Performance Evaluation of Various Filter Media for Multi-Functional Purposes to Urban Constructed Wetlands

Abstract

The escalating demand for innovative solutions is driven by the challenges posed by water quality degradation and the pervasive impacts of climate change. As such, this study evaluated the performance of filter media to mitigate these challenges through multi-functional applications in urban constructed wetlands (UCWs). Column testing of organic filter media, including biochar (BC), woodchip (WC), anthracite (AT), and activated carbon (AC), as well as inorganic filter media, such as ceramic balls (CB), basalt (BS), and porous sand (PS), with synthetic stormwater runoff influent was conducted to assess their performance through water quality parameters and Scanning Electron Microscopy with Energy Dispersive X-ray (SEM-EDX) analysis for carbon storage potential. Among the media tested, AC exhibited high pollutant removal efficiencies amounting to 84%, 54%, 56%, and 44% for total suspended solids (TSS), chemical oxygen demand (COD), total nitrogen (TN), and total phosphorus (TP), respectively. For inorganic media, PS had the highest TSS and TN removal, whereas BS exhibited the highest COD and TP removal. Overall performance rating revealed that AC and BS, despite their efficient pollutant removal, are associated with higher costs, resulting in a lower ranking compared to AT and PS. SEM-EDX analysis identified PS and AC as standout media for potential carbon storage, attributed to their extensive surface areas and pore structures providing abundant adsorption sites. The results from this study highlighted the potential applications of various filter media in UCW designs with the aim of achieving carbon neutrality and sustainable urban development.

1. Introduction

The continuous emission and accumulation of greenhouse gases (GHGs) in the atmosphere, such as carbon dioxide (CO₂), methane, and nitrous oxide (N₂O), both from natural and anthropogenic activities, results in climate change [1]. Alongside other pollutants, GHGs can cause degradation of the quality of stormwater runoff during rainfall events [2]. CO₂ and N₂O both react with water to form carbonic acid and nitric acid, respectively. High acidity in rivers and lakes can cause fish kill and increased solubility of elements toxic to plants. In addition, nitrogen deposition from atmospheric emissions and runoff, along with phosphorus, can trigger toxic algal blooms that are harmful to marine life and cause oxygen deficiency [3]. Extreme weather conditions caused by climate change, such as tropical cyclones, floods, heat waves, and droughts, also significantly affect water quality. Typhoons, extreme rainfall, and floods increase suspended solids and other pollutants due to high flow velocities. Heat waves and droughts, on the other hand, cause high temperatures and reduce stream flows, which results in high pollutant concentrations and low oxygen levels [4]. Therefore, these concerns have driven international agreements and global goals such as reduction in greenhouse gas emissions, climate change mitigation and adaptation, availability and sustainable management of water and sanitation, and sustainability of terrestrial and aquatic ecosystems.

In response to the pressing global challenges of climate change, the demand for innovative solutions, particularly nature-based solutions (NBS), to enhance environmental sustainability and climate change resilience is increasing. NBS’s solution lies in the strategic integration of urban constructed wetlands (UCWs), given its potential contribution to achieving carbon neutrality, biodiversity conservation, and sustainable development. Carbon, which has the highest level of greenhouse gas emissions in the atmosphere, can be removed by sequestration. Carbon sequestration is the process of capturing, transporting, and storing atmospheric carbon dioxide, which can be performed biologically by storing carbon in soil, aquatic environments, and vegetation, as well as geologically through the storage of carbon in rocks underground. UCWs are composed mainly of shallow, densely vegetated, and man-made ponds rooted over filter media that provide a substrate for nutrients and plant support, which help filter the stormwater through physical and biological processes such as settling, sedimentation, adsorption, and biological degradation. UCWs promote carbon neutrality by acting as carbon sinks, sequestering atmospheric CO₂ in biomass [5]. Moreover, urban wetlands designed to provide multiple functions, including water quality improvement, have successfully established diverse herbaceous vegetation and supported a variety of bird, fish, and aquatic macroinvertebrate species [6]. Generally, by addressing water quality management, biodiversity, and climate resilience, UCWs have the potential to contribute to sustainable development in line with the sustainable development goals (SDGs), including SDG 6 (Clean Water and Sanitation), SDG 11 (Sustainable Cities and Communities), SDG 13 (Climate Action), and SDG 15 (Life on Land).

