microbial films play a crucial role in removing pollutants in constructed wetlands [7], microorganisms, through their cellular metabolic functions and structural components, adsorb, decompose, and transform pollutants such as nitrogen and phosphorus in water, achieving water purification [8]. The adsorption and decomposition capacity of microbial films is closely related to their community structure and diversity [9,10]. The biodiversity of microbial films in constructed wetlands significantly influences pollutant removal efficiency by providing a diverse range of functional genes and metabolic pathways, enhancing adaptability to different pollutants, and increasing catalytic efficiency [4].

This study selected five commonly used substrates in artificial wetlands: zeolite, volcanic rock, gravel, magic rack, and ceramic pellets. Zeolite was chosen for its unique porous structure and ion exchange capacity, effectively adsorbing and exchanging cations in wastewater treatment, especially showing potential effects on the removal of ammonia nitrogen [7,11]. Volcanic rock was selected for its porous nature and high specific surface area, providing surfaces for microbial attachment and helping to increase the system’s oxygen transfer efficiency, thus promoting aerobic and anaerobic processes [12]. Gravel, as a traditional filtering medium, has become an ideal material choice due to its physical stability and extensive application history [13]. Magic rack, as a new type of environmentally friendly material, was selected for its potential special chemical components and structure, and we aim to explore the potential of this new material in improving the efficiency of nitrogen and phosphorus removal [14]. Ceramic pellets were chosen for their durability and adjustable pore characteristics, which can withstand different environmental conditions, making them suitable for long-term operation in constructed wetland systems [15]. The selection of these five substrates covers a range of different types and characteristics of materials, allowing for a comprehensive assessment of the application characteristics of commonly used materials in constructed wetland systems.

Although studies have focused on the pollutant removal efficiency of substrates in constructed wetland systems, few have compared the performance differences in commonly used substrates in engineering and deeply explored the impact of the substrate–microbe systems they form on the treatment of slightly polluted river water. The innovation of this study lies in systematically comparing the effects of five different substrates on nitrogen and phosphorus removal and exploring how substrate characteristics affect the structure and function of microbial communities, thereby influencing water purification effects. This is of great significance for gaining a deeper understanding of the purification mechanisms of constructed wetland systems. The study will help optimize the design and management of constructed wetlands, improving the efficiency and sustainability of nitrogen and phosphorus removal in urban rivers. The research not only provides important insights into urban river pollution control technologies but also plays a significant role in improving urban water quality and ecological health, thereby promoting sustainable urban development [16].

2. Materials and Methods

2.1. Materials

2.1.1. Test Substrates

This study selected five commercially available products as experimental substrates, including zeolite, volcanic rock, gravel, magic rack, and ceramic pellets. In the study, the zeolite used was sodium zeolite, sourced from Henan Province, China, with the main component being aluminosilicate minerals. The volcanic rock was collected from Heilongjiang Province, China, with main components including alkali feldspar and plagioclase feldspar. The selected gravel was granite gravel, sourced from Hebei Province, China, with the main component being silicon dioxide. The magic rack is an artificially synthesized composite material, manufactured in Shanghai, China, made from the mixture of agricultural and forestry straw biomass power generation by-products and granite gravel, processed through sintering, with main components including silicon dioxide and calcium oxide. The ceramic pellets selected were made from kaolin and feldspar, sourced from Jiangxi Province, China, with the main component being silicon dioxide. The particle size was screened to fall within the range of 1.0–1.5 cm to ensure the consistency and comparability of the text [12].

2.1.2. Test Solution

The test water was prepared in accordance with the “Urban Sewage Discharge Standard GB18918-2002” Class B standard, using the effluent from an urban sewage treatment plant as the base [7], and adding certain amounts of (NH₄)₂SO₄, KNO₃, and K₂HPO₄ to the base water to create the test water with a C/N ratio of 10 [17], maintaining the ammonia nitrogen content at 15.0 ± 1.0 mg/L, the total nitrogen content within the range of 20.0 ± 1.0 mg/L, and the total phosphorus content within the range of 1.5 ± 0.5 mg/L. The pH was controlled at 7.0 ± 0.2, and the dissolved oxygen level was maintained between 6.0 and 8.0 mg/L. The detailed initial water quality parameters, including ammonia nitrogen, are presented in Figure 3, total nitrogen in Figure 4, and total phosphorus in Figure 5.

