Bioretention Columns Research Articles

Enhancing nitrate and phosphorus removal from stormwater in a fold-flow bioretention system with saturated zones.

The traditional bioretention systems possess a remarkably low nitrogen and phosphorus removal effect. The removal rate fluctuates greatly, and even appears as negative removal of nitrogen and phosphorus. Four simulated bioretention experimental columns with different bilayer media, packing composition and structure were constructed. Based on the traditional fillers, the modified composite fillers with hydroxy-aluminum and modified vermiculite sludge particle (HAVSP) were added. The traditional filler (C1) and the modified composite filler (C2) were added respectively, moreover saturated zones were set up to enhance the effect of nitrogen and phosphorus removal. Removal of nutrients from experimental columns by simulated runoff efficiency was evaluated and compared. In addition, the effect of media depth on phosphorus retention and denitrifying enzyme activity in bioretention columns was also evaluated. The experimental column #2 filled with C2 had the optimum removal effect on TP (93.70%), however, the removal effect of TP by filling C1 experimental columns was insufficient (57.36%). Designed to remove nitrate (NO3--N) and total nitrogen, the experimental column #4 showed the best performance (83.54% and 92.15%, respectively). In this study, we propose a fold-flow bioretention system by filling HAVSP in combination with saturated zones. The runoff water quality can be effectively improved, and a new bioretention cell configuration can be provided for efficient stormwater treatment.

Nutrients and solids removal in bioretention columns using recycled materials under intermittent and frequent flow operations

This research investigated the fate and removal of nitrite (NO2–N), nitrate (NO3–N), orthophosphate (PO4–P), and total suspended solids (TSS) in two bioretention columns, which were designed with three recycled materials. The first column was packed with Recycled Concrete Aggregate (RCA). The second column was a Layered Media (LM), which has layers of RCA with crushed glass and rice husks. The columns were tested under intermittent and frequent operations of synthetic runoff with low and high feed concentrations. The effect of inflow concentration, antecedent dry days (ADD), column age, and the anticipated number of events (EN) was also statistically analyzed on the performance of columns. Depending on column types, nutrient removal was significantly (p < 0.05) increased under frequent flow operations by 26–53% over intermittent. However, TSS removal was notably (p < 0.05) increased by 23–35% under intermittent operations over frequent. Overall, LM showed an increased NO2–N (92 ± 2%) and NO3–N (88% ± 2%) removal under low feed frequent operations and TSS removal (97% ± 2%) under initial intermittent operations. On the contrary, RCA showed a maximum of 99% PO4–P removal under high feed frequent operations. Results showed that the nutrient outflow concentration was found to have a negative correlation with EN and column age and a positive correlation with ADDs throughout the experiments.

Woodchips bioretention column for stormwater treatment: Nitrogen removal performance, carbon source and microbial community analysis

This study chose Oak woodchips and gravel as media filter to enhance the denitrification in the bioretention system (saturated zone 7.7 L) treating synthetic stormwater runoff. It revealed that the denitrification process mainly occurred during the drying phase and enlarging volume of saturated zones to retain more stormwater during storm event was the direct method to promote nitrogen removal of the bioretention system. Nevertheless, it was noted that the nitrogen and dissolved organic carbon would be released into the effluent during the wetting period. The denitrification rate with different nitrate nitrogen (NO3–N) concentrations did not show the obvious change with zero order kinetics constant of 2.91 mg/L∙d on average. Furthermore, it confirmed that woodchips were degraded and converted to volatile fatty acids (VFAs), especially acetic acid as carbon source, further utilized by the denitrifying bacteria, such as Dechloromonas, Acidoborax, Pseudomonas, Denitratisoma and Acinetobacter. In addition, genera of Lachnospiraceae and Lactobacillus, which had the ability to degrade the macromolecular organic components into low molecular VFAs, were observed in the woodchips bioretention system.

