Human health risk associated with direct potable reuse – evaluation through quantitative relative risk assessment case studies

Abstract

Quantitative relative risk assessment (QRRA) evaluations were conducted for two hypothetical direct potable reuse (DPR) case studies. The goal of a risk assessment is to estimate the severity and likelihood of harm to human health or the environment occurring from exposure to a risk agent. Each case study compares a No Project Alternative (raw water that has undergone drinking water treatment) with a potential DPR Alternative (treated wastewater that has undergone advanced water treatment and drinking water treatment). Neither DPR scenario accounts for blending with raw drinking water prior to drinking water treatment, blending after drinking water treatment in the potable distribution system, or blending via directly distributing the purified reclaimed water into a drinking water distribution system. The QRRA focuses on chemicals that are currently regulated and chemicals that are not yet regulated but are of broad interest, such as pharmaceuticals and personal care products. Pathogen risk evaluation was not conducted but could follow a similar approach. The results illustrate how QRRA can inform water supply decisions that are made with respect to industrial pretreatment/source control, wastewater treatment, drinking water treatment, and advanced treatment for DPR. In these case studies, the DPR alternatives are projected to provide protection from regulated constituents and constituents of emerging concern that are comparable to or better than the No Project Alternatives. The results also indicate that future QRRA studies would benefit from specific information that could be obtained through targeted research.

---

Water impact

To assist in understanding and communicating the potential associated with direct potable reuse (DPR), one important tool is the quantitative relative risk assessment (QRRA) approach. Exposure to chemical constituents is very difficult to precisely estimate. A QRRA eliminates this issue by using a hypothetical, standardized exposure and provides a more health protective approach than traditional risk assessments because it uses observed concentrations for developing exposure that are greater than those that would actually occur. This paper provides examples of how QRRAs can evaluate the relative risks of two different DPR schemes for regulated constituents and constituents of emerging concern. The results can inform decisions regarding industrial pretreatment/source control and wastewater treatment, drinking water treatment, and advanced treatment schemes for DPR.

Introduction

Reclaimed water is domestic or municipal wastewater, which has been treated to a suitable quality for a specific use and takes the place of potable and/or raw water that would otherwise be needed from another source. Water reuse is an important component of the future water supply portfolio for many locations throughout the world to support population growth and continued economic development and to address increasingly prevalent drought conditions.

Initial reclaimed water uses were primarily for irrigation of agriculture. Today, reclaimed water is used for a wide range of beneficial purposes including power plant cooling water, commercial and municipal irrigation, river and stream flow enhancement, natural gas exploration activities, and augmentation of drinking water supplies (potable reuse). For example, water reuse will provide approximately 1.53 million acre-feet per year of water supply in the State of Texas (USA) by the year 2060 and will meet approximately 18% of the projected water needs for the State.¹ There is significant potential for additional development of water reuse as a water management strategy. Much of this potential may be realized through the development of potable reuse projects, particularly as progress is made in communicating the advantages, benefits and safety of potable reuse to the public.

Indirect potable reuse (IPR) is the use of reclaimed water for potable purposes by discharging to a water supply source, such as a surface water or groundwater. The mixed reclaimed and natural waters then receive additional treatment before entering the drinking water distribution system. On the other hand, direct potable reuse (DPR) is the introduction of reclaimed water either directly into the potable water distribution system or into the raw water supply entering a drinking water treatment plant. Whereas numerous IPR projects have been successfully implemented, DPR implementation is much less common. However, for various reasons including severe droughts, increased population, and increased confidence in water treatment technologies, many municipalities are now considering DPR.

Herein, we conduct demonstration quantitative relative risk assessment (QRRA) evaluations for two hypothetical DPR projects (case studies) to illustrate how these types of assessments can be conducted and how the information can be applied to provide a human health context to monitoring data for chemicals that are currently regulated or are unregulated.

Methods

Case studies

Two separate case studies were selected to illustrate real situations that could occur in terms of wastewater treatment, advanced water treatment and water treatment to produce reclaimed water for DPR in the State of Texas, USA. For water treatment, we deliberately included additional treatment processes (ozone and biologically activated carbon (BAC)) at the water treatment plant (WTP) for one of the case studies to represent a treatment scheme that addresses taste and odor, iron and manganese, and/or the need to reduce disinfection by-product (DBP) formation, which are common issues. This scheme was designated as “enhanced” water treatment to distinguish it from conventional water treatment. For DPR advanced water treatment, we deliberately included an advanced treatment facility (AWTF) with reverse osmosis (RO) for one of the case studies and an AWTF without RO for the other case study. It is of significant interest to identify and evaluate treatment schemes that do not include RO due to the difficulty and costs associated with disposal of brine concentrate generated by RO, particularly in inland areas. Each case study is comprised of a No Project Alternative and a DPR Alternative.

Case study 1 (non-RO AWTF/enhanced WTP). The No Project Alternative is defined as raw source water which is treated by an enhanced WTP, consisting of ozone, BAC, flocculation-sedimentation, media filtration, and chlorination with free chlorine.