In serving the potential functions of UCWs, the innovative design should employ filter media that can significantly reduce pollutants for sustainable development and sequester carbon for climate change mitigation while serving as base material for organisms and vegetation for the conservation of biological diversity. As such, this study evaluated the performance of organic and inorganic filter media, using synthetic urban runoff as an influent, based on the removal efficiency and potential carbon storage potential. The results of this study can also contribute to the future integration, design considerations, and applications of filter media in UCWs. By understanding the potential functions of UCWs, particularly in addressing carbon neutrality and sustainable development, UCWs aim to contribute to fostering resilient and sustainable urban environments in a world challenged by climate change.

2. Materials and Methods

2.1. Physical Properties of Filter Media

The selection of filter media in the study was based on the physical properties and characteristics, as well as the availability of organic and inorganic filter material obtained from local suppliers in South Korea. The cost, availability, pollutant removal capacity, and carbon storage potential were determined to identify new filter media to be applied in UCWs, which is one of the novelties of this study. Biochar (BC), woodchip (WC), anthracite (AT), and activated carbon (AC) were selected as organic filter media that contain carbon–hydrogen covalent bonds originating from living organisms. BC is characterized by its high porosity, offering a substantial surface area for microorganism attachment and contaminant adsorption [7]. WC is well-known for its nutrient retention capabilities [8]. Meanwhile, AT is a high-grade, hard coal material with a porous structure providing an expansive surface area for microorganism attachment and facilitating the removal of fine particles, organic matter, and certain dissolved pollutants [9]. Similar to BC and AT, AC is also a porous material with a large internal surface area that has excellent adsorption capacity for organic compounds. On the other hand, ceramic balls 1 (CB1), ceramic balls 2 (CB2), basalt (BS), and porous sand (PS) were selected as inorganic filter media, which lacks carbon–hydrogen bonds and does not or seldom contain carbon atoms originating from non-living components. CB1 and CB2 provide large specific surface areas due to high porosity and demonstrate potential in removing organic matter and nitrogen by promoting biofilm formation by microorganisms. BS characteristics include durability, inertness, and capacity to provide a surface for biofilm formation suitable in UCWs application, while PS consists of porous granules with diverse pore sizes and has excellent moisturizing properties.

The basic and physical properties of filter media are important factors in designing constructed wetlands for efficient pollutant removal, as well as cost-effectiveness.

Table 1. Physical properties of the different types of filter media.

Filter Media	BC	WC	CB1	CB2	PS	BS	AT	AC
Image
Particle Size (mm)	3–5	30–50	20–60	10–20	2–5	5–10	2–5	1–2
Porosity (%)	44	60	62	53	40	56	44	52
Void Ratio	0.3056	0.2958	0.3827	0.3464	0.2857	0.359	0.3056	0.3421
Bulk Density (g/cm3)	0.2724	0.2202	0.8222	0.9721	1.0318	1.0264	1.0045	0.7844
Specific Gravity	0.2724	0.2202	0.8222	0.9721	1.0318	1.0264	1.0045	0.7844
Permeability (m/s)	0.0279	0.4127	0.128	0.0686	0.0397	0.0675	0.0298	0.0120

presents physical properties that influence the performance and suitability of organic and inorganic filter media. Physical properties, including porosity, bulk density, and permeability, were determined through standard laboratory tests [10,11]. While most of the media exhibited similar porosities (e.g., WC and CB1), the permeabilities can differ due to variations in the shape of pores and the arrangement of grains. The permeability is influenced not only by porosity but also by the intricate characteristics of pore structure. Furthermore, both porosity and permeability determine the hydraulic capacity and treatment performance of the filter system. These properties can be influenced by factors such as grain size distribution, bulk density, and particle size. In general, the selection of filter media affects the hydraulic characteristics and extent of clogging in wetland systems.

