Role of wastewater treatment in COVID-19 control

Coronavirus Disease 2019 or COVID-19 is the infectious disease caused by the newly discovered coronavirus, Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). The new virus and disease were first reported in December 2019 in Wuhan, China (WHO 2020a). The disease outbreak was declared by the World Health Organization (WHO) as a Public Health Emergency of International Concern on 30 January 2020 and recognised as a pandemic on 11 March 2020. While the evidence to date indicates that the virus spreads through respiratory droplets from infected people and direct contact with a surface that has the viable virus on it, existing knowledge to date suggest a possibility – although not definitive – for the faecal–oral transmission route (Sun et al. 2020; Xiao et al. 2020). Recent evidence has shown that the virus can be detected and persists in patients stool samples after pharyngeal swabs became negative (Chen et al. 2020a; Wang et al. 2020a), while infectious SARS-CoV-2 has been isolated from urine (Sun et al. 2020) and faeces of COVID-19 patients (Xiao et al. 2020). Further research is required to scientifically validate the faecal–oral pathway risk and the survival of the SARS-CoV-2 virus in sewage and other human faecal-origin wastes. Aerosol transmission potential also needs to be validated (Santarpia et al. 2020). With insufficient evidence with regards to faecal–oral pathways and aerosol transmission from wastewater systems, it is imperative that there be appropriate control measures put in place for wastewater treatment processes. This paper presents an overview of existing evidence-based research regarding the relationship between COVID-19 and wastewater, and presents an overview of appropriate control measures that could be used for disease transmission mitigation.

The virus that causes COVID-19 (SARS-CoV-2) has been detected in the faeces of a number of patients diagnosed with COVID-19. Recent reports revealed that 2–10% of COVID-19 patients had gastrointestinal symptoms, including diarrhoea (Chen et al. 2020b; Wang et al. 2020b) and one study showed that stool viral ribonucleic acid (RNA) was positive in 9 (15.3%) patients, and the median viral load was 4.7 (range: 3.4–7.6) log10 copies/mL (Cheung et al. 2020). Subsequently, SARS-CoV-2 RNA was detected in the faeces of 81.8% cases even with a negative result of throat swab test (Ling et al. 2020). SARS-CoV-2 RNA also could be detected in faeces of asymptomatic individuals (Tang et al. 2020), which may account for a significant proportion (17.9–30.8%) of infected individuals (Mizumoto et al. 2020; Nishiura et al. 2020). Shedding of SARS-CoV-2 RNA could be up to 7 weeks after the first symptom (Jiehao et al. 2020; Wu et al. 2020a). While human colonic fluid has been shown to inactivate the virus as shown from the analysis of faecal samples of COVID-19 patients (Zhang et al. 2020), in severe cases, it appears that this may not occur (Xiao et al. 2020).

Generally, the information about the percentage of patients shedding virus, the amount of virus discharged from a patient in stool, the length of shedding period, etc. are still very limited, and evidence whether the virus in stool is infectious is scant (Xiao et al. 2020). Although the risks caused by the virus in sewage and its removal in wastewater treatment processes have not been systematically assessed, the WHO has indicated that there is no evidence to date that SARS-CoV-2 has been transmitted via sewerage systems, with or without wastewater treatment (WHO 2020b).

Recent research has indicated that faecal shedding of SARS-CoV-2 is common and that the genetic material of viral SARS-CoV-2 can be detected in wastewater and sludges. The detection of viral SARS-CoV-2 RNA in wastewater and sludges does not mean that the virus is infectious. Infectious SARS-CoV-2 virus has yet to be recovered from wastewater. There have been quite a few preliminary reports of detection of SARS-CoV-2 viral RNA in wastewater using quantitative reverse-transcription Polymerase Chain Reaction (RT-qPCR), a method used in the Netherlands (Medema et al. 2020), USA (Nemudryi et al. 2020; Wu et al. 2020b), France (Wurtzer et al. 2020), Australia (Ahmed et al. 2020), Spain (Randazzo et al. 2020), Italy (La Rosa et al. 2020; Rimoldi et al. 2020), and Israel (Bar-Or et al. 2020). The maximum level of SARS-CoV-2 virus RNA untreated wastewater has been as high as 3.0 × 10⁶ copies/L. Only one study in France detected SARS-CoV-2 virus RNA in treated wastewater as well, with concentrations of up to nearly 10⁵ copies/L (Wurtzer et al. 2020). In addition to these reports, monitoring of SARS-CoV-2 virus in wastewater in different countries is ongoing following similar approaches and will further contribute to our understanding of the load and removal of the virus in wastewater treatment systems. Currently, little is known about the survival of the SARS-CoV-2 virus in sewage although previous studies showed that the most similar virus tested, SARS-CoV, could persist in wastewater for up to 2 days at 20 °C, and at least 14 days at 4 °C with around 2 Log₁₀ inactivation (Rosa et al. 2020). Similarly, an earlier study evaluating the survivability of enveloped viruses – of which SARS-COV-2 belongs to – in municipal wastewater using two model enveloped viruses (MHV and ϕ6) in unpasteurised and pasteurised wastewater at 10 and 25 °C showed significantly slower inactivation at the lower temperature (Ye et al. 2016). Furthermore, in comparison to model non-enveloped viruses, a larger fraction of the model enveloped viruses partitioned to the wastewater solids than non-enveloped viruses indicating that enveloped viruses are removed to a greater extent than non-enveloped viruses during primary wastewater treatment (Ye et al. 2016).