The DPR Project Alternative is defined as secondary/tertiary wastewater treatment plant (WWTP) effluent which is the feed water to an AWTF that consists of microfiltration (MF) or ultrafiltration (UF), ozone, BAC, and chlorination (Fig. 1). This product water is then treated by the enhanced WTP (described above) consisting of ozone, BAC, flocculation-sedimentation, media filtration, and chlorination.

Fig. 1 Case study 1 water treatment unit process overview.

Case study 2 (membrane AWTF/conventional WTP). The No Project Alternative is defined as raw source water which is treated by a conventional WTP consisting of flocculation-sedimentation, media filtration, and chlorination with free chlorine.

The DPR Project Alternative is defined as secondary/tertiary WWTP effluent which is the feed water to an AWTF that consists of MF or UF, RO, and advanced oxidation (ultraviolet (UV) irradiation and hydrogen peroxide) (Fig. 2). This product water is then treated by a WTP consisting of flocculation-sedimentation, media filtration, and chlorination.

Fig. 2 Case study 2 water treatment unit process overview.

Risk assessment

The goal of a risk assessment is to estimate the severity and likelihood of harm to human health or the environment occurring from exposure to a risk agent.² The QRRA described herein uses accepted risk assessment methodologies to provide a health protective, relative comparison of health risks for the two case studies from consumption of existing drinking water (No Project Alternative) for a representative time period compared to consumption of drinking water under a hypothetical DPR alternative. In this evaluation the absolute risk from ingestion of water “at the tap” was not assessed, but rather a relative comparison was made based on an assumed quantity of water consumed and the estimated water quality associated with the scenarios under consideration. The methodology employed in this assessment is based on an approach that has been used successfully for previous IPR assessments.^3–7 Simplified, hypothetical exposure scenarios are used to streamline the analysis and to reduce uncertainty associated with modelling fate and transport, mixing, and potential confounders such as bottled water use and population transience.

To simulate the case studies for the QRRA, it was first necessary to collect water quality data. Monthly samples were collected from two raw drinking waters in the State of Texas and secondary/tertiary effluent from two WWTPs for the period December 2013 through May 2014. Samples were analyzed for regulated constituents (Clean Water Act priority pollutants, constituents with federal drinking water maximum contaminant levels (MCLs), and constituents with other regulatory recommendations or guidelines), and unregulated constituents (for example, prescription drugs, over-the-counter drugs, and ingredients in personal care products). For the QRRA, “detected compounds” are those that were found in at least one sample at or above the compound-specific Minimum Reporting Level (MRL). The MRL represents an estimate of the lowest concentration of a compound that can be quantitatively measured. For each constituent, if the concentration in at least one sample was at or above the MRL it was deemed to be “detected.” If the other sample concentrations were reported to be below the MRL, for calculation of the average concentration for the QRRA, the constituent was assumed to be present at the MRL. This simple approach is likely to overestimate the concentration of any observation reported below the MRL compared to a more rigorous statistical treatment.⁸ However, this straightforward approach is parsimonious, builds off of prior successful QRRA applications, and provides a reasonable and defensible basis from which risk managers can make practical decisions.^3–7

For the QRRA, four fundamental steps were carried out during the course of this assessment. Those steps follow the general guidance provided by the U.S. Environmental Protection Agency (U.S. EPA) for chemical risk assessment^9,10 and are as follows: (1) evaluate data and identify detected chemicals that can be used to represent the potential carcinogenic and noncarcinogenic hazard posed by the test waters; (2) conduct a toxicity assessment of the potential carcinogenicity and noncarcinogenic effects of the chemicals of concern; (3) conduct an exposure assessment, which for this study involves calculating potential doses based on estimated concentrations and an assumed standard intake of water; and 4) characterize the potential health risks associated with the test waters.

For the data evaluation, detected constituents were divided into two categories. Constituents of Potential Concern (CPCs) are detected compounds that are regulated or currently under consideration for regulation and had associated health-based criteria that could be used to quantify the estimated relative potential health risk. Constituents of Emerging Concern (CECs) are detected compounds that are unregulated with published toxicity information to evaluate their health significance. The Eurofins Eaton Analytical method was used for analysis because it is capable of reliably testing for more than 90 CECs in a single method at low levels (ng L⁻¹).

For the No Project alternatives, estimated WTP unit process removal efficiencies were applied to the CPCs and CECs in the raw waters to estimate resultant drinking water concentrations. For the DPR alternatives, estimated AWTF unit process and WTP unit process removal efficiencies were applied to the CPCs and CECs in the secondary wastewaters to estimate resultant drinking water concentrations.¹¹ This assessment did not account for formation of DBPs, such as trihalomethanes or N-nitrosodimethylamine through the various water treatment processes.