2.2. Laboratory Column Testing

Each transparent acrylic column has a diameter and height of 10 cm and 30 cm, respectively, and was filled to a height of 15 cm filter medium. The perforated outlet bottom, with a total area of 0.2 cm², has an outlet diameter smaller than the size of the filter media to prevent any possibility of the media passing through the column. The schematic layout of the experimental setup employed for column testing is illustrated in Figure 1. Filter media washing was performed to ensure effective filtration and removal of pollutants [12]. Following the method of Reddy et al. (2014), the washed media was dried in an oven at 105 °C for 24 h [13].

A synthetic stormwater runoff mixture comprising road sediments and tap water was prepared in a 200 L tank with a mixer, producing a total solid concentration of 150 mg/L. The mixer was continuously running throughout the duration of the experiment to maintain a homogenous mixture of the synthetic stormwater. The concentration was prepared to achieve the average total suspended solids (TSS) data from stormwater samples collected from the catchment areas of the low-impact development (LID) facilities at Kongju National University, spanning from 2011 to 2023. The sediments, obtained from roadsides and parking lots and passing through a No. 60 (250 μm) sieve, were used to create this mixture. The TSS concentration and particle size utilized in this study were aligned with the values reported in previous studies [14,15]. Synthetic stormwater runoff water was pumped to the acrylic columns at a flow rate of 11 cm³/s. The flow rate used for column tests was determined using the rational Equation (1):

Q = ciA,

(1)

where c is the estimated capture rate of 90% of stormwater runoff by the treatment facility; i is the average recorded rainfall intensities over a 25-year period in South Korea; and A is the catchment area. In addition to the long-term average rainfall in the area, the permeability of the selected filter media was also considered in the design flow rate with allowance for clogging to prevent overtopping.

Influent and effluent flow rates were recorded at five-minute intervals during the initial hour and, subsequently, at 15 min intervals for the remaining duration. Along with influent–effluent flow rate data collection, effluent water samples were also collected every 15 min for the first hour and every hour for the succeeding runtime. The run ends when the 200 L synthetic runoff has been filtrated into the media. Hence, the total run duration varied between five and eight hours per filter media, depending on its permeability and clogging characteristics. Water samples were collected and tested for various parameters, including pH, conductivity, TSS, chemical oxygen demand (COD), nitrite (NO₂), nitrate (NO₃), ammonium nitrogen (NH₄-N), total nitrogen (TN), phosphate phosphorus (PO₄-P), and total phosphorus (TP), following the standard methods for water and wastewater examination [16]. Analyses were conducted using method codes 2540-D, 4500-N, 4500-NO₂, 4500-NO₃, and 4500-P for TSS, TN, NO₂, NO₃, and TP, respectively. Additionally, NO₃ and COD analyses were performed using method numbers 2220 and 1000 from the HS-1000 plus user guide, respectively. Removal efficiencies were calculated by dividing the change in influent and effluent concentrations with the influent concentration multiplied by 100%.

Moreover, after the experimental runtime, a sample of the filter medium from the top part of the column was tested for SEM-EDX to determine the potential carbon storage by the filter medium. The carbon from the synthetic stormwater runoff comes from organic carbon derived primarily from plant matter, as well as inorganic carbon extracted from soil minerals obtained from road sediments. Road dust consisted primarily of soil-derived minerals, such as quartz, clay-forming minerals, and organic matter, and toxic pollutants from brake and tire wear, combustion emissions, and fly ash [17]. The elemental composition of the synthetic stormwater runoff in the study consisted mainly of C, O, P, N, Si, Fe, Al, Na, K, Mn, and other heavy metals.

3. Results and Discussion

3.1. Changes in Water Quality Parameters

The average ± standard deviation for the different water quality parameters obtained from the experiment is presented in

Table 2. Comparison of inflow and outflow water quality concentrations for different filter media.

