Retention of heavy metals by stormwater filtration systems: breakthrough analysis

Biofiltration systems are widely used to mitigate the impacts of stormwater on receiving waters, however their long-term capacity to retain heavy metals has not previously been assessed. Accelerated-dosing laboratory experiments were used to assess the likelihood of breakthrough occurring for three different types of soil-based filter media that are commonly used in stormwater biofilters. In all cases, breakthrough of zinc (Zn) was observed, but not of cadmium (Cd), copper (Cu) and lead (Pb). If biofiltration systems are sized so that they are large relative to their catchment (at least 2-3% of its area) or have a deep filter layer (at least 0.5 m deep), then breakthrough will not occur for at least ten years and probably longer. However, after the equivalent of 12-15 years of operation, Cd, Cu and Zn had accumulated in the filter media to levels that exceeded human health and/or ecological guidelines. Further, depending on the design, it is possible that spent filter media may be classified as contaminated soil and thus require special disposal.     

Filter media for stormwater treatment and recycling: the influence of hydraulic properties of flow on pollutant removal.

Improved urban water management in Australia is of national importance. Water resources are stretched and urban runoff is a recognized leading cause of degradation of urban waterways. Stormwater recycling is an option that can contribute to easing these problems. Biofilters are effective structural stormwater pollution control measures with the potential for integration into stormwater treatment and recycling systems. However, premature clogging of biofilters is a major problem, with resulting decreased infiltration capacity (and hence the volume of stormwater the system can detain) and increased detention time. This paper presents preliminary findings with respect to the effect of clogging on pollutant removal efficiency in conventional stormwater filter media. A one-dimensional laboratory rig was used to investigate the impact of clogging on pollutant removal efficiency in a conventional biofiltration filter media (gravel over sand). Both the individual gravel layer and the overall multi-filter were highly efficient at removing suspended solids and particulate-associated pollutants. This removal efficiency was consistent, even as the filters became clogged. Removal of dissolved nutrients was more variable, with little reduction in concentrations overall. Although preliminary, these results challenge the concept that increased detention time improves the treatment performance of stormwater filtration systems.     

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

Biofiltration systems are an effective stormwater treatment technology. However, their robustness is yet to be tested, particularly their performance following extended dry periods. The hydraulic and treatment performance of five different non-vegetated, soil-based filters under varying periods of inundation and drying was assessed. The infiltration capacity of the filters decreased during wet periods and increased following dry periods, most probably due to swelling and shrinkage of the filter media. Treatment of sediment, heavy metals and phosphorus was not influenced by the wetting and drying regime. However, outflow concentrations of nitrogen were significantly higher upon re-wetting following extended dry periods compared with wet periods. This result has implications for current design practices, as these nitrogen pulses could negatively impact the ecological health of downstream receiving waters.     

Assessment of the Impact of Stormwater Characteristics on Clogging in Stormwater Filters

Hydraulic conductivity of granular filter media and its evolution over time is a key design parameter for stormwater filtration and infiltration systems that are now widely used in management of polluted urban runoff. In fact, clogging of filter media is recognised as the main limiting factor of these stormwater treatment systems. This paper focuses on the effect of stormwater characteristics on the clogging of stormwater filters. Effect of five different operational regimes has been tested in this study of sediment concentration; pollutant concentrations; stormwater sediment size; loading rate and stormwater loading/dosing regime and compared with the Base case. For each operational condition, five column replicates were tested. Results suggest that sediment concentration in stormwater is a significant parameter affecting hydraulic and treatment performance, eventually affecting longevity of these stormwater treatment systems. Further, the size of sediments (and their relation to the size of filter media grains) in stormwater was found to be an important parameter to be considered in design of coarse filters with high infiltration rates that are used for stormwater treatment. As expected, the addition of metals and nutrients had limited or no contribution to changes in hydraulic or sediment removal performance of the studied stormwater filters. Whilst loading rate was found to be an important parameter affecting the hydraulic and treatment performance of these systems, any variation in the stormwater loading regime had a limited effect on their performance. This study therefore develops an understanding of the effect of catchment characteristics on design of filters and hence their longevity and maintenance needs.     