In treatment plants, sludge collected from primary, secondary, and tertiary wastewater treatment are combined and then treated with biological (i.e., aerobic digestion, anaerobic digestion, composting), chemical (i.e., lime), and thermal (i.e., heat drying, incineration) solids treatment processes. Effective sludge thermal treatment processes would likely kill viruses as well as other pathogens in sludge. Therefore, this paper does not place much focus on this process. SARS-CoV-2 may potentially be present in biologically and chemically treated sludge, and there are currently no published studies on the occurrence of SAR-CoV-2 in treated sludge (biosolids) as well as its fate after land application of biosolids.

As mentioned previously, RNA of SARS-CoV-2 has been found in the guts and faeces of some COVID-19 patients (Wu et al. 2020a) and in raw sewage (Medema et al. 2020). In the USA, SARS-CoV-2 RNA has been reported in primary sludge, detectable in 100% of the samples the concentrations ranged from 1.7 × 10⁷ virus RNA copies/L to 4.6 × 10⁸ copies/L, which is several orders of magnitude higher than in untreated wastewater (Peccia et al. 2020). It has also been shown that the SARS-CoV-2 RNA concentrations were highly correlated with the COVID-19 epidemiological curve and local hospital admissions indicating the potential of the technique as a disease outbreak surveillance tool (Peccia et al. 2020). In Spain, SARS-CoV-2 RNA was higher in some sludge sampling spots, specifically in primary sludge and thickened sludge, compared with the influent samples while none was detected in the final effluent (Balboa et al. 2020). In Turkey, SARS-CoV-2 RNA was detected not only in primary sludge but also in waste activated sludge collected from two treatment plants in Istanbul (Kocamemi et al. 2020).

It is important to note that the detection of viral RNA does not mean that infectious virus particles are present, and the stability and infectivity of SARS-CoV-2 in faeces and raw sewage still need to be investigated. Knowledge from previous studies, particularly from those on coronaviruses and other enveloped viruses, may provide insight into the fate and persistence of SARS-CoV-2 through wastewater and sludge treatment processes.

Coronaviruses are known to survive for a few days in the environment outside of a host cell (Kampf et al. 2020) and in wastewater between 2 and 4 days (Gundy et al. 2009). Temperature, organic matter, and antagonistic bacteria were reported to play a major role in the inactivation of coronaviruses (Gundy et al. 2009). Viruses are sensitive to temperature increases due to increased denaturation of their proteins, and the activity of extracellular polymers secreted by wastewater bacteria (John & Rose 2005). Gundy et al. (2009) reported that the survival time of coronavirus in filtered tap water increased from 10 days at 23 °C to over 100 days at 4 °C. Coronavirus inactivation was very quick in wastewater, likely due to the presence of detergents and solvents that caused damage to their envelope. Thus, sludge itself would not be a good environment for coronavirus survival. SARS-CoV was also reported to have low stability in the environment and to be very sensitive to high temperatures and oxidants (Wang et al. 2005; Pinon & Vialette 2018).

