Evaluation of three full-scale stormwater treatment systems with respect to water yield, pathogen removal efficacy and human health risk from faecal pathogens

Application of a microbial and pathogen source tracking toolbox to identify infrastructure problems in stormwater drainage networks: a case study

Water scarcity and increasing urbanization are forcing municipalities to consider alternative water sources, such as stormwater, to fill in water supply gaps or address hydromodification of receiving urban streams. Mounting evidence suggests that stormwater is often contaminated with human feces, even in stormwater drainage systems separate from sanitary sewers. Pinpointing sources of human contamination in drainage networks is challenging given the diverse sources of fecal pollution that can impact these systems and the non-specificity of traditional fecal indicator bacteria (FIB) for identifying these host sources. As such, we used a toolbox approach that encompassed microbial source tracking (MST), FIB monitoring, and bacterial pathogen monitoring to investigate microbial contamination of stormwater in an urban municipality. We demonstrate that human sewage frequently contaminated stormwater (in >50% of routine samples), based on the presence of the human fecal marker HF183, and often exceeded microbial water quality criteria. Arcobacter butzleri, a pathogen of emerging concern, was also detected in >50% of routine samples, with 75% of these pathogen-positive samples also being positive for the human fecal marker HF183, suggesting human municipal sewage as the likely source for this pathogen. MST and FIB were used to track human fecal pollution in the drainage network to the most likely point source of contamination, for which a sewage cross-connection was identified and confirmed using tracer dyes. These results point to the ubiquitous presence of human sewage in stormwater and also provide municipalities with the tools to identify sources of anthropogenic contamination in storm drainage networks.IMPORTANCEWater scarcity, increased urbanization, and population growth are driving municipalities worldwide to consider stormwater as an alternative water source in urban environments. However, many studies suggest that stormwater is relatively poor in terms of microbial water quality, is frequently contaminated with human sewage, and therefore could represent a potential health risk depending on the type of exposure (e.g., irrigation of community gardens). Traditional monitoring of water quality based on fecal bacteria does not provide any information about the sources of fecal pollution contaminating stormwater (i.e., animals/human feces). Herein, we present a case study that uses fecal bacterial monitoring, microbial source tracking, and bacterial pathogen analysis to identify a cross-connection that contributed to human fecal intrusion into an urban stormwater network. This microbial toolbox approach can be useful for municipalities in identifying infrastructure problems in stormwater drainage networks to reduce risks associated with water reuse.

Occurrence, Sources and Virulence Potential of Arcobacter butzleri in Urban Municipal Stormwater Systems

A. butzleri is an underappreciated emerging global pathogen, despite growing evidence that it is a major contributor of diarrheal illness. Few studies have investigated the occurrence and public health risks that this organism possesses from waterborne exposure routes including through stormwater use. In this study, we assessed the prevalence, virulence potential, and primary sources of stormwater-isolated A. butzleri in fecally contaminated urban stormwater systems. Based on qPCR, A. butzleri was the most common enteric bacterial pathogen [25%] found in stormwater among a panel of pathogens surveyed, including Shiga-toxin producing Escherichia coli (STEC) [6%], Campylobacter spp. [4%], and Salmonella spp. [<1%]. Concentrations of the bacteria, based on qPCR amplification of the single copy gene hsp60, were as high as 6.2 log10 copies/100 mL, suggesting significant loading of this pathogen in some stormwater systems. Importantly, out of 73 unique stormwater culture isolates, 90% were positive for the putative virulence genes cadF, ciaB, tlyA, cjl349, pldA, and mviN, while 50-75% of isolates also possessed the virulence genes irgA, hecA, and hecB. Occurrence of A. butzleri was most often associated with the human fecal pollution marker HF183 in stormwater samples. These results suggest that A. butzleri may be an important bacterial pathogen in stormwater, warranting further study on the risks it represents to public health during stormwater use.

