Growing populations and increasingly scarce new water sources have spurred a variety of water management measures over the last few decades, including the processing and reuse of water for many purposes. In a small but growing number of communities, these measures include the use of highly treated municipal wastewater to augment the raw water supply. This trend is motivated by need, but made possible by advances in treatment technology.

However, important questions remain regarding the levels of treatment, monitoring, and testing needed to ensure the safety of such ''potable reuse." A 1982 National Research Council (NRC) report, Water Quality Criteria for Reuse, initially explored some of these questions. The significant advances, interest, and research in potable reuse since then, however, have spurred a reevaluation of these issues and this current report.

This study assesses the public health implications of using reclaimed water as a component of the potable water supply. It examines the different types of water reuse, discusses considerations for ensuring reliability and for evaluating the suitability of water sources augmented with treated wastewater, and seeks to identify future research needs regarding potable reuse safety testing and health effects

When considering potable reuse as an option for public water supplies, a critical distinction must be made between "direct" and "indirect" reuse. Direct potable reuse refers to the introduction of treated wastewater (after extensive processing beyond usual wastewater treatment) di-

rectly into a water distribution system without intervening storage. Direct use of reclaimed wastewater for human consumption, without the added protection provided by storage in the environment, is not currently a viable option for public water supplies. Instead, this report focuses on planned indirect potable reuse, which refers to the intentional augmentation of a community's raw water supply with treated municipal wastewater. The reclaimed water might be added to a water course, lake, water supply reservoir, or underground aquifer and then withdrawn downstream after mixing with the ambient water and undergoing modification by natural processes in the environment. The mix of reclaimed and ambient water is then subjected to conventional water treatment before entering the community's distribution system.

Planned indirect potable reuse cannot be considered in isolation from more general drinking water issues. Many communities currently use water sources of varying quality, including sources that receive significant upstream discharges of wastewater. In this sense, cities upstream of drinking water intakes are already providing water reclamation in their wastewater treatment facilities—for they treat the water, then release it into the raw water supply used by downstream communities. For example, more than two dozen major water utilities use water from rivers that receive wastewater discharges amounting to more than 50 percent of the stream flow during low flow conditions. Although most water systems using such raw water supplies meet current drinking water regulations, many of the concerns about planned, indirect potable reuse raised in this report also apply to these conventional water systems. The focus of this report, however, is planned indirect potable reuse of treated wastewater.

The several indirect potable reuse projects currently operating in the United States generally produce reclaimed water that meets or exceeds the quality of the raw waters those systems would use otherwise, as measured by current standards. In some instances the reclaimed water meets or exceeds federal drinking water standards established by the Safe Drinking Water Act. Current potable reuse projects and studies have demonstrated the capability to produce reclaimed water of excellent measurable quality and to ensure system reliability. In communities using reclaimed water where analytical testing, toxicological testing, and epidemiological studies have been conducted, significant health risks have not been identified. This suggests that reclaimed water can likely be used safely to supplement raw water supplies that are subject to further treatment in a drinking water treatment plant. However, these projects

raise some important questions: Can data from these projects safely be generalized to apply elsewhere? If not, what additional data are required? Do we know enough to establish criteria by which treated wastewater can be judged suitable for human consumption?

Our general conclusion is that planned, indirect potable reuse is a viable application of reclaimed water—but only when there is a careful, thorough, project-specific assessment that includes contaminant monitoring, health and safety testing, and system reliability evaluation. Potable reuse projects should include multiple, independent barriers that address a broad spectrum of microbiological and organic chemical contaminants. They should also conduct continuous toxicological monitoring if, as a result of the reclaimed water, the drinking water supply contains significant levels of organic contaminants of wastewater origin. Further, indirect potable reuse is an option of last resort. It should be adopted only if other measures—including other water sources, nonpotable reuse, and water conservation—have been evaluated and rejected as technically or economically infeasible.

It is important to recognize that although indirect potable reuse can be considered a viable option, many uncertainties are associated with assessing the potential health risks of drinking reclaimed water. These uncertainties are especially significant in toxicological and epidemiological studies. However, similar concerns also apply to the adequacy of these sciences for evaluating the safety of potable water from conventional sources, particularly the large number of sources already exposed to sewage contamination. These uncertainties are not an adequate reason for rejecting indirect potable reuse because the best available current information suggests that the risks from indirect potable reuse projects are comparable to or less than the risks associated with many conventional supplies.