2.2. Research Methods

2.2.1. Pore Observation of Substrates

The nanoVoxel-2000 X-ray CT scanner from Sanying Precision Instrument Co., Ltd., located in Tianjin, China, was used to observe the pore structure of the substrate with a maximum resolution of 1 μm and a rapid scanning mode. X-ray CT was used to observe the pore structure of the substrates. By irradiating the sample with X-rays, different components are distinguished based on their density and atomic coefficients’ ability to absorb X-rays, creating a contrast that allows for the observation of the three-dimensional distribution and size of the pores [18,19]. These data support the subsequent analysis of the impact of the substrates on microbial attachment and pollutant removal efficiency by calculating the material’s porosity and specific surface area.

2.2.2. Analysis of Microbial Community Structure

Each of the five substrates, 50 kg each, was placed into square tanks with a volume of 63 L, with dimensions of length (cm) × width (cm) × height (cm) = 35 × 30 × 60, and filled with test water to the brim [20]. Following 48 h of adsorption culture at (25.0 ± 2.0) °C, static culture was conducted for 12 days. After culturing, portions of substrates from both upper and lower tank layers were taken, ultrasonically cleaned with physiological saline, and centrifuged at 160 r/pm for 15 min to collect mixed sediment for a community classification analysis (NGS). NGS technology was utilized to analyze microbial DNA in the substrates, providing detailed information on community structure presented in a bar graph format for clarity [21].

2.2.3. Denitrification and Phosphorus Removal Efficiency Test

Once the substrates were coated with biofilm, the tanks were interconnected with adjacent tanks to form a system, and 150 L of test water was added. A push flow device maintained uniform water flow within the system, with experimental water introduced from top to bottom into the substrate tanks. Complete water replacement in the tanks occurred every 24 h, ensuring a constant system water volume of 150 L. The system operated for 7 days at room temperature, with changes in pH, ammonia nitrogen, and total phosphorus content in the water recorded before and after treatment to assess the impact of different substrates on nitrogen and phosphorus removal effects. Through the above methods, each substrate was tested in independent tanks to eliminate interference from variables other than the substrate type. For a detailed visualization of the experimental setup, please refer to the schematic diagram provided in Figure 1 below.

2.2.4. Data Analysis Methods

Normality and homogeneity of variance were assessed using the Kolmogorov–Smirnov test and Levene’s test, respectively. When data met the criteria for a normal distribution and homogeneity of variance, one-way ANOVA (Analysis of Variance) was used to compare the removal efficiency differences between different substrates [22], followed by Tukey’s HSD (Honestly Significant Difference) test for multiple comparisons. When data did not meet the criteria for a normal distribution or homogeneity of variance, the non-parametric Kruskal–Wallis H test was employed for comparison. A correlation analysis was conducted using Spearman’s rank correlation. Experimental data were analyzed using SPSS statistical software, version 29.0.1, with the level of statistical significance set at p < 0.05.

In the article, we utilized an advanced language processing AI developed by Moonshot AI, known as Kimi, version 1.0, to assist in the translation of certain parts of the content.