The purpose of the toxicity assessment is to weigh available evidence regarding the potential for a particular chemical to cause adverse health effects in exposed individuals and to provide, where possible, an estimate of the relationship between the extent of exposure to a chemical and the increased likelihood and/or severity of adverse health effects.⁴ Detected chemical constituents were evaluated for their carcinogenic and noncarcinogenic potential based on a hazard identification and a dose–response evaluation. From this evaluation, toxicity values for CPCs (characterized in terms of reference doses [RfDs] for noncarcinogenic effects and carcinogenic oral slope factors [SFs] for carcinogenic effects) were identified to estimate the potential for adverse effects as a function of human exposure to a given constituent. In addition, risk based action levels (RBALs) were collated from the literature for CECs. The health-based criteria were used as input to the QRRA to quantify the estimates of relative potential risk from the No Project and DPR alternatives.

The objective of an exposure assessment is to estimate the type and magnitude of exposure to the constituents of concern. For this relative risk assessment study, which strictly focused on exposure through drinking water, a hypothetical exposure was calculated based on observed average concentrations of the CPCs and CECs in raw water or wastewater and their predicted concentrations in drinking water taking into consideration removal efficiencies of treatment processes through water treatment and combined advanced treatment and water treatment, and a standard ingestion volume to compare the various scenarios under investigation. For the purposes of this exposure assessment, the daily volume of water ingested is assumed to be a constant 1.2 L.^12,13

Risk assessment methods used in this study compare water sources where the hypothetical exposures used in the assessment are not expected to actually occur, but are used to “normalize” exposure between the existing drinking water and the hypothetical alternative. Although stochastic methods may be used to estimate situational exposure effectively, the hypothetical alternative scenarios present a high degree of uncertainty since there is no method of determining long-term, realistic consumption. There are also many confounding factors that impact exposure and public health that cannot be quantified, such as bottled water usage, smoking, diet, exercise regimen, etc. Other routes of exposure, such as dermal absorption and inhalation may also be valid, but are not the focus of this investigation.

For CPCs, QRRAs were conducted for noncarcinogenic and carcinogenic risk. For noncarcinogenic risk, the QRRA evaluated the cumulative hazard index ([HI] – the sum of hazards for each CPC) for each case study alternative, where a result greater than or equal to 1 is the threshold for potential adverse health effects. Carcinogenic risks are estimated as the incremental probability of an individual developing cancer over a lifetime as a result of exposure to a potential carcinogen based on its cancer SF. Cancer SFs are based on experimental animal data and limited epidemiological studies, when available. The model generally used by the U.S. EPA to calculate numerical cancer potency values over predicts risk in comparison to average population risk.

For CECs, the 2012 National Research Council (NRC) risk exemplar approach was utilized to assess risk (NRC, 2012). The risk exemplar approach relies on estimates of the amount of a substance in drinking water that can be ingested daily over a lifetime without appreciable risk. These “safe” levels are called Drinking Water Equivalent Levels (DWELs), Predicted No Effect Concentrations (PNECs), or Drinking Water Guidelines (DWGs). For each of the detected CECs, potential lifetime health risks were assessed by calculating margins of safety (MOSs).¹⁴ A MOS is the ratio of a risk-based action level (RBAL) based on a DWEL, PNEC, DWG or other available heath benchmark, divided by the estimated concentration of the constituent in water. In using the risk exemplar approach, the NRC opined that an MOS lower than 1 for a specific CEC posed a potential concern from that CEC. This interpretation was made in light of the multiple safety factors, such as the application of uncertainty factors, included in the derivation of the RBALs.

Results

Water quality monitoring and data consolidation

An overview of the water quality monitoring program results is provided in Table 1. As shown in Table 1, two source waters and two wastewater effluents were monitored. In total, over 7500 observations were included. Raw waters had fewer observations at or above MRLs than effluent waters, and fewer constituents with observations at or above MRLs than effluent waters. Taken together, there were a total of 119 unique analytes that were detected at least once. The exposure to any particular constituent is computed based on the concentrations of CPCs and CECs reported above reduced by the expected removals across advanced water treatment and/or drinking water treatment, as appropriate for each scenario evaluated. The estimated reductions of each constituent across each treatment unit process are provided in TWDB¹¹ along with associated references on how they were derived. Raw water from both case studies contained low concentrations of nitrate (significantly lower than the MCL). The presence of nitrate in the raw waters is attributed to the influence of upstream wastewater discharges and/or land application of fertilizers in the watershed.

Table 1 Overview of water quality monitoring program results

Water	Number of analytes	Total number of observations	Number of observations > MRL	Number of constituents with observations > MRL
Case study 1 raw source water	347	1868	212	55
Case study 1 WWTP effluent	367	1898	351	97
Case study 2 raw source water	308	1861	55	17
Case study 2 WWTP effluent	373	1901	321	90

---

QRRA results for CPCs

The risk assessment results for noncarcinogenic risks are shown in Table 2. For each No Project alternative and DPR alternative, the cumulative hazard index was less than 1. However, for the case study 1 DPR Alternative (the non-RO AWTF), the cumulative hazard index was close to 1, with most of the contributions coming from nitrate and fluoride. None of the CPCs were detected at levels that exceeded MCLs for any of the alternatives. Fluoride can occur naturally in drinking water or can be added to public drinking water supplies as a public health measure to reduce cavities. For case study 1, fluoride is added to water for the prevention of tooth decay. The MCL for fluoride is based on prevention of dental fluorosis. The higher cumulative hazard index for the case study 1 DPR Alternative in comparison to the case study 2 DPR Alternative illustrates the role of RO membranes in removing nitrogen and its risk contribution. This result also suggests that better removal of nitrogen at the WWTP or an added nitrogen barrier as part of the AWTF would reduce risk for both DPR alternatives.