That said, however, the intentional reuse of treated wastewater raises issues that must be addressed to ensure protection of public health. Drinking water standards cover only a limited number of contaminants. They are intended for water obtained from conventional, relatively uncontaminated sources of fresh water, not for reclaimed water, and therefore cannot be relied on as the sole standard of safety. The requirements for indirect potable reuse systems thus should exceed the requirements that apply to conventional drinking water treatment facilities.

The major recommendation of this report is that water agencies considering potable reuse fully evaluate the potential public health impacts from the microbial pathogens and chemical contaminants found or likely to be found in treated wastewater through special microbiological, chemical, toxicological, and epidemiological studies, monitoring programs, risk assessments, and system reliability assess-

ments. This report provides guidelines and suggestions regarding how such evaluations should be carried out. Thorough evaluation of the risks of a proposed potable reuse project, in addition to full consideration of other options for potable water supply augmentation, is essential for a sound decision about whether the project is viable.

Municipal wastewater contains chemical contaminants of three sorts: (1) inorganic chemicals and natural organic matter that are naturally present in the potable water supply; (2) chemicals created by industrial, commercial, and other human activities in the wastewater service area; and (3) chemicals that are added or generated during water and wastewater treatment and distribution processes. Any project designed to reclaim and reuse such water to augment drinking supplies must adequately account for all three categories of contaminants.

The organic chemicals in a wastewater present one of the most difficult challenges a public health engineer or scientist faces in considering potable reuse. The challenge arises from the large number of compounds that may be present, the inability to analyze for all of them, and the lack of toxicity information for many of the compounds. Efforts to account for the total mass of organic carbon in water are further frustrated by the fact that the bulk of this material is aquatic humus, which varies slightly in structure and composition from one molecule to the next and cannot be identified like conventional organic compounds. These challenges are not unique to potable reuse systems. In fact, the most protected water supplies are those for which the smallest fraction of the organic material can be identified. For potable reuse systems, however, anthropogenic organic compounds pose the greatest concern and should be the major focus of monitoring and control efforts.

The following recommendations suggest several important guidelines to account for chemical contaminants of potential concern in potable reuse systems:

• The research community should study in more detail the organic chemical composition of wastewater and how it is affected by treatment, dilution, soil interaction, and injection into aquifers. The composition of wastewater and the fate of the organic compounds it contains need to be better understood to increase the certainty that health risks of reclaimed water have been adequately controlled through treatment and storage in the environment.
• Proposed potable reuse projects should include documentation of all major chemical inputs from household, industrial, and agricul-

• tural sources. Reuse project managers can estimate household chemical inputs on a per capita basis from general information about domestic wastewaters. Project managers should undertake a major effort to quantify the inputs of industrial chemicals, paying special attention to chemicals of greatest health concern.
• Stringent industrial pretreatment and pollutant source control programs should be used to reduce the risk from the wide variety of synthetic organic chemicals (SOCs) that may be present in municipal wastewater and consequently in reclaimed water. Guidelines for developing lists of wastewater-derived SOCs that should be controlled through industrial pretreatment and source control should be prepared by the Environmental Protection Agency (EPA) and modified for local use. Potable reuse operations should include a program to monitor for these chemicals in the treated effluent, tracking those that occasionally occur at measurable concentration with greater frequency than those that do not.
• Although Safe Drinking Water Act regulations cannot alone ensure the safety of drinking water produced from treated wastewater, potable reuse projects must nonetheless bring contaminant concentrations within those regulations' guidelines. Potable reuse projects can manage regulated contaminants by a combination of secondary or tertiary wastewater treatment processes, dilution or removal in the receiving water, and removal in the drinking water treatment plant.
• The risks posed by unidentifiable or unknown contaminants in reuse systems should be managed by a combination of reducing concentrations of general contaminant classes, such as total organic carbon, and conducting toxicological studies of the water. Reducing the level of organic matter to the lowest practical concentration will reduce but not necessarily eliminate the need for toxicological studies and monitoring. The nature of the organic carbon in the water will influence what the appropriate total organic carbon limit should be. This judgment should be made by local regulators, integrating all the available information concerning a specific project.
• The research community should determine whether chlorination of wastewater creates harmful levels of unique disinfection by-products that might pose concerns in potable reuse systems. Whether reclaimed water forms significantly different by-products than natural waters upon disinfection is not yet clear.
• Finally, every reuse project should have a rigorous and regularly updated monitoring system to ensure the safety of the product water. This program should be updated periodically as inputs to the system change or as its results reveal areas of weakness. Pretreatment requirements, wastewater treatment processes, and/or monitoring requirements

• may need to be modified to protect public health from exposure to specific chemical contaminants.