3. Results and Analysis

3.1. Pore Analysis of Substrates

By conducting a detailed characterization of the pore structures of the five substrates, we obtained data on the porosity and specific surface area of each, as detailed in Supplementary Table S1. Combined with the comparative analysis results from Figure 2, volcanic rock has the highest specific surface area and a relatively large porosity, exhibiting a complex pore structure. Its intricate pore structure and good internal pore connectivity facilitate the flow of liquids along the pore walls, thereby promoting microbial attachment and the adsorption and removal of pollutants. In contrast, ceramic pellets have the highest porosity, but a significant proportion of their pores are closed, and their specific surface area is much lower than that of volcanic rock. This reduces the flow and contact of liquids within the substrate, thus affecting its efficiency in pollutant removal. The magic rack has a specific surface area similar to that of ceramic pellets, but its porosity is much lower than that of volcanic rock and ceramic pellets, and its large pore size and poor connectivity reduce the amount of water flowing through, limiting its potential for microbial attachment and pollutant removal. Zeolite has a relatively high specific surface area second only to volcanic rock, but its porosity is very low, similar to that of gravel, with its numerous small internal pores and poor connectivity leading to reduced water flow. Gravel has the lowest porosity and specific surface area among the five materials. For a detailed three-dimensional structural characterization and pore properties of the substrates, please refer to Figures S1–S5 in Supplementary Materials.

3.2. Microbial Biofilm Analysis

The sequence read counts and base pair numbers of the biofilms on the five types of substrates are presented in Supplementary Table S2, while the relative abundance, phylogenetic affiliations, and ecological niche information of the microbial communities are detailed in Supplementary Figures S6–S8.

Combining Supplementary Table S2 and Figures S6–S8, it can be concluded that the sequence number and base count of the biofilm on the zeolite substrate are relatively low, with Aeromonas and Planctomyces nitrifying bacteria showing a high degree of ecological niche overlap. The sequence number and base count of the biofilm on the volcanic rock substrate are relatively high, indicating a rich biomass and diversity of the attached microbial community, with Pseudomonas, which enhances denitrification, being highly abundant and beneficial for improving the denitrification efficiency of the biofilm. The sequence number and base count of the biofilm on the ceramic pellet substrate are the highest, showing greater diversity and biomass than the biofilm on volcanic rock, with Pseudomonas, Ardenticatenia_norark, and Aeromonas all exhibiting a high degree of ecological niche overlap. The sequence number and base count of the biofilm on the magic rack substrate are relatively low, similar to those on zeolite and gravel, but there are significant differences in ecological niches compared to microbial communities on other substrates, attracting specific microbial communities such as salt-tolerant denitrifying bacteria (Alishewanella), which is beneficial for enhancing the denitrification efficiency of the biofilm–substrate system. The gravel substrate has almost no internal pore structure, and the sequence number and base count of its biofilm are relatively low, indicating that microbial attachment and growth on its surface are not as good as on other substrates, thus affecting its microbial diversity and the functionality of the biofilm.

3.3. Treatment Effects of Substrate Membrane Hanging

The study investigated the purification efficiency of five substrate–microbe systems in artificially prepared wastewater in a simulated wetland environment. After a 12-day cultivation, the systems were used to treat the wastewater, and the effluent water quality in terms of pH, nitrogen, and phosphorus content was monitored and assessed.

3.3.1. pH

In the simulated wetland system, the pH values of the effluent were measured for the five substrate–microbe systems after a 12-day cultivation period. According to the data presented in Figure S9 of the Supplementary Materials, the pH values of the effluent from the five substrate–microbe systems remained stable within the neutral range of 7–8 after 24 h of wastewater treatment. This indicates that the tested systems have a significant positive impact on maintaining the stability of water body pH, which is crucial for preserving the balance of aquatic ecosystems and facilitating subsequent water purification processes.

3.3.2. Ammonia Nitrogen Removal Effect

In the simulated wetland system, we cultivated the five substrate–microbe systems for 12 days and subsequently measured the pH levels of the effluent. According to the data shown in Figure S9 of the Supplementary Materials, the pH levels of the effluent from these substrate–microbe systems remained stable within the neutral range of 7 to 8 after 24 h of wastewater treatment. This result indicates that the tested systems have a significant positive effect on maintaining the stability of water pH, which is crucial for preserving the balance of aquatic ecosystems and facilitating subsequent water purification processes.

All data points in the figures are represented as the mean ± standard error of the mean (SEM), with error bars included to display the variability of the data. In some figures, the confidence intervals (95% CIs) are also shown to further assess the reliability of the results. The same applies below.