Adsorption of coronavirus on suspended solids and organic matter protected them from inactivation, and also assisted with their removal from the liquid phase through sedimentation (Gundy et al. 2009). The hydrophobicity of the viral envelope increases partitioning to solids resulting in their accumulation in sludge. Enveloped viruses were reported to associate more strongly with wastewater solids compared with non-enveloped viruses, which would mean their higher removal, particularly in primary sedimentation tanks (Ye et al. 2016). The above findings indicate that coronaviruses would be present in sludge but would not likely be able to survive sludge treatment processes which typically involve high temperatures (i.e., mesophilic and thermophilic aerobic and anaerobic digestion, composting), pH changes (i.e., anaerobic digestion, lime treatment), and diverse microbial populations and predatory microorganisms (i.e., biological treatment processes). Wang et al. (2005) reported that it was easier to inactivate SARS-CoV than Escherichia coli with chlorine, and when the total inactivation of SARS-CoV was achieved, E. coli was still not inactivated. E. coli is one of the easier microorganisms to inactivate in sludge compared with enteric viruses, protozoa, and other bacteria, and sludge stabilisation processes are typically designed to achieve the desired reduction in E. coli. Based on the findings of Wang et al. (2005), it can be concluded that current sludge treatment processes would likely be effective against coronaviruses.

Viruses with lipid envelopes, such as coronaviruses, lose their infectivity in water environments more quickly than non-enveloped viruses. However, significant variations in inactivation rates may exist between related strains. For example, the T₉₀ value (time to reach 90% inactivation) for SARS-CoV-1 was reported to be 9 days and less than 1 day for MHV (Wigginton et al. 2015). Similarly, it has been predicted SARS-CoV-2 to be among the sturdiest in the coronavirus family based on the rigidity of its outer shell and to be more resilient in the environment in comparison to SARS-CoV-1, MERS-CoV, and other coronaviruses (Goh et al. 2020). Therefore, it is important to have the actual inactivation data from sludge treatment processes for SARS-CoV-2 and other coronavirus strains to better understand the occurrence, fate, persistence, and infectivity of these emerging viruses.

Recent studies that used viral metagenome analysis provided new insights to viral pathogen diversity in sewage sludge. In biosolids treated with mesophilic anaerobic digestion, 43 different human viruses were identified, and surprisingly emerging viruses, including Coronavirus HKU1, were found to have a high abundance in comparison to Enteroviruses with relatively minor abundance (Bibby & Peccia 2013). These results suggest that coronaviruses might be present in biosolids, and their concentrations and infectivity need to be investigated. Previous studies almost exclusively focused on non-enveloped enteric viruses, and there is an urgent need for studies on SARS-CoV-2 and other enveloped viruses.

Conventional municipal wastewater treatment has preliminary, primary, and secondary treatment processes, while some municipal wastewater treatment plants also employ tertiary treatment. Basically, except for preliminary treatment with screening and grit chamber, all sewage treatment processes could remove or destroy viruses including the SARS-CoV-2 to some extent, although none is likely to remove all of the viruses present in sewage.

Viruses can be taken as fine particles with colloidal characteristics and could be absorbed by suspended particles. Therefore, some percentage removal of viruses is usually accompanied by the removal of suspended solids. In primary treatment, mechanical screening and the grit chamber only remove coarse solid particles; thus, the removal of virus associated with coarse particles usually is less than 0.3 Log₁₀ (Table 1). Primary sedimentation settles 50–75% of total suspended solids, removing a portion of the viruses in sewage, especially for those associated with the suspended solids. By sedimentation, virus is only transported to the separated sludge but not fully inactivated yet. The sludge and virus in it should be further treated to reduce the risk. Generally, primary treatment only removes a small percentage of viruses. Combining coagulation–flocculation and sediment – Chemically Enhanced Primary Treatment (CEPT) – a treatment process which involves the adding of chemicals to primary sedimentation tanks to facilitate improved settling properties – could be more effective regarding virus removal from the liquid stream.

Among different secondary treatment processes, activated sludge is one of the best biological methods for removing viruses from sewage, with well-controlled biodegradation conditions for virus decay in the aeration tank combined with secondary sedimentation. Conventional activated sludge process could remove 11 different enteric viruses by 0.65–2.85 Log₁₀ (Kitajima et al. 2014), and while another study conducted at a full-scale Wastewater Treatment Plant (WWTP) in Canada showed that activated sludge could remove 1.0–2.6 Log₁₀ of seven viruses (Qiu et al. 2015). The removal percentage could be even higher for Membrane Bioreactors (MBR), up to 6.8 Log10 reduction, due to further removal by membrane filtration (see Table 1).