Microbial contaminants in reclaimed water include bacteria, viruses, and protozoan parasites. Even though classic waterborne bacterial diseases such as dysentery, typhoid, and cholera have dramatically decreased in the United States, Campylobacter, nontyphoid Salmonella, and pathogenic Escherichia coli still cause a significant number of illnesses, and new emerging diseases pose potentially significant health risks.

Historically, coliforms, which serve as an adequate treatment indicator or "marker" for many bacterial pathogens of concern, have been used as general indicators of the levels of microbial contamination in drinking water. Today, however, most outbreaks of waterborne disease in the United States are caused by protozoan and viral pathogens in waters that meet coliform standards. Yet few drinking water systems, either conventional ones or those involving potable reuse, monitor for the full range of such pathogens, and little information exists regarding the efficacy of water and wastewater treatment processes in removing them. In addition, wastewater may contain a number of newly recognized or "emerging" waterborne enteric pathogens or potential pathogens.

To ensure the safety of drinking water, planners, regulators, and operators of potable reuse systems should take steps to further reduce the various existing and potential health risks posed by these microbial contaminants:

• Potable reuse systems should continue to employ strong chemical disinfection processes to inactivate microbial contaminants even if they also use physical treatment systems to remove these contaminants. Some new membrane water filtration systems can almost completely remove microbial pathogens of all kinds, but experience with them is not yet adequate to depend on them alone for protection against the serious risks posed by these pathogens. Therefore, strong chemical disinfectants, such as ozone or free chlorine, should also be used, even in systems that include membrane filters.
• Managers of current and future potable reuse facilities should assess and report the effectiveness of their treatment processes in removing microbial pathogens so that industry and regulators can develop operational guidelines and standards. Reuse project managers should provide data on number of barriers, microbial reduction performance, and reliability or variation.

• The potable reuse industry and the research community should establish the performance and reliability of individual barriers to microorganisms within treatment trains and should develop performance goals appropriate for planned potable reuse. Existing microbial standards for drinking water systems assume that the water source is natural surface or ground water. Treatment standards and goals more appropriate for potable reuse projects need to be developed.

Any effort to augment potable water supplies with reclaimed water must include an evaluation of the potential health risks. Such assessment is complicated by several factors, including uncertainties about the potential contaminants and contaminant combinations that may be found in reclaimed water and about the human health effects those contaminants may cause. Any such effort must evaluate health risks from both microbial and chemical contaminants.

The lack of information nationwide on the levels of viral and protozoan pathogens in all waters and the efficacy of both conventional water treatment and wastewater treatment for water reclamation in removing those pathogens poses challenges to estimating risks from microbial contamination in potable reuse systems. The Information Collection Rule promulgated in 1996 by the EPA should help provide the exposure data needed for more effective risk assessments, but additional steps are needed to improve methods for assessing risks posed by microbial pathogens in water reuse projects:

• Potable reuse project managers should consider using some of the newer analytical methods, such as biomolecular methods, as well as additional indicator microorganisms, such as Clostridium perfringens and the F-specific coliphage virus, to screen drinking water sources derived from treated wastewaters. These screening methods should cover some of the gaps in analysis left by the bacterial and cell culture methods currently used for detecting pathogens in water. Additional research should be sponsored to improve methods for detecting emerging pathogens in environmental samples.
• The EPA should include data on the concentrations of waterborne pathogens in source water in the new Drinking Water National Contaminant Occurrence Data Base and should develop better data on

• the reductions of pathogens accomplished by various levels of treatment. The lack of monitoring data for evaluating exposure remains the greatest single barrier to adequate risk assessment for microbial pathogens.
• The research community should conduct more research to document removal rates of relevant protozoa by natural environmental processes. Indirect potable reuse projects may rely on dilution in the environment, die-off in ambient waters, and removal by soil infiltration to reduce concentrations of microbial pathogens. While reductions of bacteria and virus concentrations in natural environments have been well documented, information on protozoa survival in ambient waters remains inadequate.
• Risk estimates should consider the effects pathogens may have on sensitive populations and the potential for secondary spread of infectious disease within a community. This precaution is necessary to prevent pathogens from infecting sensitive populations (the elderly or very young, or those with suppressed immune systems) in whom mortality may be high and from whom diseases might spread to others.
• The research community should carry out further studies to document the removal of pathogens of all types by membrane filtration systems. Certain membrane filtration systems show the potential for nearly complete removal of pathogens. More research is needed to demonstrate the suitability of these systems for potable reuse applications and to develop monitoring methods capable of continuously assessing system performance.