Assessment of Impact of Filter Design Variables on Clogging in Stormwater Filters

Stormwater filters are widely used in stormwater management, sometimes as standalone structures (e.g. stormwater filter beds), or as part of porous pavements, soak ways, infiltration basins and trenches. Due to the high levels of sediment present in stormwater, clogging is the main operational issue for these systems. A laboratory-based study was conducted to investigate the effect of filter bed design variables on the clogging phenomenon in non-vegetated stormwater filters with high infiltration rates. Design parameters studied include: filter media particle sizes (0.5 mm, 2 mm, 5 mm); depth of the filter bed (100 mm, 300 mm and 500 mm); and filter media packing configurations (layered or mixed). The size of filter media particles significantly impact the clogging process, as well as the overall sediment removal performance of the filters; filters with smaller particles had better sediment removal efficiency, but subsequently shorter lifespan. Deeper systems had longer lifespan compared with shallower ones, notwithstanding deeper systems removed more sediment over their life span. Having two layers of distinct sized media in the filter bed improved performance (e.g. volume of water treated; sediment removed) over the single-layered systems. However, the three-layered systems behaved similarly to two-layered systems. Mixed systems also showed improved performance, as compared with single-layered systems, and were similar to the three-layered systems. This study therefore suggests that simple modifications to a stormwater filtration system can help improve sediment removal performance and/or reduce maintenance intervals significantly, while only slightly affecting sediment removal performance.     

Performance of enviss? stormwater filters: results of a laboratory trial

An experimental study was undertaken by Monash University to develop and test enviss? stormwater treatment and harvesting technologies - non-vegetated filtration systems with an extremely low footprint. This paper focuses on the water quality and hydraulic performance of two systems tested over a 'year' of operation in a Melbourne climate: (1) REUSE enviss? filters, designed for stormwater harvesting systems for non-potable supply substitution, and (2) WSUD enviss? filters, developed to treat urban stormwater prior to discharge to downstream systems. The presence of chlorine as a disinfection agent proved to be very efficient for the removal of microorganisms in REUSE enviss? filters. WSUD enviss? filters had the benefit of providing an elevated nutrient treatment performance, due to an extended depth of filter media. However, nutrient outflow concentrations (total nitrogen (TN) in particular) were found to increase during the testing period. Also, extended dry weather periods were found to have a detrimental effect on the treatment performance of almost all pollutants for both filters (nutrients, Escherichia coli and heavy metals). Although hydraulic conductivity results indicated two or three sediment trap replacements per year are required to maintain filtration rates, it is expected that the compressed loading rate schedule overestimated this maintenance frequency.     

Zinc-sulphate-heptahydrate coated activated carbon for microbe removal from stormwater

There is a need to develop effective stormwater filters for passive (without any addition of chemicals or energy) and effective removal of pathogens in order to mainstream stormwater harvesting. This study focuses on the development of coated granular activated carbon (GAC) filtration material in order to develop filters for effective removal of pathogens from urban stormwater. Several laboratory trials were performed to gauge the effectiveness of the filters, which use a mixture of the zinc-sulphate-heptahydrate coated GAC and sand, on the removal of Escherichia coli (E. coli) from semi-natural stormwater. On average, a 98% removal of the inflow concentration of E. coli was achieved. Furthermore, there was also an improvement of approximately 25% in the removal of phosphorous. However, it was found that the treated material was leaching zinc. It was important to determine whether the observed removal of E. coli was indirectly caused by the sampling methodology. The results showed that the inactivation of the E. coli in the collected sample was small compared with the inactivation which actually occurred within the filter. This provides much promise to the filter, but the presence of zinc in the outflow demonstrates the need for further investigation into the stabilisation of the coating process.     

Removal of dissolved nitrogen, phosphorus and carbon from stormwater by biofiltration mesocosms.

Biofiltration systems are becoming a popular stormwater treatment device in water sensitive urban design for the removal of fine particulate and dissolved pollutants from stormwater. However, there is limited published data on the effectiveness of these systems for nutrient removal. We constructed biofiltration mesocosms to assess nutrient removal (nitrogen, phosphorus and carbon) under experimental conditions. Different types of media were compared (gravel, sand, and sandy-loam) in vegetated and non-vegetated mesocosms (six treatments in total). Five plant species were used. Vegetated sand and vegetated sandy-loam provided the best overall treatment. Vegetated mesocosms were very effective in removing nitrogen (63-77% removal) and phosphorus (85-94% removal) from synthetic stormwater, and removed substantially more nutrients than the non-vegetated treatments. All treatments removed a substantial portion of the carbon from the stormwater (28-66%). When flushed with tap water, nitrogen and phosphorus were retained by the vegetated mesocosms, but leached from the non-vegetated mesocosms. Plant growth was most vigorous in the sandy-loam media, indicating that this is a good growth media, even without the addition of organic matter.     

