Environ Eng Res > Volume 28(3); 2023 > Article
Osman, Takaijudin, Massoudieh, and Goh: Effects of Vegetation and Saturated Zone in Cascaded Bioretention on Enhancing Nutrient Removal

### Abstract

Bioretention is a water management practice that is increasingly being applied for runoff quality control. Although previous Bioretention studies have used some techniques to improve nutrient removal, some nutrients still leach out. Therefore, this study used Cascaded Bioretention (CB) by connecting three Bioretention columns in series. The planted Bioretention Column was retrofitted by adding a subsurface drainage module (SDM) below the gravel layer to create a dual saturated zone. This study aimed to investigate the effect of the number of treatments, vegetation, and modified saturated zone on enhancing nutrient removal from agricultural runoff and to understand nutrient removal mechanisms. The removal efficiencies of NH3-N, NO3-N, NO2-N, and TN improved to 89.8%, 49.7%, 49.2%, and 53.4%, respectively. The only negative removal was ON, which significantly decreased by incorporating vegetation and a modified saturated zone. Increasing the number of treatments significantly enhanced TN and ON while maintaining stable removal for other nitrogen compounds. However, phosphorus was less sensitive to increasing the number of treatments. Nitrogen removal could be enhanced by different removal processes such as nitrification, denitrification, mineralization, assimilation by plant uptake, and Anammox. However, phosphorus removal was less complicated, as adsorption and infiltration are likely to be the main removal mechanisms.

### 1. Introduction

In recent years, modern agricultural practices started to cause considerable agricultural pollution. It contributes to ecosystem degradation, land, and the environment due to the modern by-products of agriculture [1, 2]. Farmers use fertilizers and pesticides on their fields; these fertilizers provide the necessary elements for crop growth. However, when the plants do not fully use fertilizer, they can transport from agricultural fields. Hence, surface runoff from agricultural lands (agricultural runoff) transports fertilizers and pesticides to different water bodies, making them harmful to human health [3]. Fertilizers and pesticides are very rich in nutrients, especially phosphorus and nitrogen. These nutrients can be an environmental problem when found in excess range due to their harmful impact [4, 5]. Excessive nutrients can have dramatic effects on water quality and aquatic life. They cause algae growth and increase the level of nitrates and phosphates, which negatively affect the quality of water bodies [6, 7]. When these nutrients leach to rivers, lakes, streams, or oceans during rainy periods, they alter the cycle of marine and freshwater nutrients [8, 9]. The most common result of this agricultural pollution is eutrophication, which depletes the dissolved oxygen in the water and thus kills fish and other aquatic life [10, 11]. Agricultural pollution and the control of nutrients are significant subjects for global consideration. Therefore, considerable attention has been paid to water management in agricultural areas, especially water quality control. Green infrastructure is a water management strategy intended to preserve and restore the natural water cycle [12, 13]. It is reliable, sustainable, and increases the city’s protection and quality of life. Bioretention is one technique that is increasingly being applied for runoff quantity and quality control. It can be a good alternative for treating runoff because it maximizes water storage so that water can be treated [14, 15]. However, in conventional Bioretention systems, nitrogen removal is generally low and unstable, and phosphorus removal is inconsistent. The high infiltration rate and limited retention time for transformation processes lead to incomplete nutrient removal [16, 17]. Therefore, retention time in the Bioretention system needs to be adjusted accordingly to prevent nutrient leaching [17, 18]. Increasing retention time can significantly enhance nutrient transformation processes especially plant uptake [19]. For instance, the denitrification process needs a longer retention time to remove nitrate [2022]. The nitrification process also needs a longer retention time to be accomplished [23]. Retention time can be increased by incorporating the saturated zone [17] or plant [24]. Glaister et al. [25] highlighted the benefit of using vegetation with saturated zone to maintain nutrient treatment. Incorporating both plant species and the saturated zone was key for promoting complementary nitrogen removal mechanisms. Plants can enhance the removal through plant uptake, increasing microbial activity, and improving soil structure [26]. The saturated zone can increase retention time; thus, biological uptake and chemical reaction improve [25]. Davis et al. [27] reported that vegetation and saturated zone significantly improved phosphorus removal. Another study was conducted by Zhang et al. [28] along the same line, but it has shown different results; incorporating the plant and the saturated zone enhanced nitrogen removal. However, phosphorus removal substantially improved with the inclusion of a saturated zone regardless of the presence or absence of the plant. On the other hand, Davis et al. [27] observed poor nitrogen removal with the leaching from a single Bioretention box with a saturated zone planted with different plant spices and suggested that the removal of nitrogen is often insufficient due to a lack of denitrification. Zinger et al. [29] and Dietz et al. [30] also observed phosphorus leaching from a single Bioretention column with vegetation and a saturated zone. Although saturated zone and plants have been used to improve nutrient removal in Bioretention, some nutrients can still be leached from these systems. In these studies, single Bioretention, one layer of the saturated zone, and different plant species were used. Some evidence suggests that with single Bioretention and one layer of the saturated zone, nitrogen removal was low, and phosphorus removal was unstable. In addition, much inconsistency still exists about the effect of vegetation and saturated zone on enhancing nutrient removal in the Bioretention system. Furthermore, nitrogen and phosphorus removal mechanisms within these systems remain unclear. As such, there is a need to modify the Bioretention system in terms of the saturated zone and assess the effect of the plant and modified saturated zone on promoting nutrient removal. Moreover, the nutrient removal mechanisms in this modified Bioretention system must be highlighted. Therefore, this study uses a new technique called Cascaded Bioretention (CB); in this CB, three Bioretention columns planted with Cordyline Fruticosa (Ti) plant were connected in series and retrofitted by adding a subsurface drainage module (SDM) as an additional layer to create a dual saturated zone. The objectives of this paper are to (1) assess the use of Cascaded Bioretention by evaluating the effect of the number of treatments on nutrient removal; (2) investigate the effect of vegetation and the number of saturated zone layers on enhancing nutrient removal of agricultural runoff. (3) understand the impact of this modification on nutrient removal mechanisms in the system.