Water Quality Parameters	Unit	Inflow	Outflow
			BC	WC	CB1	CB2	PS	BS	AT	AC
pH	-	7.65 ± 0.19	7.33 ± 0.05	7.38 ± 0.06	7.53 ± 0.09	8.41 ± 0.11	7.37 ± 0.04	7.48 ± 0.08	7.25 ± 0.87	8.67 ± 0.35
Conductivity	μs/cm	388 ± 604.4	162.91 ± 63.67	221.79 ± 218.65	140.38 ± 7.14	140.33 ± 2.97	121.99 ± 4.62	173.01 ± 124.13	129.88 ± 11.97	127.37 ± 17.38
TSS	mg/L	93 ± 15.65	14.39 ± 9.16	29.07 ± 18.88	36.73 ± 12.35	60.75 ± 33.43	5.92 ± 3.78	20.69 ± 17.24	20.92 ± 21.20	11.42 ± 7.87
COD	mg/L	7.06 ± 5.57	11.42 ± 8.93	3.29 ± 3.26	0.89 ± 0.70	4.32 ± 1.09	3.12 ± 1.97	2.01 ± 1.50	3.28 ± 3.66	3.27 ± 1.50
NO2	mg/L	0.007 ± 0.011	0.005 ± 0.003	0.006 ± 0.002	0.007 ± 0.003	0.002 ± 0.001	0.002 ± 0.001	0.049 ± 0.002	0.003 ± 0.002	0.003 ± 0.001
NO3	mg/L	1.90 ± 0.60	1.11 ± 0.64	1.63 ± 0.43	2.03 ± 0.52	1.94 ± 0.59	1.73 ± 0.33	1.62 ± 0.37	2.05 ± 0.41	0.61 ± 0.46
NH4-N	mg/L	0.05 ± 0.019	0.044 ± 0.011	0.021 ± 0.010	0.029 ± 0.017	0.058 ± 0.033	0.035 ± 0.018	0.022 ± 0.011	0.029 ± 0.013	0.025 ± 0.013
TN	mg/L	3.98 ± 1.33	3.30 ± 1.01	3.21 ± 0.73	2.91 ± 0.91	4.02 ± 0.14	2.97 ± 0.27	3.29 ± 0.39	2.66 ± 0.13	1.75 ± 0.90
PO4-P	mg/L	0.04 ± 0.024	0.180 ± 0.283	0.028 ± 0.026	0.029 ± 0.012	0.044 ± 0.029	0.067 ± 0.005	0.026 ± 0.026	0.026 ± 0.028	0.037 ± 0.046
TP	mg/L	0.076 ± 0.062	0.204 ± 0.285	0.039 ± 0.029	0.068 ± 0.029	0.069 ± 0.038	0.090 ± 0.029	0.034 ± 0.035	0.047 ± 0.034	0.042 ± 0.054

Note: The values in the table represent the mean ± standard deviation, n = 12.

. The influent water had a pH of 7.65 ± 0.19, which showed a slight decrease ranging from 0.12 to 0.4 after passing through different types of filter media except for CB2 and AC. CB2, in particular, may have a buffering effect on pH due to the presence of materials like clay or ceramics in composition, whereas AC is an excellent adsorbent and can remove acidic or basic components from the water. In general, these pH variations can be attributed to the different chemical compositions and buffering capacities of the filter media [18]. On the other hand, the reductions in conductivity levels (166–266 μs/cm) observed in the outflow from all types of filter media suggested efficient adsorption and removal of dissolved ions, which can have positive implications for water quality.

Synthetic stormwater runoff applied to the columns has TSS, COD, TN, and TP concentrations amounting to 77–110 mg/L, 3–22 mg/L, 3–7 mg/L, and 0.02–0.3 mg/L, respectively. All filter media resulted in lower TSS concentrations (32 to 87 mg/L) in the effluent. In particular, PS exhibited the highest TSS removal efficiency of 94%, which can be attributed to its diverse pore spaces, allowing for effective trapping and retention of suspended particles. Among different types of filter media, BC exhibited an increase in pollutant levels for COD, PO₄-P, and TP concentrations. This can be attributed to its porous nature, which, despite providing adsorption sites for pollutants, can also lead to competitive adsorption and the release of organic compounds into the water. Additionally, BC typically repels phosphate anions, resulting in limited adsorption capacities for inorganic phosphorus [19].