Influence of rainfall on the performance of bioretention systems modified with activated carbon and biochar

The use of bioretention areas is common in urban stormwater management, but their performance varies significantly depending on rainfall characteristics and design conditions. In this study, a pilot experiment using bioretention columns with different media (commercial activated carbon and river sediment-derived biochar) investigated the influence of rainfall on bioretention performance. The results indicated that the runoff volume retention ratio (Rv), which included the runoff purified and discharged at the bottom of the column, and the runoff retained in media during rainfall event, decreased significantly with increases in the rainfall event return period (p < 0.05). The Rv of the activated carbon and biochar columns decreased with a 2-yr return period and then fell further with a 50-yr return period. Porous material has been shown to improve the water-holding capacity of bioretention media, but it did not result in an improved Rv under heavy rain that exceeded the 2-yr return period. With the increase of the return period from two to 50 yr, the mass removal efficiency (RL) of total phosphorus and phosphate illustrated a clear decreasing trend in all columns. The total nitrogen, ammonia and nitrate removal did not show a clear trend with return periods because of transformations among different forms of nitrogen and similar saturation periods during the different rainfall events. The influence of the return period on chemical oxygen demand (COD) removal was related to whether the inflow COD reached maximum COD removal capacity of the bioretention media. Under a rainfall event with a specific return period, there were no significant differences in the RL of all nitrogen species and COD among the different columns (p > 0.05). The addition of adsorptive material, such as activated carbon and biochar, may not be the key factor for improving nitrogen and COD removal under heavy rain that exceeds the 2-yr return period. The bioretention performance of phosphorus removal from urban stormwater runoff could be improved by replacing or adding media with high adsorption capacity, but these improvements would not be significant under heavy rain that exceeds the 2-yr return period. The results provide some reference for evaluating bioretention performance and optimizing bioretention design in the future.

Nitrogen removal in stormwater bioretention facilities: Effects of drying, temperature and a submerged zone

Removal of ammonium‑nitrogen (NH4-N), nitrite/nitrate‑nitrogen (summarized as NOxN) and total nitrogen (TN) was examined in pilot-scale bioretention columns with and without a submerged zone under varied temperature and length of antecedent dry periods. The experiment was divided into wet and dry periods and samples were collected to represent continuously wet conditions and conditions with antecedent dry periods. Removal percentages and main effects of the examined factors as well as the effect of their two-way interactions were evaluated. Generally, low temperature favored NOx-N and thus TN removal independent of bioretention column design. Interestingly, bioretention columns with a submerged zone achieved the highest nitrogen removal at low temperature (1.5 °C). For standard bioretention columns, outflow concentrations of NOx-N and TN increased with increasing length of antecedent dry period, whereas a submerged zone mitigated this effect.

Organics removal and microbial interaction attributes of zeolite and ceramsite assisted bioretention system in copper-contaminated stormwater treatment

Bioretention has been increasingly used recently to treat heavy metals contaminated stormwater. However, less is known about how metal accumulation influences microbial performance and organics removal mechanisms in different layers of the bioretention system. Two lab-scale bioretention columns (i.e., control and Cu treatment) were designed and filled with soil and fillers (zeolite and ceramsite). The results obtained from the time-series experiment of 121 days showed that the removal of organics markedly affected by Cu accumulation and microbial activities, varied between soil and filler layers of bioretention system. The overall organics removal rate was higher in filler than soil. However, at the individual level, the chemical oxygen demand (COD) removal rate was higher than total organic carbon (TOC) in the soil, while the opposite trend was observed in fillers. Mixed media (soil + fillers) significantly reduced the bio-available and labile fractions of Cu from 33.5 to 8% and 67.5 to 33.4%, respectively. The bioretention column treated with Cu lost 14% more microbial biomass in soil than filler over the 121 days study period. Therefore mixed media in bioretention system can offset the substantial negative impacts of long-term metal accumulation on pollutant removal and microbial degradation function in the bioretention. The present study advanced our understanding to resolve the complex metals-impacted microbial pollutant biodegradation mechanisms and highlight importance of mixed media in the long-term maintenance of the bioretention system, which is imperative for developing effective and stormwater-specific remediation strategies.

Conventional bioretention column with Fe-hydrochar for stormwater treatment: Nitrogen removal, nitrogen behaviour and microbial community analysis

An FeCl3-modified rice husk hydrochar (‘Fe-hydrochar’) was used as the filler in a conventional bioretention column to remove nitrogen from synthetic stormwater. When the ammonia nitrogen (NH4-N) and nitrate nitrogen (NO3-N) concentrations of the influent were both 20 mg/L, the average removal rates of NH4-N and total nitrogen (TN) were approximately 97% and 50%, respectively. Nitrogen was mainly removed by microbial nitrification and denitrification, with 25% of NH4-N being adsorbed by the Fe-hydrochar. The remaining NH4-N was converted into NO3-N by nitrification in the upper layer, and NO3-N was mainly converted to nitrogen gas (N2) by denitrification in the lower layer. The organic matter released by the Fe-hydrochar was degraded and used as the carbon source for denitrification. The dominant bacteria were Pseudomonas, Rhizobium, and Flavobacterium at the genus level. Pseudomonas and Rhizobium were responsible for heterotrophic nitrification–aerobic denitrification, while Flavobacterium was related to the degradation of organic matter.