As shown in Figure 3, the volcanic rock–microbe system outperformed others in ammonia nitrogen removal efficiency, with an average removal rate of 89.86%, demonstrating good stability. This high efficiency is attributed to the high porosity and specific surface area of volcanic rock, which provide abundant attachment sites and metabolic space for microbes, especially nitrifying and denitrifying bacteria that play a key role in the transformation of ammonia nitrogen, thereby enhancing the system’s overall ammonia nitrogen removal capacity. Ceramic pellets have the highest porosity among the five substrates, a characteristic that endows them with excellent liquid permeability and low air resistance, allowing for faster flow rates. Although their specific surface area is slightly lower than that of volcanic rock, limiting the contact area between the substrate surface and flowing liquid and affecting mass transfer efficiency, the high porosity provides ample attachment space for microbes. With extended cultivation time, the ammonia nitrogen removal rate of the ceramic pellet–microbe system gradually increased, reaching an average removal rate of 88.45%. The porous structure of ceramic pellets facilitates the transport of oxygen and nutrients, promoting the activity of nitrifying and denitrifying bacteria, thereby improving ammonia nitrogen removal efficiency. In the experiment, the magic rack–microbe system had an average ammonia nitrogen removal rate of 82.66%, showing its potential in ammonia nitrogen removal. Despite the magic rack’s specific surface area being similar to that of ceramic pellets and its lower porosity, which limits the contact area with flowing liquid and the transport of oxygen and nutrients, it still attracts specific microbial communities. In particular, the salt-tolerant denitrifying bacteria Alishewanella formed a dominant community on the magic rack, playing a positive role in ammonia nitrogen removal, highlighting the unique value of the magic rack–microbe system in the denitrification application of constructed wetlands. In this study, the zeolite–microbe system had an average ammonia nitrogen removal rate of 78.46%. Zeolite, with its high specific surface area, provides abundant attachment and growth space for microbes, which is conducive to enhancing the formation and function of biofilms. Although zeolite has a relatively low porosity, its unique pore structure, composed of silicate and aluminate tetrahedra, forms regular channels with molecular sieve effects. This structure, in conjunction with the adsorption–desorption of ammonia nitrogen and the activity of nitrifying bacteria, highlights the importance of the zeolite–microbe system in ammonia nitrogen removal. The gravel–microbe system, due to its physical stability and wear resistance, may provide stable ammonia nitrogen removal effects in long-term operations. In the experiment, the gravel–microbe system had an average ammonia nitrogen removal rate of 74.27%, which, although lower than the volcanic rock and ceramic pellet systems, still shows a certain removal capacity.

3.3.3. Total Nitrogen Removal Effect

As shown in Figure 4, the volcanic rock–microbe system demonstrated the highest total nitrogen removal efficiency, with an average removal rate of 78.78%. The porous structure of volcanic rock is conducive to the attachment of microbial communities, among which Pseudomonas, abundant in the system, possesses heterotrophic nitrification–aerobic denitrification (HN-AD) capabilities that are beneficial for enhancing the system’s total nitrogen removal efficiency. The total nitrogen removal efficiency of the ceramic pellet–microbe system averaged 74.97%. The porous structure of ceramic pellets favors the denitrification activities of dominant bacterial groups, improving the removal efficiency of total nitrogen. The magic rack–microbe system had an average total nitrogen removal rate of 70.35%. Specific salt-tolerant denitrifying bacteria Alishewanella communities in its biofilm also played a positive role in total nitrogen removal. The zeolite–microbe system had an average total nitrogen removal rate of 67.39%, and its unique pore structure did not significantly contribute to total nitrogen removal. The gravel–microbe system had an average total nitrogen removal rate of 65.19%, showing a certain removal capacity, although its removal efficiency was not as high as that of the volcanic rock and ceramic pellet systems.