For the tertiary treatment, coagulation using aluminium or iron coagulant could remove fine suspended particle and colloidal particles, resulting in less than 1–3 Log₁₀ virus removal. The removal efficiency of virus using membrane filtration could be a very large range, depending on membrane pore size, i.e., 0.2–5 Log₁₀ for Microfiltration (MF), >3 Log₁₀ for Ultrafiltration (UF), >5 Log₁₀ for nanofiltration, and >6 Log₁₀ for Reverse Osmosis (RO).

Two general classes of polymeric membranes are commonly used for advanced water treatment plants. These are low-pressure membranes (MF and UF) and high-pressure membranes (nanofiltration and reverse osmosis). Rejection of viruses can be achieved by each of these types of membranes and involves a number of mechanisms, including size exclusion and electrostatic repulsion from the membrane surface. Size exclusion is a function of the physical porosity of the membrane material, while electrostatic repulsion is a function of the surface charge on both the membrane surface and the virus particle, both of which are dependent upon ambient water pH.

Under challenge testing conditions, low-pressure membranes can achieve 4–6 Log₁₀ reduction for viruses with MF providing removal at the lower end of this range and UF at the higher end (WHO 2017). Many of the challenge tests undertaken to derive these Log₁₀ reduction values were undertaken using MS-2 bacteriophage since these are non-pathogenic to humans and can be easily cultivated in a laboratory.

MS-2 virions are approximately 30 nm in diameter (Strauss & Sinsheimer 1963), which is as small or smaller than many important waterborne viruses, such as Hepatitis A virus (30 nm), Rotavirus (80 nm), and Adenovirus (90 nm). Comprising an external lipid envelope, SARS-CoV-2 is larger again at around 120 nm. This is slightly larger than the typically quoted porosity of MF membranes (∼100 nm) and an order of magnitude larger than the typically quoted porosity of UF membranes (∼10 m).

MS-2 virions have an isoelectric point of 3.9 (Dowd et al. 1998), indicating that at pH > 3.9, the virion poses a net-negative charge and would experience electrostatic repulsion from a net-negatively charged membrane. A published isoelectric point for SARS-CoV-2 has not yet been identified, but a similar value (or even up to 6) would indicate a similar degree of electrostatic repulsion as a rejection mechanism for MS-2.

Under challenge testing conditions, RO membrane filtration can achieve 6 Log₁₀ reduction for viruses (Pype et al. 2016) and this figure is accepted in WHO Guidelines for Potable Reuse (WHO 2017). However, fully validated credit for RO removal for viruses is commonly limited to 1.5–2 Log₁₀ reduction since this is typically the limit of sensitivity for currently available online monitoring methods (WHO 2017).

As the last step of wastewater treatment, disinfection by chlorination, Ultraviolet (UV) or ozone could sufficiently inactivate viruses and provide a solid barrier to prevent them from entering the environment with the effluent, and ozone and UV seem to be more effective than chlorine. Very limited information about SARS-CoV-2 removal is available by different disinfection processes although generally what we do know is that the virus that causes COVID-19 is enveloped and is expected to be more sensitive to disinfection than non-enveloped viruses such as coxsackievirus, Hepatitis A and adenovirus which had been studied before. The details of disinfection will be covered in the following section.

Standard water disinfection technologies, such as free chlorine, chlorine dioxide, ozone, and UV have been reported to be effective for viral inactivation. These disinfection technologies played an important role in fighting against previous global epidemics, including the SARS epidemic in 2003, the influenza A virus subtype H1N1 (Swine flu) epidemic in 2009, the Middle East Respiratory Syndrome (MERS) epidemic in 2012, and the Coronavirus pandemic since December 2019, although water, wastewater, and sludge are not the main routes of transmission. As the SARS-CoV-2 is an enveloped virus, it is regarded to be sensitive to typical disinfection technologies.

Chlorine disinfection is widely used for virus inactivation, with applied doses described in terms of the product of chlorine concentrations and Contact Time (CT). When free chlorine is used as a disinfectant, the efficiency of inactivation is also influenced by the temperature and pH of the water. At pH 7.5, CT required for 3 Log₁₀ reduction of Giardia cysts can range from 50 to 300 mg min/L, depending on temperature, whereas for 4 Log₁₀ reduction of viruses, CT <12 mg min/L is typically required (United States Environmental Protection Agency 2003). Thus, when chlorine doses are selected for effective control of Giardia cysts, even greater control of waterborne viruses can generally also be assumed. Furthermore, virus CT requirements have typically been derived from experiments with non-enveloped viruses, such as Hepatitis A virus, which is known to persist in water under a variety of conditions (Sobsey et al. 1988). Inactivation of Hepatitis A virus is used since it is known to require higher CT values than those for inactivation of polio and rotaviruses under similar conditions of pH and temperature. Experiments with enveloped viruses (as SARS-CoV-2 is) have revealed that the envelope provides no additional protection from free chlorine disinfection, as free chlorine readily penetrates the lipid envelope to react with proteins in the nucleocapsid and polymerase complex which is internal to the nucleocapsid (Ye et al. 2018). As the weakest link in the overall virus structure to withstand the impact of free chlorine, it is these rapidly oxidised proteins that determine the overall inactivation rate of the virus during chlorination. There is no apparent reason to assume that CT values based on data from Hepatitis A virus inactivation will not be equally effective for reliable inactivation of SARS-CoV-2.