Hydraulic properties of stormwater biofilters during dry phase

ABSTRACT Stormwater biofilters operate in alternating wet-dry cycles and subjects processes to distinct environmental conditions, particularly, moisture content. Pollutants in residual water may undergo chemical or biological transformation during the dry phase and affect pollutant removal. The study aims to analyse filter water content during dry-cycle. Ten Perspex bioretention columns (94 mm internal diameter) were fed with tapwater at 100 mL/min for three hours and the filters were left to dry for 40 days. Filter material samples were obtained at different depths after different number of dry days (0–40 days) and were tested for moisture content. Top 5 cm had moisture content decreased to wilting point (6% degree of saturation) while deeper layers stabilised at higher degrees (40%). A mathematical model developed indicates significant amounts of moisture being held in the filter during the dry-phase, indicating potential impact on pollutant removal performance of the filter.

Study on bioretention for stormwater management in cold climate, Part I: Hydraulics

Simulated storm events were applied to four large bioretention columns to approximate 1.6 years of equivalent volume in Edmonton, Alberta’s typical climate. Summer, winter, and spring runoff were simulated in temperature-controlled laboratories with a range of −20 °C to +20 °C. During summer less porous bioretention media (i.e. loam soil) effectively weakened peak flows by >83% for 1:2 year events while more porous bioretention media (i.e. sandy loam soil) maintained hydraulic conductivities >9.1 cm/h. Winter operation consisted of all columns being subjected to −20 °C and then 1 °C repeatedly. Events were applied at an air temperature of 1 °C and, although frozen initially, more porous media experienced faster water breakthrough and ponding disappearance in winter indicating that hydraulic performance during intermittent warming periods in winter may be achievable. All columns’ hydraulic performance rebounded quickly in the subsequent summer. All columns successfully managed 1:2 year events in terms of infiltration rate, ponding depth and duration. Preliminary results also showed that both media have the potential to manage less frequent (1:5 and 1:10 year) events.

Effects of lead (Pb) in stormwater runoff on the microbial characteristics and organics removal in bioretention systems

Bioretention systems have been proved to be a natural approach for effectively reducing stormwater runoff pollution loads. However, the effects of heavy metals in stormwater runoff on microbial characteristics and organics removal in bioretention systems are unclear. In this study, two lab scale bioretention columns including the control and lead (Pb) treatment with the soil and filler layer were established. The changes of organic matter and lead in the effluent water and the soil (or fillers) were monitored during 121 operation days. The soil (or fillers) microbial characteristics were also analyzed. The results showed that most of Pb was intercepted by soil, while a small amount accumulated in fillers after 121 days. The long-term Pb accumulation in the biorentention system negatively affected the microbial biomass and microbial activity, while positively affected the community diversity. Pb accumulation killed some microorganisms, but simultaneously stimulated the growth of some Pb-tolerance microorganisms. The abundance of bacteria with COD degradation function in soil layer decreased, while that in fillers increased, indicating the effect of Pb on the community structure of these two layers was different. The COD removal in the soil and filler layer was promoted and inhibited by Pb contamination respectively. Moreover, Pb affected the removal of organic matter by chelating organic matters and changing their composition. The results suggested that the long-term accumulation of heavy metals in bioretention system would affect microbial degradation function and pollutants removal, causing our concern for the long-term maintenance of the bioretention system.

A bilayer media bioretention system for enhanced nitrogen removal from road runoff

Bioretention has been widely used in urban non-point source (NPS) pollution management for effectively reducing downstream pollution loads and peak flows. However, nitrogen (N) removal in conventional bioretention systems has been uniformly unstable and highly variable due to a lack of anaerobic denitrification. To improve the stability and effectiveness of N removal, two bioretention columns with bilayer media (C1 and C2) were designed. High permeability quartz sand (~2 mm diameter) was used as the upper media, and low permeability modified media (~0.6 mm diameter, adding 5% organic substance) as the lower media. The bilayer media structure formed an anaerobic zone for promoting denitrification processes. The results showed that the retrofitted columns performed well and that the removal efficiencies of various forms of N were considerably enhanced to 76.8%–95.3%, 85.1%–98.3%, and 87.5%–97.4% for TN, NH4+-N, and NO3−-N, respectively. Additionally, copying numbers of the denitrification functional genes detected via FQ-PCR in the lower media of C1 and C2 were accounted for 46.06% and 44.16% of the 16S rDNA gene, respectively. These results confirmed the presence of anaerobic denitrification processes. Consequently, bilayer media bioretention systems are worth promoting in cities where nitrogen in urban runoff poses a threat to the receiving surface water, due to the systems' remarkable performance in nitrogen removal, simple structure, and easy implementation.