Table 2 Summary of noncarcinogenic risk assessment

	Case study 1		Case study 2
	No project alternative	DPR alternative	No project alternative	DPR alternative
Hazard index (HI)	0.13	0.89	0.20	0.05
# CPCs present with RfD	16	27	9	22
Any single constituent with HI > 1	No	No	No	No
Major contributors to overall HI	Fluoride (58%) Nitrate (34%) Aluminum (5%)	Nitrate (73%) Fluoride (22%) Monochloroacetic acid (2%) Strontium (2%)	Fluoride (61%) Nitrate (31%) Arsenic (6%)	Fluoride (37%) Nitrite (28%) Manganese (11%) Molybdenum (7%) Cyanide (7%)
Any constituent > MCLs	No	No	No	No

---

The results of the carcinogenic QRRA are shown in Table 3. Similar to the noncarcinogenic risk assessment, concentrations of CPCs were below MCLs and health advisory levels. The carcinogenic risks for the case study 1 No Project Alternative, case study 1 DPR Alternative, and case study 2 No Project Alternative were in approximately the same range. The carcinogenic risk for the case study 2 DPR Alternative (the membrane AWTF), however, is about an order of magnitude lower. For each alternative, arsenic and DBPs are the major contributors to risk. For the case study 2 DPR Alternative, RO and UV/AOP play an important role in reducing risk through removal of these CPCs. These results highlight the need to consider prevention of DBP formation or removal of DBPs as part of a DPR treatment scheme (beginning with the WWTP through AWTF and WTP).

Table 3 Summary of carcinogenic risk assessment

	Case study 1		Case study 2
	No project alternative	DPR alternative	No project alternative	DPR alternative
NDMA – N-nitrosodimethylamine. BDCM – Bromodichloromethane. CDBM – Chlorodibromomethane. TCA – Trichloroacetic acid.
Drinking water risk (point estimate)	1.3 × 10^−6	3.9 × 10^−6	7.0 × 10^−6	7.3 × 10^−7
# CPCs with SF	4	8	4	3
Major risk contributors	Arsenic (24%) NDMA (74%)	Arsenic (12%) BDCM (20%) CDBM (18%) NDMA (60%)	Arsenic (85%) BDCM (8%) CDBM (13%)	Arsenic (59%) NDMA (22%) TCA (19%)
Any constituent present at levels > MCLs or advisory levels	No	No	No	No

---

Risk exemplar results for CECs

A list of the RBALs for the detected CECs is presented in Table 4. A summary of the CEC risk for both Case Studies is presented in Table 5. With one exception, for all of the alternatives, all of the CECs have MOSs greater than 1. The exception is quinoline with an MOS in the range of 1 for the case study 1 No Project Alternative and DPR Alternative. For case study 2, quinoline was not found in the No Project Alternative raw drinking water, but was found in the secondary wastewater for the DPR Alternative, and is removed by RO.