### 2.1. Characterization of Agricultural Runoff

Agricultural runoff water samples were collected from nearby paddy and oil palm plantation fields in the Perak Tengah area. Three (3) samples of agricultural runoff water were collected from various sites in Felcra, Seberang Perak, Kampung Gajah, Perak’s paddy fields. Similarly, three (3) samples of agricultural runoff water were taken from different oil palm plantation farm sites. The samples were collected in 500 mL bottles, transferred to the laboratory, and kept at 4°C in a cold storage room before the analysis.

The Bioretention column was designed according to Malaysia’s Urban Stormwater Management Manual (MSMA) [31]. The columns used in this study were made of high-density polyethylene and shaped truncated cone cylinders with the following dimensions: 315 mm upper diameter, 265 mm lower diameter, and 595 mm depth. The CB was set up by connecting three (3) Bioretention columns in series, as shown in Fig. 1.
Three types of CB were used in this study; the first set (CBC) was CB without plant and with a single saturated zone which consisted of one layer of gravel, which was used as a control set. The second set (CBS) was CB with the plant (Cordyline Fruticosa) and a single saturated zone consisting of one layer of gravel. The third set (CBD) was CB with the plant (Cordyline Fruticosa) and a dual saturated zone consisting of two layers of the saturated zone (gravel and SDM), as shown in Table S1. SDM is a water infiltration storage tank for stormwater control and management, and it is also known as the Modular Rainwater. The main purpose of the modular tank is to allow natural water infiltration on-site, which reduces surface runoff [32]. However, in this study, the SDM was used as an additional layer of the saturated zone creating an anaerobic condition to enhance nutrient removal by increasing retention time and promoting more biological removal processes. Four filter layers were set up from top to bottom, including ponding, engineered soil layer, transition layer (sand), and drainage layer, as shown in Fig. 2. The engineered soil mixture used in this study was composed of 60% medium sand, 30% topsoil, and 10% compost. The saturated zone could be provided by controlling the Bioretention outlet with a simple adjustment using a control valve as recommended by Lucas and Greenway. [19, 33].

### 2.3. Synthetic Agricultural Runoff Preparation

Several researchers have implemented this strategy to maintain targeted concentration levels during the experimental stage [3436]. Therefore, synthetic agricultural runoff water was used instead of the real agricultural runoff due to restrictions associated with the use of natural agricultural runoff. Synthetic agricultural runoff targeted concentrations were based on the highest concentrations obtained from the collected agricultural runoff, as shown in Fig. S1. The characteristics of synthetic agricultural runoff are shown in Table S1.

### 2.4. Design of Storm Event

The storm event was designed according to MSMA. [31], using 15 minutes storm duration and a 3-month Annual Recurrence Interval (ARI) because the system was designed as a minor system.
The catchment area (paddy field) was divided into sub-catchments by assuming the sub-catchment area of 400 m2 (0.04 ha). The rational formula (1) was used to estimate the peak flow as this formula gives satisfactory results for the small catchment.
##### (1)
$Q=CiA360$
Where Q = Peak Flow Discharge (m3/s) for 3-month ARI, C = Runoff Coefficient of the minor system, i = 0.4 by selecting the coefficient for grass cover, A = Rainfall Intensity (mm/ hr), A = Drainage Area (ha).
The flow rate that maintains 100 mm ponding (0.014 L/s) was achieved by scaling down Q using the appropriate scale.
The rainfall intensity () was calculated using the following formula:
##### (2)
$i=λTk(d+θ)η$
Where, i = Rainfall Intensity (mm/ hr), T = Average Recurrence Interval (year) = 0.25, d = Storm Duration (hr), = λ, κ, θ and η are coefficients.
Felcra is located at Perak ID 4010001 JPS Teluk Intan; thus, according to this location λ, κ, θ η and equal 54.017, 0.198,0.84, and 0.79 respectively.