Developing an Integrated "Regression-QMRA method" to Predict Public Health Risks of Low Impact Developments (LIDs) for Improved Planning

Worldwide Low Impact Developments (LIDs) are used for sustainable stormwater management; however, both the stormwater and LIDs carry microbial pathogens. The widespread development of LIDs is likely to increase human exposure to pathogens and risk of infection, leading to unexpected disease outbreaks in urban communities. The risk of infection from exposure to LIDs has been assessed via Quantitative Microbial Risk Assessment (QMRA) during the operation of these infrastructures; no effort is made to evaluate these risks during the planning phase of LID treatment train in urban communities. We developed a new integrated "Regression-QMRA method" by examining the relationship between pathogens' concentration and environmental variables. Applying of this methodology to a planned LID train shows that the predicted disease burden of diarrhea from Campylobacter is highest (i.e. 16.902 DALYs/1000 persons/yr) during landscape irrigation and playing on the LID train, followed by Giardia, Cryptosporidium, and Norovirus. These results illustrate that the risk of microbial infection can be predicted during the planning phase of LID treatment train. These predictions are of great value to municipalities and decision-makers to make informed decisions and ensure risk-based planning of stormwater systems before their development.

Highly variable removal of pathogens, antibiotic resistance genes, conventional fecal indicators and human-associated fecal source markers in a pilot-scale stormwater biofilter operated under realistic stormflow conditions

Green stormwater infrastructure systems, such as biofilters, provide many water quality and other environmental benefits, but their ability to remove human pathogens and antibiotic resistance genes (ARGs) from stormwater runoff is not well documented. In this study, a field scale biofilter in Southern California (USA) was simultaneously evaluated for the breakthrough of a conservative tracer (bromide), conventional fecal indicators, bacterial and viral human-associated fecal source markers (HF183, crAssphage, and PMMoV), ARGs, and bacterial and viral pathogens. When challenged with a 50:50 mixture of untreated sewage and stormwater (to mimic highly contaminated storm flow) the biofilter significantly removed (p < 0.05) 14 of 17 microbial markers and ARGsin descending order of concentration reduction: ermB (2.5 log(base 10) reduction) > Salmonella (2.3) > adenovirus (1.9) > coliphage (1.5) > crAssphage (1.2) > E. coli (1.0) ∼ 16S rRNA genes (1.0) ∼ fecal coliform (1.0) ∼ intl1 (1.0) > Enterococcus (0.9) ∼ MRSA (0.9) ∼ sul1 (0.9) > PMMoV (0.7) > Entero1A (0.5). No significant removal was observed for GenBac3, Campylobacter, and HF183. From the bromide data, we infer that 0.5 log-units of attenuation can be attributed to the dilution of incoming stormwater with water stored in the biofilter; removal above this threshold is presumably associated with non-conservative processes, such as physicochemical filtration, die-off, and predation. Our study documents high variability (>100-fold) in the removal of different microbial contaminants and ARGs by a field-scale stormwater biofilter operated under transient flow and raises further questions about the utility of human-associated fecal source markers as surrogates for pathogen removal.

Risk-Based Evaluation of Improvements in Drinking Water Treatment Using Cost-Benefit Analysis

Reliable and safe drinking water supply requires adequate risk management. Decision support models can aid decisionmakers to effectively evaluate risk mitigation measures and allocate societal resources. Here, a Swedish case study illustrates how the installation of ultrafiltration membranes can be evaluated by combining risk assessment and cost-benefit analysis. Quantitative microbial risk assessment was used to assess several contamination sources and estimate the achieved risk reduction from waterborne pathogens using Campylobacter, Norovirus, and Cryptosporidium as reference pathogens. The societal value of the improved water quality was estimated in the cost-benefit analysis by monetising the gained quality adjusted life years and aesthetic water quality improvements. The calculated net present value (mean of 7 MEUR) indicated that the installation of the ultrafiltration membranes was a sound investment from a societal economic perspective. The ultrafiltration membranes reduced the annual probability of infection from 3 × 10−2 to 10−7, well below the U.S. EPA’s acceptable level, as well as improving the aesthetic quality of the drinking water. The results provide a novel example of the importance for water distributors to consider not only health-related metrics when evaluating treatment options or monitoring the drinking water quality, but to also consider the aesthetic quality of the drinking water.