Table 4 List of RBALs for detected CECs

CECs	RBAL (ng L^−1)	Category	Reference	CECs	RBAL (ng L^−1)	Category	Reference
DCPA – Dimethyl tetrachloroterephthalate. DEET – N,N-diethyl-meta-toluamide. FDA – Food and Drug Administration. MRTD – Maximum Recommended Therapeutic Dose. NSAID – Non-steroidal anti-inflamatory drug. OTC – Over the counter. TCEP – Tris(2-chloroethyl) phosphate. TCPP – Tris(2-chloroisopropyl) phosphate. TDCPP – Tris(1,3-dichloro-2-propyl) phosphate.
1,7-Dimethylxanthine	7.0 × 10^2	Caffeine metabolite	Environment Protection and Heritage Council et al., 2008 (ref. 15)	Iopromide	7.5 × 10^5	Imaging contrast agent	Environment Protection and Heritage Council et al., 2008 (ref. 15)
3-Hydroxycarbofuran	4.2 × 10^2	Metabolite of carbofuran, pesticide	U.S. EPA, 2009 (ref. 16)	Ketoprofen	1.3 × 10^4	NSAID	Nellor et al., 2010 (ref. 17)
4-tert-Octylphenol	5.0 × 10^4	Intermediate for phenolic resins	Environment Protection and Heritage Council et al., 2008 (ref. 15)	Ketorolac	8.8 × 10^7	NSAID	U.S. FDA MRTD Database (ref. 18)
Acesulfame-K	5.3 × 10^8	Non calorie sweetner	U.S. FDA, 1988 (ref. 19)	Lidocaine	1.1 × 10^8	OTC pain reliever	Donald & Derbyshire, 2004 (ref. 20)
Acetaminophen	1.2 × 10^7	Analgesic	Intertox, 2009 (ref. 21)	Lincomycin	3.7 × 10^5	Antibiotic	Schwab et al., 2005 (ref. 22)
Albuterol	4.1 × 10^4	Bronchodilator	Schwab et al., 2005 (ref. 22)	Lopressor	2.3 × 10^8	Beta blocker; blood pressure medication	U.S. FDA MRTD Database (ref. 18)
Amoxicillin	1.5 × 10^3	Antibiotic	Environment Protection and Heritage Council et al., 2008 (ref. 15)	Meprobamate	2.6 × 10^5	Skeletal muscle relaxant	Snyder et al., 2008 (ref. 23)
Atenolol	7.0 × 10^4	Beta blocker; blood pressure medication	Snyder et al., 2008 (ref. 23)	Metolachlor	7.0 × 10^6	Herbicide	U.S. EPA, 2009 (ref. 16)
Azithromycin	3.9 × 10^3	Anitbiotic	Environment Protection and Heritage Council et al., 2008 (ref. 15)	Naproxen	2.2 × 10^5	NSAID	Environment Protection and Heritage Council et al., 2008 (ref. 15)
Bezafibrate	3.0 × 10^5	Chloesterol medication	Environment Protection and Heritage Council et al., 2008 (ref. 15)	Nifedipine	1.1 × 10^9	Blood pressure and angina medication	U.S. FDA MRTD Database (ref. 18)
Bisphenol A	3.5 × 10^5	Production of polycarbonate plastics and epoxy resins	U.S. EPA, 2009 (ref. 16)	Pentoxifylline	1.3 × 10^5	Blood circulation medication	Nellor et al., 2010 (ref. 17)
Butalbital	1.8 × 10^8	Barbiturate	U.S. FDA MRTD Database (ref. 18)	Primidone	8.4 × 10^2	Anti-seizure medication	Intertox, 2009 (ref. 21)
Caffeine	8.7 × 10^7	Stimulant	Intertox, 2009 (ref. 21)	Quinoline	1.0 × 10^1	Intermediate in the manufacture of other products	U.S. EPA, 2009 (ref. 16)
Carbamazepine	1.2 × 10^4	Anti-seizure medication	Intertox, 2009 (ref. 21)	Sucralose	1.7 × 10^5	Artificial sweetner	U.S. FDA, 1999 (ref. 24)
Carisoprodol	7.0 × 10^8	Skeletal muscle relaxant	U.S. FDA MRTD Database (ref. 18)	Sulfadimethoxine	3.5 × 10^4	Anitbiotic	Environment Protection and Heritage Council et al., 2008 (ref. 15)
Cimetidine	2.0 × 10^5	Heartburn medication, histamine H2-blocker	Environment Protection and Heritage Council et al., 2008 (ref. 15)	Sulfamerazine	2.3 × 10^6	Antibiotic	U.S. FDA MRTD Database (ref. 18)
Cotinine	1.0 × 10^4	Metabolite of nicotine	Environment Protection and Heritage Council et al., 2008 (ref. 15)	Sulfamethazine	7.7 × 10^4	Antibacterial	Nellor et al., 2010 (ref. 17)
Diethanolamine	7.5 × 10^5	Ingredient in soaps, cosmetics, shampoos	Schriks et al., 2009 (ref. 25)	Sulfamethoxazole	1.8 × 10^7	Antibiotic	Intertox, 2009 (ref. 21)
DEET	8.1 × 10^4	Insecticide	Intertox, 2009 (ref. 21)	Sulfathiazole	7.3 × 10^5	Antibacterial	Schwab et al., 2005 (ref. 22)
Dehydronifedipine	2.0 × 10^4	Metabolite of nifedipine, calcium channel blocker	Environment Protection and Heritage Council et al., 2008 (ref. 15)	TCEP	4.4 × 10^3	Flame retardant	Nellor et al., 2010 (ref. 17)
Deisopropylatrazine	1.5 × 10^2	Metabolite of atrazine	Nellor et al., 2010 (ref. 17)	TCPP	1.0 × 10^4	Flame retardant	World Health Organization, 2004 (ref. 26)
Diclofenac	2.3 × 10^6	NSAID	Snyder et al., 2008 (ref. 23)	TDCPP	2.7 × 10^3	Flame retardant	California Department of Toxic Substances Control, 2014 (ref. 27)
Dilantin	6.7 × 10^3	Anti-seizure medication	Intertox, 2009 (ref. 21)	Theobromine	2.8 × 10^10	Ingredient in chocolate	U.S. Department of Health and Human Services, 2006 (ref. 28)
Diltiazem	6.0 × 10^4	Blood pressure medication; calcium channel blocker	Environment Protection and Heritage Council et al., 2008 (ref. 15)	Theophylline	3.0 × 10^8	Bronchodilator	U.S. FDA MRTD Database (ref. 18)
Erythromycin	4.9 × 10^3	Antibiotic	U.S. EPA, 2009 (ref. 16)	Total DCPA Mono & Diacid Degradate	3.5 × 10^5	Degradation products of herbicide DCPA	U.S. EPA, 2008 (ref. 29)
Estrone	4.6 × 10^2	Estrogenic hormone	Snyder et al., 2008 (ref. 23)	Triclocarban	8.8 × 10^8	Antibacterial	California Office of Environmental Health Hazard Assessment, 2010 (ref. 30)
Fluoxetine	1.0 × 10^4	Anti-depressant	Environment Protection and Heritage Council et al., 2008 (ref. 15)	Triclosan	2.6 × 10^6	Antibacterial	Intertox, 2009 (ref. 21)
Gemfibrozil	4.6 × 10^4	Chloesterol medication	Snyder et al., 2008 (ref. 23)	Trimethoprim	6.1 × 10^4	Antibacterial	Schwab, 2005 (ref. 22)
Ibuprofen	3.4 × 10^4	NSAID	Nellor et al., 2010 (ref. 17)	Warfarin	2.3 × 10^3	Blood thinner	Schwab, 2005 (ref. 22)
Iohexal	7.2 × 10^5	Imaging contrast agent	Environment Protection and Heritage Council et al., 2008 (ref. 15)

