Water safety planning application for aesthetic quality of drinking water

The water safety plan (WSP) was introduced by the World Health Organization (WHO) in the 3rd edition of the Guidelines for Drinking-water Quality (WHO 2004) based on comprehensive risk assessment and management approaches to ensure the safety of drinking water (Hrudey et al. 2024). The WSPs are a tool that utilities can implement as a best management practice to complement government regulations, along with internal operational guidelines. The water safety planning framework consists of (1) system assessment to understand the entire water supply system from the source to the customer to identify potential hazards and assess risk, (2) control measures to effectively reduce the risks to acceptable levels, (3) monitoring to check the performance of the control measures, (4) management and communication, and (5) documentation and review (Davison et al. 2005).

The usefulness of WSPs lies in the fact that they are flexible and can be tailored to each individual utility's needs (Burlingame & Bartrand 2023). WSPs incorporate centralized controls (e.g., chemical dosing points within the water treatment process to reduce source water odors) and decentralized controls (e.g., flushing water mains throughout the distribution system to alleviate customer complaints). These controls can be used to mitigate single or multiple issues and risk scores can be assigned to issues that can then be prioritized for effective management of events (Ferrero et al. 2019; Tsitsifli & Tsoukalas 2021; WHO 2023). In the past decades, WSPs have been successfully developed and implemented to control chemical quality (Gunnarsdóttir et al. 2012), to assess and improve microbial quality in rural water utilities (Pundir et al. 2021; Souter et al. 2024), and to evaluate the possibility of improving consumer relations and challenging water quality issues (Rodriguez-Alvarez et al. 2022).

Utilities can use WSPs for health-related contaminants, as well as aesthetic issues, including customer complaints. Burlingame et al. (2024) described how water safety planning can be applied to such guidelines as the United States Environmental Protection Agency secondary maximum contaminant levels (US EPA SMCLs) which include aesthetic and cosmetic parameters, such as aluminum, chloride, odor, and taste in a more integrated manner. The WHO includes aesthetic quality as important for a safe water supply (WHO 2023) and the hazard category ‘Acceptability’ covers aspects that affect customer acceptance of the water (e.g., taste, odor, color, and appearance). These factors are crucial because if consumers find the water unacceptable, it could lead to a lack of confidence and might push them to seek alternative water sources that may not be safe (Turgeon et al. 2004; Doria et al. 2009). Consequently, aesthetic aspects are considered in water safety planning due to their potential indirect health implications. However, to date, there is a lack of case studies describing actual WSPs created and implemented by the water utilities to monitor and improve aesthetic quality and customer confidence (Ferrero et al. 2019).

This article outlines how the City of Wichita Falls Public Water System (CWF PWS) developed a WSP for aesthetic issues ranging from organic taste-and-odor (T&O) compounds to inorganics like chloride and iron. The WSP complements adherence to US EPA SMCLs by providing additional guidance on how to manage and prevent events caused by aesthetic issues. CWF PWS was already addressing aesthetic issues but in silos – not in an integrated way. A water safety planning approach brings together the goals, controls, monitoring, and management of the various aesthetic issues that could affect the water utility, whether currently being addressed or not, into one risk management plan – a WSP. The WSP becomes the standard procedure for controlling aesthetic water quality for the CWF PWS. This paper presents the water safety planning process that was used and shows various elements of the WSP that were developed.

The development of a WSP (i.e., water safety planning) follows certain steps in an organized manner, from identifying the hazards (in this case, the aesthetic concerns or SMCLs) and the means for managing or controlling them to the validation that controls actually work (WHO 2023). This can be summarized in the following steps as used for this WSP for the CWF PWS:

• 1. Identify and list the hazards (contaminants) to be included in the WSP, covering both actual and potential hazards.

• 2. Inventory the conditions or events that result in hazardous events, such as customer complaints.

• 3. Assign ratings of the likelihood of occurrence and consequence of occurrence for each hazardous event in #2 above.

• 4. Calculate risk scores based on the ratings for #3 above, for each hazardous event.

• 5. Prioritize the risk scores to identify where improvement plans can reduce either the likelihood or consequence ratings and thus the risk scores.

• 6. For each hazardous event, identify the critical controls that are in place to prevent an event or mitigate an event. Also identify control limits, monitoring plans, and response plans.

• 7. Design validation monitoring to confirm that the controls are working as intended (such as actual measurement of SMCL parameters in the treated drinking water).

• 8. Finalize the WSP and provide periodic reviews and updates.

