Hydraulic and pollutant removal performance of stormwater filters under variable wetting and drying regimes

Evaluation of nitrogen removal, functional gene abundance and microbial community structure in a stormwater detention basin

Stormwater control measures such as detention basins are used to mitigate the negative effects of urban stormwater resulting from watershed development. In this study, the performance of a detention basin in mitigating nitrogen pollution was examined and the abundance of N-cycling genes (amoA, nirK, nosZ, hzsB and Ntsp-amoA) present in the soil media of the basin was measured using quantitative PCR. Results showed a net export of nitrogen from the basin, however, differences between in- and outflow concentrations were not significant. Furthermore, the quantitative PCR showed that nirK (denitrification gene) was more abundant in the winter season, whereas amoA (nitrification gene) was more abundant in the summer season. The abundance of nirK, Ntsp-amoA and hzsB genes also varied with the sampling depth of soil and based on 16S rRNA gene sequencing of soil samples, Actinobacteria and Proteobacteria were the most dominant phyla. Species diversity appeared higher in summer, while the top and bottom layer of soil clustered separately based on the bacterial community structure. These results underline the importance of understanding nitrogen dynamics and microbial processes within stormwater control measures to enhance their design and performance.

Nitrogen process in stormwater bioretention: effect of the antecedent dry days on the relative abundance of nitrogen functional genes

In this study, we evaluated the relative abundance of nitrogen functional genes (amoA, nirK and nirS) involved in ammonia oxidation and denitrification bacteria in laboratory-scale bioretention columns in response to environmental factors (e.g., moisture content, pH, soil organic matter, soil nitrogen) under different antecedent dry days (ADDs). We observed a decrease tendency of the relative abundance of ammonia-oxidizing bacteria at first and then increased when increasing ADDs from 1 to 22 day, while the relative abundance of denitrifying bacteria showed a downward trend. The abundance of bacteria gene amoA was positively associated with soil ammonia nitrogen concentration (r² = 0.389, p < 0.05) and soil organic matter concentration (r² = 0.334, p < 0.05), while the abundance of bacteria gene nirS was positively correlated with soil ammonia nitrogen (r² = 0.730, p < 0.01), soil organic matter (r² = 0.901, p < 0.01) and soil total nitrogen (r² = 0.779, p < 0.01). Furthermore, gene counts for bacteria gene nirS were correlated negatively with plant root length (r² = 0.364, p < 0.05) and plant biomass (r² = 0.381, p < 0.05). Taken together, these results suggest that both nitrification and denitrification can occur in bioretention systems, which can be affected by environmental factors.

Bioretention system mediated by different dry-wet alterations on nitrogen removal: Performance, fate, and microbial community

In laboratory experiments, the nitrogen migration and transformation in the stormwater bioretention system under different dry-wet alterations were studied. The removal efficiency showed that nitrogen could be removed efficiently in bioretention system under all dry-wet alterations, and the shorter antecedent dry days (ADDs) (1-5 days) were beneficial to the removal of nitrogen before plants decay, compared to the longer ADDs (7-22 days). Using a new method combined with Hydrus-1D model, water transport was simulated and nitrogen migration in bioretention system was quantified, indicating that NH₄⁺-N was mainly removed in the planting layer, and the removal of NO₃^--N was occurred in the submerged layer. Fate experiment showed the main fate of the nitrogen was microorganisms (1-5 ADDs) and soil immobilization (7-22 ADDs). Microbial analysis showed that shorter ADDs (1-5 days) were suitable for Firmicutes growth, while Proteobacteria and Actinobacteria accounted for greater abundance under longer ADDs (7-22 days). Canonical correlation analysis (CCA) revealed the relationships between microbial community and environmental factors. Soil moisture content, soil organic matter (SOM), TN (water), root length, and NO₃^--N (water) were significantly correlated with bacterial community. This work may give new insights into nitrogen migration and transformation, and can provide a reference for the further mechanism study and construction of stormwater bioretention systems.

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.

Characterizing laboratory-scale clinoptilolite bio-columns for removal and nitrification of ammoniacal nitrogen in simulated stormwater

Due to the diverse speciation and biochemical characteristics of nitrogen in urban runoff, excess nitrogen continues to be a major source of eutrophication in receiving waters. The performance of a nitrifying-sorptive Clinoptilolite (ZT) was examined for use in a media-based stormwater control measure (SCM) for ammonium removal. Results suggested that columns operated under continuous feed showed more nitrification as the media approached ammonium exhaustion. Influent concentrations of 2.5 and 5 mg NH₄ ⁺ -N/L tested under continuous flow regimes both showed steady-state operation after media exhaustion, with the average effluent [NO₃ ^- N] of 1.2 and 1.7 mg/L, respectively. The performance of the media under intermittent flow regime showed lower effluent ammonium, nitrification between simulated saturated periods, and could treat an additional 70 bed volumes of simulated runoff when compared to a column receiving identical continuous feed. However, nitrification was not sufficient to prevent desorption of ammonium during drops in influent NH₄ ⁺ -N concentrations. Use of Clinoptilolite for ammonium sorption/nitrification is a systematic approach for capture and transformation of incoming/mineralized ammonium to nitrate prior to reaching an anoxic/denitrifying zone within SCMs. PRACTITIONER POINTS: Clinoptilolite can accumulate stormwater ammonium, allowing it to be nitrified. Ammonium nitrification will regenerate exchange sites on the clinoptilolite. Intermittent flow conditions allowed more nitrification between stormwater events.