Biofiltration as pre-treatment to water harvesting and recycling

This paper presents the results of the long term biofilter experiments conducted with raw stormwater collected from a canal at Carlton, in Sydney. Anthracite and granular activated carbon (GAC) were used as a single filter media in biofilter columns. Media heights of 75 and 40 cm were used. The filter columns were operated at filtration velocities of 0.12 and 0.25 m/h. The removal efficiency for turbidity and DOC for the GAC filter media were found to be 75% and almost 100% respectively. The removal efficiency for the anthracite filter was much lower. Molecular weight distribution analysis showed an almost similar trend to the DOC removal. Compared with anthracite filter media, the GAC biofilter removed a much larger range of organic compounds present in the stormwater. The GAC biofilter removes organic matter earlier as compared to anthracite. Based on a limited sample of stormwater, the removal efficiency for phosphorus was upto 74% and that of nitrogen was up to 30%. In general GAC filter shows higher heavy metal removal efficiency than anthracite. The removal of zinc, iron, lead and nickel were good. However the concentration of heavy metal in the raw surface water sample was low.     

Life-cycle energy and CO2 analysis of stormwater treatment devices

Environmental impacts associated with the construction, maintenance, and disposal of low-impact stormwater management devices are one aspect that should be considered during decision-making and life-cycle assessment (LCA) is a suitable method for quantifying such impacts. This paper reports a pilot study that employs LCA to compare life-cycle energy requirements and CO2 emissions of two stormwater devices in New Zealand. The two devices are a raingarden servicing an urban feeder road, and a sand filter that could have been installed in its stead. With an assumed life-time of 50 years, the life-cycle energy requirements of the built raingarden were almost 20% less than for the sand filter, while the CO2 emissions were 30% less. Our analysis shows that given the difference between the infiltration rates used in the raingarden design (0.3 m/day) and measured during monitoring (3 m/day) there was potential to make significantly greater life-time savings using a smaller design for the raingarden that would have also met the treatment efficiency expectations. The analysis highlights the significant contribution of transportation-of both materials and staff-and ongoing maintenance to a treatment device's life-cycle energy and CO2 profiles.     

Achieving multiple benefits from stormwater harvesting.

As the concept of integrated urban water management is incorporated into the practice of urban water servicing, new options, such as stormwater harvesting, which can have multiple benefits, are of increasing interest. The multi-functional benefits of stormwater harvesting include the potential to enhance urban stream health through improvements to the flow regime as well as providing a valuable water supply source. This paper synthesises a current research programme being undertaken to assess the viability of, and develop recommendations for, stormwater harvesting. The design of the collection, treatment, storage, flood protection, and distribution components of an integrated system are each discussed, along with the environmental flow consequences of urban stormwater harvesting. The incorporation of swales and biofilters into the collection system was not found to lead to significant exfiltration and evaporation losses in most circumstances and so can be employed as part of the treatment train. Further treatment can be provided by WSUD-type biophysical measures such as ponds, wetlands or novelly designed biofilters or physio-chemical treatment processes. Depending on the design, the stormwater storage component may or may not provide flood protection. In many circumstances, the storage capacity requirements are not considered to be a barrier to stormwater harvesting.     

Comparison of the nutrient removal efficiency of onsite wastewater treatments systems: applications of conventional sand filters and sequencing batch reactors (SBR).

When domestic wastewater was treated with different onsite applications of buried sand filters and sequencing batch reactors, good organic matter removal was common and effluent BOD7 concentrations from 5 to 20 mg/l were easily achievable. For total nitrogen, effluent concentrations were usually between 20 and 80 mg/l. Good phosphorus removal, even using special adsorption or precipitation materials, was difficult to achieve and large variations occurred. The median effluent concentration of total phosphorus in the most successful sand filter application was less than 0.1 mg/l and other sand filters and SBRs had the median concentrations varying from 1.7 to 6.7 mg/l. These results are based on one year in situ monitoring of 2 conventional buried sand filters, 6 sand filter applications with special phosphorus adsorbing media within the filter bed, 5 sand filters with separate tertiary phosphorus filtration and 11 small SBRs of three different types. The study was carried out in southern Finland during 2003-05. The whole project included monitoring of more than 60 plants of 20 different treatment types or methods, used in normal conditions to treat domestic wastewater. Evaluation of the different systems was made by comparing the measured effluent concentrations. In addition the effluent concentrations were compared to the discharge limits calculated according to the new Finnish regulation.     

Characterisation of the biosand filter for E. coli reductions from household drinking water under controlled laboratory and field use conditions.