The CWF PWS began building its WSP by using contaminants listed in Table 1 to create a risk matrix, drawing from decades of customer complaints that have been collected and stored. These strengthened the WSP, building it based on real water quality issues over time. However, no PWS can predict every possible scenario, so WSPs must be reviewed and revised accordingly as new data are collected. Aesthetic hazards were identified for each process step from source to distribution, along with the type of event the hazard could cause. The hazardous events that result in a contaminant causing a water quality problem were rated based on their likelihood of occurrence versus the severity of their consequence if they did occur. The inventory of hazardous events was combined with risk scores and ranked to determine where any immediate focus should be given. Then controls were identified for each hazardous event, as well as how to monitor each control and verify the controls' effectiveness at mitigating/reducing the hazard by the use of internally established action levels. The following sections describe how each part of the WSP was developed using examples from the CWF PWS.

The US EPA's list of SMCLs, as well as several additional parameters, was compared with CWF PWS's experience and history with aesthetic issues and challenges to identify the existing hazards to be addressed by a water safety planning approach (Table 1). Each parameter is listed with its associated SMCL, including updates and customer complaint issues that are routinely observed. Many of these parameters are in need of update by the US EPA, as detailed in Dietrich & Burlingame (2015). Parameters that are known aesthetic issues for CWF PWS are labeled ‘Actual’. Parameters that have not caused issues but could be a concern are labeled ‘Potential’ and parameters that are unlikely to become issues are labeled ‘N/A’.

The parameters in Table 1 that are listed as centrally controlled are the parameters that are managed by source water controls or water treatment processes. Parameters that are controlled by multiple, distributed means include parameters controlled centrally but also within the distribution system, such as by valve operations, flushing, storage tank management, and corrosion control strategies.

Water safety planning uses terminology and procedures as used in risk management programs (Alberta Environment and Parks 2013; WHO 2023). For example, a hazardous event is an event that could happen due to a hazard being present, such as a chemical contaminant in the water source due to an industrial spill. The hazard is a quantifiable parameter that poses a risk such as the concentration of a chemical contaminant. A risk is determined by the likelihood of exposure to the hazard and the consequence of exposure, such as skin irritation or sickness. The risk factors that lead to a hazardous event could be extrinsic (imposed by conditions outside of the utility's control such as a flood) or intrinsic (due to conditions or actions within the utility's control such as a transmission main break or a loss of chlorine residual in a storage tank due to failure in following proper tank operations). The following items were used to quantify and prioritize the risks:

• Likelihood: A rating of the estimated or quantified frequency of occurrence of a hazardous event. The likelihood scoring is shown in Table 2 (Alberta Environment and Parks 2013).

• Consequence: A rating of the impact or quantifiable result of a hazardous event.

• Uncertainty: The confidence that the above two factors are accurately reflected.

Uncertainty expresses the level of confidence in the likelihood or consequence rating. One way to change a risk score is to gather better data and information by which to show that a likelihood or consequence rating should be changed. For example, the likelihood of a geosmin odor event happening for a reservoir supply may be based on cyanobacterial bloom occurrence data. Therefore, the likelihood rating has a good degree of uncertainty associated with it since not all cyanobacterial blooms produce geosmin. The collection of actual geosmin concentrations during blooms will better determine the likelihood that any particular bloom produces geosmin, and therefore will better determine the likelihood of a hazardous event due to geosmin which could adjust the risk score for this aesthetic hazard.

The consequence scoring is shown in Table 3 (Alberta Environment and Parks 2013). These ratings have been adjusted because Texas has no regulatory requirements for SMCLs. For systems in locations that do, they would need to include regulatory and possible public health-related requirements.

The risk severity score is the product of the consequence rating multiplied by the likelihood rating. This score is used to prioritize the risks (Alberta Environment and Parks 2013). The risk score matrix is shown in Table 4.

Table 4

Risk matrix with risk severity scores as used in CWF PWS's WSP

Note. The scores reflect priorities that should be given to the hazardous events and the colors represent the actions that should be taken: Green represents review periodically; Yellow represents introduce actions to manage the risk or monitor for changes; and Red represents act to reduce the risk. The priority placed on a hazardous condition would follow the color code with red representing high priority for attention. This list of priorities and scores should be evaluated annually and updated.

View Large

Based on the identification of aesthetic hazards or parameters of concern, the next step was for the CWF PWS to identify the conditions under which the aesthetic hazards would create a potential event, such as customer complaints. Such events can be actual ones that have been experienced or near misses, or ones that could theoretically occur but have not yet been experienced. Table 5 provides an inventory of hazards and hazardous events pertinent to CWF PWS.

Table 6 expands upon the inventory of hazardous events to provide the ratings for the likelihood and consequence of each event, which are then used to derive the risk scores. These are listed in order from high to low in risk scores. The ratings are based on both expert opinion from tenured staff and consulting scientists and actual historical experience, and a confidence rating is shown for how confident CWF PWS is with the ratings that were given. Confidence levels can be based on professional opinion, historical data within the utility, or data from utilities within the watershed or region. For example, a high level of confidence in the ratings for likelihood and consequence shows that CWF PWS has experienced these hazardous events and has information associated with them. A low level of confidence would suggest that research or evaluations should be done to better determine the ratings in order to estimate a reliable risk score.