---

Table 5 Summary of constituents of emerging concern risk exemplar

	Case study 1		Case study 2
	No. project alt.	DPR	No. project alt.	DPR
# CECs present > MRL	32	46	5	53
MOS range	1.6–10 500 000 000	0.9–59 000 000 000	3600–16 000 000	13–6 000 000 000
# CECs with MOS 1–10	1	1	0	0
CECs with MOS 1–10	Quinoline	Quinoline	—	0

---

The RBAL for quinoline, a probable human carcinogen, is based on U.S. EPA's PNEC of 10 ng L⁻¹. Quinoline has specific industrial sources (it is used in the production of dyes, paints, pharmaceuticals, and fragrances), but also has ubiquitous sources including automobile exhaust. Quinoline is biodegradable, removed by RO, and can be photolysized. Thus, if the case study 1 DPR Alternative utilized UV photolysis or RO, it is likely that the concentration would have been further reduced and the MOS would be greater than 1.

For CEC assessments it is important to acknowledge that over time new and updated RBALs are likely to be developed that would further inform risk evaluations, as well as additional information on advanced treatment process performance from research, piloting, or full-scale operations.

Discussion

Wastewater effluent quality

A crucial consideration for DPR projects is quality of the treated wastewater that undergoes advanced treatment. The current focus of operating WWTPs is to meet discharge or non-potable reuse requirements. Because a higher quality wastewater can improve the quality of the final DPR product water and the operations of the AWTF, a shift in thinking about the function of the WWTP is worthwhile as it now is part of an integrated treatment system to produce a potable drinking water supply. A number of process modifications can be implemented at existing WWTPs or WRPs to improve the final effluent quality, including (1) influent wastewater flow equalization; (2) improved primary treatment performance via chemical addition such as alum or polymers; (3) improved secondary treatment performance via increased solids retention times (SRTs), and the addition of microbial selectors to achieve nitrification, denitrification, and/or biological phosphorus removal; (4) enhanced secondary particle settling or phosphorus removal with chemical addition such as alum or polymers; and (5) alternative management of return flows from solids processing facilities including flow equalization, treatment, and/or elimination. For nitrogen removal, a project sponsor may want to consider using de-nitrifying filters or Membrane Bioreactors (MBR) as an additional barrier for not just nitrogen but also for pathogens. For any systems change, such as chemical addition using polymers, consideration should be given to the type of polymer used since it could contain precursors for formation of DBPs, including NDMA.

Constituents not monitored

To provide a relative comparison of health risks related to the quality of representative water from the two No Project alternative/DPR alternative case studies, this investigation was based on the use of data that could be collected relatively quickly and efficiently. Those data included analyses for approximately 350 constituents and 2000 individual data points for each of the test waters. The constituents analysed represent a broad range of chemicals and chemical categories, including compounds with drinking water MCLs and those identified by U.S. EPA as “priority pollutants.” These analyses provide a reasonable basis for completing a relative comparison and to support risk management considerations, which often requires action within reasonable time frames and budgets. Such decisions must always balance the remaining level of uncertainty against the need and value associated with additional monitoring.

This study was not intended to be exhaustive, nor was the intent to try to eliminate all uncertainty regarding the level of risk attributable to drinking water. There are potential pollutants for which data were not available, however the CECs evaluated in this investigation included many of the constituents that have been observed in water supplies or wastewater, and were evaluated based on a reliable analytical method. As new methods mature and other CECs emerge as potential constituents of concern, the methods employed herein can be extended to include additional constituents.

Disinfection by-product formation

The generation or formation of DBPs (trihalomethanes, haloacetic acids, and NDMA) can occur during conventional drinking water treatment, conventional wastewater treatment, and advanced water treatment. This assessment did not account for formation of DBPs during treatment. These compounds can contribute to health risk and depending on the source water and treatment methods utilized could be major contributors to risk. This suggests that formation of DBPs for DPR is an area deserving scrutiny for projects on a case-by-case basis. Bench and pilot studies should be considered for DPR systems that employ technologies (such as ozone and/or chlorination) that have high potential for DBP formation.