One valuable disinfection study with SARS-CoV virus-containing wastewater revealed that free chlorine was found to inactivate SARS coronavirus better than chlorine dioxide (Wang et al. 2005). Free residue chlorine over 0.5 mg/L for chlorine or 2.2 mg/L for chlorine dioxide in wastewater ensures complete inactivation of SARS coronavirus while it does not inactivate completely E. coli and f₂ phage at that dosage. Thus, the E. coli and f₂ phage could be potentially used as the indicators for SARS-Cov-2 inactivation in WWTPs.

The Ministry of Ecology and Environment of China issued guidelines for hospital and domestic wastewater disinfection on 1 February 2020 after the start of the epidemic. It was suggested to use liquid chlorine, chlorine dioxide, or bleach for disinfection with the dosage of 50 mg/L (as available chlorine). The CT for disinfection should be no less than 1.5 h, the residual chlorine should be over 6.5 mg/L (as free chlorine), and the faecal coliform should be less than 100 CFU/L in the finished water. For the facilities not able to satisfy the CT, a higher residual dosage of free chlorine is required. For example, over 10 g/L residual free chlorine is required in the case of 1 h contact. Considering the quick reaction of free chlorine and ammonia in hospital and domestic wastewater to form chloramines with lower inactivation ability, higher dosage of free chlorine was applied to achieve the breakpoint and maintain the sufficient free chlorine residual. Ozone disinfection cannot be interfered by ammonia which is quite useful for virus inactivation in wastewater. The guideline requires the utilities to reduce the suspended solids to less than 20 mg/L, to add ozone over 50 mg/L and to keep the CT over 0.5 h. The 4 Log₁₀ removal of total coliform and the residual faecal coliform less than 100 CFU/L are used as the indicators for good performance also.

This guideline echoed the previous reports on virus inactivation experiments and the US EPA (1999) Disinfection profiling and benchmarking guidance manual for drinking water which is included in Table 2.

Ozone disinfection requirements are also described by standard CT values for Giardia (CT_{3-log, Giardia}) and viruses (CT_{4-log, Virus}) (US EPA 2003). At temperatures up to 25 °C, ozone CT 0.5–3.0 mg min/L is typically required for 3 Log₁₀ reduction Giardia, whereas doses in a similar range (0.3–1.8 mg min/L) are applied to target 4 Log₁₀ reduction for viruses. These were based on experiments for the inactivation of poliovirus 1 (Roy et al. 1982), noting that they were also conservative for the inactivation of rotavirus (Vaughn et al. 1987). While few data are available to compare the ozone susceptibility of coronaviruses with poliovirus 1, it is assumed that it will be at least similar. Thus, ozonation CT values selected to target virus or (especially) Giardia disinfection should be expected to be similarly effective for SARS-CoV-2.

UV radiation in the short-wavelength UV (UVC) range (200–280 nm) is known to be effective for inactivating many types of microorganisms, including viruses, bacteria, and protozoa. Thus, it is increasingly used in water reuse applications for this purpose. There are two main types of lamps used to produce the UVC radiation for water treatment plants, some of which produce a narrow band of UV wavelengths, and some produce a broader range of wavelengths. But in all cases, UV radiation at 254 nm is produced, and this wavelength is known to be highly effective for microbial inactivation. Thus, UV of 254 nm wavelength (UV₂₅₄) is the focus of practically all UV-related water treatment, process control, regulation, and scientific studies.

When UV radiation is used to cause direct chemical disruptions to microorganisms (or to chemical contaminants), this is referred to as direct UV photolysis. Often, more indirect pathways are also important, whereby the UV causes reactions with other chemicals present in the water, the products of which then go on to inactivate the microorganisms. This process is referred to as indirect UV photolysis. UV inactivation of microorganisms is primarily achieved by the UV radiation causing chemical disruptions by direct UV photolysis of the microorganisms' genetic material (DNA or RNA), known as nucleic acids.