Among the filter media tested, PS, AC, and WC achieved the lowest effluent concentrations of 0.002 mg/L for NO₂, 1.31 mg/L for NO₃, and 0.029 mg/L for NH₄-N, respectively. AC, with its large surface area, appropriate pore diameter, and surface functional groups, had the highest reduction in TN, removing an average of 2 mg/L. This performance can be attributed to its effective adsorption capacity, particularly for heavy metals, including nitrogen compounds [20]. In general, each media exhibited varying trends in pollutant reductions. For specific types of pollutants and the desired water quality target, a specific filter media may be more suitable than the others.

Time series plots of effluent concentration over influent concentration (C_out/C_in) for TSS, COD, TN, and TP were shown in Figure 2. The experimental runtime for each filter media varied due to differences in the permeability and clogging characteristics of each material. All filter media, except for CB2, exhibited an immediate decrease in C_out/C_in values for TSS, achieving removal efficiencies of approximately 70% after an hour. Increasing removal efficiencies were observed throughout, obtaining high removal efficiencies of about 90% for AC, BC, and PS, which is partly attributed to high porosity and large surface areas. Other factors that may influence the removal efficiencies may include the reaction of attractive forces between the adsorbate (pollutant) and the adsorbent (filter media). Among the filter media, WC, CB1, PS, BS, and AC obtained high COD removal efficiencies of 70 to 80% after an hour, while BC, CB2, and AT reached 50% removal efficiency after three hours. In this study, COD, which is composed of biodegradable and refractory organics, is removed mainly by extraction of TSS. Bacterial oxidation of biodegradable organics is minimal, in this case, due to the short hydraulic retention time ranging from 43 to 66 s within the column. Hence, filter media that exhibited high TSS removal efficiency also showed a high COD removal rate. At the end of the run duration, all filter media exhibited a declining trend in the C_out/C_in ratio and high removal rate, with WC and CB1 achieving the highest removal efficiency of approximately 95%, while the rest had an average of around 80%. The range of final removal efficiencies is 60–95% and 79–95% for TSS and COD, respectively.

In TN removal, only AC reached 50% efficiency after an hour, with a slight downtrend in C_out/C_in values and a final efficiency of 65%. Only BS and BC joined AC with a final removal efficiency of >50%, while WC, CB1, PS, and AT exhibited almost constant removal efficiency of 40 to 50%. The low removal efficiency in TN is due to the required time to perform nitrification and denitrification, as well as the absence of plants to utilize the nutrient product. TN removal in this study may be attributed to the possible adsorption of ammonia nitrogen [21], NO₂, and NO₃ present in the influent, which are all components of TN in addition to organic nitrogen. Finally, for TP removal efficiency, WC, BS, AT, and AC achieved a high removal efficiency of around 70% after the first hour, whereas BC, CB1, and CB2 only reached 50% efficiency after three hours. At the end of the run, WC, BS, and AC achieved around 80% efficiency, while BC, CB1, CB2, and AT achieved 55% efficiency. Similar to nitrogen, the natural way of removing phosphorous in water is through biomass incorporation. However, due to the absence of plants and bacteria in this study, the removal of TP is attributed mainly to adsorption and filtration through filter media. The high porosity and large surface area of AC, BS, and WC, known for their high nutrient-retention capabilities, facilitated the adsorption and filtration of particulate phosphorous, thus exhibiting high removal rates [22].

Generally, most of the filter media exhibited a downtrend in C_out/C_in values or increasing removal efficiency for various water quality parameters. Filter media that attained high removal efficiency immediately at the beginning of the runtime showed a lower decline rate in C_out/C_in values as compared to filter media with low starting removal efficiency. Increasing removal efficiencies can be attributed to increasing hydraulic retention times due to clogging throughout the run duration. Although clogging may increase the removal efficiency in the early stages of flow due to trapped particles, it may also cause a reduction in the effluent rate that may result in reduced hydraulic efficiency, as observed in the conducted laboratory column tests. Effluent rates were reduced by 1.86 cm³/s or 17% of the original flow rate throughout the run duration, with WC and BC exhibiting the highest flow rate reduction due to blockage of the outlet drain. Moreover, the accumulation of particles may also eventually lead to declining removal efficiencies due to clogged pore spaces; thus, increased test run durations may result in proper investigation and assessment of clogging effects on filter media.