More than a billion people in the developing world lack access to safe and reliable sources of drinking water. Point of use (POU) household water treatment technology allows people to improve the quality of their water by treating it in the home. One emerging POU technology is the biosand filter (BSF), a household-scale, intermittently operated slow sand filter. Laboratory and field studies examined Escherichia coli reductions achieved by the BSF. During two laboratory studies, mean E. coli reductions were 94% and they improved over the period of filter use, reaching a maximum of 99%. Field analysis conducted on 55 household filters near Bonao, Dominican Republic averaged E. coli reductions of 93%. E. coli reductions by the BSF in laboratory and field studies were less than those typically observed for traditional slow sand filters (SSFs), although as for SSFs microbial reductions improved over the period of filter use. Further study is needed to determine the factors contributing to microbial reductions in BSFs and why reductions are lower than those of conventional SSFs.     

Coarse filters for pond effluent polishing: comparison of loading rates and grain sizes.

A system comprising a UASB reactor, shallow polishing ponds and shallow coarse filters, treating actual wastewater from the city of Belo Horizonte, Brazil, has been evaluated. The main focus of the research was to compare grain sizes and hydraulic loading rates in the coarse filters. Two filters operating in parallel were investigated, with the following grain sizes: Filter 1: 3 to 10 cm; Filter 2: 8 to 20 cm. Two hydraulic loading rates were tested: 0.5 and 1.0 m3/m3.d. The filter with the lower rock size had a better performance than the filter with the larger rock size in the removal of SS and, as a consequence, BOD and COD. A better performance was obtained with the hydraulic loading rate of 0.5 m3/m3.d, as compared to the rate of 1.0 m3/m3.d. The effluent quality during the period with the lower loading rate was very good for discharge into water bodies or for agricultural reuse (median effluent concentrations from Filter 1: BOD: 20 mg/L; COD: 106 mg/L; SS: 28 mg/L; E. coli: 528 MPN/100 mL).     

Pathogen removal efficiency from UASB + BF effluent using conventional and UV post-treatment systems.

The aim of this study was to verify the efficiency of removal of microorganisms in effluents of a Wastewater Treatment Plant (WWTP) comprising an association of a UASB reactor followed by three submerged aerated biofilters (BAF) and one tertiary filter. The WWTP designed to treat domestic wastewater from a population of 1,000 inhabitants showed high removal efficiency for organic matter and suspended solids. Helminth eggs were also efficiently removed from the tertiary effluent and were found in the sludge from the UASB reactor; however, removal of bacteria in this system was very low. To enhance the efficiency of the system, the effluent from tertiary filters was submitted to UV disinfection in a real scale reactor. Our results showed that UV irradiation was very effective at lowering the concentrations of E. coli, thermotolerant coliforms and coliphages to acceptable levels for agricultural reuse. Salmonella spp. and helminth eggs were seeded into the tertiary effluent before passing through the UV reactor. Salmonella was not found in the final effluent, but helminth eggs were not completely inactivated by UV irradiation and viable eggs were detected after 28 d of incubation.     

Removal of geosmin and 2-methylisoborneol by biological filtration.

The quality of drinking water is sometimes diminished by the presence of certain compounds that can impart particular tastes or odours. One of the most common and problematic types of taste and odour is the earthy/musty odour produced by geosmin (trans-1, 10-dimethyl-trans-9-decalol) and MIB (2-methylisoborneol). Taste and odour treatment processes including powdered activated carbon, and oxidation using chlorine, chloramines, potassium permanganate, and sometimes even ozone are largely ineffective for reducing these compounds to below their odour threshold concentration levels. Ozonation followed by biological filtration, however, has the potential to provide effective treatment. Ozone provides partial removal of geosmin and MIB but also creates other compounds more amenable to biodegradation and potentially undesirable biological instability. Subsequent biofiltration can remove residual geosmin and MIB in addition to removing these other biodegradable compounds. Bench scale experiments were conducted using two parallel filter columns containing fresh and exhausted granular activated carbon (GAC) media and sand. Source water consisted of dechlorinated tap water to which geosmin and MIB were added, as well as, a cocktail of easily biodegradable organic matter (i.e. typical ozonation by-products) in order to simulate water that had been subjected to ozonation prior to filtration. Using fresh GAC, total removals of geosmin ranged from 76 to 100% and total MIB removals ranged from 47% to 100%. The exhausted GAC initially removed less geosmin and MIB but removals increased over time. Overall the results of these experiments are encouraging for the use of biofiltration following ozonation as a means of geosmin and MIB removal. These results provide important information with respect to the role biofilters play during their startup phase in the reduction of these particular compounds. In addition, the results demonstrate the potential biofilters have in responding to transient geosmin and MIB episodes.     

