Data fitting approach more critical than exposure scenarios and treatment of censored data for quantitative microbial risk assessment

Virus surrogates throughout a full-scale advanced water reuse system

Water reuse as an alternative water supply is increasing throughout the world due to water stress and scarcity; however, there are no standard practices for monitoring virus pathogens in such systems. This study aimed to identify suitable surrogates for virus fate, transport, and removal throughout a water reuse scheme. Various microbial targets (11 viruses, two phage, and three bacteria) were monitored using molecular and culture methods across all treatment stages in a wastewater reclamation facility and advanced water treatment facility. Criteria were established for identifying suitable surrogates, which included reliable detection, observable fate and transport, calculable log-reduction values (LRVs), correlations with other targets, and various morphological types. In total, five viruses (PMMoV, AiV, GII NoV, AdV, FRNA GII) met these stringent criteria and were suggested as potential virus surrogates. These surrogates enabled successful comparison of assigned versus actual LRVs throughout a water reuse scheme. Results suggest that virus pathogens are effectively removed throughout water reuse treatment and the suggested surrogates can be utilized for monitoring treatment performance and ensuring public health safety. This study provides a framework that water utilities across the world can reference for establishing virus monitoring practices.

Data-driven systematic analysis of waterborne viruses and health risks during the wastewater reclamation process

Waterborne viral epidemics are a major threat to public health. Increasing interest in wastewater reclamation highlights the importance of understanding the health risks associated with potential microbial hazards, particularly for reused water in direct contact with humans. This study focused on identifying viral epidemic patterns in municipal wastewater reused for recreational applications based on long-term, spatially explicit global literature data during 2000-2021, and modelled human health risks from multiple exposure pathways using a well-established quantitative microbial risk assessment methodology. Global median viral loads in municipal wastewater ranged from 7.92 × 104 to 1.4 × 106 GC L-1 in the following ascending order: human adenovirus (HAdV), norovirus (NoV) GII, enterovirus (EV), NoV GI, rotavirus (RV), and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Following secondary or tertiary wastewater treatment, NoV GI, NoV GII, EV, and RV showed a relatively higher and more stable log reduction value with medians all above 0.8 (84%), whereas SARS-CoV-2 and HAdV showed a relatively lower reduction, with medians ranging from 0.33 (53%) to 0.55 (72%). A subsequent disinfection process effectively enhanced viral removal to over 0.89-log (87%). The predicted event probability of virus-related gastrointestinal illness and acute febrile respiratory illnesses in reclaimed recreational water exceeded the World Health Organization recommended recreational risk benchmark (5% and 1.9%, respectively). Overall, our results provided insights on health risks associated with reusing wastewater for recreational purposes and highlighted the need for establishing a regulatory framework ensuring the safety management of reclaimed waters.

A mass balance approach for quantifying the role of natural decay and fate mechanisms on SARS-CoV-2 genetic marker removal during water reclamation

The recent SARS-CoV-2 outbreak yielded substantial data regarding virus fate and prevalence at water reclamation facilities (WRFs), identifying influential factors as natural decay, adsorption, light, pH, salinity, and antagonistic microorganisms. However, no studies have quantified the impact of these factors in full scale WRFs. Utilizing a mass balance approach, we assessed the impact of natural decay and other fate mechanisms on genetic marker removal during water reclamation, through the use of sludge and wastewater genetic marker loading estimates. Results indicated negligible removal of genetic markers during P/PT (primary effluent (PE) p value: 0.267; preliminary and primary treatment (P/PT) accumulation p value: 0.904; and thickened primary sludge (TPS) p value: 0.076) indicating no contribution of natural decay and other fate mechanisms toward removal in P/PT. Comparably, adsorption and decomposition was found to be the dominant pathway for genetic marker removal (thickened waste activated sludge (TWAS) log loading 9.75 log10 GC/day); however, no estimation of log genetic marker accumulation could be carried out due to high detections in TWAS. PRACTITIONER POINTS: The mass balance approach suggested that the contribution of natural decay and other fate mechanisms to virus removal during wastewater treatment are negligible compared with adsorption and decomposition in P/PT (p value: 0.904). During (P/PT), a higher viral load remained in the (PE) (14.16 log10 GC/day) compared with TPS (13.83 log10 GC/day); however, no statistical difference was observed (p value: 0.280) indicting that adsorption/decomposition most probably did not occur. In secondary treatment (ST), viral genetic markers in TWAS were consistently detected (13.41 log10 GC/day) compared with secondary effluent (SE), indicating that longer HRT and the potential presence of extracellular polymeric substance-containing enriched biomass enabled adsorption/decomposition. Estimations of total solids and volatile solids for TPS and TWAS indicated that adsorption affinity was different between solids sampling locations (p value: <0.0001).