3.2. Carbon Storage Potential through SEM-EDX Analysis

The carbon storage potential of each filter medium is analyzed in this study using SEM-EDX and evaluated over other filter media. SEM-EDX microanalysis is a technique of elemental analysis that reveals the composition of a specimen. The results from SEM-EDX tests were used to identify the elemental composition of filter media materials before and after filtration, as summarized in Figure 3. The same results were also used to determine the carbon storage potential of each filter medium or the ability of the material to store carbon. An increase in carbon content after filtration implies stored carbon within the filter media. The chemical composition of the filter media also defines the type of material, whether organic or inorganic in nature, depending on the carbon and mineral composition. Specifically, high carbon and low mineral composition depicts organic material, whereas minimal to no carbon and high mineral composition depicts inorganic material. Moreover, chemical characteristics such as total metal content can affect the sorption capacity of the media for pollutants, i.e., phosphorus [23]. Thus, determining such characteristics is important to assess the filtration capabilities of each medium.

The elemental composition of organic media, including BC, WC, AT, and AC, revealed that the major elements found on the surface are C and O. After filtration, all organic media demonstrated an increase in carbon content with compositional differences by weight ranging from 0.60% to 1.78%. On the other hand, inorganic media exhibited distinct compositions with variations in the elemental content after filtration. CB1, CB2, and PS, consisting of a composition rich in O, Si, and A but low levels of C, showed an increase in carbon content after filtration, though their initial carbon content was relatively low. Basalt, a complex mixture of elements, predominantly including metallic components like Si, Ca, Fe, Al, Na, Mg, K, and Ti, demonstrated noticeable shifts after filtration, including an increase in carbon content. This suggested that basalt’s structural features, including fractures and pores, could provide substantial surface area for carbonate mineral formation, increasing its capacity for carbon sequestration [24]. These changes in elemental composition for both organic and inorganic filter media demonstrated its potential ability to capture, adsorb, or sequester certain elements or compounds from the water as it passed through the media.

Figure 4 displayed the surface morphologies or micrographs and surface elemental composition data of inorganic medium, PS and organic medium, and AC from SEM-EDX results. The two media obtained the highest carbon concentration potential with differences in surface compositions by weight of 1.77% and 1.57%, respectively. The elemental composition of other filter media materials before and after filtration is presented in Figure 3. Zhang et al. (2016) reported that the composition of PS, primarily Si and O, is conducive to carbon adsorption [25]. PS, though inorganic, can have a potential for carbon sequestration in UCWs by effectively trapping and storing CO₂ molecules within its porous structure [26]. Moreover, this medium acts as an efficient bioreactor, allowing for rapid remineralization of organic material by sediment microbial communities [27]. AC’s porous structure, created during the activation process, results in an exceptionally large surface area relative to its volume. The high carbon content, exceeding 80% by weight, along with its strong adsorption capabilities, enables nutrient retention, soil properties improvement, and an increase in soil organic matter content [28].

3.3. Performance Evaluation of Filter Media

Based on the removal efficiencies of filter media, the performances of each material for pollutant reduction, including TSS, COD, TN, and TP, as well as the carbon storage potential, were evaluated using a three-point rating scale. The rating scale categorized the performance of each media for each water quality parameter as high (H), moderate (M), and low (L), which corresponds to the range of class intervals of removal efficiency. Moreover, factors aside from water quality, such as cost and application frequency, were also considered in the performance evaluation of the filter media. Low cost, moderate cost, and high cost corresponded to high (H), moderate (M), and low (L) performance ratings of the filter media, which are based on the actual average cost of filter media from local suppliers. On the other hand, for the application frequency, high (H), moderate (M), and low (L) performance ratings corresponded to generally used, frequently used, and seldomly used filter media material. All values of pollutant reduction, carbon storage potential, cost-effectiveness, and application frequency were normalized to achieve all scales between zero and one. Normalization placed the data points within the range proportional to the minimum and maximum range; thus, rating scales were classified as L for 0–0.33, M for 0.33–0.66, and H for 0.66–1.00.