Key components of DNA and RNA nucleotides are nucleobases. Some nucleobases strongly absorb UV₂₅₄ radiation, leading to photochemical reactions, including the formation of chemical bonds between two consecutive nucleobases (a process known as dimerisation). The most studied are the formation of thymine–thymine dimer products, but thymine–cytosine, cytosine–cytosine, and uracil–uracil dimer products have also been reported (Douki 2013). These dimerisation reactions disrupt essential genetic processes, such as transcription and replication, and ultimately lead to inactivation of the microorganism. Other photochemical reactions can also contribute to inactivation, such as hydration, protein–nucleic acid linkages, covalent cross-links between complementary strands, and nucleic acid backbone breakages (Görner 1994).

Experiments comparing UV-induced dimerisation of virus genetic material have revealed that external components of virus particles, including lipid enveloped and protein nucleocapsids, provide no protection since UV₂₅₄ is able to penetrate these components and reach the genetic material within (Qiao et al. 2018). Thus, variability in UV₂₅₄ susceptibility among viruses is assumed to be primarily a function of genome size, sequence, and structure. The genetic material of viruses can be composed of single-stranded RNA, double-stranded RNA, single-stranded DNA, or double-stranded DNA.

During UV₂₅₄ photolysis experiments, single-stranded DNA was observed to undergo the most rapid photolysis reactions, followed by single-stranded RNA and double-stranded DNA. The slowest photolysis reactions were observed for double-stranded RNA, indicating a higher level of resistance for these types of viruses to UV₂₅₄ photolysis.

SARS-CoV-2 is a single-stranded RNA virus, and thus, it is possible to make a rough estimation of the likely susceptibility of SARS-CoV-2 to UV₂₅₄ photolysis, compared with other known waterborne viruses, which are regularly targeted for inactivation by UV₂₅₄ photolysis. Doing that, we could assume that SARS-CoV-2 has:

• at least the same susceptibility to UV photolysis as other single-stranded RNA viruses (e.g., Hepatovirus A, Polioviruses, Noroviruses, Coxsackieviruses, and MS-2 bacteriophage), and likely higher susceptibility to resistance since the coronaviruses have very large genomes compared with other single-stranded RNA viruses;

• roughly the same susceptibility to UV photolysis as double-stranded DNA viruses (e.g., Adenoviruses); and

• greater susceptibility to UV photolysis than double-stranded RNA viruses (e.g., Rotaviruses).

These suppositions will need to be validated. However, until such direct validation is available, we can assume that they are based on the best available information. A logical conclusion is that current UV disinfection practices, targeting waterborne viruses such as those listed above, can be assumed to be (at least) similarly effective for the inactivation of SARS-CoV-2.

Following conventional wastewater treatment, treated effluents may be reused for various applications. Conventional approaches to water reuse include reuse for municipal irrigation, agricultural irrigation, household reuse (toilet flushing, garden watering, etc.), industrial reuse, and reuse as a component of drinking water supplies, known as potable reuse. Potable reuse may occur by the planned and purposeful augmentation of a groundwater or surface water system, or may simply be a consequence of de facto potable reuse, whereby treated effluents are discharged to waterways which are subsequently used as a source of raw water for a drinking water supply, regardless of whether such a connection is formally acknowledged. A small but an increasing number of cities also practise direct potable reuse, whereby highly purified water is returned directly to a drinking water supply system without first being used to augment the groundwater or surface water body.

Depending on the nature of particular water reuse practices, various types and degrees of human exposure may be anticipated. While some practices would normally involve relatively low levels of exposure, others such as potable reuse imply high levels of water exposure by oral ingestion, dermal contact, and inhalation of aerosols. This anticipated degree of exposure is a key factor in determining risks associated with waterborne contaminants, including viruses. As a general principle, reuse practices involving higher levels of human exposure will require greater levels of water treatment to ensure greater effectiveness and reliability of pathogen removal, including viruses. In such cases, advanced water treatment processes may be applied, including additional chemical disinfection (e.g., chlorine or ozonation), UV disinfection, membrane filtration, and advanced oxidation processes (AOPs).