### 2.5. Water Quality Analysis

Water quality sampling was performed using manual mode. For each storm event, water quality samples were collected from the inflow and outflows of the CB at 30-minutes intervals. A total of 10 samples were collected; one sample was collected from the influent, and three samples were collected from each column outflow. All water samples were collected in 100 ml bottles and stored at 4 °C prior to water quality analysis. Water quality parameters include total nitrogen (TN), ammonia (NH3-N), nitrate (NO3-N), nitrite (NO3-N), organic nitrogen (ON), total phosphorus (TP), and phosphate (PO4) were analyzed according to HACH. [37] using Hach DR3900 Spectrophotometer. There are typical pollutants characterizing runoff in Malaysia, according to MSMA. [31]. Therefore, these parameters were selected as representative indicators of nutrients in the agricultural runoff. The persulfate digestion method (10071) was used to determine TN concentration. PhosVer 3 with acid persulfate digestion method (8190) was used to determine TP concentration. A fresh potassium persulfate digestion was prepared for every test as recommended by APHA. [38]. Nessler method (8038) was used to determine NH3-N concentration. Cadmium Reduction (8171) was used to determine NO3-N concentration, and the diazotization method (8507) was used to determine NO2-N concentration. PhosVer 3 (8048) was used to determine PO4 concentration. Organic nitrogen (ON) was calculated by subtracting NO3-N, NO2-N, and NH3-N from TN [17]. The average removal efficiency of each nutrient was calculated using formula (3) [39].
##### (3)
$Average Removal Efficiency=∑i=1ncin-cout∑i=1ncin×100%$
Where Cin and Cout = inflow and outflow concentrations at i period, and n = the number of samples during the runoff event.

### 2.6. Statistical Analysis

A one-way analysis of variance (ANOVA) test with a significance threshold of 95 percent and an α-value of 0.05 [40] was performed using Minitab 19 to evaluate the statistical significance of two data sets.

### 3.1. The Effect of The Number of Treatments on Nitrogen Removal

Fig. 3 shows the effect of CB on nitrogen removal by evaluating the effect of the number of treatments (columns numbers) on nitrogen removal efficiency. It can be seen that in In Fig. 3 (a), (b), the removal efficiencies of NH3-N, NO3-N, NO2-N, ON, and TN gradually decrease in the second column and then drop in the third column.
It is apparent that the negative impact of the increasing number of treatments is more pronounced on the ON as a result of the release of ON from the particular column is extremely high, which causes an increase in ON leaching over time. There is no significant correlation (p > 0.05) between the number of treatments and the removal efficiency in (a) CBC and (b) CBS. In Fig. 3 (c), increasing the number of treatments enhances TN and ON removal while maintaining stable removal for other nitrogen compounds and decreasing the leaching. It was observed that the removal efficiencies in the last columns for NH3-N, NO3-N, NO2-N, and TN are 89.8%, 49.7%, 49.2%, and 53.4%, respectively. The only negative removal is ON, which significantly decreases in the last column. The low ON removal can be attributed to the organic matter (compost) used within the soil media. It can be assumed that there is a deficit in the mineralization process which eventually converts ON to ammonia. Thus, organic matters decompose and leach into the water, aligning with Hunt [41] and Huo et al. [42]. It may also be linked to the assimilation by plant uptake, which converts the inorganic forms of nitrogen (NH4 and NO3) into ON. This accords with recent observations by Dig et al. [43], which showed that 50% of total leached nitrogen was ON, and the assimilation of inorganic nitrogen forms was a main contributing factor. Despite the improvement in the removal efficiencies of NH3-N, NO3-N, and NO2-N in the last column compared to the first column in the CBD, the relationship between the removal efficiencies and the number of treatments is not statically significant (p > 0.05). However, there is a significant correlation between the removal efficiencies of ON and TN and the number of treatments (p < 0.05).

### 3.3. The Effect of the number of Treatments on Phosphorus Removal

As shown in Fig. 4 (a) and (c), it is evident that the removal efficiencies of PO4 and TP slightly improve from one column to another by increasing the number of treatments. However, Fig. 4 (b) shows some fluctuations that imply the system does not respond to increasing the number of treatments. The best removal efficiencies for PO4 and TP are 94.78% and 95.18% obtained by CBD. TP removal is similar to 95–96 % TP removal obtained from a laboratory-scale constructed wetland study by Žibienė et al. [44]. Although most removal efficiencies of PO4 and TP improve from one column to another, there is a slight difference between these removal efficiencies, which is not statistically significant (p > 0.5). The majority of removal efficiencies are above 90%, even with a single treatment, indicating that phosphorus may not be sensitive to increasing the number of treatments.
In general, the evidence shows that using the CB leads to a minor improvement in phosphorus removal in some cases; nevertheless, phosphorus appears to be less sensitive to increasing the number of treatments. However, applying the CB shows a negative effect on nitrogen removal efficiency in CBC and CBS and a positive effect on CBD. Regarding the negative impact of using CB on nitrogen removal, it is feasible to say that this negative effect can be attributed to the accumulation of nutrients from the individual column.