Waste-derived compost and biochar amendments for stormwater treatment in bioretention column: Co-transport of metals and colloids

Bioretention systems, as one of the most practical management operations for low impact development of water recovery, utilize different soil amendments to remove contaminants from stormwater. For the sake of urban sustainability, the utilization of amendments derived from waste materials has a potential to reduce waste disposal at landfill while improving the quality of stormwater discharge. This study investigated the efficiency of food waste compost and wood waste biochar for metal removal from synthetic stormwater runoff under intermittent flow and co-presence of colloids. Throughout intermittent infiltration of 84 pore volumes of stormwater, columns amended with compost and biochar removed more than 50–70% of influent metals, whereas iron-oxide coated sand was much less effective. Only a small portion of metals adsorbed on the compost (< 0.74%) was reactivated during the drainage of urban pipelines that do not flow frequently, owing to abundant oxygen-containing functional groups in compost. In comparison, co-existing kaolinite enhanced metal removal by biochar owing to the abundance of active sites, whereas co-existing humic acid facilitated mobilization via metal-humate complexation. The results suggest that both waste-derived compost and biochar show promising potential for stormwater harvesting, while biochar is expected to be more recalcitrant and desirable in field-scale bioretention systems.

A pilot bioretention system with commercial activated carbon and river sediment-derived biochar for enhanced nutrient removal from stormwater.

Bioretention is an effective technology for urban stormwater management, but the nutrient removal in conventional bioretention systems is highly variable. Thus, a pilot bioretention column experiment was performed to evaluate the nutrient control of systems with commercial activated carbon and river sediment-derived biochar. Significant chemical oxygen demand (COD) and total phosphorus (TP) leaching were found with the addition of activated carbon and biochar, but total nitrogen (TN) leaching was significantly improved when activated carbon was used as the medium. During a semi-synthetic runoff experiment, the bioretention systems containing two types of fluvial biochar showed relatively better COD and TN control (average mass removal efficiencies and cumulative removal efficiencies) than commercial activated carbon. However, the average TP mass removal efficiency with commercial activated carbon (95% ± 3%) was significantly higher than biochar (48% ± 20% and 56 ± 14%). The addition of biochar in the media increased the nitrogen removal efficiency, and the addition of activated carbon significantly increased the phosphorous removal efficiency. Therefore, both biochar and activated carbon are effective materials for bioretention, and fluvial biochar provides an alternative approach to comprehensively utilize river sediment.

Bioretention cell incorporating Fe-biochar and saturated zones for enhanced stormwater runoff treatment

Nitrogen (N) and phosphorus (P) removal in conventional bioretention systems is highly variable. Therefore, five novel experimental columns with different media configurations and constituents, and incorporating a saturated zone were developed and assessed to optimize the removal of N, P and other nutrients. Three types of media composed of the conventional mixed sand and soil media (T1), biochar-amended media (T2), and iron-coated biochar (ICB)-amended media (T3) were evaluated. Two of the experimental columns were designed with double-layer configurations, while the other three were of a single-layer structure. Removal efficiencies of nutrients in the experimental columns were evaluated and compared using simulated runoff. Also, the effect of media depth on the retention of P and denitrifying enzyme activity (DEA) in the bioretention columns were evaluated. The experimental column only filled with T3 showed the best performance for COD, ammonia (NH4+-N) and total phosphorus (TP) removal (94.6%, 98.3% and 93.70%, respectively), whereas columns filled with T2 performed poorly for TP removal (57.36%). For the removal of nitrate (NO3−-N) and total nitrogen (TN), the columns using a single-layer and only filled with either T3 or T2 exhibited the best performance (93% and 97% TN removal, respectively). Overall, this study demonstrates that our proposed single-layered bioretention cell only filled with T3 and incorporating a saturated zone effectively improves the runoff quality, and can provide a new bioretention cell configuration for efficient stormwater treatment.