Nitrogen process in stormwater bioretention: the impact of alternate drying and rewetting on nitrogen migration and transformation

Nitrogen migration and transformation in the stormwater bioretention system were studied in laboratory experiments, in which the effects of drying-rewetting were particularly investigated. The occurrence and distribution of nitrogen in the plants, the soil, and the pore water were explored under different drying-rewetting cycles. The results clearly showed that bioretention system could remove nitrogen efficiently in all drying-rewetting cycles. The incoming nitrogen could be retained in the topsoil (0-10 cm) and accumulated in the planted layer. However, the overlong dry periods (12 and 22 days) cause an increase in nitrate in the pore water. In addition, nitrogen is mostly stored in the plants' stem tissues. Up to 23.26% of the inflowing nitrogen can be immobilized in plant tissues after a dry period of 22 days. In addition, the relationships between nitrogen reductase activity in the soil and soil nitrogen content were explored. The increase of soil TN content could enhance the activity of nitrate reductase. Meanwhile, the activity of hydroxylamine reductase (HyR) could be enhanced with the increase of soil NO₃^- content. These results provide a reference for the future development of nitrogen transformation mechanism and the construction of stormwater bioretention systems.

Pilot and Field Studies of Modular Bioretention Tree System with Talipariti tiliaceum and Engineered Soil Filter Media in the Tropics

Stormwater runoff management is challenging in a highly urbanised tropical environment due to the unique space constraints and tropical climate conditions. A modular bioretention tree (MBT) with a small footprint and a reduced on-site installation time was explored for application in a tropical environment. Tree species used in the pilot studies were Talipariti tiliaceum (TT1) and Sterculia macrophylla (TT2). Both of the MBTs could effectively remove total suspended solids (TSS), total phosphorus (TP), zinc, copper, cadmium, and lead with removal efficiencies of greater than 90%. Total nitrogen (TN) removal was noted to be significantly higher in the wet period compared to the dry period (p < 0.05). Variation in TN removal between TT1 and TT2 were attributed to the nitrogen uptake and the root formation of the trees species. A field study MBT using Talipariti tiliaceum had a very clean effluent quality, with average TSS, TP, and TN effluent EMC of 4.8 mg/L, 0.04 mg/L, and 0.27 mg/L, respectively. Key environmental factors were also investigated to study their impact on the performance of BMT. It was found that the initial pollutant concentration, the dissolved fraction of influent pollutants, and soil moisture affect the performance of the MBT. Based on the results from this study, the MBT demonstrates good capability in the improvement of stormwater runoff quality.

The effect of intermittent drying and wetting stormwater cycles on the nutrient removal performances of two vegetated biofiltration designs

Vegetated biofiltration systems (biofilters) are now a well-established technology for treatment of urban stormwater, typically showing high nutrient uptake. However, the impact of high temporal variability of rainfall events (further exacerbated by climate change) on nitrogen and phosphorus removal processes, within different biofiltration designs, is still unknown. Hence, a laboratory-based study was conducted to uncover mechanisms behind nutrient removal in biofilters across different drying and wetting regimes. Two sets of experimental columns were based on (1) the standard biofiltration design (unsaturated zone only), and (2) combination of unsaturated and saturated (submerged) zone (SZ) with additional carbon source. Columns were watered with synthetic stormwater according to three drying and wetting schemes, exploring 1, 2, 3, 4 and 7-week drying. Hydraulic performance, soil moisture and pollutant removal were monitored. The results show that hydraulic conductivity of SZ design experiences less change over time compared to standard design, due to slower media drying, crack formation and lower plant die-off. Varied drying lengths challenged both designs differently, with 2-week drying resulting in significant drop of performance across most pollutants in standard design (except ammonia), while SZ design was able to retain high performance for up to four weeks of drying, sustaining microbial and plant uptake. Increased oxygenation of SZ columns during short-term drying was beneficial for ammonia and phosphorus removal. While SZ design showed better performance and quicker recovery for nitrogen removal, in regions with inter-rain event shorter than two weeks, the standard design (no saturated zone, no carbon source) can achieve similar if not better results.

Bioretention for removal of nitrogen: processes, operational conditions, and strategies for improvement

As one of the low-impact development measures, bioretention plays an important role in reducing the runoff peak flow and minimizing runoff pollutants, such as heavy metals, suspended solids, and nutrients. However, the efficiency of nitrogen removal in the bioretention system is unstable, owing to the different chemical properties of various forms of nitrogen and the limitations of current bioretention system for nitrogen transformation. This review article summarizes the recent advances in bioretention system in treatment of urban stormwater and agricultural runoff for nitrogen removal. The microbial characteristics and main processes of nitrogen transformation in bioretention are reviewed. The operational conditions affecting nitrogen removal, including climatic conditions, pH, wet-dry alternation, influent loads and nitrogen concentration, and hydraulic residence time are discussed. Finally, measures or strategies for increasing nitrogen removal efficiency are proposed from the perspectives of structural improvement of the bioretention system, optimization of medium composition, and enhancement of the nitrogen removal reaction processes.