### 3.3. The Combined Effect of the number of Treatments and Vegetation on Nutrient Removal

To further understand the effect of CB on nutrient removal, this part will discuss the effect of the number of treatments combined with vegetation. The contour plot in Fig. 5 shows that the highest nitrogen removal occurs across a wide area within a single treatment, regardless of whether vegetation is used or not. This is due to the negative effect of increasing the number of treatments in the unplanted system. The combination of three treatments and vegetation seems to be the most effective for all nitrogen compounds except NO2-N and NO3-N. However, the combination of three treatments and the absence of vegetation results in the lowest nitrogen removal for all nitrogen compounds. This may be justified by the fact that the retention time may be extended as a consequence of increasing the number of treatments, as well as the plant may promote other removal processes. Vegetation can improve nitrogen removal through direct plant uptake or promoting microbial activity, which consequently enhances nitrification, denitrification, and mineralization processes [45, 46]. The single treatment seems to be the most effective for NO2-N and NO3-N removals; however, the use of vegetation does not appear to be relevant. It is most likely because nitrite and nitrate removals are less affected by plants as the plant prefers ammonium for assimilation by plant uptake than nitrate and nitrite [45, 47]. The Contour plot in Fig. 5 shows that the highest PO4 removal occurs in a small area within two treatments and close to the vegetation. In comparison, the highest TP removal occurs when the vegetation is used throughout all treatments, with a little bias toward two treatments. Thus, vegetation affects phosphorus removal; hence, plant uptake plays some role in phosphorus removal [48].

### 3.4. The Combined Effect of the number of Treatments and Number of Saturated Zone Layers on Nutrient Removal

To further understand the effect of CB on nutrient removal, this part will discuss the effect of the number of treatments combined with the number of saturated zone layers. The contour plot in Fig. 6 indicates that the highest nitrogen removal mostly occurs with the combination of three treatments and a dual saturated zone for all nitrogen compounds except NO2-N. However, the lowest nitrogen removal occurs with the combination of three treatments and a single saturated zone for all nitrogen compounds except ON. Remarkably, a combination of three treatments and a dual saturated zone results in better nitrogen removal. It is likely explained by increasing retention time due to increasing the number of treatments and promoting more biological removal. Moreover, a dual saturated zone creates anaerobic conditions, leading to a completely saturated layer. Consequently, it enhances bacterial activities as a consequence of expanding the saturated zone, which is essential for nitrate removal by denitrification. It is expected to be responsible for the significant improvement in ammonia removal. Increasing the number of treatments seems to enhance the nitrification process. The dual saturated zone is likely to provide a suitable anaerobic condition for the Anammox process as the Anammox process converts ammonia to N2 gas under anaerobic conditions [49]. The improvement in TN removal with the combination of three treatments and a dual saturated zone can be attributed to the successful removal processes such as nitrification and denitrification. The highest removal of NO2-N occurs with the combination of a single treatment and a dual saturated zone. Increasing nitrite concentration with the number of treatments can be assigned to the nitrification process, as nitrite is the first nitrification product (Paul and Clark. [50]). The lowest removal of ON obtains with a single treatment where the number of saturated zone layers does not appear relevant. A feasible explanation can be attributed to the release of organic matter from compost which is expected to exceed ON removal by mineralization. It can also be observed that there is low removal of TN with a single saturated zone, where the number of treatments does not seem to be a major contributor. As shown in Fig. 6, the highest PO4 removal occurs in a wide region within two treatments, while the saturated zone has no apparent effect. In contrast, the highest TP removal occurs in a narrower region within two treatments, slightly tilted to the dual saturated zone.
Essentially, the findings of these results demonstrate that increasing the number of treatments combined with using vegetation is effective for nitrogen removal. Similarly, increasing the number of treatments is effective for nitrogen removal only when combined with the dual saturated zone. In contrast, a combination of two treatments and vegetation is preferable for phosphorus removal. However, using two treatments is better for phosphorus removal for any number of saturated zone layers. Moreover, the results reveal that the influence of the number of treatments on nitrogen removal does not differ between the vegetation or the number of saturated zone layer factors. However, for phosphorus removal, the results show that the combined effect of the number of treatments and vegetation seems to be stronger than the combined effect of the number of treatments and the number of saturated zone layers.