Per- and poly-fluoroalkyl substances removal in multi-barrier advanced water purification system for indirect potable reuse

The study evaluated the removal efficacy of per- and poly-fluoroalkyl substances (PFAS) across various advanced water treatment (AWT) processes in a field-scale AWT train using secondary effluent samples from a full-scale water reclamation facility (WRF). Samples collected from April to October 2020 revealed PFCAs as the dominant PFAS compounds in the WRF secondary effluent, with PFPeA having the highest average concentration and PFSAs in notably lower amounts. Temporal fluctuations in total PFAS concentrations peaked in September 2020, which may reflect the seasonality in PFAS discharges related to applications like AFFFs and pesticides. In assessing AWT processes, coagulation-flocculation-clarification-filtration system showed no notable PFAS reduction, while ozonation resulted in elevated PFBS and PFBA concentrations. Biological activated carbon (BAC) filtration effectively removed long-chain PFAS like PFOS and PFHxS but saw increased concentrations of short-chain PFAS post-treatment. Granular activated carbon (GAC) filtration was the most effective treatment, reducing all PFSAs below the detection limits and significantly decreasing most PFCAs, though short-chain PFCAs persisted. UV treatment did not remove short-chain PFCAs such as PFBA, PFPeA, and PFHxA. The findings highlight the efficacy of AWT processes like GAC in PFAS reduction for potable reuse, but also underscore the challenge presented by short-chain PFAS, emphasizing the need for tailored treatment strategies. PRACTITIONER POINTS: Secondary effluents showed higher concentrations of PFCAs compared to PFSAs. Advanced water treatment effectively removes long-chain PFAS but not short-chain. Ozonation may contribute to formation of short-chain PFAS. BAC is less effective on short-chain PFAS, requiring further GAC treatment.

Distribution-based maximum likelihood estimation methods are preferred for estimating Salmonella concentration in chicken when contamination data are highly left-censored

Salmonella is a common chicken-borne pathogen that causes human infections. Data below the detection limit, referred to as left-censored data, are frequently encountered in the detection of pathogens. The approach of handling the censored data was regarded to affect the estimation accuracy of microbial concentration. In this study, a set of Salmonella contamination data was collected from chilled chicken samples using the most probable number (MPN) method, which consisted of 90.42% (217/240) non-detect values. Two simulated datasets with fixed censoring degrees of 73.60% and 90.00% were generated based on the real-sampling Salmonella dataset for comparison. Three methodologies were applied for handling left-censored data: (i) substitution with different alternatives, (ii) the distribution-based maximum likelihood estimation (MLE) method, and (iii) the multiple imputation (MI) method. For each dataset, the negative binomial (NB) distribution-based MLE and zero-modified NB distribution-based MLE were preferable for highly censored data and resulted in the least root mean square error (RMSE). Replacing the censored data with half the limit of quantification was the next best method. The mean concentration of Salmonella monitoring data estimated by the NB-MLE and zero-modified NB-MLE methods was 0.68 MPN/g. This study provided an available statistical method for handling bacterial highly left-censored data.

Tracking SARS-CoV-2 in rivers as a tool for epidemiological surveillance

The aim of this work was to evaluate if rivers could be used for SARS-CoV-2 surveillance. Five sampling points from three rivers (AR-1 and AR-2 in Arenales River, MR-1 and MR-2 in Mojotoro River, and CR in La Caldera River) from Salta (Argentina), two of them receiving discharges from wastewater plants (WWTP), were monitored from July to December 2020. Fifteen water samples from each point (75 in total) were collected and characterized physico-chemically and microbiologically and SARS-CoV-2 was quantified by RT-qPCR. Also, two targets linked to human contributions, human polyomavirus (HPyV) and RNase P, were quantified and used to normalize SARS-CoV-2 concentration, which was compared to reported COVID-19 cases. Statistical analyses allowed us to verify the correlation between SARS-CoV-2 and the concentration of fecal indicator bacteria (FIB), as well as to find similarities and differences between sampling points. La Caldera River showed the best water quality; FIBs were within acceptable limits for recreational activities. Mojotoro River's water quality was not affected by the northern WWTP of the city. Instead, Arenales River presented the poorest water quality; at AR-2 was negatively affected by the discharges of the southern WWTP, which contributed to significant increase of fecal contamination. SARS-CoV-2 was found in about half of samples in low concentrations in La Caldera and Mojotoro Rivers, while it was high and persistent in Arenales River. No human tracers were detected in CR, only HPyV was found in MR-1, MR-2 and AR-1, and both were quantified in AR-2. The experimental and normalized viral concentrations strongly correlated with reported COVID-19 cases; thus, Arenales River at AR-2 reflected the epidemiological situation of the city. This is the first study showing the dynamic of SARS-CoV-2 concentration in an urban river highly impacted by wastewater and proved that can be used for SARS-CoV-2 surveillance to support health authorities.