3.3.4. Total Phosphorus Removal Effect

As shown in Figure 5, the volcanic rock–microbe system demonstrated the best phosphorus removal efficiency from polluted water, with an average total phosphorus removal rate of 92.67%. The magic rack–microbe system and ceramic pellet–microbe system also showed high removal rates, with average removal rates of 87.36% and 80.82%, respectively. In contrast, the zeolite–microbe system and gravel–microbe system had lower average removal rates, at 73.44% and 71.55%, respectively. Volcanic rock, magic rack, and ceramic pellets, due to their porous structure and high specific surface area, have strong phosphorus adsorption capabilities. Pseudomonas, which is abundant in the biofilm on these substrates, participates not only in nitrification and denitrification but also in phosphorus removal processes [23]. The structure of the substrates provides a rich array of attachment points for microbes, promoting microbial metabolic activities and enhancing the transformation and release rates of phosphorus, thereby improving phosphorus removal efficiency. Zeolite and gravel, with their lower porosity, have limited phosphorus adsorption capabilities, which also restrict microbial attachment and metabolic activities, further affecting phosphorus removal efficiency.

4. Conclusions and Discussion

4.1. The Impact of Substrate Characteristics on Microbial Community Composition

The three-dimensional structural characterization results reveal that volcanic rock and ceramsite stand out due to their rich pore distribution, high porosity, and large specific surface area. Specifically, the porosity of volcanic rock is 37.8%, with a specific surface area of 0.349 × 10⁶ m²/m³; the porosity of ceramsite is 44.9%, with a specific surface area of 0.131 × 10⁶ m²/m³. These characteristics provide numerous attachment points and ample activity space for microorganisms, affecting their access to oxygen and nutrients, and promoting the diversification and complexity of microbial communities, consistent with the research findings of Kadlec and Brix (2016) [15]. The larger specific surface area of volcanic rock and ceramsite increases the contact area between microorganisms and the substrate, potentially enhancing biofilm formation and pollutant adsorption capabilities [24]. In contrast, zeolite and gravel, with their lower porosity and specific surface area, limit microbial attachment and activity, resulting in a relatively simple microbial community structure [15]. Although the porosity of the magic rack (18.7%) is not high, its specific surface area (0.137 × 10⁶ m²/m³) is similar to that of ceramsite, providing suitable growth conditions for certain specific types of microorganisms, leading to unique structures and ecological niche characteristics of their microbial communities [16].

The interaction between the substrate and microorganisms is complex. The substrate serves as a physical support for microorganisms, and its chemical and biological properties may affect the metabolic pathways and expression of functional genes of microorganisms, thereby affecting the efficiency of pollutant removal. The ion exchange capacity of zeolite may attract specific types of microorganisms that play a role in the removal of nitrogen and phosphorus [25]. The biodegradability of the substrate can provide additional carbon sources for microorganisms, affecting the metabolic activities and functions of microbial communities [26]. At the same time, the stability of the substrate, as provided by gravel, may maintain the stability and persistence of microbial communities in long-term operations [27].

4.2. Impact of Substrate–Microbe Systems on Water Quality

The study results reveal the significant effects of different substrate–microbe systems in nitrogen and phosphorus removal. The volcanic rock–microbe system excels in ammonia nitrogen removal, achieving an average removal rate of 89.86%, with total nitrogen and phosphorus removal efficiencies reaching up to 78.78% and 92.67%, respectively. The ceramic pellet–microbe system also performs well, with an ammonia nitrogen removal rate of 88.45%, and total nitrogen and phosphorus removal rates of 74.97% and 80.82%, respectively. The magic rack–microbe system shows high potential in phosphorus removal, with an average removal rate of 87.36%, indicating its application value under specific conditions. Although the gravel–microbe system may not match the phosphorus removal performance of volcanic rock and ceramic pellets, its stability and wear resistance suggest that it may provide a more stable phosphorus removal effect in the long term.

The five substrates each have distinct characteristics in phosphorus removal within constructed wetland systems. Volcanic rock and ceramic pellets, due to their physical structural properties, facilitate the adsorption, precipitation, and biodegradation of nitrogen compounds. The mineral components in volcanic rock can form insoluble precipitates with phosphorus, thereby enhancing the phosphorus removal effect [28]. Zeolite’s ion exchange capacity allows it to effectively adsorb phosphates from water, with its porous structure providing additional surface area for phosphorus adsorption [11]. Although gravel has a lower porosity, its physical stability may provide stable phosphorus removal over the long term. The magic rack, despite its lower porosity, may provide a suitable growth environment for specific types of microorganisms due to its porosity and specific surface area, affecting its performance in phosphorus removal.