### 3.5. The Role of Incorporating Vegetation and SDM in Nutrient Removal Mechanisms

Fig. 7 shows the comparison between CBC and CBD to assess the effect of incorporating vegetation and SDM on nutrient removal. The results show that CBD has a much higher removal of all nutrients than CBC. There is a significant difference between the two systems for all nutrients (p < 0.05) except PO4, as there is no significant difference between the two systems in PO4 (p > 0.05). Interestingly, despite limited aerobic conditions, this system could achieve efficient ammonia removal due to a smaller soil filter media depth (200 mm) than the previous system (300 mm). It suggests that only partial nitrification can be achieved under limited aerobic conditions. It is likely that incorporating SDM could provide a suitable anaerobic condition for the Anammox process. Also, we note that nitrate removal is somewhat improving by incorporating SDM, as it may lead to a completely saturated zone by creating anaerobic conditions during and after the storm event. The created anaerobic zone is essential for enhancing nitrate removal by the denitrification process. This finding broadly supports the previous studies by Hunt. [41], Huo et al. [42], and Ding et al. [43], which reported that denitrification is positively correlated with anaerobic conditions. It was not surprising that ON removal only slightly improves by incorporating the SDM; meanwhile, it highly improves by incorporating the SDM and plant. The most obvious finding to emerge from this analysis is that despite incorporating the SDM and plant providing a significant improvement in ON removal, there is still significant leaching of ON. It seems that ON leaching is higher than the mineralization process, as evidenced by the lower ammonia concentration compared to ON concentration. This is similar to the finding by Palmer et al. [51], which reported ON leaching from Bioretention even with the use of plant and saturated zone. The observed ON leaching could be attributed to the fact that ON could not be used directly by the plant unless converted to inorganic form (ammonia or nitrate) [45, 50]. The plant essentially absorbs inorganic nitrogen, either nitrate or ammonium [50]. Moreover, it suggests that the release of ON exceeds the conversion by the mineralization process. This is according to the assumption that various complex physical, chemical, and biological mechanisms are responsible for the removal of ON; however, mineralization seems to be a dominant removal process for ON [17, 27]. These findings are consistent with Huo et al. [42], who demonstrated that ON is very difficult to be removed by a normal infiltration process. Interestingly, this system could provide appropriate conditions for some nitrogen removal processes. Hence, TN removal could be enhanced by different removal processes such as nitrification, denitrification, mineralization, assimilation by plant uptake, and Anammox. This outcome is comparable with Qiu et al. [52] and Michael et al. [53], as their results demonstrated that the inclusion of the saturated zone significantly enhanced TN removal. However, it is contrary to that of Hunt et al. [15], who revealed that the effect of the saturated zone was not apparent in enhancing TN removal. It also completely opposed the view of Goh et al. [54], who stated that incorporating the saturated zone in the Bioretention system under tropical conditions is ineffective for nitrogen removal. On the other hand, the results show a slight improvement in TP removal. This observation supports the hypothesis that the improvement in phosphorus removal under anaerobic conditions may be due to increased microbial uptake, as the microbial uptake is likely to increase in the anaerobic zone [55, 56]. This observation differed from studies by other researchers, as Fan et al. [57] showed that phosphorus is negatively affected by anaerobic conditions. Also, Dietz et al. [30] and Søberg et al. [58] revealed that anaerobic conditions are not important for phosphorus removal processes. There is no significant difference in PO4 removal between the two systems, and both systems could successfully remove PO4. It is well known that PO4 is the most prevalent type of phosphorus in the water [59]. Hence, TP is mostly in PO4 form, consisting of 86% of PO4, as reported by Kiiza. [60]. Nevertheless, PO4 removal mechanisms seem to be similar in some respects to those of TP as it occurs through the same processes such as adsorption, infiltration, and plant and microbial uptake; however, in this study, it seems that PO4 removal is less affected by the microbial process. In addition, adsorption and infiltration seem to be the main removal mechanisms of PO4 and TP. It is worth noting that there may be some sources of errors during the experimental investigation as the temperatures were ranged between 28 to 32°C ±1.9 °C, which may affect the nutrient removal as the climatic conditions such as temperature and relative humidity can also affect the performance of the system [61, 62]. In addition, plant uptake was not measured, and the performance evaluation of vegetation in this study was based on comparing the planted and unplanted systems, which may also be another contributing factor to the uncertainty.