Evaluating the relationships between specific drainage area characteristics and soil metal concentrations in long-established bioswales receiving suburban stormwater runoff

Bioswales are used to attenuate stormwater pollution, but their long-term sustainability regarding sequestered metals is relatively unknown, and a clear rationale for prioritizing soil management is lacking. Impervious areas draining into four 14-year-old suburban bioswales were delineated, for which surface soils (top 10 cm; 72 samples) were sampled; soils from 4 adjacent reference sites were also sampled. Total and water soluble metals (Cd, Cu, Pb, Zn) were quantified, and the relationships between metal concentrations and drainage area characteristics evaluated. Annual metal loads were estimated using regional runoff data to simulate current and future metal concentrations; risks to soil biota were assessed by comparing metal concentrations to ecological screening levels. The drainage areas' percent imperviousness (37-71%) and ratios of impervious drainage area to bioswale area (2.0-5.7) varied, owing to differing proportions of rooftops, paved surfaces, lawns, and natural soils. Total Cu and Zn ranged from 10.0 to 43.2 mg/kg dry soil, and 15.6 to 129.5 mg/kg dry soil, respectively. Across all bioswales, total Zn was positively correlated to percent impervious area (r = 0.32, p = 0.0073), the ratio of connected impervious drainage area to infiltration area (r = 0.32, p = 0.0073), and percent drainage area as paved surfaces (r = 0.46, p = 5.6 E-05), but negatively correlated to percent drainage area as lawns (r = -0.48; p = 2.4 E-05). Water soluble metal concentrations were orders of magnitude lower than total metals. Given annual metal loads (0.2-0.4 mg Cu/kg dry soil; 1.5-3.1 mg Zn/kg dry soil), replacing bioswale soils to constrain metal concentrations would be unnecessary for decades. Taken together, this study proposes a transferable approach of estimating, then verifying via sampling and analysis, bioswale soil metal concentrations, such that soil management decisions can be benchmarked to ecological screening levels.

The impact of stormwater biofilter design and operational variables on nutrient removal - a statistical modelling approach

Biofiltration systems can help mitigate the impact of urban runoff as they can treat, retain and attenuate stormwater. It is important to select the optimal design characteristics of biofilters (e.g., vegetation, filter media depth) to ensure high treatment performance. Operational conditions (e.g., infiltration rate) can also lead to significant changes in biofilter treatment performance over time. The impact of specific operational conditions on water quality treatment performance of stormwater biofilters is still not well understood. Furthermore, despite the importance of design characteristics and operational conditions on biofilter treatment performance, there is a lack of models that can be used to determine the optimal design and operation. In this paper, we developed a series of statistical models to predict the Total Phosphorus (TP) and Total Nitrogen (TN) removal performance of stormwater biofilters using various numbers of design characteristics and operational conditions. These statistical models were tested using data collected from four extensive laboratory-scale biofilter column studies. It was found that all models performed relatively well with a Nash-Sutcliffe Efficiency (NSE) of 0.42 - 0.61 for TP and 0.37 - 0.63 for TN. The most important design characteristics were filter media type and depth for TP treatment, and vegetation type and submerged zone depth for TN treatment. In addition, infiltration rate and inflow concentrations were the operational conditions that greatly influence outflow TP and TN concentrations from stormwater biofilters. As such, these variables need to be carefully considered when designing and operating stormwater biofilters. Sensitivity analysis results indicate that the model was quite sensitive to all regression coefficients and intercepts. Additional modelling exercises show that the model could be further simplified by reducing the number of cross-correlated parameters. These models can be used by practitioners for not just optimising the design, but also operating biofilters using real-time monitoring and control to achieve optimum performance.

The impact of media, plants and their interactions on bioretention performance: A review

Bioretention systems have gained considerable popularity as a more natural approach to stormwater management in urban environments. The choice of bioretention media is frequently cited as one of the critical design parameters with the ultimate impact on the performance of the system. The goal of this review is to highlight data that challenge the importance of media as being the dominant design parameter and argue that the long-term performance is shaped by the interactions between media and the living components of a bioretention system, especially vegetation. Some of the key interactions are related to the impact of plant roots on media pore structure, which has implications on infiltration, storage capacity, and treatment. Another relevant interaction pertains to evapotranspiration and the associated impacts on the water balance and the water quality performance of bioretention systems. The impacts of vegetation on the media are highlighted and actual, as well as potential, impacts of plant-media interactions on bioretention performance are presented.

Seasonal operation of dual-mode biofilters: The influence of plant species on stormwater and greywater treatment

The use of stormwater biofilters (also known as bioretention systems and raingardens), in tropical and semi-arid areas is hindered by seasonal rainfall patterns which cause extended dry periods. These periods can result in plant die-off, long-term damage to system health and leaching of pollutants when stormwater inflows resume. Using an additional polluted water source during dry periods could minimise system stress and eliminate the need to irrigate biofilters with potable water during dry spells. As such, the presented laboratory study tested the seasonal operation of biofilters, by switching from stormwater treatment in wet months to greywater treatment in dry months. Forty-five single planted biofilter columns, incorporating sedges, grasses, understory ornamentals and vines, were subjected to four months of stormwater inflows, followed by three months of greywater inflows. We also investigated the impact of including a carbon source in the saturated zone on treatment performance. The results showed plant species selection to be critical for nitrogen and phosphorus removal, with consistently effective species such as Carex appressa and Canna x generalis able to maintain low outflow concentrations (e.g. total nitrogen of 0.2-0.3 mg/L and 0.3-0.6 mg/L, respectively) across both water sources. Low outflow phosphorus concentrations were associated with plant species that had high filterable reactive phosphorus removal across both water sources. Similarly, higher removal of ammonia and oxidised nitrogen was associated with lower overall nitrogen concentrations. In contrast, high removal of total suspended sediment (>94%), biochemical oxygen demand (>98%) and some heavy metals (e.g. lead >98% and copper >93%) was reported across all designs. The results suggest that with the careful selection of plant species, biofilters can be operated seasonally as a feasible and practical solution to maintaining system health during extended dry periods.