Investigating the public health risks of low impact developments at residential, neighbourhood, and municipal levels

Low Impact Developments (LIDs) employ a series of vegetative techniques to retain rainfall close to the site of origin. Although LIDs offer sustainable runoff management, these infrastructures can be considered a risk to public health due to the presence of pathogens in the runoff and human exposure to contaminated water held in and transported by LIDs. The objective of this study is to examine the disease burden of Gastrointestinal illness (GI) from exposure to LIDs at the residential, neighbourhood, and municipal levels. The authors conducted a meta-analysis of literature on three water features: (1) harvested rainwater obtained from LIDs, (2) surface water, and (3) floodwater. A set of 32 studies were systematically selected to collect values of risks of infection and expressed as the disease burden, i.e. disability adjusted life years (DALYs). The results showed that the percentage of GI illness exceeding the health guidelines were high for harvested rainwater, i.e. 22% of annual disease burden exceeded the WHO guidelines (0.001 DALYs/1000 persons), and 2% exceeded the US EPA guidelines (5.75 DALYs/1000 bathers). Among the six exposures for harvested rainwater, exposure to spray irrigation, exceeded US EPA guidelines whereas; five exposures, i.e. flushing, hosing, daily shower, spray irrigation, and children playing, surpassed the WHO guidelines. Considering LID treatment, the values of annual disease burden from all the selected barriers were below US EPA guidelines however, these values exceeded the WHO guidelines for three barriers i.e. water plaza, grass swale, and open storage ponds. These findings provide a broader perspective of the disease burden associated with LIDs and emphasise to consider the type of exposures and required treatment barriers for developing LID infrastructures in urban areas.

A review on microbial contaminants in stormwater runoff and outfalls: Potential health risks and mitigation strategies

Demands on global water supplies are increasing in response to the need to provide more food, water, and energy for a rapidly growing population. These water stressors are exacerbated by climate change, as well as the growth and urbanisation of industry and commerce. Consequently, urban water authorities around the globe are exploring alternative water sources to meet ever-increasing demands. These alternative sources are primarily treated sewage, stormwater, and groundwater. Stormwater including roof-harvested rainwater has been considered as an alternative water source for both potable and non-potable uses. One of the most significant issues concerning alternative water reuse is the public health risk associated with chemical and microbial contaminants. Several studies to date have quantified fecal indicators and pathogens in stormwater. Microbial source tracking (MST) approaches have also been used to determine the sources of fecal contamination in stormwater and receiving waters. This review paper summarizes occurrence and concentrations of fecal indicators, pathogens, and MST marker genes in urban stormwater. A section of the review highlights the removal of fecal indicators and pathogens through water sensitive urban design (WSUD) or Best Management Practices (BMPs). We also discuss approaches for assessing and mitigating health risks associated with stormwater, including a summary of existing quantitative microbial risk assessment (QMRA) models for potable and non-potable reuse of stormwater. Finally, the most critical research gaps are identified for formulating risk management strategies.

Bioretention performance under different rainfall regimes in subtropical conditions: A case study in São Carlos, Brazil

Low Impact Development practices have emerged as alternative solutions for traditional urban drainage by restoring the pre-development hydrologic regime. In subtropical climate areas, the performance of these systems is still poorly understood. This study aims to assess the performance of a bioretention basin in a subtropical climate area during an entire hydrological year in order to analyze the differences between dry and rainy seasons. The main climatic factors and conditions influencing the runoff retention efficiency and peak attenuation were also analyzed in order to support bioretention design for flood control purposes. Data of 29 precipitation events were collected over three years (2016-2018). The results show that the bioretention system retained between 9% and 100% of the runoff volume with an average efficiency of 65% during a whole hydrological year. The average runoff retention efficiency was of 73% and 61% for dry and rainy seasons, respectively. This difference is explained by the climatic factors which affected the bioretention performance. During dry periods, the antecedent soil moisture condition and runoff generation rate were found to be more important than the total precipitation depth, while the runoff retention efficiency was primarily influenced by the total rainfall depth and the maximum rainfall intensity during the wet period. Future research should focus on each of these periods in more detail, including water quality aspects.