### 4. Conclusion

The ability of the system to effectively remove nitrogen requires successful removal processes such as mineralization, nitrification, denitrification, assimilation by plant uptake, as well as some secondary removal processes such as Anammox. Similarly, phosphorus can be removed through some processes such as adsorption, infiltration, and plant and microbial uptake. Incorporating vegetation and modified saturated zone using a SDM could enhance nutrient removal by maintaining the appropriate nitrogen and phosphorus removal conditions. Vegetation could improve nutrient removal through direct plant uptake and increasing microbial activity, which consequently enhances nitrification, denitrification, and mineralization processes. The dual saturated zone is likely to provide an extending anaerobic condition; accordingly, some biological removal processes can be facilitated. It was possible to enhance nitrogen removal by combining aerobic and anaerobic conditions in one system. Successful phosphorus removal can achieve in both aerobic and anaerobic conditions. On the other hand, increasing the number of treatments in the modified Bioretention (CBD) enhances TN removal and decreases ON leaching while maintaining stable removal for other nitrogen compounds. However, phosphorus appears to be less sensitive to increasing the number of treatments. It suggests that ammonia removal in this system could be achieved through partial nitrification, plant uptake, and Anammox. Nitrate removal could be achieved by denitrification and plant uptake. However, ON seems to be very difficult to remove, and various complex chemical and biological mechanisms seem to be responsible for the removal of ON. TN removal could be enhanced by different removal processes such as nitrification, mineralization, denitrification, assimilation by plant uptake, and Anammox. On the other hand, phosphorus removal seems less complicated, as adsorption and infiltration are likely to be the primary removal mechanisms. Other removal processes such as plant and microbial uptake also contribute to phosphorus removal but less than the major removal mechanisms. Hence, using Cascaded Bioretention and incorporating vegetation and SDM can be an effective technique for enhancing nutrient removal from agricultural runoff.

### Acknowledgments

This work was funded by the Fundamental Research Grant Scheme (FRGS), FRGS/1/2019/TK01/UTP/03/2, entitled “The Mechanisms of Nutrients and Pesticides Removal from Agricultural Runoff in Treatment Train”, Ministry of Higher Education Malaysia.

### Notes

Conflict of Interest

The authors declare they have no conflict of interest.

Author Contributions

M.O (Ph.D. student) conducted the experiments, collected data, investigation process, and formal analysis, and wrote the original draft. H.T (Senior Lecturer) administrated the project, made funding acquisitions and resources, supervised the student, and reviewed and edited the manuscript. A. M (professor) supervised, reviewed, and edited the manuscript. H.W.G (Senior Lecturer) supervised the student.

### References

1. Whittinghill LJ, Hsueh D, Culligan P, et al. Stormwater performance of a full scale rooftop farm: Runoff water quality. Ecol. Eng. 2016;91:195–206.

2. Hereher M, El-Kenawy A. Assessment of Land Degradation in Northern Oman Using Geospatial Techniques. Earth Syst. Environ. 2021;6:469–482.

3. Andersen Jesper H, Laamanen Maria, et al. Eutrophication in the Baltic Sea area –an integrated thematic assessment of the effects of nutrient enrichment and eutrophication in the Baltic Sea region. Science and Integrated Coastal Management. 2009;1–148.

4. Barry B, Namara R, Bank W, Bahri A. Better Rural Livelihoods through Improved Irrigation Management: Office du Niger (Mali). Integr. Water Resour. Manag. Pract. 2009;6:71–88.

5. Mander Ü, Kuusemets V, Lõhmus K, Mauring T. Efficiency and dimensioning of riparian buffer zones in agricultural catchments. Ecol. Eng. 1997;8(4)299–324.

6. Yan T, Wang J, Huang J. Urbanization, agricultural water use, and regional and national crop production in China. Ecol. Model. 2015;318:226–35.

7. Buaisha M, Balku S, Özalp-Yaman Ş. Heavy metal removal investigation in conventional activated sludge systems. Civ. Eng. J. 2020;6(3)470–7.

8. You H. Impact of urbanization on pollution-related agricultural input intensity in Hubei, China. Ecol. Indic. 2016;62:249–58.

9. Sharma Y. Water Pollution Control - A Guide to the Use of Water Quality Management Principles: Case Study I - The Ganga, India. World Health Organization E; 1997. p. 1–459.

10. Owa FD. Water pollution: Sources, effects, control and management. Mediterr. J. Soc. Sci. 2013;4(8)65–8.

11. Carpenter SR. Eutrophication of aquatic ecosystems: Bistability and soil phosphorus. In : Proceedings of the National Academy of Sciences of the United States of America 2005; USA. 10229p. 10002–5.

12. Dietz ME, Arnold CL. Can Green Infrastructure Provide Both Water Quality and Flood Reduction Benefits? J. Sustain. Water Built Environ. 2018;4(2)02518001

13. Jayasooriya VM, Ng AWM, Muthukumaran S, et al. Optimal Sizing of Green Infrastructure Treatment Trains for Stormwater Management. Water Resour. Manag. 2016;30(14)5407–20.

14. Davis AP, Shokouhian M, Sharma H, et al. Laboratory Study of Biological Retention for Urban Stormwater Management. Water Environ. Res. 2001;73(1)5–14.

15. Hunt WF, Jarrett AR, Smith JT, et al. Evaluating Bioretention Hydrology and Nutrient Removal at Three Field Sites in North Carolina. J. Irrig. Drain. Eng. 2006;132(6)600–8.

16. Sarazen JC, Faulkner JW, Hurley SE. Evaluation of nitrogen and phosphorus removal from a denitrifyingwoodchip bioreactor treatment system receiving silage bunker runoff. Appl. Sci. 2020;10(14)1–16.

17. Wang C, Wang F, Qin H, et al. Effect of saturated zone on nitrogen removal processes in stormwater bioretention systems. Water. 2018;10(2)1–13.