Validation and uncertainty analysis of a stormwater biofilter treatment model for faecal microorganisms

Stormwater biofilters, also known as rain gardens or bioretention systems, are effective stormwater treatment systems. This paper presents the validation, sensitivity and uncertainty analyses of a model for microbial removal in stormwater biofilters. The model, previously developed based on a rather limited laboratory study, was fully validated using the data collected in extensive laboratory experiments and field tests. The lab-scale and field-scale systems used for validation were of various designs (e.g., system size, plant type, media type), and have been operated under a wide range of operational conditions (e.g., length of antecedent dry period, and the inflow volume and concentration). For each tested biofilter design, the predicted E. coli concentrations in biofilters' outflow showed relatively good agreement with the measured ones: e.g., Nash-Sutcliffe Efficiency (E_c) ranged from 0.50 to 0.60 for the laboratory tests, and E_c = 0.55 for the field system. The results from sensitivity analysis confirmed the significance of adsorption and desorption processes, and also revealed the impact of temperature on microbial die-off (which was not fully represented in the model development stage). Finally, parameter transferability from one system to another with similar design was examined, achieving generally promising E_c values (0.04-0.56 with the best-fit parameter set for the other system; maximum value: 0.46-0.63) and acceptable uncertainties (intersection between prediction uncertainty band and observation: 50%-97%). Most importantly, the prediction of E. coli outflow concentrations from the field system was reasonably good when laboratory-determined parameter values were adopted: with the best-fit parameter set for the lab-scale system, E_c = 0.39; maximum E_c = 0.55; intersection between prediction and observation = 83%. These results suggested that the very rare biofilter model for microbial removal could provide reliable prediction for large scale field systems, by simply calibrating parameters with limited laboratory-scale experiments.

Evaluation of a Low-Cost Ceramic Filter for Sustainable Reuse of Urban Stormwater in Arid Environments

Sustainable reuse of urban stormwater is inevitable in the fight against water crises in arid regions. This research aimed to evaluate the effectiveness of a low-cost ceramic filtration process for reuse applications of urban stormwater. Stormwater was collected from a storage pond located in Buraydah (Qassim, Saudi Arabia) for laboratory experiments. The filtration tests were performed in a continuous mode with constant pressure using a low-cost ceramic filter made of clay soil and rice bran. The removal rates of the contaminants (heavy metals) as well as the turbidity, suspended solids, and nutrients of the stormwater were assessed. High removal efficiencies for turbidity (97.4%), suspended solids (97.0%), BOD5 (78.4%), and COD (76.1%) were achieved while low removals were achieved for the nutrients: 19.7% for total nitrogen, 25.3% for nitrate, and 8.6% for phosphate. Removal efficiencies ranged between 36.2% and 99.9% for the heavy metals, i.e., iron, manganese, lead, zinc, nickel, copper, cadmium, selenium, and barium. Contaminant removal rates observed for the ceramic filter were also compared with the alum coagulation process operated in a continuous mode at an optimum alum dose of 50 mg/L. Similar removal behaviors for removal of turbidity, suspended solids, organics, nutrients, and heavy metals suggested that both ceramic filtration and alum coagulation can be effectively used for stormwater treatment. Effluent qualities of both the ceramic filter and alum coagulation met the standards, for recycling/reuse of wastewater, set by the Kingdom of Saudi Arabia and World Health Organization for unrestricted irrigation and toilet flushing. The study results revealed that ceramic filtration is a low-cost, energy efficient, and easy to maintain technology which can be complimentary to best management practices for stormwater.

Real time control of biofilters delivers stormwater suitable for harvesting and reuse

Stormwater biofilters have great potential to treat stormwater for harvesting and reuse, but their variable performance in pathogen removal requires further optimisation prior to widespread uptake. This paper provides the first evidence that real time control (RTC) of stormwater biofilters can mitigate the impact of operational characteristics that result in poor microbial removal. We developed two RTC strategies and validated them using long-term laboratory experiments, utilising biofilters with a raised outlet pipe that creates a submerged zone. The first RTC strategy focuses on delivering the best water quality for harvesting and reuse or for recreational waterways. It has two components which ensure adequate treatment (microbial die-off): (1) it retains water in the biofilter for at least two days before allowing any further inputs into the system, and (2) the input volume is restricted to the submerged zone's pore volume. This strategy was effective and significantly improved water quality in the biofilter effluent. However, since the system favours bypassing influent to ensure good quality effluent, only 28.4% of the stormwater was treated. This still resulted in a 62.3% reduction in the influent E. coli load because the system was effective at removing E. coli under controlled conditions. The second RTC strategy builds upon the first strategy, and focuses on delivering a balance between good water quality for harvesting and protecting the environment (i.e., lower bypass). Three hours before the next rainfall event begins, the water that has remained in the biofilter's submerged zone for at least two days is drained and collected for harvesting through a bottom pipe. When stormwater inflow begins, the bottom pipe is closed and the biofilter operates without control, with water leaving the biofilter to the environment via the raised outlet pipe. The harvested effluent of this RTC strategy met the Australian stormwater harvesting guideline requirements for dual reticulation with indoor and outdoor use and irrigation of commercial food crops. Although only 5.4% of stormwater was collected for harvesting in this strategy, the environment was better protected because of a significantly reduced bypass volume. Our experiments also showed that the nutrient and sediment removal was high for both RTC strategies. This study presents the first stepping stone toward RTC of stormwater biofilters, demonstrating that these systems can deliver safe stormwater for harvesting and reuse, and for active recreational uses.