Tracking SARS-CoV-2 in rivers as a tool for epidemiological surveillance

The aim of this work was to evaluate if rivers could be used for SARS-CoV-2 surveillance. Five sampling points from three rivers (AR-1 and AR-2 in Arenales River, MR-1 and MR-2 in Mojotoro River, and CR in La Caldera River) from Salta (Argentina), two of them receiving discharges from wastewater plants (WWTP), were monitored from July to December 2020. Fifteen water samples from each point (75 in total) were collected and characterized physico-chemically and microbiologically and SARS-CoV-2 was quantified by RT-qPCR. Also, two targets linked to human contributions, human polyomavirus (HPyV) and RNase P, were quantified and used to normalize SARS-CoV-2 concentration, which was compared to reported COVID-19 cases. Statistical analyses allowed us to verify the correlation between SARS-CoV-2 and the concentration of fecal indicator bacteria (FIB), as well as to find similarities and differences between sampling points. La Caldera River showed the best water quality; FIBs were within acceptable limits for recreational activities. Mojotoro River's water quality was not affected by the northern WWTP of the city. Instead, Arenales River presented the poorest water quality; at AR-2 was negatively affected by the discharges of the southern WWTP, which contributed to significant increase of fecal contamination. SARS-CoV-2 was found in about half of samples in low concentrations in La Caldera and Mojotoro Rivers, while it was high and persistent in Arenales River. No human tracers were detected in CR, only HPyV was found in MR-1, MR-2 and AR-1, and both were quantified in AR-2. The experimental and normalized viral concentrations strongly correlated with reported COVID-19 cases; thus, Arenales River at AR-2 reflected the epidemiological situation of the city. This is the first study showing the dynamic of SARS-CoV-2 concentration in an urban river highly impacted by wastewater and proved that can be used for SARS-CoV-2 surveillance to support health authorities.

Comparison of the risk of infection of human rotavirus and astrovirus according to fishing and swimming activities at Las Cañas beach, Uruguay

AIMS

To estimate the risk of human Rotavirus (RV) and Astrovirus (HAstV) infections for swimmers and fishers at Las Cañas beach, Uruguay.

METHODS AND RESULTS

Surface water samples were collected monthly for one year. The dose response models used were β Poisson and ₁ F₁ hypergeometric for RV and HAstV, respectively. The probabilities of infection were calculated using a kernel density estimate to fitting the data and then sampling from this distribution (Monte Carlo simulation). The probability of RV infection for fishers was between 0 and 65% and for swimmers was between 0 and 50% (<18 years old) and between 0 and 38% (>18 years old). For HAstV, the probability of infection for fishers was between 0% and 45% and for swimmers was between 0 and 38% (<18 years old) and between 0 and 18% (>18 years old).

CONCLUSIONS

This study suggests that fishers are at higher risk of infection for both viruses when compared with swimmers mainly due to higher viral frequency and concentration at the site for fishing activities.

SIGNIFICANCE AND IMPACT OF THE STUDY

This study highlights the different risk of infection for Rotavirus and Astrovirus, determined according to two popular activities such as swimming and fishing evidencing their negative impact in public health when contaminated waters are used mainly in developing countries like Uruguay.

Genotype-dependent kinetics of enterovirus inactivation by free chlorine and ultraviolet (UV) irradiation

Inactivation kinetics of enterovirus by disinfection is often studied using a single laboratory strain of a given genotype. Environmental variants of enterovirus are genetically distinct from the corresponding laboratory strain, yet it is poorly understood how these genetic differences affect inactivation. Here we evaluated the inactivation kinetics of nine coxsackievirus B3 (CVB3), ten coxsackievirus B4 (CVB4), and two echovirus 11 (E11) variants by free chlorine and ultraviolet irradiation (UV). The inactivation kinetics by free chlorine were genotype- (i.e., susceptibility: CVB5 < CVB3 ≈ CVB4 < E11) and genogroup-dependent and exhibited up to 15-fold difference among the tested viruses. In contrast, only minor (up to 1.3-fold) differences were observed in the UV inactivation kinetics. The differences in variability between the two disinfectants could be rationalized by their respective inactivation mechanisms: inactivation by UV mainly depends on the genomic size and composition, which was similar for all viruses tested, whereas free chlorine targets the viral capsid protein, which exhibited critical differences between genogroups and genotypes. Finally, we integrated the observed variability in inactivation rate constants into an expanded Chick-Watson model to estimate the overall inactivation of an enterovirus consortium. The results highlight that the distribution of inactivation rate constants and the abundance of each genotype are essential parameters to accurately predict the overall inactivation of an enterovirus population by free chlorine. We conclude that predictions based on inactivation data of a single variant or reference pathogen alone likely overestimate the true disinfection efficiency of free chlorine.