18. Fowdar HS, Hatt BE, Breen P, et al. Evaluation of sustainable electron donors for nitrate removal in different water media. Water Res. 2015;85:487–496.

19. Lucas WC, Greenway M. Hydraulic Response and Nitrogen Retention in Bioretention Mesocosms with Regulated Outlets: Part II-Nitrogen Retention. Water Environ. Res. 2011;83(8)703–13.

20. Hunt WF, Asce M, Davis AP, et al. Meeting Hydrologic and Water Quality Goals through Targeted Bioretention Design. J. Environ. Eng. 2012;138(6)698–707.

21. Osman M, Yusof KW, Takaijudin H, et al. A review of nitrogen removal for urban stormwater runoff in bioretention system. Sustain. 2019;11(19)1–21.

22. Peterson IJ, Igielski S, Davis AP. Enhanced Denitrification in Bioretention Using Woodchips as an Organic Carbon Source. J. Sustain. Water Built Environ. 2015;9(2010)1–9.

23. Vymazal J. The role of natural and constructed wetlands in nutrient cycling and retention on the landscape. Springer International Publishing; Switzerland: 2015. p. 1–326.

24. Muerdter C, Özkök E, Li L, et al. Vegetation and Media Characteristics of an Effective Bioretention Cell. J. Sustain. Water Built Environ. 2016;2(1)04015008-8–11.

25. Glaister BJ, Fletcher TD, Cook PLM, et al. Co-optimisation of phosphorus and nitrogen removal in stormwater biofilters: The role of filter media, vegetation and saturated zone. Water Sci. Technol. 2014;69(9)1961–9.

26. Li L, Yang J, Davis AP, et al. Dissolved Inorganic Nitrogen Behavior and Fate in Bioretention Systems: Role of Vegetation and Saturated Zones. J. Environ. Eng. 2019;145(11)04019074-1–9.

27. Davis AP, Shokouhian M, Sharma H, et al. Water Quality Improvement through Bioretention Media : Nitrogen and Phosphorus Removal. Water Environ. Res. 2006;78(3)284–93.

28. Zhang Z, Rengel Z, Liaghati T, et al. Influence of plant species and submerged zone with carbon addition on nutrient removal in stormwater biofilter. Ecol. Eng. 2011;37:1833–41.

29. Zinger Y, Blecken GT, Fletcher TD, et al. Optimising nitrogen removal in existing stormwater biofilters: Benefits and tradeoffs of a retrofitted saturated zone. Ecol. Eng. 2013;51:75–82.

30. Dietz ME, Clausen JC. Saturation to Improve Pollutant Retention in a Rain Garden. Environ. Sci. Technol. 2006;40(4)1335–40.

31. Department of Irrigation and Drainage. Government of Malaysia Department of Irrigation and Drainage Urban Stormwater Management Manual for Malaysia MSMA. 2 nd Edition2012. p. 1–806.

32. Abdurrasheed AS, Yusof KW, Alqadami EHH, et al. Modelling of Flow Parameters through Subsurface Drainage Modules for Application in BIOECODS. Water. 2019;11:1–15.

33. Lucas WC, Greenway M. Hydraulic Response and Nitrogen Retention in Bioretention Mesocosms with Regulated Outlets: Part I-Hydraulic Response. Water Environ. Res. 2011;83(8)692–702.

34. Ergas SJ, Sengupta S, Siegel R, et al. Performance of Nitrogen-Removing Bioretention Systems for Control of Agricultural Runoff. J. Environ. Eng. 2010;136(10)1105–12.

35. You Z, Zhang L, Pan SY, et al. Performance evaluation of modified bioretention systems with alkaline solid wastes for enhanced nutrient removal from stormwater runoff. Water Res. 2019;161:61–73.

36. Li M-H, Swapp M, Kim MH, et al. Comparing Bioretention Designs With and Without an Internal Water Storage Layer for Treating Highway Runoff. Water Environ. Res. 2017;86(5)387–97.

37. HACH 1989. Water Analysis Guide Edition1. HACH Company; Loveland, CO, USA: 2013. p. 1–57.

38. Baird Rodger B, Eaton Andrew D. Standard Methods for the Examination of Water and Wastewater. 23rd EditionWashington, D.C: American Public Health Association; 2017. p. 1–11545.

39. Zahari NM, Chua KHM, Sidek L. Removal Efficiency Evaluation for Constructed Wetland in Tropical Climate NA. J Energy Environ. 2014;18–23.

40. Stahle L, Wold S. Analysis of variance. Chemom. Intell. Lab. Syst. 1989;6:259–272.

41. Hunt WF. Pollutant removal evaluation and hydraulic characterization for bioretention stormwater treatment devices [dissertation] USA:. The Pennsylvania State University; 2003.