Investigating the hydrologic and water quality performance of trees in bioretention mesocosms

Cities across the world are increasingly utilizing green infrastructure practices as part of their stormwater management programs. Bioretention areas have become a popular green infrastructure practice due to their widespread success in improving water quality and reducing runoff generated from impervious surfaces. Several studies have demonstrated that pollutant removal performance can be improved when plants are included in bioretention design; however, while numerous benefits of trees in urban areas have been identified, little knowledge of their contributions to stormwater management in green infrastructure currently exists. To address this need, a controlled mesocosm experiment was conducted to characterize the degree of stormwater treatment provided by bioretention columns planted with one of three native tree species commonly found across the eastern United States (Acer rubrum – red maple, Pinus taeda – loblolly pine, and Quercus palustris – pin oak). Tree pollutant removal performance was compared to nonvegetated mesocosms using a semi-synthetic stormwater mixture applied to the mesocosms over a period of 17 weeks. The hydrologic benefits of each species were characterized using data-logging scales placed below the mesocosms to compare evapotranspiration (ET) rates and drainage in each configuration. Differences in pollutant removal between tree species were largely not significant, indicating the dominant role of the bioretention media in mitigating dissolved and particle-bound constituents. Mesocosms planted with red maple (Acer rubrum) had significantly greater average ET rates (3.2 mm d−1) than all other configurations, attributable to plant development and increased growth and canopy size. All mesocosms planted with trees had significantly higher ET rates than the nonvegetated mesocosms, illustrating the role of transpiration in bioretention hydrology which, depending on species, accounted for 8.2–37.5% of average daily water losses from the mesocosms during testing. These results suggest that trees contribute to bioretention hydrology through evapotranspiration, and that significant differences between species exist and are likely related to growth rate.

Development of an empirical formula for estimation of bioretention outflow rate

Urbanization of a watershed affects both surface water and groundwater resources. When impervious area increases, the excess runoff and volume of water collected at the downstream end of the watershed also increases, due to the decrease in groundwater recharge, depression storage, infiltration and evapotranspiration. Low-impact development (LID) methods have been developed in order to diminish adverse effects of excess stormwater runoff. Bioretention is one of the LID types which is used to prevent flooding by decreasing runoff volume and peak flow rate, and to manage storm-water by improving water quality. In this study, an empirical formula is derived to predict the peak outflow out of a bioretention column as a function of the ponding depth on bioretention, hydraulic conductivity, porosity, suction head, initial moisture content and height of the soil mixture used in the bioretention column. Coefficients of the empirical formula are determined by using metaheuristic algorithms. For analyses, the experimental data obtained from rainfall-watershed-bioretention (RWB) system are used. The reliability of the empirical formula is evaluated by calculating the absolute per cent error between the peak value ofthe measured outflow and the calculated outflow of the bioretention columns. The results show that the performance of the empirical formula is satisfactory.

Nutrient removal from urban stormwater runoff by an up-flow and mixed-flow bioretention system.

Bioretention is one of the most popular technical practices for urban runoff pollution control. However, the efficiency of nutrient removal from urban stormwater runoff by bioretention systems varies significantly. To improve the nutrient removal performance, innovative up-flow and mixed-flow bioretention systems were proposed in this study, and a laboratory study was conducted to investigate the runoff retention and nutrient removal performance. During the leaching experiment using tap water as the inflow, turbidity, chemical oxygen demand (COD), total nitrogen (TN), and total phosphorus (TP) leaching phenomenon was obvious. COD and TN leaching controls were obviously improved when the up-flow and mixed-flow bioretention systems were adopted comparing with the conventional bioretention. During the semi-synthetic runoff experiments, after the leaching experiments' performance (accumulated 2.78 times of empty bed volume), there were no significant differences in COD mass removal efficiencies of conventional and up-flow bioretention processes (p > 0.05); however, the COD mass removal efficiencies of the mixed-flow bioretention processes increased by 10% when compared with conventional bioretention. The TN mass removal efficiencies of the up-flow and mixed-flow bioretention increased obviously from 17% ± 13% (conventional) to 41% ± 23% (up-flow) and 31% ± 16% (mixed-flow). However, there were no significant differences in TP mass removal or runoff reduction among the three bioretention columns (p > 0.05). Both up-flow and mixed-flow bioretention can effectively improve TN mass removal, and the mixed-flow bioretention did not show a better TN removal performance than the up-flow bioretention because these two bioretention had almost the same volume of the saturated zone. Overall, the results indicate the mixed-flow bioretention proposed in this study can effectively improve TN mass removal and slightly improve COD mass removal relative to conventional methods via increases in hydraulic retention time and in-flow paths, respectively.