Removal of SARS-CoV-2 viral markers through a water reclamation facility

There have been multiple reports of COVID-19 virus, SARS-CoV-2 RNA presence in influent wastewater of water reclamation facilities (WRFs) across the world. In this study, the removal of SARS-CoV-2 RNA was investigated in a WRF by collecting samples from various stages relayed to hydraulic retention time (HRT) and analyzed for viral RNA (N1 and N2) gene markers and wastewater characteristics. SARS-CoV-2 RNA was detected in 28 out of 28 influent wastewater and primary effluent samples. Secondary effluent showed 4 out of 9 positive samples, and all tertiary and final effluent samples were below the detection limit for the viral markers. The reduction was significant (p value < 0.005, one-way analysis of variance [ANOVA] test) in secondary treatment, ranging from 1.4 to 2.0 log10 removal. Adjusted N1 viral marker had a positive correlation with total suspended solids, total Kjeldahl nitrogen, and ammonia concentrations (Spearman's ρ = 0.61, 0.67, and 0.53, respectively, p value < 0.05), while demonstrating a strongly negative correlation with HRT (Spearman's ρ = -0.58, p value < 0.01). PRACTITIONER POINTS: Viral RNA was present in all samples taken from influent and primary effluent of a WRF. SARS-CoV-2 gene marker was detected in secondary effluent in 4 out of 9 samples. Tertiary and final effluent samples tested nondetect for SARS-CoV-2 gene markers. Up to 0.5 and 2.0 log10 virus removal values were achieved by primary and secondary treatment, respectively.

Assessing the microbial risk of faecal sludge use in Ugandan agriculture by comparing field and theoretical model output

Reuse of faecal sludge in agriculture has many potential benefits, but also poses risks to human health. To better understand the potential risks, Quantitative Microbial Risk Assessment (QMRA) was performed for three population groups in Kampala, Uganda: wastewater and faecal sludge treatment plant workers; farmers using faecal sludge; and consumers of faecal sludge-fertilised vegetables. Two models were applied for farmers and consumers, one based on pathogen concentrations from field sampling of sludge, soils and vegetables, and one based on theoretical pathogen contribution from the last sludge application, including decay and soil to crop transfer of pathogens. The risk was evaluated for two pathogens (enterohaemorrhagic E. coli (EHEC) and Ascaris lumbricoides). The field data on sludge, soil and vegetables indicated that the last application of faecal sludge was not the sole pathogen source. Correspondingly, the model using field data resulted in higher risks for farmers and consumers than the theoretical model assuming risk from sludge only, except when negligible for both. For farmers, the yearly risk of illness, based on measured concentrations, was 26% from EHEC and 70% from Ascaris, compared with 1.2% and 1.4%, respectively, considering the theoretically assumed contribution from the sludge. For consumers, the risk of illness based on field samples was higher from consumption of leafy vegetables (100% from EHEC, 99% from Ascaris) than from consumption of cabbages (negligible for EHEC, 26% from Ascaris). With the theoretical model, the risk of illness from EHEC was negligible for both crops, whereas the risk of illness from Ascaris was 64% and 16% for leafy vegetables and cabbage, respectively. For treatment plant workers, yearly risk of illness was 100% from EHEC and 99.4% from Ascaris. Mitigation practices evaluated could reduce the relative risk by 30-70%. These results can help guide treatment and use of faecal sludge in Kampala, to protect plant workers, farmers and consumers.

Navigating Data Uncertainty and Modeling Assumptions in Quantitative Microbial Risk Assessment in an Informal Settlement in Kampala, Uganda

Decision-makers in developing communities often lack credible data to inform decisions related to water, sanitation, and hygiene. Quantitative microbial risk assessment (QMRA), which quantifies pathogen-related health risks across exposure routes, can be informative; however, the utility of QMRA for decision-making is often undermined by data gaps. This work integrates QMRA, uncertainty and sensitivity analyses, and household surveys in Bwaise, Kampala (Uganda) to characterize the implications of censored data management, identify sources of uncertainty, and incorporate risk perceptions to improve the suitability of QMRA for informal settlements or similar settings. In Bwaise, drinking water, hand rinse, and soil samples were collected from 45 households and supplemented with data from 844 surveys. Quantified pathogen (adenovirus, Campylobacter jejuni, and Shigella spp./EIEC) concentrations were used with QMRA to model infection risks from exposure through drinking water, hand-to-mouth contact, and soil ingestion. Health risks were most sensitive to pathogen data, hand-to-mouth contact frequency, and dose-response models (particularly C. jejuni). When managing censored data, results from upper limits of detection, half of limits of detection, and uniform distributions returned similar results, which deviated from lower limits of detection and maximum likelihood estimation imputation approaches. Finally, risk perceptions (e.g., it is unsafe to drink directly from a water source) were identified to inform risk management.