Because of the uncertainty of the organic chemical composition of reclaimed water, toxicological testing should be the primary component of chemical risk assessments of potable reuse systems. However, recent experience and research have shown that the conventional toxicological testing strategies developed in the food and drug industries, as well as the similar testing protocols recommended by the NRC in its 1982 report, are not adequate for evaluating risks from the complex chemical mixtures found in reclaimed wastewater. These testing protocols, which stress the use of concentrates of representative organic chemicals in both in vitro (cell culture) and in vivo (whole-animal) tests, have several critical limitations. These limitations include uncertainty as to whether the concentrates used for testing are truly representative of those in the wastewater; higher than expected occurrences of false negative results; long lag times between sample collection and the availability of results; difficulty in tracing results to particular constituents; and lack of suitability

for continuous monitoring. In addition, a truly thorough application of the NRC protocol, which would involve extensive testing of concentrates on live animals, is both expensive and time-consuming.

Given these complications, in waters where toxicological testing appears to be important for determining health risks, emphasis should be placed on live animal test systems that are capable of expressing a wide variety of toxicological effects. Chapter 5 presents a suggested approach using fish populations in unconcentrated treated wastewater.

Further, toxicological testing standards for reclaimed water should be supplemented by strict regulation of the processes for "manufacturing" the water. Regulators should review the processes for manufacturing the reclaimed water (that is, the treatment systems and environmental storage employed) on a plant-by-plant basis.

The few studies that have examined the health effects of drinking reclaimed water suggest that the current approaches to safety testing of reclaimed water, derived mainly from consumer product testing protocols, are inadequate for evaluating reclaimed water and should be replaced by a more appropriate method. Even a brief look at these studies makes clear the need for a new approach.

This report includes a review of six planned potable reuse projects that tried to analyze and compare the toxicological properties of reclaimed water to those of the communities' current drinking water supplies. In most of the six studies, testing was limited to assessing whether the water caused genetic mutations in bacterial systems. Some studies also used in vitro systems derived from mammalian cells, and two projects also used chronic studies in live mammal systems. Only two studies, carried out in Denver and Tampa, addressed a broad range of toxicological concerns. Those studies suggested that no adverse health effects should be anticipated from the use of Denver's or Tampa's reclaimed water as a source of potable drinking water. However, these studies, drawn from two discrete points in time and conducted only at a pilot plant level of effort, provide a very limited database from which to extrapolate to other locations and times.

Overall, the intent of toxicological testing can be grouped into (1) chemical screening and identification studies; (2) surveys to determine genetic mutation potentials; and (3) integrated toxicological testing. In theory, all three stages will be applied when needed. In practice, the

application of the third and crucial stage, that of integrated testing, has been both uneven and impractically expensive.

Screening studies merely identify chemicals that may be causing mutations; the mutagenic activity may or may not cause health problems. Gauging the actual health risk such chemicals pose requires test systems that can more directly measure a complete range of health hazards and can define dose-response relationships that allow an estimate of the risks associated with various levels of exposure. The most common method to accomplish this goal in conventional product safety testing is to test the chemicals or contaminants in question at doses approaching the maximally tolerated dose so as to establish the margin between environmental levels and those that produce adverse effects. For reclaimed water, however, the high cost and methodological problems inherent in this approach make it both unreliable and inefficient.

Accordingly, a new, alternative testing approach, such as one using fish in source water, should be developed to allow continuous toxicological testing of reclaimed water at reasonable cost. The system, an example of which is discussed in detail in Chapter 5, should employ a baseline screening test using a whole-animal rather than an in vitro approach and should be modified as results and research suggest improvements. The tests should use water samples at ambient concentrations in order to reduce the uncertainty and high costs of using concentrates. Any losses in sensitivity from not using high doses should be offset by the increased statistical power brought by using larger numbers of whole-animal test subjects, such as fish.

Research efforts should investigate the qualitative and quantitative relationships among responses in whole-animal test species, such as fish, and adverse health effects in humans. In vitro short-term testing using concentrations of chemicals should be confined to qualitative evaluations of particular toxicological effects found in the product water in order to identify potential sources of contaminants and to quickly guide remedial actions.

For any toxicological test used for reclaimed water, a clear decision path should be followed. Testing should be conducted on live animals for a significant period of their life span. If an effect is observed, risk should be estimated using state-of-the-art knowledge about the relative sensitivity of the animal and human systems, and, if warranted, further defined by more research. This decision path is quite workable if the underlying basis of the biological response in question is understood (for example, endocrine disruption). For some health outcomes, such as carcinogenesis, the mechanism is less well understood, and an observed effect may have to be accepted as implying an impact on human health.