Pollutant removal efficiency of alternative filtration media in stormwater treatment.

Sorption experiments were used to assess the ability of various materials (sand, compost, packing wood, ash, zeolite, recycled glass and Enviro-media) to remove heavy metal contaminants typically found in stormwater. Compost was found to have the best physicochemical properties for sorption of metal ions (Cu, Zn and Pb) compared with sand, packing wood, ash, zeolite and Enviro-media. The compost sorption of these metal ions conformed to the linear form of the Langmuir adsorption equation with the Langmuir constants (q,) for Zn(ll) being 11.2 mg/g at pH 5. However, compost was also found to leach a high concentration of dissolved organic carbon (DOC, 4.31 mg/g), compared with the other tested materials. Various combinations of sand, compost and other materials were observed to have excellent heavy metal removal (75-96% of Zn and 90-93% of Cu), with minimal DOC leaching (0.0013-2.43 mg/g). The sorption efficiency of the different Enviro-media mixes showed that a combination of traditional (sand) and alternative materials can be used as an effective medium for the treatment of dissolved metal contaminants commonly found in stormwater. The application of using recycled organic materials and other waste materials (such as recycled glass) also provides added value to the products life cycle.     

Microbial challenge-testing of treatment processes for quantifying stormwater recycling risks and management.

Pathogenic microorganisms have been identified as the main human health risks associated with the reuse of treated urban stormwater (runoff from paved and unpaved urban areas). As part of the Smart Water initiative (Victorian Government, Australia), a collaborative evaluation of three existing integrated stormwater recycling systems, and the risks involved in non-potable reuse of treated urban stormwater is being undertaken. Three stormwater recycling systems were selected at urban locations to provide a range of barriers including biofiltration, storage tanks, UV disinfection, a constructed wetland, and retention ponds. Recycled water from each of the systems is used for open space irrigation. In order to adequately undertake exposure assessments, it was necessary to quantify the efficacy of key barriers in each exposure pathway. Given that none of the selected treatment systems had previously been evaluated for their treatment efficiency, experimental work was carried out comprising dry and wet weather monitoring of each system (for a period of 12 months), as well as challenging the barriers with model microbes (for viruses, bacteria and parasitic protozoa) to provide input data for use in Quantitative Microbial Risk Assessment.     

Development of a confined water sensitive urban design (WSUD) system using engineered soils.

Innovative Water Sensitive Urban Design (WSUD) systems are being investigated at three locations to the north and south of Sydney, Australia. These systems contain porous concrete pipes that are designed so that stormwater exfiltrates through the permeable walls of the pipes into the surrounding substrate media material. The porous pipes and media material treat the passing stormwater. The primary aim of the overall project is to develop a model to describe the treatment effectiveness of confined WSUD systems. This paper focuses on the system located at the Weathertex Industrial Site, Heatherbrae. Due to wood processing operations that occur at this site, it is recognised that the surface runoff will carry a heavy organics loading. Granulated Activated Carbon (GAC) is recognised for its ability to reduce the concentration of dissolved organics present in both wastewater and stormwater. GAC was therefore chosen as a filtration medium to be investigated at this site. To maximise the effectiveness of the GAC, extensive laboratory batch studies were undertaken prior to the field system being constructed to determine the optimum GAC/sand ratio. The purpose of the experimental work was to assess the dissolved organic removal potential through sorption of various concentrations of GAC. The aim of this paper is to describe these laboratory experiments and discuss how they related to the field system. Through these experiments it was determined that a sand/GAC ratio of 25:1 was ideal for the media material at the Heatherbrae site.     

Granular media filtration: old process, new thoughts.

The design of granular media filters has evolved over many years so that modern filters have larger media sizes and higher filtration velocities than in earlier times. The fundamental understanding of filtration has also improved over time, with current models that account reasonably for all characteristics of the media, the suspension and the filter operation. The methodology for design, however, has not kept pace with these improvements; current designs are based on pilot plants, past experience, or a simple guideline (the ratio of the bed depth to media grain size). We propose that design should be based universally on a characteristic removal length, with the provision of a bed depth that is some multiple of that characteristic length. This characteristic removal length is calculated using the most recent (and most complete) fundamental model and is based on the particle size with the minimum removal efficiency in a filter. The multiple of the characteristic length that yields the required bed depth has been calibrated to existing, successful filters.