Table 3. Summary of the performance rating of filter media.

Factors	Filter Media
	BC	WC	CB1	CB2	PS	BS	AT	AC
Pollutant Reduction
Suspended Solids	H	M	M	L	H	H	H	H
COD	L	M	H	M	M	H	M	M
TN	L	L	M	L	M	L	M	M
TP	L	M	L	L	L	M	M	M
Carbon storage potential (from SEM-EDX Analysis)	M	L	M	M	H	M	M	H
Cost-effectiveness	M	H	M	M	M	L	H	L
Frequently Applied	M	M	L	L	H	L	M	M
Overall Rating	M (1.7)	M (1.9)	M (1.9)	L (1.4)	H (2.3)	M (1.9)	H (2.3)	M (2.1)

    H     = high (3 points);     M     = moderate (2 points);     L     = low (1 point).

summarizes the performance rating of different filter media based on pollutant reduction, potential carbon storage, cost-effectiveness, and application frequency. Assigning a three-point numerical rating scale of three points, two points, and one point for high (H), moderate (M), and low (L) performance, respectively, the overall performance of filter media, considering all the factors, was obtained. Among the organic media investigated, AT obtained the highest overall performance rating (2.3 points), followed closely by AC (2.1 points). On the other hand, for inorganic filter media, PS (2.3 points) exhibited the highest performance rating, followed by BS (1.9 points).

In assessing the effectiveness of filter media in UCWs, it is essential to contextualize the findings within the broader landscape of existing research. A comparative analysis with previous studies from different countries, each examining filter media performance in varied applications, was conducted.

Table 4. Filter media performance in comparison with the other reported literature.

Country	Application	Technology	Filter Media	Performance	Reference
South Korea	Synthetic wastewater	Constructed wetlands	Biochar	Average removal efficiency of 91%, 58%, and 80% for COD, TN, and TP, respectively.	[29]
China	Wastewater treatment	Constructed wetlands	Biochar	Improved removal of nitrogen (>20% on average), phosphorus, organic contaminants, heavy metals, and pathogens from wastewater. Enhanced macrophyte growth and mitigated greenhouse gas emissions.	[30]
South Korea	Stormwater treatment	Constructed wetlands	Woodchip	TSS Removal efficiency amounting to 89–100%. Improved nitrogen removal through leaching organic matter as carbon sources.	[8]
Sweden	Stormwater treatment	Constructed wetlands	Sand	Reported total Cu and Zn removal of 67% and 93%, respectively.	[31]
Australia	Stormwater treatment	Stormwater sand filters	Coarse and fine sand	Removal reduction for Zn of 80%. High nutrient removal and superior suspended solids removal. Reduction in fecal coliforms in the stormwater was 65% (fine filter media) and 79% (coarse media).	[32]
Ethiopia	Water treatment	Water treatment	Ceramic filter media	Effective in maintaining hydraulic flow and reducing clogging. Showed an average removal efficiency of 60% to 89% for turbidity, total coliform, E. coli, Ca, Mg, S, P, Fe, and N.	[36]
Australia	Stormwater treatment	Permeable pavement system	Basalt	Achieved 38% to 67% removal of heavy metals from synthetic stormwater.	[34]
Malaysia	Groundwater pollutants	Groundwater remediation	Anthracite	Capability of adsorbing organic pollutants due to their hydrophobic characteristics. Alternative to remove pollutants in groundwater.	[35]
Canada	Drinking water treatment	Drinking water treatment	Activated Carbon	Exhibited removal rate of over 90% for ammonia. High adsorption capabilities for volatile organic compounds.	[37]
India	Textile effluent treatment	Constructed wetlands	Activated Carbon	BOD and COD removal efficiencies were greater than 40%. Effective in adsorbing and reducing organic matter and pollutants.	[33]
South Korea	Synthetic stormwater runoff	Application to UCWs	Biochar	TSS and TN removal efficiency of 79% and 17%, respectively.	This study
			Woodchip	TSS, COD, TN, and TP removal efficiency of 58%, 53%, 19%, and 49%, respectively.
			Ceramic balls 1	TSS, COD, TN, and TP removal efficiency of 47%, 87%, 27%, and 10%, respectively.
			Ceramic balls 2	TSS, COD, and TP removal efficiency of 13%, 39%, and 8%, respectively.
			Porous sand	TSS, COD, and TN removal efficiency of 92%, 56%, and 26%, respectively.
			Basalt	TSS, COD, TN, and TP removal efficiency of 70%, 72%, 17%, and 55%, respectively.
			Anthracite	TSS, COD, TN, and TP removal efficiency of 70%, 54%, 33%, and 38%, respectively.
			Activated carbon	TSS, COD, TN, and TP removal efficiency of 84%, 54%, 56%, and 44%, respectively.