A field and laboratory investigation of kerb side inlet pits using four media types

Kerb side inlets with adjacent leaky wells are an emerging tool to harvest stormwater and to reduce runoff volumes and peak flow rates. This is achieved by collecting the first flush runoff into kerb side storages and infiltrating this water into the surrounding soil, thereby also reducing stormwater pollutant loadings. The hydraulic performance of the kerb side inlet, filter media and surrounding soil are key factors in the performance of these systems. However, no field or laboratory data are currently available for the hydraulic performance of a kerb side tree inlet pit. In this study, 12 tree inlet pits were constructed and filled with various media types including gravel, water treatment solids (a recycled waste product), sandy loam and clay to examine (1) leaky well infiltration rates (2) emptying times of the wells and (3) the well capacity (runoff storage volume) before and after runoff filtering through the wells. Using a laboratory model, the water harvesting performance of the kerb side inlet plate was also examined for various road longitudinal slopes. Using the field and laboratory data, simulation of the well performance was undertaken using the Model for Urban Stormwater Improvement Conceptualisation (MUSIC) to assess the capacity of these systems to reduce runoff volumes at the residential street scale. It was hypothesised that the type of filter media used in leaky well systems has a significant impact on the infiltration rate, regardless of the native soil type through which the stormwater eventually infiltrates. The results showed that the infiltration rates of systems filled with gravel were significantly higher than for the other media types, and this was followed by water treatment solids, sandy loam and clay. The results of the MUSIC modelling indicated that 2.8% of the mean annual runoff volume in the catchment could be harvested by the systems at the case study site. It was found that selection of high infiltration rate media and regular maintenance are the key factors for maintaining long-term performance of these systems.

Influences of stormwater concentration infiltration on soil nitrogen, phosphorus, TOC and their relations with enzyme activity in rain garden

Rain garden is a typical facility with many applications in urban low impact development (LID). It plays an important role in regulating runoff water quantity and quality. Two rain gardens with the discharge ratios of 20:1 and 15:1 were used as studied facilities. Seven soil sampling events were conducted from April 2017 to February 2019 to study the influences of stormwater concentration infiltration in rain gardens on soil nitrogen (N), phosphorus (P) and TOC and their relations with enzymes. The results showed that the contents of soil TN and NO₂-N + TON in gardens gradually decreased with time, while those of NH₃-N and TP increased with time. The content of NO₃-N varied greatly with time, and there was no obvious rule. TOC increased first and then decreased. Vertical distributions of N, P and TOC showed that the contents of NH₃-N, NO₂-N + TON and TN at 0-50 cm were high, so the upper soil was the sensitive area to the influence of stormwater concentration infiltration in rain gardens. The content of NH₃-N decreased gradually with the increase of soil depth, but those of NO₃-N and TP increased with the soil depth. Therefore, NO₃-N and TP migrated down with water infiltration in soil, and preventing NO₃-N and P leaching was critical for effective N and P removal though rain gardens. Soil urease (SU), sucrose (SS), protease (SP) and acid phosphatase (SAP) had a good linear relationship with N, P and TOC, and R²were all greater than 0.5.

Predicting bioretention pollutant removal efficiency with design features: A data-driven approach

The objective of this study is to synthesize previous research findings from bioretention experiments and identify design features that lead to the best performance of bioretention pollutant removal with a data-driven approach. A bioretention database was built from 79 bioretention publications, composed of 182 records of bioretention cells with their design features and the corresponding pollutant removal efficiency data. Non-parametric correlation analysis, multiple linear regression (MLR), and decision tree classifiers were applied to investigate the relationships between bioretention design features and pollutant removal efficiencies. Non-parametric statistics and MLR results indicated that bioretention surface area, media depth, the presence of an internal water storage (IWS) layer, soil composition, and vegetation cover are all significantly correlated with pollutant removal efficiencies. The impacts of design features are significantly different under different climate and inflow conditions. Decision tree classifiers showed that non-vegetated bioretention cells with sand filter media generally have higher than 80% total suspended solid (TSS) mass removal efficiencies; bioretention cells with minimum organic matter and greater than 0.58 m soil media depth tend to remove more than 51% of total nitrogen (TN); and vegetated bioretention cells with minimum organic matter remove more than 67% of total phosphorus (TP). The overall accuracy of decision tree classifiers in the test set is around 70% to predict TSS, TN, and TP mass removal efficiency classes. This study suggests that the data-driven approach provides insights into understanding the complex relationship between bioretention design features and pollutant removal performance.

Urban Stormwater to Enhance Water Supply

The capture, treatment, and recharge of urban runoff can augment water supplies for water-scarce cities. This article describes trends in urban stormwater capture for potable water supply using examples from the U.S. and Australia. In water-limited climates, water supply potential exists for large scale stormwater harvesting and recharge, such as neighborhood-scale and larger projects. The beneficial use of urban stormwater to meet nonpotable water demands has been successfully demonstrated in the U.S. and internationally. However, in terms of potable water use in the U.S., the lack of a regulatory framework and uncertainty in treatment and water quality targets are barriers to wide-scale adoption of urban stormwater for recharge, which is not so evident in Australia. More data on urban stormwater quality, particularly with respect to pathogens and polar organic contaminants, are needed to better inform treatment requirements. New technologies hold promise for improved operation and treatment, but must be demonstrated in field trials. Stormwater treatment systems may be needed for large-scale recharge in highly urbanized areas where source control is challenging. The co-benefits of water supply, urban amenities, and pollution reduction are important for financing, public acceptance and implementation-but are rarely quantified.