A range of AOPs has been developed for water treatment, including processes based on ozone and a range of heterogeneous catalysts. However, by far the most commonly applied to potable reuse projects are high-dose UV processes (Khan et al. 2017). Compared with a typical drinking water UV disinfection dose (40–180 mJ/cm²), the applied UV fluence in a UV-AOP (usually >500 mJ/cm²) is many times greater. The United States Environmental Protection Agency (US EPA) has developed UV fluence requirements for potable water systems to receive credit for inactivation of viruses (US EPA 2006). These range from 39 mJ/cm² for 0.5 Log₁₀ virus inactivation to 186 mJ/cm² for 4.0 Log₁₀ virus inactivation (US EPA 2006).

These UV inactivation credits were developed in the context of a US drinking water regulation (The Long-Term 2 Enhanced Surface Water Treatment Rule). Consequently, they have mostly been applied to drinking water applications and do not account for the much higher UV fluences used for AOPs in potable reuse projects. It is generally recognised that the much higher energy applied for UV AOPs is more than sufficient to achieve effective disinfection to the limits at which regulatory agencies commonly credit disinfection performance. For example, Californian potable reuse regulations provide a maximum credit of 6 Log₁₀ reduction for viruses by any advanced water treatment process. The WHO Guidelines for Potable Reuse concur that UV-AOP can achieve 6 Log₁₀ reduction under challenge testing conditions, and also indicate that UV AOPs can be validated to this level of performance (WHO 2017).

In order to fully assess risks associated with SARS-CoV-2 in water reuse projects, including potable reuse projects, some form of Quantitative Microbial Risk Assessment (QRMA) will be necessary (Kitajima et al. 2020). Such an assessment will not be possible until estimates are available for some key parameters, including excretion rates or wastewater concentrations of viable infectious virions. Once such data are available, it will be possible to provide treatment-based Log₁₀ reduction values by conservatively applying existing virus-related Log₁₀ reduction values for processes such as additional chemical disinfection (e.g., chlorine or ozonation), UV disinfection, membrane filtration, and AOPs.

However, risks will remain unquantifiable until estimates for dose–response relationships for infection are available. Whether existing dose–response relationships for other coronaviruses are applicable (Watanabe et al. 2010) that needs to be determined. The WHO drinking water guidelines and potable reuse guideless have set tolerable levels of risk for microbial pathogens in terms of Disability-Adjusted Life Years (DALYs) (WHO 2011, 2017). To do this, same for SARS-CoV-2 will require the establishment of a numerical value for the number of DALYs-per-case of infection with SARS-CoV-2, taking into account anticipated rates of death and other illnesses for various age groups in the community.

This section provides guidance for COVID-19 disease transmission regarding small WWTPs. The definition of what constitutes a small city (as compared with a large city) can vary, but it is important to recognise that there are wastewater collection systems and treatment technologies available for treating small flows that are not feasible for large flows. The demand for decentralised small systems is being considered as an alternative or complement to large, centralised collection and treatment systems. This is much easier to adopt in developments outskirts of the city or suitable for small community development and has been applied in both developed and developing world contexts.

Wastewater treatment in small populations is not straightforward. The high variability of wastewater characteristics together with the fact that these small towns often present a lack of financial, technical, and human resources, has led to the conclusion that the implementation of the treatment solutions commonly used in larger populations has not produced the desired results (Aragón et al. 2011). The design capacity of several WWTPs in small communities is significantly less than 10,000 person equivalents. As the population in the area increases, the operation capacity of the plant is increased. One of the challenges with small systems is that they are not designed for large fluctuations in flow, either they over flow or end up with small flow. Small systems can, therefore, be easily overloaded beyond their limits or hydraulically and/or organically underloaded. The technical alternatives range from mechanical and simple biological systems such as ponds, sand filters, and reed beds to complex high rate suspended and fixed biomass reactors.

Most small WWTPs apply not more than secondary treatment process to the wastewater. This usually involves relatively low technology or basic treatment processes. Where the plant is discharging to a sensitive area, tertiary treatment may be applied. Treatment processes in many small WWTPs do not usually possess automated instrumentation or telemetric system. One of the challenges in understanding the risk associated with COVID-19 from small WWTPs is the general lack of scientific information regarding virus removal from these options. As small WWTPs come in a variety of technical options and dependent on local operating conditions, there is no single COVID-19 control measure that is applied for all systems. Rather, the appropriate action would be to ensure proper operation and maintenance. This would involve regulation and control of the treatment process, i.e., clean the pre-treatment screen, remove debris, monitor equipment, and attend to operational challenges immediately. The inclusion of tertiary treatment processes as part of the treatment train would provide an additional control measure as evidence presented earlier has shown. The major challenge for developing countries is the effective operation and maintenance of systems. While there is a large body of evidence available that illustrates this challenge with large WWTPs (Eales 2008; Hawkins et al. 2013; United Nations (UN)-Water 2015), it is assumed that similar challenges may be experienced for small WWTPs. Best practices for protecting the health of workers at sanitation treatment facilities should be followed. Workers should wear appropriate Personal Protective Equipment (PPE), which includes protective outerwear, gloves, boots, goggles or a face shield, and a mask; they should perform hand hygiene frequently; and they should avoid touching eyes, nose, and mouth with unwashed hands.