Biotreatment technologies for stormwater harvesting: critical perspectives

Biotreatment technologies offer many advantages for passive stormwater treatment before harvesting, but performance can be variable and sensitive to system design, construction, operation and maintenance. While there is substantial research underpinning pollutant removal, hydraulic function, internal processes and optimal design, specific focus upon stormwater harvesting is relatively limited. Recent advances in system design include testing media amendments for targeted pollutant removal, enhanced pathogen removal using antimicrobial plants, and broadening technology application. However, the production of reliable fit-for-purpose water requires the development of robust validation methodologies to meet public safety expectations. While foundation studies exist, more needs to be done to extend the validation framework, monitor and control system performance and operation in real-time, and apply standards and regulatory checks.

Risk-based cost-benefit analysis for evaluating microbial risk mitigation in a drinking water system

Waterborne outbreaks of gastrointestinal diseases can cause large costs to society. Risk management needs to be holistic and transparent in order to reduce these risks in an effective manner. Microbial risk mitigation measures in a drinking water system were investigated using a novel approach combining probabilistic risk assessment and cost-benefit analysis. Lake Vomb in Sweden was used to exemplify and illustrate the risk-based decision model. Four mitigation alternatives were compared, where the first three alternatives, A1-A3, represented connecting 25, 50 and 75%, respectively, of on-site wastewater treatment systems in the catchment to the municipal wastewater treatment plant. The fourth alternative, A4, represented installing a UV-disinfection unit in the drinking water treatment plant. Quantitative microbial risk assessment was used to estimate the positive health effects in terms of quality adjusted life years (QALYs), resulting from the four mitigation alternatives. The health benefits were monetised using a unit cost per QALY. For each mitigation alternative, the net present value of health and environmental benefits and investment, maintenance and running costs was calculated. The results showed that only A4 can reduce the risk (probability of infection) below the World Health Organization guidelines of 10^-4 infections per person per year (looking at the 95th percentile). Furthermore, all alternatives resulted in a negative net present value. However, the net present value would be positive (looking at the 50^th percentile using a 1% discount rate) if non-monetised benefits (e.g. increased property value divided evenly over the studied time horizon and reduced microbial risks posed to animals), estimated at 800-1200 SEK (€100-150) per connected on-site wastewater treatment system per year, were included. This risk-based decision model creates a robust and transparent decision support tool. It is flexible enough to be tailored and applied to local settings of drinking water systems. The model provides a clear and holistic structure for decisions related to microbial risk mitigation. To improve the decision model, we suggest to further develop the valuation and monetisation of health effects and to refine the propagation of uncertainties and variabilities between the included methods.

Fit-for-purpose wastewater treatment: Conceptualization to development of decision support tool (I)

This article is the first in a series of two papers. Paper I focuses on model conceptualization and development, and Paper II in the series focuses on model validation and implementation. The amount of water reuse has been increasing across the globe. Wastewater can be treated based on the intended end use of reclaimed water. Fit-for-purpose wastewater treatment (WWT) simultaneously considers intended end use, economic viability, and environmental sustainability. WWT technologies differ mainly in terms of treatment efficiency, cost, energy use, and associated carbon emissions. The planning and evaluation of water reuse projects requires a decision support tool (DST) to evaluate alternative WWT trains and water reuse applications. However, such a DST is not available in the publically accessible literature. A DST, FitWater, has been developed for the evaluation of WWT for various urban reuses. The evaluation is based on the following criteria: amount of reclaimed water production, health risk of water reuse, cost, energy use, and carbon emissions. The cost is estimated as annualized life cycle cost and health risk is estimated using quantitative microbial risk assessment. The uncertainty analysis has been performed using probabilistic and fuzzy-based methods. A multi-criteria decision analysis, using fuzzy weighted average, is employed to aggregate different criteria and generate a final score. FitWater ranks alternative WWT trains based on the resulting final score. The proposed FitWater DST is user-friendly, and its application is demonstrated using an example. The DST can be enhanced to include additional treatment technologies and carbon emissions of different treatment processes.