Evaluation of pilot-scale biochar-amended woodchip bioreactors to remove nitrate, metals, and trace organic contaminants from urban stormwater runoff

Stormwater is increasingly being valued as a freshwater resource in arid regions and can provide opportunities for beneficial reuse via aquifer recharge if adequate pollutant removal can be achieved. We envision a multi-unit operation approach to capture, treat, and recharge (CTR) stormwater using low energy, cost-effective technologies appropriate for larger magnitude, less frequent events. Herein, we tested nutrient, metal, and trace organic contaminant removal of a pilot-scale CTR system in the laboratory using biochar-amended woodchip bioreactors following eight months of aging under field conditions with exposure to real stormwater. Replicate columns with woodchips and biochar (33% by weight), woodchips and straw, or woodchips only were operated with continuous, saturated flow for eight months using water from a watershed that drained an urban area consisting of residential housing and parks in Sonoma, California. After aging, columns were challenged for five months by continuous exposure to synthetic stormwater amended with 50 μg L^-1 of six trace organic contaminants (i.e., fipronil, diuron, 1H-benzotriazole, atrazine, 2,4-D, and TCEP) and five metals (Cd, Cu, Ni, Pb, Zn) frequently detected in stormwater in order to replicate the treatment unit operation of a CTR system. Throughout the eight-month aging and five-month challenge experiment, nitrate concentrations were below the detection limit after treatment (i.e., <0.05 mg N L^-1). The removal efficiencies for metals in all treatments were >80% for Ni, Cu, Cd, and Pb. For Zn, about 50% removal occurred in the woodchip-biochar systems while the other systems achieved about 20% removal. No breakthrough of the trace organic compounds was observed in any biochar-containing columns. Woodchip columns without biochar removed approximately 99% of influent atrazine and 90% of influent fipronil, but exhibited relatively rapid breakthrough of TCEP, 2,4-D, 1H-benzotriazole, and diuron. The addition of straw to the woodchip columns provided no significant benefit compared to woodchips alone. Due to the lack of breakthrough of trace organics in the biochar-woodchip columns, we estimated column breakthrough with a diffusion-limited sorption model. Results of the model indicate breakthrough for the trace organics would occur between 10,000 and 32,000 pore volumes. Under ideal conditions this could be equivalent to decades of service, assuming failure by other processes (e.g., clogging, biofouling) does not occur. These results indicate that multiple contaminants can be removed in woodchip-biochar reactors employed in stormwater treatment systems with suitable flow control and that the removal of trace organic contaminants is enhanced significantly by addition of biochar.

Stormwater biofilter treatment model for faecal microorganisms

This paper presents a new model to simulate long-term microbial removal in stormwater biofilters. The water flow module uses a 'three-bucket' approach to describe the flow processes in biofilters, while the microbial quality module employs the one-dimensional advection-dispersion equation to represent microbial transport and fate under different design and operational conditions. Three governing processes for microbial removal, adsorption, desorption and die-off, are included; temperature is also incorporated as a key factor for die-off. The model was tested using long term monitoring data collected from laboratory columns in which five different biofilter configurations were studied over a period of 44weeks. A multi-objective calibration with the balance of instantaneous ponding levels and event outflow volumes was implemented on the water flow module, and the Nash-Sutcliffe Efficiency (E) values ranged from 0.82 to 0.95. The microbial quality module was tested using the effluent Escherichia coli concentration data, and the E values obtained for different configurations were between 0.46 and 0.68. The optimized parameter values agreed with those presented in literature. However, sensitivity analyses suggested that the model's prediction was not sensitive to all parameters, the explanation for which was hypothesized to be data paucity rather than model structural uncertainties. Model validation was also conducted by splitting the data into calibration and validation datasets. The results further reinforced the needed for more data for model calibration.

Effective nitrogen removal during different periods of a field-scale bioretention system

Bioretention systems are designed to manage stormwater, but they are always subject to substantial nitrate leaching. In previous laboratory-scale experiments, nitrate removal was enhanced using wood chips as a medium. However, the ability of field-scale systems to transform and remove nitrogen species is not well understood. Here, we report a wood-chip bioretention system that accepts the natural runoff from an elevated highway. The transformation of different nitrogen species indicates that the field-scale system with wood chips also inhibits nitrogen leaching. Nitrate removal was mainly found during the wet period, and was affected by the hydraulic loading rate of the system. Overall, this study showed that the field performance of the wood-chip bioretention system effectively removes nitrogen. In addition, the main period associated with the nitrogen removal process was identified.

Experimental study and simulation of phosphorus purification effects of bioretention systems on urban surface runoff

Excessive phosphorus (P) contributes to eutrophication by degrading water quality and limiting human use of water resources. Identifying economic and convenient methods to control soluble reactive phosphorus (SRP) pollution in urban runoff is the key point of rainwater management strategies. Through three series of different tests involving influencing factors, continuous operation and intermittent operation, this study explored the purification effects of bioretention tanks under different experimental conditions, it included nine intermittent tests, single field continuous test with three groups of different fillers (Fly ash mixed with sand, Blast furnace slag, and Soil), and eight intermittent tests with single filler (Blast furnace slag mixed with sand). Among the three filler combinations studied, the filler with fly ash mixed with sand achieved the best pollution reduction efficiency. The setting of the submerged zone exerted minimal influence on the P removal of the three filler combinations. An extension of the dry period slightly promoted the P purification effect. The combination of fly ash mixed with sand demonstrated a positive purification effect on SRP during short- or long-term simulated rainfall duration. Blast furnace slag also presented a positive purification effect in the short term, although its continuous purification effect on SRP was poor in the long term. The purification abilities of soil in the short and long terms were weak. Under intermittent operations across different seasons, SRP removal was unstable, and effluent concentration processes were different. The purification effect of the bioretention system on SRP was predicted through partial least squares regression (PLS) modeling analysis. The event mean concentration removal of SRP was positively related to the adsorption capacity of filler and rainfall interval time and negatively related to submerged zones, influent concentration and volume.