Table 6

Development of risk scores

¹Likelihood of occurrence rating: (1) rare, (2) unlikely, (4) possible, (8) likely, and (16) almost certain.

²Level of confidence: high, moderate, and low. States the level of confidence with which the rating can be given. Consider: Confidence level might be based on professional opinion, historical data within the utility, or data from utilities within the watershed or region.

³Consequence of occurrence rating: (1) Insignificant, (2) Minor, (4) Moderate, (8) Major, and (16) Severe.

⁴Explanation of consequence: Based on SMCL explanations and customer complaints.

⁵The risk score is the product of the likelihood and consequence ratings.

View Large

Table 7 provides possible improvement plan actions to reduce the risk scores for six of the highest hazardous events. Such action could reduce the likelihood that an event would occur, or the consequence when an event does occur, or both. The reduction in one or both ratings would then lower the risk score.

Table 7

Improvement plans to consider for reducing higher risk scores focused on aesthetic quality risk management

View Large

The next step was to list all the controls that are in place within the water utility for managing aesthetic quality (Table 8). Controls can be water treatments, standard operating procedures, flushing of mains, cross-connection control, valve inspection, and water main disinfection practices as laid out in AWWA C651 Standard (AWWA 2023). Table 8 shows examples of contaminant controls, and how each control is monitored, or itself controlled. Examples include dosing a reservoir with copper sulfate (the control) to reduce blooms of alga/cyanobacteria (the hazardous event), which would reduce the production of geosmin or 2-methylisoborneol (MIB) (the hazards). How the effectiveness of the control is monitored is accomplished by performing organism counts before and after dosing copper sulfate. Finally, verification that this control works would be demonstrated by low to non-detectable levels of geosmin or MIB in the water entering the treatment plants.

Verification monitoring can be seen in Table 9, which shows how the WSP keeps the hazard below action levels. For example, algal/cyanobacteria-related T&O verification monitoring would include quantification of T&O compounds (geosmin/MIB/haloanisoles) to ensure they are kept below action levels. If action levels are exceeded, controls and their monitoring should be re-evaluated to ensure the best methods are being utilized.

Although this article addresses maintaining aesthetic water quality within the utility, when fully implemented, WSPs on aesthetic quality should include safeguarding aesthetic quality all the way to the ultimate customer. Meeting this goal will be challenging because many aesthetic water quality issues arise in premise plumbing systems that are outside the control of water utilities. As such, safeguarding aesthetic water quality all the way to the customer and ultimate user involves cooperation between the water supplier and the building owners/operator since all have a stake in providing high-quality water to the consumer. This is an emerging area that requires the cooperation of organizations and professionals who often operate independently. The International Association of Plumbing and Mechanical Officials is working in conjunction with utilities and the drinking water industry to produce guidance on reopening building water systems after extended stagnation (Rhoads et al. 2020) and guidance on water management during construction with an appendix on water quality for plumbers and plumbing engineers (IAPMO 2024). Collaborations such as these are the key to meaningfully addressing water quality aspects for safe and aesthetically pleasing water quality at the customer's tap and to comprehensively implementing water safety planning for aesthetics.

The CWF PWS has completed the key documentation for water safety planning covering aesthetic water quality and US EPA SMCLs, namely known T&O compounds, taste-related metals and ions, turbidity, total dissolved solids (TDS), color, pH, and ammonia. The resultant integrates the various source water, water treatment, and distribution system preventive and mitigative controls, providing context and perspective on why the controls are necessary. Water quality staff also have clear explanations as to why analyses and reporting are done and why the triggers for action are in place. As water quality events that affect aesthetic parameters occur, whether successfully managed or not, the WSP will be reviewed for updates. As improvement plans are acted on, the WSP will be a living document that is reviewed and updated regularly.

While not a regulatory requirement in the US, proactive utilities use water safety planning to strengthen monitoring programs already in place and build rapport with customers by ensuring the production and delivery of safe and aesthetically pleasing water. A WSP integrates information and the management of controls from across a water utility, providing knowledge transfer and information exchange, which helps to build cooperation and unity within the water utility. For WSPs to be effective tools, PWSs must continue to develop and revise them as infrastructures and treatment changes are made to maintain them. Response to customer complaints and ongoing research with the PWS must also be included as the WSP is revised over time.

H. A.: Conceptualization, methodology design, project administration, writing – review and editing; G. A. B.: Conceptualization, methodology design, writing – review and editing; and M. S., A. M. D., T. B., and K. I.: Writing – review and editing.

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

The authors declare there is no conflict.