Fit-for-purpose wastewater treatment: Testing to implementation of decision support tool (II)

This paper is the second in a series of two papers. In Paper I, a decision support tool (DST), FitWater, was developed for evaluating the potential of wastewater treatment (WWT) trains for various water reuse applications. In the present paper, the proposed DST has been tested and implemented. FitWater has been tested with several existing WWT plants in Canada and the USA, demonstrating FitWater's effectiveness in estimating life cycle cost (LCC), health risk, and energy use. FitWater has also been implemented in a newly planned neighbourhood in the Okanagan Valley (BC, Canada) by developing 12 alternative WWT trains for water reuse in lawn and public parks irrigation. The results show that FitWater can effectively rank WWT train alternatives based on LCC, health risk, amount of reclaimed water, energy use, and carbon emissions. Moreover, functions have been developed for the variation of unit annualized LCC and energy intensity per unit log removal of microorganisms in different treatment technologies with varying plant capacities. The functions have power relations, showing the economies of scale. FitWater can be applied to identify a cost-effective, risk-acceptable, and energy efficient wastewater treatment train with a plant capacity of 500m³/day or more. Furthermore, FitWater can be used to assess potential economic impacts of developing microbiologically stringent effluent standards. The capability of FitWater can be enhanced by including physio-chemical quality of wastewater, additional treatment technologies, and carbon emissions from wastewater decomposition processes.

Current Stormwater Harvesting Guidelines Are Inadequate for Mitigating Risk from Campylobacter During Nonpotable Reuse Activities

Campylobacter is a pathogen frequently detected in urban stormwater worldwide. It is one of the leading causes of enteric disease in many developed countries and is the leading cause of enteric disease in Australia. Prior to harvesting stormwater, adequate treatment is necessary to mitigate risks derived from such harmful pathogens. The goal of this research was to estimate the health risks associated with the exposure to Campylobacter when harvesting urban stormwater for toilet flushing and irrigation activities, and the role treatment options play in limiting risks. Campylobacter data collected from several urban stormwater systems in Victoria, Australia, were the inputs of a Quantitative Microbial Risk Assessment model. The model included seven treatment scenarios, spanning wetlands, biofilters, and more traditional treatment trains including those recommended by the Australian Guidelines for Water Recycling. According to our modeling and acceptable risk thresholds, only two treatment scenarios could supply water of sufficient quality for toilet flushing and irrigation end-uses: (1) using stormwater biofilters coupled with UV-treatment and (2) a more conventional coagulation, filtration, UV, and chlorination treatment plant. Importantly, our modeling results suggest that current guidelines in place for stormwater reuse are not adequate for protecting against exposure to Campylobacter. However, more research is required to better define whether the Campylobacter detectable in stormwater are pathogenic to humans.

Simulation of enteric pathogen concentrations in locally-collected greywater and wastewater for microbial risk assessments

As decentralized water reuse continues to gain popularity, risk-based treatment guidance is increasingly sought for the protection of public health. However, effort s to evaluate pathogen risks and log-reduction requirements have been hindered by an incomplete understanding of pathogen occurrence and densities in locally-collected wastewaters (i.e., from decentralized collection systems). Of particular interest is the potentially high enteric pathogen concentration in small systems with an active infected excreter, but generally lower frequency of pathogen occurrences in smaller systems compared to those with several hundred contributors. Such variability, coupled with low concentrations in many source streams (e.g., sink, shower/bath, and laundry waters), has limited direct measurement of pathogens. This study presents an approach to modeling pathogen concentrations in variously sized greywater and combined wastewater collection systems based on epidemiological pathogen incidence rates, user population size, and fecal loadings to various residential wastewater sources. Pathogen infections were modeled within various population sizes (5-, 100-, and 1,000-person) for seven reference pathogens (viruses: adenoviruses, Norovirus, and Rotavirus; bacteria: Campylobacter and Salmonella spp.; and protozoa: Cryptosporidium and Giardia spp.) on each day of 10,000 possible years, accounting for intermittent infection and overlap of infection periods within the population. Fecal contamination of fresh greywaters from bathroom sinks, showers/baths, and laundry, as well as combined greywater and local combined wastewater (i.e., including toilets), was modeled based on reported fecal indicators in the various sources. Simulated daily infections and models of fecal contamination were coupled with pathogen shedding characteristics to generate distributions of pathogen densities in the various waters. The predicted frequency of pathogen occurrences in local wastewaters was generally low due to low infection incidence within small cohort groups, but increased with collection scale (population size) and infection incidence rate (e.g., Norovirus). When pathogens did occur, a decrease in concentrations from 5- to 100- and from 100- to 1,000-person systems was observed; nonetheless, overall mean concentrations (i.e., including non-occurrences) remained the same due to the increased number of occurrences. This highlights value of the model for characterizing scaling effects over averaging methods, which overestimate the frequency of pathogen occurrence in small systems while underestimating concentration peaks that likely drive risk periods. Results of this work will inform development of risk-based pathogen reduction requirements for decentralized water reuse.