Investigation into the long-term stormwater pollution removal efficiency of bioretention systems

In recent years, there has been a steady increase in the number of bioretention systems installed worldwide. However, there has only been limited research on the long-term effectiveness of these sustainable urban drainage system devices. This paper presents the results of a series of controlled field experiments investigating the pollutant removal efficiency of three, 10-year-old, bio-filtration systems that have been in service in the Sunshine Coast in Australia. The results of this study suggest that the long-term pollution removal performance of these systems may not be as effective as previously thought and further research is needed.

The pollution removal and stormwater reduction performance of street-side bioretention basins after ten years in operation

This study evaluated the pollution removal and hydrologic performance of five, 10-year old street-side bioretention systems. The bioretention basins were subjected to a series of simulated rainfall events using synthetic stormwater. Four different pollution concentrations were tested on three of the bioretention basins. The four concentrations tested were: A) no pollution; B) typical Australian urban pollutant loads; C) double the typical pollution loads, and; D) five times the typical pollution loads. Tests were also undertaken to determine the levels of contaminant and heavy metals build-up that occurred in the filter media over the 10 year operational life of the bioretention systems. Although highly variable, the overall hydrological performance of the basins was found to be positive, with all basins attenuating flows, reducing both peak flow rates and total outflow volumes. Total suspended solids removal performance was variable for all tests and no correlation was found between performance and dosage. Total nitrogen (TN) removal was positive for Tests B, C and D. However, the TN removal results for Test A were found to be negative. Total phosphorus (TP) was the only pollutant to be effectively removed from all basins for all four synthetic stormwater tests. The study bioretention basins were found to export pollutants during tests where no pollutants were added to the simulated inflow water (Test A). Heavy metal and hydrocarbon testing undertaken on the bioretention systems found that the pollution levels of the filter media were still within acceptable limits after 10 years in operation. This field study has shown bioretention basin pollution removal performance to be highly variable and dependant on a range of factors including inflow pollution concentrations, filter media, construction methods and environmental factors. Further research is required in order to fully understand the potential stormwater management benefits of these systems.

Comparison of filter media materials for heavy metal removal from urban stormwater runoff using biofiltration systems

The filter media in biofiltration systems play an important role in removing potentially harmful pollutants from urban stormwater runoff. This study compares the heavy metal removal potential (Cu, Zn, Cd, Pb) of five materials (potting soil, compost, coconut coir, sludge and a commercial mix) using laboratory columns. Total/dissolved organic carbon (TOC/DOC) was also analysed because some of the test materials had high carbon content which affects heavy metal uptake/release. Potting soil and the commercial mix offered the best metal uptake when dosed with low (Cu: 44.78 μg/L, Zn: 436.4 μg/L, Cd, 1.82 μg/L, Pb: 51.32 μg/L) and high concentrations of heavy metals (Cu: 241 μg/L, Zn: 1127 μg/L, Cd: 4.57 μg/L, Pb: 90.25 μg/L). Compost and sludge also had high removal efficiencies (>90%). Heavy metal leaching from these materials was negligible. A one-month dry period between dosing experiments did not affect metal removal efficiencies. TOC concentrations from all materials increased after the dry period. Heavy metal removal was not affected by filter media depth (600 mm vs. 300 mm). Heavy metals tended to accumulate at the upper 5 cm of the filter media although potting soil showed bottom-enriched concentrations. We recommend using potting soil as the principal media mixed with compost or sludge since these materials perform well and are readily available. The use of renewable materials commonly found in Singapore supports a sustainable approach to urban water management.

Urban stormwater runoff nitrogen composition and fate in bioretention systems

Multiple chemical forms of nitrogen in urban stormwater make its management challenging. Sixteen storm events were monitored and analyzed for total nitrogen (TN), particulate organic nitrogen (PON), nitrate (NO3-N), nitrite (NO2-N), ammonium (NH3-N), and dissolved organic nitrogen (DON) in stormwater runoff and in treated discharge through a conventional bioretention cell. Influent PON can be effectively removed via bioretention sedimentation/filtration, NH3-N by ion exchange/sorption, and NO2-N by oxidation. However, significant DON and NO3-N leached from the bioretention cell, resulting in only 9% net overall TN concentration reduction. Captured PON and vegetation detritus in the bioretention cell can be leached as DON or mineralized into NO3-N. The effluent N is dominated by NO3-N (46%) and DON (42%). Therefore, in addition to creating denitrification conditions for NO3-N, preventing DON leaching is also critical for effective nitrogen removal though bioretention systems. The bioretention cell exhibited a moderate mass load reduction for TN (41%), which mainly results from runoff volume reduction.

A review of ion and metal pollutants in urban green water infrastructures

In urban environments, the breakdown of chemicals and pollutants, especially ions and metal compounds, can be favoured by green water infrastructures (GWIs). The overall aim of this review is to set the basis to model GWIs using deterministic approaches in contrast to empirical ones. If a better picture of chemicals and pollutant input and an improved understanding of hydrological and biogeochemical processes affecting these pollutants were known, GWIs could be designed to efficiently retain these pollutants for site-specific meteorological patterns and pollutant load. To this end, we surveyed the existing literature to retrieve a comprehensive dataset of anions and cations, and alkaline and transition metal pollutants incoming to urban environments. Based on this survey, we assessed the pollution load and ecological risk indexes for metals. The existing literature was then surveyed to review the metal retention efficiency of GWIs, and possible biogeochemical processes related to inorganic metal compounds were proposed that could be integrated in biogeochemical models of GWIs.