The QRRA results clearly illustrate that for the specific constituents in the risk assessment, RO and UV/AOP can be important treatment processes for the removal and control of DBPs including NDMA.

DPR implementation other than those considered for the QRRA

This investigation comprised two case studies, each of which considered a No Project alternative (use of surface water as the drinking water source) and a DPR alternative. In both DPR alternatives, the product water from the AWTF was assumed to be subsequently treated at a drinking water facility (although the specific treatment unit processes at the drinking water facilities varied significantly between the case studies).

There are numerous locations worldwide that are considering DPR as an alternative/supplemental source for their water supplies. Each of these communities has specific financial and logistical constraints that are aspects of their risk management considerations. In some cases, communities may be considering DPR options in which the product water from the AWTF is being considered as a direct component of the drinking water supply rather than as a source water to be subsequently treated at a drinking water facility prior to distribution. This study does not address the relative risks of implementing DPR in this way. However, the approach described and illustrated here could serve as a template for evaluating the relative risks of implementing DPR in ways not specifically evaluated herein.

Practical applications

The results of this investigation indicate that a QRRA can inform decisions that are made with respect to source control, wastewater treatment, water treatment, and advanced treatment for DPR. Information from a QRRA can be used to: (1) assist with decisions on the need for bench scale and/or pilot testing of advanced treatment technologies, potentially including evaluation of CPCs (for example DBP removal efficiency and DBP formation during water/wastewater treatment) or CECs; (2) assist with decisions on the components to include in a DPR treatment scheme – this could be done as a screening framework similar to the approach used for this QRRA or as a site-specific study based on the results of bench scale or pilot testing – for example removal of nitrogen to limit exposure to nitrate and nitrite; (3) modify or tailor monitoring programs to ensure that data for the most relevant contaminants are collected rather than compounds that have little impact on evaluating overall risk; (4) focus on specific source control and/or treatment options in cases where the relative risk may increase over time or reach a level of potential concern – additional information on source control options and effectiveness can be found in TWDB;¹¹ (5) inform the public about the safety of DPR by using the results of a QRRA for public outreach efforts – this will become more important over time as analytical methodology becomes more sensitive and more constituents are found in water even after advanced treatment; and (6) assess the risks and benefits of using DPR as a short-term drought mitigation measure as opposed to a long-term water supply solution by comparing acute and chronic health risks. This analysis did not include a sensitivity analysis to evaluate the relative impact of the conservative assumptions employed because the focus was to compare the relative level of risk associated with No Project and DPR alternatives, rather than estimating “absolute risk”. Future studies could include sensitivity analyses to add further insight relative to the topics highlighted above. Sensitivity analyses are likely to provide significant insights for site specific studies.

Research needs

For DPR, the output of a QRRA could advance and benefit from more reliable information on the following topic areas, which could be obtained through targeted research to: develop better-defined and additional RBALs for CECs that are not removed by treatment; develop a better understanding of the removal or formation of CPCs and CECs through advanced treatment of reclaimed water and through water treatment facilities; and enhance the methods for quantifying health effects of chemical mixtures as part of risk assessments.

Conclusions

A QRRA is a straightforward, easy to understand, and useful tool for decision makers when evaluating water supply options.

Acknowledgements

This project was funded in part by the Texas Water Development Board as part of the project “Evaluating the Potential for Direct Potable Reuse in Texas.” Other participants and co-sponsors were Alan Plummer Associates, Inc., the Brazos Valley Groundwater Conservation District, City of College Station, TX, El Paso Water Utilities, City of Houston, TX, City of Irving, TX, City of Lewisville, TX, City of Lubbock, TX, San Antonio Water System, and the Upper Trinity Compact (Trinity River Authority, North Texas Municipal Water District, Cities of Dallas and Fort Worth, TX), and WateReuse Texas. The authors would like to thank Dr. Adam Olivieri, Dr. Shane Trussell, and Dr. George Tchobanoglous for their input on the QRRA.