presents the comparison of the performance outcomes with other studies. Gupta et al. (2016) and Deng et al. (2021) investigated BC’s application in synthetic wastewater and wastewater treatment, respectively, emphasizing its effectiveness in nitrogen removal [29,30]. In alignment with these findings, this study achieved a comparable 17% TN removal. Sand, frequently applied for stormwater treatment, particularly in countries like Sweden and Australia, was reported to have high nutrient removal, superior suspended solids removal, and heavy metals removal such as Cu and Zn [31,32]. The results of this study for PS were consistent with these findings, achieving a TSS removal efficiency of 92% and a TN removal efficiency of 26%. On the other hand, Jayabalan et al. (2020) reported that AC achieved removal efficiencies greater than 40% for BOD and COD, highlighting its effectiveness in adsorbing and reducing organic matter and pollutants [33]. Meanwhile, BS exhibited 38% to 67% removal of heavy metals from synthetic stormwater [34], whereas AT demonstrated the capability to adsorb organic pollutants due to its hydrophobic characteristics [35]. Overall, this analysis identified the unique contributions, potential advantages, and applications of the filter media evaluated in this study. Such comparisons are invaluable for guiding future applications of filter media in UCWs, serving as an innovative solution to address water quality and sustainable challenges.

4. Conclusions

UCWs, as NBS, have emerged as one of the innovative solutions to the escalating challenges posed by water quality degradation and the pervasive impacts of climate change. This study identified and evaluated the strengths, potential advantages, and overall performance of selected organic and inorganic filter media in terms of pollutant removal efficiency, carbon storage potential, cost-effectiveness, and applicability for possible contribution and guidance to future applications of filter media in UCWs. From the water quality analysis, AC obtained the highest removal efficiency, with TSS, COD, TN, and TP having 84%, 54%, 56%, and 44%, respectively, for organic media. On the other hand, PS had the highest TSS and TN removal efficiency at 92% and 26%, whereas BS exhibited the highest COD and TP removal efficiency at 72% and 55% for inorganic media. From the SEM-EDX analysis, the results revealed high carbon storage potential for PS and AC, showing more surface carbon composition from EDX spectrum results after filtration. The high pollutant removal efficiency and carbon storage potential of PS and AC can be attributed to their extensive surface areas and pore structures that provide abundant adsorption sites, which enhances the interaction of attractive forces between the pollutant and filter media. In terms of cost-effectiveness and application frequency, PS has emerged as the most utilized filter media in UCWs due to its low cost, abundance, availability, and pollutant-removal capabilities. Though AC and BS manifested to be highly efficient in terms of pollutant removal, they are ranked behind PS and AT in the overall performance rating due to high cost. While this study has demonstrated the potential of certain filter media for carbon removal, the accuracy of the results is still to be validated using other tests since SEM-EDX only identified and characterized the elemental composition of the material before and after filtration. Additional assessments, such as adsorption isotherms and leaching tests, can provide a more comprehensive understanding of the filter media’s carbon removal capacity and the potential release of adsorbed carbon over time.