The community structure analysis shows that dominant bacterial groups such as Pseudomonas, Ardenticatenia_norark, and Aeromonas transform organic phosphorus into inorganic phosphorus through their metabolic activities, increasing the precipitability of phosphorus and thus improving phosphorus removal efficiency, playing a crucial role in the system’s denitrification process [23]. The high ecological niche overlap in the biofilm on zeolite indicates that Planctomyces plays a significant role in the decomposition of organic matter and adsorption of phosphorus [29]. The increased proportion of salt-tolerant denitrifying bacteria (Alishewanella) on the magic rack biofilm may fix atmospheric nitrogen, providing a usable nitrogen source for wetland plants and promoting plant growth [30]. This diverse microbial community structure is essential for pollutant removal as it increases the diversity of functional genes and metabolic pathways in the microbial community, enhancing the system’s adaptability and treatment efficiency for various pollutants [29,31].

This finding suggests that in practical applications, to promote the formation of microbial community structures and achieve the best comprehensive treatment effects, we need to select substrate characteristics that match the target pollutant removal effects and adjust the height and structure of the substrate layer according to the physical characteristics of the substrate and the needs of the microbial community, ensuring even water distribution and oxygen supply. The design of multi-layer structures can combine the advantages of different substrates, using substrates with high specific surface area to enhance biofilm formation and substrates with high porosity to support deep biological treatment. In addition, the configuration of the substrate should consider the diversity and functional redundancy of the microbial community to enhance the system’s adaptability and stability to environmental changes. By comprehensively considering these factors, the design of constructed wetland systems can more accurately meet specific wastewater treatment needs while improving the ecological and economic benefits of the system.

It is particularly noted that nitrogen and phosphorus, as essential nutrients for the growth of aquatic plants and algae, are crucial for maintaining the health and productivity of aquatic ecosystems. However, when the input of these nutrients exceeds the self-purification capacity of the water body, it leads to eutrophication issues. While constructed wetland treatment measures are significantly effective for the management of slightly polluted water bodies, the required wetland area and treatment costs will increase with the rising concentrations of nitrogen and phosphorus, which is influenced by various factors and difficult to achieve. Therefore, to control the eutrophication of rivers due to nitrogen and phosphorus, a comprehensive set of measures needs to be taken, including reducing point source pollution, managing non-point source pollution, ecological restoration, and water quality monitoring and assessment. The scientific basis for these measures includes assessments of environmental capacity, ecological threshold studies, and long-term water quality monitoring data, along with policy support to ensure the scientific and rational nature of nitrogen and phosphorus discharge limits. The findings of this study contribute to the protection and restoration of water body health and ecological balance.

4.3. Innovation

This study is the first to systematically compare the effects of five commonly used engineering substrates—zeolite, volcanic rock, gravel, magic rack, and ceramic pellets—on nitrogen (N) and phosphorus (P) removal in constructed wetland systems, and to delve into the impact of substrate characteristics on the structure and function of microbial communities, an area that has received less attention in previous research. Furthermore, this study explores the differences in water purification effects of various substrate–microbe systems, filling a gap in the existing literature that lacked a synchronous comprehensive comparative analysis. The detailed comparative results of substrate characteristics, microbial community structures, and treatment effects provided by this study have significant practical application value for the design and optimization of constructed wetland systems.

On the basis of this study, considering the economy and sustainability of the substrates is of great significance for the design and optimization of constructed wetland systems. In-depth research on the functional genes and metabolic pathways of microbial communities can help develop efficient microbial ecological engineering technologies, thereby further improving the quality of urban water bodies. These innovations in findings and methodology provide new perspectives and tools for the management and restoration of urban water environments.