In developing countries, a mixture of sanitation systems may be implemented in a single city. Technologies range from sewerage-based systems to non-sewered sanitation systems, such as pour-flush latrines, conventional dry latrines, and septic tanks. The coverage of each sanitation system is largely dependent on the household, community, town, or city preferences. Regarding COVID-19, a large percentage of human faecal waste in developing countries is not adequately treated. For WWTPs, a critical shortage of skilled staff to operate and maintain the treatment works is a common theme as well as the inability of the treatment works to deal with increasing pollution loads, unstable water and energy supply, and the lack of investment in the current infrastructure. The challenge would be to place more emphasis of process control and include additional treatment steps, such as chlorination, to manage COVID-19 risk.

Non-sewered sanitation systems can also pose a challenge. In India, for example, large parts of that country do not have underground drainage systems that connect directly to a treatment plant. More than 60% of urban households are dependent on non-sewered sanitation systems such as septic tanks. These systems eventually require safe emptying, transportation, and treatment of the faecal sludge collected. Similar patterns exist throughout the developing world. It is estimated that close to 3 billion people use non-sewered sanitation systems globally (Strande et al. 2014) with the systems found to be the predominate technology in developing cities of varying population size (Chowdhry & Koné 2012). Challenges occur throughout the sanitation value chain with respect to the safe collection, emptying, treatment, and disposal of accumulated faecal sludges. Little is known so far on the infectious nature in faecal sludges although recent evidence suggests that this may be possible (Xiao et al. 2020). In a study conducted in China, SARS-CoV-2 viral RNA was observed in septic tank effluents from Wuchang Fangcang Hospital after disinfection with 800 g/m³ sodium hypochlorite, the recommended disinfection guideline by the WHO and Chinese Center for Disease Control and Prevention (Zhang et al. 2020). Viral RNA was not detected in limited influent samples (3-day sampling). The researchers have noted that the study was limited with sample size and sample transport due to outbreak restrictions at the time of the study. The researchers indicated that the need for further exploration of the optimal dosage of sodium hypochlorite required for complete removal of SARS-CoV-2 viral RNA and understanding viral escape from disinfection procedures in faecal sludge (Zhang et al. 2020).

Strengthening of cleaning and disinfection protocols and occupational health and safety are recommended for the sanitation value chain. The WHO recommends that utility gloves or heavy-duty, reusable plastic aprons are cleaned with soap and water, and then decontaminated with 0.5% sodium hypochlorite solution each time they are used. Single-use gloves made of nitrile or latex, and gowns should be discarded as infectious waste after each use and not reused; hand hygiene should be performed after PPE is removed. In addition, wearing masks is suggested due to the potential formation of bioaerosols from contaminated wastewater. If greywater includes disinfectants used in prior cleaning, it may not need to be chlorinated or treated again. However, it is important that such water is disposed of in drains connected to a septic system, a sewer or in a soak-away pit. If greywater is disposed of in a soak-away pit, the pit should be fenced off within the health facility grounds to prevent tampering and to avoid possible exposure in the case of overflow.

This review paper considered various scenarios for both the developed world and the developing world and provided recommendations for managing risk related to COVID-19 and other viruses. While research into the virus that causes COVID-19 and its implications for wastewater process control is, at the time of this publication, in its infancy, there is a wide body of knowledge related to virus control through the wastewater process chain that could prove applicable for the control of COVID-19. This review paper summarised this knowledge to provide guidance to the water sector on the various process control options available and their efficiencies in virus control.

The authors thank Dr Charles Haas (Civil, Architectural and Environmental Engineering, Drexel University) and Dr Kyle Bibby (Department of Civil & Environmental Engineering & Earth Sciences, University of Notre Dame) for reviewing the paper before submission as part of IWA COVID-19 Task Force Scientific Committee.

All relevant data are included in the paper or its Supplementary Information.