The influence of design parameters on clogging of stormwater biofilters: a large-scale column study

A large-scale laboratory study was conducted to test the influence of design and operating conditions on the lifespan of stormwater biofilters. The evolution of hydraulic conductivity over time was studied in relation to a number of key design parameters (media type, filter depth, vegetation type, system sizing, etc). The biofilters were observed to clog over time, with average hydraulic conductivity decreasing by a factor of 3.6 over the 72 weeks of testing. The choice of plant species appears to have a significant effect on the rate of decrease in permeability, with plants with thick roots (e.g. Melaleuca) demonstrating an ability to maintain permeability over time. Other species studied, with finer roots, had no such beneficial effects. As expected, small systems relative to their catchment (and thus which are subjected to high loading rates) are more prone to clogging, as increases in hydraulic and sediment loading can lead to extremely low hydraulic conductivities. Sizing and the appropriate choice of vegetation are thus key elements in design because they can limit clogging, and therefore, indirectly increase annual load treated by limiting the volume of water bypassing the system.

The effects of antecedent dry days on the nitrogen removal in layered soil infiltration systems for storm run-off control

The effects of antecedent dry days (ADD) on nitrogen removal efficiency were investigated in soil infiltration systems, with three distinguishable layers: mulch layer (ML), coarse soil layer (CSL) and fine soil layer (FSL). Two sets of lab-scale columns with loamy CSL (C1) and sandy CSL (C2) were dosed with synthetic run-off, carrying chemical oxygen demand of 100 mg L(-1) and total nitrogen of 13 mg L(-1). The intermittent dosing cycle was stepwise adjusted for 5, 10 and 20 days. The influent ammonium and organic nitrogen were adsorbed to the entire depth in C1, while dominantly to the FSL in C2. In both columns, the effluent ammonium concentration increased while the organic nitrogen concentration decreased, as ADD increased from 5 to 20 days. The effluent of C1 always showed nitrate concentration exceeding influent, caused by nitrification, by increasing amounts as ADD increased. However, the wash-out of nitrate in C1 was not distinct in terms of mass since the effluent flow rate was only 25% of the influent. In contrast, efficient reduction (>95%) of nitrate loading was observed in C2 under ADD of 5 and 10 days, because of insignificant nitrification in the CSL and denitrification in the FSL. However, for the ADD of 20 days, a significant nitrate wash-out appeared in C2 as well, possibly because of the re-aeration by the decreasing water content in the FSL. Consequently, the total nitrogen load escaping with the effluent was always smaller in C2, supporting the effectiveness of sandy CSL over loamy FSL for nitrogen removal under various ADDs.

Influence of intermittent wetting and drying conditions on heavy metal removal by stormwater biofilters

Biofiltration is a technology to treat urban stormwater runoff, which conveys pollutants, including heavy metals. However, the variability of metals removal performance in biofiltration systems is as yet unknown. A laboratory study has been conducted with vegetated biofilter mesocosms, partly fitted with a submerged zone at the bottom of the filter combined with a carbon source. The biofilters were dosed with stormwater according to three different dry/wet schemes, to investigate the effect of intermittent wetting and drying conditions on metal removal. Provided that the biofilters received regular stormwater input, metal removal exceeded 95%. The highest metal accumulation occurs in the top layer of the filter media. However, after antecedent drying before a storm event exceeding 3-4 weeks the filters performed significantly worse, although metal removal still remained relatively high. Introducing a submerged zone into the filter improved the performance significantly after extended dry periods. In particular, copper removal in filters equipped with a submerged zone was increased by around 12% (alpha=0.05) both during wet and dry periods and for lead the negative effect of drying could completely be eliminated, with consistently low outflow concentrations even after long drying periods.

Removal of nitrogen by a layered soil infiltration system during intermittent storm events

The fates of various nitrogen species were investigated in a layered biological infiltration system under an intermittently wetting regime. The layered system consisted of a mulch layer, coarse soil layer (CSL), and fine soil layer (FSL). The effects of soil texture were assessed focusing on the infiltration rate and the removal of inorganic nitrogen species. The infiltration rate drastically decreased when the uniformity coefficient was larger than four. The ammonium in the synthetic runoff was shown to be removed via adsorption during the stormwater dosing and nitrification during subsequent dry days. Stable ammonium adsorption was observed when the silt and clay content of CSL was greater than 3%. This study revealed that the nitrate leaching was caused by nitrification during dry days. Various patterns of nitrate flushing were observed depending on the soil configuration. The washout of nitrate was more severe as the silt/clay content of the CSL was greater. However, proper layering of soil proved to enhance the nitrate removal. Consequently, a strictly sandy CSL over FSL with a silt and clay content of 10% was the best configuration for the removal of ammonium and nitrate.

Activated soil filters (bio filters) for the elimination of xenobiotics (micro-pollutants) from storm- and waste waters

A technical scale (0.12 m3) activated soil filter (bio filter) has been used to eliminate diverse xenobiotics (organic micro-pollutants) such as organophosphate flame retardants, and -plasticisers, musk fragrances, DEHP, benzothiazoles and triclosan from water. Model experiments to treat combined sewer overflow, storm water and a post treatment of waste water were performed in controlled laboratory experiments. The indicator compounds were typical for waste water. Diverse chemical compound groups and a wide spectrum from the lipophilic (pKow=5.9) to the hydrophilic (pKow=2.6) were included. The system consisted of a layer with high organic content (with vegetation to prevent clogging), a sand filter and a gravel drainage layer. The organic layer was spiked with activated sludge to enhance biomass and biodegradation potential. Usually the elimination rates varied from 64% to 99%, with only one compound reaching as little as 17%. For a technical suitability assessment it was calculated how long these filters would be stable in eliminating organic compounds from water. The estimated operating times for such systems was found to be about 100 years for a stack height of 2 m a year in regard to most compounds in this study.