Microbial quality of reclaimed water for urban reuses: Probabilistic risk-based investigation and recommendations

Although Canada has abundant freshwater resources, many cities still experience seasonal water shortage. Supply-side and demand-side management is a core strategy to address this water shortage. Under this strategy, reclaimed water, which the Canadian public is willing to use for non-potable purposes, is an option. However, no universal guidelines exist for reclaimed water use. Despite the federal government's long-term goal to develop guidelines for many water reuse applications, guidelines have only been prescribed for reclaimed water use in toilet and urinal flushing in Canada. At the provincial level, British Columbia (BC) has promulgated guidelines for wide applications of reclaimed water but only at broad class levels. This research has investigated and proposed probabilistic risk-based recommended values for microbial quality of reclaimed water in various non-potable urban reuses. The health risk was estimated by using quantitative microbial risk assessment. Two-dimensional Monte Carlo simulations were used in the analysis to include variability and uncertainty in input data. The proposed recommended values are based on the indicator organism E. coli. The required treatment levels for reuse were also estimated. In addition, the recommended values were successfully applied to three wastewater treatment effluents in the Okanagan Valley, BC, Canada. The health risks associated with other bacterial pathogens (Campylobacter jejuni and Salmonella spp.), virus (adenovirus, norovirus, and rotavirus), and protozoa (Cryptosporidium parvum and Giardia spp.), were also estimated. The estimated risks indicate the effectiveness of the E. coli-based water quality recommended values. Sensitivity analysis shows the pathogenic E. coli ratio and morbidity are the most sensitive input parameters for all water reuses. The proposed recommended values could be further improved by using national or regional data on water exposures, disease burden per case, and the susceptibility fraction of population.

Risk-based enteric pathogen reduction targets for non-potable and direct potable use of roof runoff, stormwater, and greywater

This paper presents risk-based enteric pathogen log reduction targets for non-potable and potable uses of a variety of alternative source waters (i.e., locally-collected greywater, roof runoff, and stormwater). A probabilistic Quantitative Microbial Risk Assessment (QMRA) was used to derive the pathogen log10 reduction targets (LRTs) that corresponded with an infection risk of either 10-4 per person per year (ppy) or 10-2 ppy. The QMRA accounted for variation in pathogen concentration and sporadic pathogen occurrence (when data were available) in source waters for reference pathogens in the genera Rotavirus, Mastadenovirus(human adenoviruses), Norovirus, Campylobacter, Salmonella, Giardia and Cryptosporidium. Non-potable uses included indoor use (for toilet flushing and clothes washing) with occasional accidental ingestion of treated non-potable water (or cross-connection with potable water), and unrestricted irrigation for outdoor use. Various exposure scenarios captured the uncertainty from key inputs, i.e., the pathogen concentration in source water; the volume of water ingested; and for the indoor use, the frequency of and the fraction of the population exposed to accidental ingestion. Both potable and non-potable uses required pathogen treatment for the selected waters and the LRT was generally greater for potable use than non-potable indoor use and unrestricted irrigation. The difference in treatment requirements among source waters was driven by the microbial quality of the water - both the density and occurrence of reference pathogens. Greywater from collection systems with 1000 people had the highest LRTs; however, those for greywater collected from a smaller population (~ 5 people), which have less frequent pathogen occurrences, were lower. Stormwater had highly variable microbial quality, which resulted in a range of possible treatment requirements. The microbial quality of roof runoff, and thus the resulting LRTs, remains uncertain due to lack of relevant pathogen data.