42. Huo S, Xi B, Yu H, et al. Characteristics and transformations of dissolved organic nitrogen in municipal biological nitrogen removal wastewater treatment plants. Environ. Res. Lett. 2013;8:1–9.

43. Ding W, Qin H, Yu S, et al. The overall and phased nitrogen leaching from a field bioretention during rainfall runoff events. Ecol. Eng. 2022;179:1–9.

44. Žibienė G, Dapkienė M, Kazakevičienė J, et al. Phosphorus removal in a vertical flow constructed wetland using dolomite powder and chippings as filter media. J. Water Secur. 2015;1(1)46–52.

45. USA Department of Agriculture (Washington). Soil Conservation Service. Chapter 6 Role of Plants in Waste Management. Agricultural Waste Management Field Handbook; 1992. p. 1–24.

46. Leverenz HL, Haunschild K, Hopes G, et al. Anoxic treatment wetlands for denitrification. Ecol. Eng. 2010;36(11)1544–51.

47. Li J, Yang X, Wang Z, et al. Comparison of four aquatic plant treatment systems for nutrient removal from eutrophied water. Bioresour. Technol. 2015;179:1–7.

48. Muerdter CP, Wong CK, Lefevre GH. Emerging investigator series: The role of vegetation in bioretention for stormwater treatment in the built environment: Pollutant removal, hydrologic function, and ancillary benefits. Environ. Sci. Water Res. Technol. 2018;4(5)592–612.

49. Tomaszewski M, Cema G, Ziembińska-Buczyńska A. Influence of temperature and pH on the anammox process: A review and meta-analysis. Chemosphere. 2017;182:203–14.

50. Paul EA, Clark FE. Soil microbiology and biochemistry. 2nd edOxford Press Academic; 1996.

51. Palmer ET, Poor CJ, Hinman C, et al. Nitrate and Phosphate Removal through Enhanced Bioretention Media: Mesocosm Study. Water Environ. Res. 2013;85(9)823–32.

52. Qiu F, Zhao S, Zhao D, et al. Enhanced nutrient removal in bioretention systems modified with water treatment residuals and internal water storage zone. Environ. Sci. Water Res. Technol. 2019;5(5)993–1003.

53. Dietz ME. Modified Bioretention for Enhanced Nitrogen Removal from Agricultural Runoff. J. Environ. Eng. 2016;142(12)06016007-1–4.

54. Goh HW, Lem KS, Azizan NA, et al. A review of bioretention components and nutrient removal under different climates—future directions for tropics. Environ. Sci. Pollut. Res. 2019;26(15)14904–19.

55. Cording A. Evaluating stormwater pollutant removal mechanisms by bioretention in the context of climate change [dissertation]. USA: Univ. of Vermont; 2016.

56. Passeport E, Hunt WF. Asphalt Parking Lot Runoff Nutrient Characterization for Eight Sites in North Carolina, USA. J. Hydrol. Eng. 2009;14(4)352–61.

57. Fan G, Ning R, Huang K, et al. Hydrologic characteristics and nitrogen removal performance by different formulated soil medium of bioretention system. J. Clean. Prod. 2021;290:1–11.

58. Søberg LC, Al-Rubaei AM, Viklander M, et al. Phosphorus and TSS Removal by Stormwater Bioretention: Effects of Temperature, Salt, and a Submerged Zone and Their Interactions. Water, Air, Soil Pollut. 2020;231:1–12.

59. Reza M. Simultaneous Removal of Ammonia and Phosphorus from Wastewater in a Continuous Flow Vertical Bioreactor [dissertation]. Canada: Univ. of Waterloo; 2017.

60. Kiiza CJ. Design and modelling of pollutant removal in stormwater constructed wetlands By School of Engineering [dissertation]. United Kingdom: Cardiff University; 2017.

61. Abderrahmane B, Naima B, Tarek M, et al. Influence of highway traffic on contamination of roadside soil with heavy metals. Civ. Eng. J. 2021;7(8)1459–71.

62. Sampson AP, Weli VE, Nwagbara MO, et al. Sensations of Air Temperature Variability and Mitigation Strategies in Urban Environments. J. Human, Earth Future. 2021;2(2)100–13.

CB set up.
##### Fig. 2
The arrangement of Layers in different CB configurations.
##### Fig. 3
Effect of the number of treatments on nitrogen removal efficiency for (a) CBC, (b) CBS, and (c) CBD.
##### Fig. 4
Effect of the number of treatments on phosphorus removal efficiency for (a) CBC, (b) CBS, and (c) CBD.
##### Fig. 5
Contour plot of the combined effect of the number of treatments and vegetation on nutrient concentrations (mg/L).
##### Fig. 6
Contour plot of the combined effect of the number of treatments and number of saturated zone layers on nutrient concentrations (mg/L).
##### Fig. 7
Boxplot of nutrient concentrations for CBC vs. CBD
TOOLS
Full text via DOI
Supplement
Print
Share:
METRICS
 0 Crossref
 0 Scopus
 485 View