The need for toxicological testing of water is inversely related to

how well the water's chemical composition has been characterized. If a water contains very few or very low concentrations of chemicals or chemical groups of concern, the need for toxicological characterization of the water may be substantially reduced. Conversely, if a large fraction or high concentrations of potentially hazardous and toxicologically uncharacterized organic chemicals are present, toxicological testing will provide an additional assurance of safety.

Numerous epidemiologic studies have examined the relationship between contaminants in drinking water and health problems. However, only three such studies apply to potable reuse of reclaimed water, and only one set of epidemiological studies (Los Angeles County) has been conducted in a setting that can be generalized to apply to other communities. These studies have used an ecologic approach, which is appropriate as an initial step when health risks are unknown or poorly documented, but negative results from such studies do not prove the safety of the water in question. These studies can only be considered as preliminary examinations of the risks of exposure to reclaimed water. Epidemiological data that can be confidently applied to the potable use of reclaimed water are lacking. Filling that knowledge gap would aid planning and help ensure the safety of such projects.

The 1982 NRC report on potable reuse concluded that ''unless epidemiological methodology is improved, it is doubtful whether it can be used to evaluate the potential carcinogenic risk of drinking reused water" but recommended monitoring for acute waterborne diseases. Since that report, at least 17 large epidemiological studies (using several designs) have examined the association between chlorinated surface water and cancer, and two large cohort studies have examined the risk of endemic waterborne disease due to infectious agents. These studies have greatly increased our experience with exposure assessment, and outcome measurements in this area could be used to help design future epidemiologic studies of reclaimed water.

Therefore, epidemiologic studies should be conducted at the national level using alternative study designs and more sophisticated methods of exposure assessment and outcome measurement to evaluate the potential health risks associated with reclaimed water. Ecologic studies should be conducted for water reuse systems using ground water and surface water in areas with low population mobility. Case-control studies or retrospective cohort studies should be undertaken to provide information on health outcomes and exposure on an individual level while controlling for other important risk factors.

To ensure that any temporary weaknesses in the treatment process or water quality are promptly detected and corrected, potable reuse systems should provide multiple barriers to contamination and should monitor both water quality and potential health effects of substandard water, according to the following guidelines:

• Potable water reuse systems should employ multiple, independent barriers to contaminants, and the barriers should be evaluated both individually and together for their effectiveness in removing each contaminant of concern. Further, the cumulative capability of all barriers to accomplish removal should be evaluated, and this evaluation should consider the levels of the contaminant in the source water.
• Barriers for microbiological contaminants should be more robust than those for forms of contamination posing less acute dangers. The number of barriers must be sufficient to protect the public from exposure to microbial pathogens in case one of the barriers fails.
• Because performance of wastewater treatment processes may vary, such systems should employ quantitative reliability assessments to gauge the probability of contaminant breakthrough among individual unit processes. "Sentinel parameters," indicators of treatment process malfunctions that are readily measurable on a rapid (even instantaneous) basis and that correlate well with high contaminant breakthrough, should be used to monitor critical processes that must be kept under tight control.
• Utilities using surface waters or aquifers as environmental buffers should take care to prevent "short-circuiting," a process by which treated wastewater influent either fails to fully mix with the ambient water or moves through the system to the drinking water intake faster than expected. In addition, the buffer's expected retention time should be long enough—probably 6 to 12 months, as outlined in recently proposed California regulations—to give the buffer time to provide additional contaminant removal. Such a lag time also allows public health authorities to take action in the event that unanticipated problems arise in the water reclamation system.
• Potable reuse operations should have alternative means for disposing of the reclaimed water in the event that it does not meet required standards. Such alternative disposal routes protect the environmental buffer from contamination.
• Every community using reclaimed waters as drinking water should implement well-coordinated public health surveillance systems

• to document and possibly provide early warning of any adverse health events associated with the ingestion of reclaimed water. Any such surveillance system must be jointly planned and operated by the health, water, and wastewater departments and should identify key individuals in each agency to coordinate planning and rehearse emergency procedures. Further, appropriate interested consumer groups should be involved with and informed about the public health surveillance plan and its purpose.
• Finally, operators of water reclamation facilities should receive training regarding the principles of operation of advanced treatment processes, the pathogenic organisms likely to be found in wastewaters, and the relative effectiveness of the various treatment processes in reducing contaminant concentrations. Operators of such facilities need training beyond that typically provided to operators of conventional water and wastewater treatment systems.