Reduction of Escherichia coli, Enterococcus faecalis and Pseudomonas aeruginosa in stormwater bioretention: Effect of drying, temperature and submerged zone

The impact of drying and temperature on the reduction of Escherichia coli, Enterococcus faecalis and Pseudomonas aeruginosa in stormwater bioretention systems with and without a submerged zone was assessed using 16 pilot-scale bioretention columns under controlled laboratory conditions. The experimental design enabled analysis of possible interactions between the factors. First outflow and event-based samples were collected. Outflow concentrations were independent of inflow concentrations and hence controlled by internal processes. Overall TSS removal was high but sensitive to bacterial synthesis. Event-based samples had significantly higher bacteria concentrations than first outflow samples, suggesting that remaining/surviving bacteria in the bioretention cells have little effect on initial peak outflow concentrations. The effect of temperature varied between bacterial species and sample types. Long dry periods seemed beneficial for bacteria reduction, but outflow bacteria concentrations peaked during the second watering after long dry periods. Submerged zones significantly reduced bacteria outflow concentrations. However, sudden temperature increases caused bioretention cells with a submerged zone to produce significantly higher bacteria outflow concentrations than before the temperature increase, which was not the case for standard cells. Bioretention cells with submerged zones may thus be poor choices for reducing bacterial concentrations in stormwater runoff in areas experiencing winter conditions. Finally, our results suggest that adsorption (e.g. further enhanced by biofilm formation) is the major mechanism governing bacteria reduction in bioretention systems.

Real time control schemes for improving water quality from bioretention cells

Abstract Extreme weather and the proliferation of impervious areas in urban watersheds increases the frequency of flood events and deepens water quality concerns. Bioretention is a type of green infrastructure practice developed to mitigate these impacts by reducing peak flows, runoff volume, and nutrient loads in stormwater. However, studies have shown inconsistency in the ability of bioretention to manage some pollutants, particularly some forms of nitrogen. Innovative sensor and control technologies are being tested to actively manage urban stormwater, primarily in open water stormwater systems such as wet ponds. Through these cyber-physical controls, it may be possible to optimize storage time and/or soil moisture dynamics within bioretention cells to create more favorable conditions for water quality improvements. A column study testing the influence of active control on bioretention system performance was conducted over a 9-week period. Active control columns were regulated based on either maintaining a specific water level or soil moisture content and were compared to free draining (FD) and internal water storage standards. Actively controlled bioretention columns performed similarly, with the soil moisture-based control showing the best performance with over 86% removal of metals and total suspended solids (TSS) while also exhibiting the highest ammonium removal (43%) and second highest nitrate removal (74%). While all column types showed mostly similar TSS and metal removal trends (median 94 and 98%, respectively), traditionally FD and internal water storage configurations promoted aerobic and anaerobic processes, respectively, which suggests that actively controlled systems have greater potential for targeting both processes. The results suggest that active controls can improve upon standard bioretention designs, but further optimization is required to balance the water quality benefits gained by retention time against storage needs for impending storms.

Purification effects of amended bioretention columns on phosphorus in urban rainfall runoff.

In order to develop bioretention fillers with better phosphorus removal capacity, we built 12 bioretention columns with six kinds of modified fillers, and analyzed the operation effects of the columns under different conditions through field tests. Results show that adding water treatment residual has optimal removal rates for total phosphorus (TP) (median = 96.80%) and soluble reactive phosphorus (SRP) (median = 97.13%). The water reduction rates of the columns with improved fillers are 1.23-2.04 times that of the bioretention soil media column. The coconut chaff column has the best water storage capacity (median = 40.54%). Among the external factors affecting column operation, influent concentration of pollutants in urban surface runoff is the biggest influence factor on the removal efficiency of TP. However, there are no significant correlations between the removal efficiency of SRP and rainfall, influent concentration, and discharge ratio. The columns modified with medical stone, vermiculite, peat soil, medical stone + peat soil, green zeolite + peat soil all have good removal for phosphorus pollutant. After entering the columns, the contents of TP and SRP in most columns increased. The recommended fillers and the accumulation performance of phosphorus can help to improve purification effects in bioretention systems.