References

1. TWDB, Water for Texas 2012 State Water Plan, Austin, TX, 2012.
2. J. J. Cohrssen and V. T. Covello, in Risk Analysis: A Guide to Principles and Methods for Analyzing Health and Environmental Risks, Council on Environmental Quality Executive Office of the President of the United States, 1989.
3. R. C. Cooper, A. W. Olivieri, J. A. Eisenberg, L. Pettegrew, R. E. Danielson, W. A. Fairchild and L. A. Sanchez, City of San Diego Total Resource Recovery Project: Health Effects Study, Western Consortium for Public Health, Berkeley, CA, 1992.
4. R. C. Cooper, A. W. Olivieri, D. M. Eisenberg, J. A. Soller, L. A. Pettegrew and R. E. Danielson, The City of San Diego Total Resource Recovery Project Aqua III San Pasqual Health Effects Study Final Summary Report, Western Consortium for Public Health, Berkeley, CA, 1997.
5. J. A. Soller, D. M. Eisenberg, A. W. Olivieri and C. Galitsky, Groundwater Replenishment System Water Quality Evaluation, prepared by EOA Inc. for the Orange County Water District and Orange County Sanitation District, California, 2000.
6. J. A. Soller and M. H. Nellor, Development and Application of Tools to Assess and Understand the Relative Risks of Regulated Chemicals in Indirect Potable Reuse Projects Task 1 - the Montebello Forebay Groundwater Recharge Project, WaterReuse Foundation, Alexandria, VA, 2011.
7. J. A. Soller and M. H. Nellor, Development and Application of Tools to Assess and Understand the Relative Risks of Regulated Chemicals in Indirect Potable Reuse Projects Task 1 - the Chino Basin Groundwater Recharge Project, WaterReuse Foundation, Alexandria, VA, 2011.
8. S. J. Khan and J. A. McDonald, Quantifying human exposure to contaminants for multiple-barrier water reuse systems, Water Sci. Technol., 2010, 61, 77–83.
9. National Research Council, Risk Assessment in the Federal Government, Managing the process, National Academy Press, Washington, D.C., 1983.
10. U.S. EPA, Risk Assessment Guidance for Superfund, U.S. Environmental Protection Agency, 1989.
11. TWDB, Final Report Direct Potable Reuse Resource Document, Austin, TX, 2015.
12. U.S Dept of Agriculture, Continuing Survey of Food Intakes by Individuals 1994-96, B.H.N.R.C, Agricultural Research Service, Food Surveys Research Group, Beltsville, MD, 1998.
13. U.S. EPA, National Primary Drinking Water Regulations: Long Term 2 Enhanced Surface Water Treatment Rule (LT2ESWTR): Final Rule, 40 CFR Parts 9, 141 and 142, Federal Register, vol. 71, 2006.
14. National Research Council, in Water Reuse: Expanding the Nation's Water Supply Through Reuse of Municipal Wastewater, National Academy Press, Washington, D.C., 2012.
15. Environment Protection and Heritage Council, National Health and Medical Research Council, and Natural Resource Management Ministerial Council, Australian Guidelines for Water Recycling Augmentation of Drinking Water Supplies, 2008.
16. U.S. EPA, Drinking Water Contaminant List 3, Federal Register, 2009, vol. 74, p. 51850.
17. M. H. Nellor, J. A. Soller, G. M. Bruce, R. C. Pleus and M. K. Peterson, Development and Application of Tools to Assess and Understand the Relative Risks of Drugs and Other Chemicals in Indirect Potable Reuse Water, WRF 06-018 Task 2, WateReuse Foundation, Alexandria, VA, 2009.
18. U.S. Food and Drug Administration, MRTD Database, http://www.fda.gov/Drugs/default.htm, (accessed June 2, 2015).
19. U.S. Food and Drug Administration, Acceptable Daily Intake Acesulfame-K, Federal Register, 1988, vol. 53, p. 28379.
20. M. J. Donald and S. Derbyshire, Case report: Lignocaine toxicity; a complication of local anaesthesia administered in the community, Emerg. Med. J., 2004, 21, 249–250.
21. Intertox, Comparison of Analytical Results for Trace Organics in the Santa Ana River at the Imperial Highway to Health Risk-based Screening Levels, prepared for Orange County Water District, Fountain Valley, CA, 2009.
22. B. W. Schwab, E. P. Hayes, J. M. Fiori, F. J. Mastrocco, N. M. Roden, D. Cragin, R. D. Meyerhoff, V. J. D'Aco and P. D. Anderson, Human pharmaceuticals in US surface waters: a human health risk assessment, Regul. Toxicol. Pharmacol., 2005, 42, 296–312.
23. S. Snyder, R. A. Trenholm, E. M. Snyder, G. M. Bruce, E. Bennett, R. C. Pleus and J. D. C. Hemming, Toxicological Relevance of EDCs and Pharmaceuticals in Drinking Water, Awwa Research Foundation and Water Research Foundation, Denver, CO, 2008.
24. U.S. Food and Drug Administration, Food Additives Permitted for Direct Addition to Food for Human Consumption, Sucralose, 21 CFR Part 172, 1999.
25. M. Schriks, M. B. Heringa, M. van der Kooi, P. De Voost and A. van Wezel, Toxicological relevance of emerging contaminants for drinking water qualit, Water Res., 2010, 44, 461–476.
26. WHO, Environmental Health Criteria 209, Flame Retardants Tris(chloropropyl) phosphate and Tris(2-chloroethyl) phosphate, 2004.
27. California Department of Toxic Substances Control, Priority Product Profile: Children's Foam-padded Sleeping Products Containing TDCPP, Sacramento, CA, 2014.
28. U.S Dept of Health and Human Services, Toxnet Toxicology Data Network, Theobromine, CASRN: 83-67-0, 2006.
29. U.S. EPA, Regulatory Determinations Support Document for Selected Contaminants from the Second Drinking Water Contaminant Candidate List (CCL 2), EPA 815-R-08-012, 2008.
30. California Office of Environmental Health Hazard Assessment, Safety & Exposure Information Related to Potential Designated Chemical (Triclocarban), May 24, 2010 Meeting of Scientific Guidance Panel Biomonitoring California, 2010.

---