69390 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

ENVIRONMENTAL PROTECTION document includes the best available review at EPA’s Drinking Water Docket: AGENCY technologies (BATs) upon which the 401 M Street, SW., Washington, DC MRDLs and MCLs are based. The set of 20460 from 9 a.m. to 4 p.m., Eastern 40 CFR Parts 9, 141, and 142 regulations promulgated today is also Standard Time, Monday through Friday, know as the Stage 1 Disinfection excluding legal holidays. For access to [WH±FRL±6199±8] Byproducts Rule (DBPR). EPA believes docket materials, please call 202/260– RIN 2040±AB82 the implementation of the Stage 1 DBPR 3027 to schedule an appointment and will reduce the levels of disinfectants obtain the room number. National Primary Drinking Water and disinfection byproducts in drinking FOR FURTHER INFORMATION CONTACT: For Regulations: Disinfectants and water supplies. The Agency believes the general information contact, the Safe Disinfection Byproducts rule will provide public health Drinking Water Hotline, Telephone protection for an additional 20 million AGENCY: Environmental Protection (800) 426–4791. The Safe Drinking households that were not previously Agency (EPA). Water Hotline is open Monday through covered by drinking water rules for ACTION: Final rule. Friday, excluding Federal holidays, disinfection byproducts. In addition, the from 9:00 am to 5:30 pm Eastern Time. rule will for the first time provide SUMMARY: In this document, EPA is For technical inquiries, contact Tom public health protection from exposure finalizing maximum residual Grubbs, Office of Ground Water and to haloacetic acids, chlorite (a major disinfectant level goals (MRDLGs) for Drinking Water (MC 4607), U.S. chlorine dioxide byproduct) and chlorine, chloramines, and chlorine Environmental Protection Agency, 401 bromate (a major ozone byproduct). dioxide; maximum contaminant level M Street SW, Washington, DC 20460; The Stage 1 DBPR applies to public goals (MCLGs) for four trihalomethanes telephone (202) 260–7270. For Regional water systems that are community water (chloroform, bromodichloromethane, contacts see SUPPLEMENTARY systems (CWSs) and nontransient dibromochloromethane, and INFORMATION. bromoform), two haloacetic acids noncommunity water systems (dichloroacetic acid and trichloroacetic (NTNCWs) that treat their water with a SUPPLEMENTARY INFORMATION: This acid), bromate, and chlorite; and chemical disinfectant for either primary regulation is effective 60 days after National Primary Drinking Water or residual treatment. In addition, publication of Federal Register Regulations (NPDWRs) for three certain requirements for chlorine document for purposes of the disinfectants (chlorine, chloramines, dioxide apply to transient Administrative Procedures Act and the and chlorine dioxide), two groups of noncommunity water systems Congressional Review Act. Compliance organic disinfection byproducts (total (TNCWSs). dates for specific components of the rule trihalomethanes (TTHMs)—a sum of the EFFECTIVE DATE: This regulation is are discussed below. Solely for judicial four listed above, and haloacetic acids effective February 16, 1999. Compliance review purposes, this final rule is (HAA5)—a sum of the two listed above dates for specific components of the rule promulgated as of 1 p.m. Eastern Time plus monochloroacetic acid and mono- are discussed in the Supplementary December 30, 1998, as provided in 40 and dibromoacetic acids), and two Information Section. The incorporation CFR 23.7. inorganic disinfection byproducts by reference of certain publications Regulated entities. Entities regulated (chlorite and bromate). The NPDWRs listed in today’s rule is approved by the by the Stage 1 DBPR are community and consist of maximum residual Director of the Federal Register as of nontransient noncommunity water disinfectant levels (MRDLs) or February 16, 1999. systems that add a disinfectant during maximum contaminant levels (MCLs) or ADDRESSES: Public comments, the any part of the treatment process treatment techniques for these comment/response document, including a residual disinfectant. In disinfectants and their byproducts. The applicable Federal Register documents, addition, certain provisions apply to NPDWRs also include monitoring, other major supporting documents, and transient noncommunity systems that reporting, and public notification a copy of the index to the public docket use chlorine dioxide. Regulated requirements for these compounds. This for this rulemaking are available for categories and entities include:

Category Examples of regulated entities

Industry ...... Community and nontransient noncommunity water systems that treat their water with a chemical disinfectant for either pri- mary of residual treatment. In addition, certain requirements for chlorine dioxide apply to transient noncommunity water systems. State, Local, Tribal, Same as above. or Federal Gov- ernments.

This table is not intended to be applicability criteria in § 141.130 of this Regional Contacts exhaustive, but rather provides a guide rule. If you have questions regarding the I. Kevin Reilly, Water Supply Section, for readers regarding entities likely to be applicability of this action to a JFK Federal Bldg., Room 203, Boston, regulated by this action. This table lists particular entity, contact one of the MA 02203, (617) 565–3616 the types of entities that EPA is now persons listed in the preceding FOR II. Michael Lowy, Water Supply Section, aware could potentially be regulated by FURTHER INFORMATION CONTACT section 290 Broadway 24th Floor, New York, this action. Other types of entities not or the Regional contacts below. NY 10007–1866, (212) 637–3830 listed in this table could also be III. Jason Gambatese, Drinking Water regulated. To determine whether your Section (3WM41), 1650 Arch Street, facility is regulated by this action, you Philadelphia, PA 19103–2029, (215) should carefully examine the 814–5759 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69391

IV. David Parker, Water Supply Section, HAN: Haloacetonitriles TNCWS: Transient noncommunity 345 Courtland Street, Atlanta, GA ICR: Information collection rule (issued water systems 30365, (404) 562–9460 under section 1412(b) of the SDWA) TWG: Technical work group V. Miguel Del Toral, Water Supply ILSI: International Life Sciences UMRA: Unfunded mandates reform act Section, 77 W. Jackson Blvd., Chicago, Institute URTH: Unreasonable risk to health IL 60604, (312) 886–5253 IESTWR: Interim Enhanced Surface WIDB: Water Industry Data Base VI. Blake L. Atkins, Drinking Water Water Treatment Rule Table of Contents Section, 1445 Ross Avenue, Dallas, LOAEL: Lowest Observed Adverse TX 75202, (214) 665–2297 Effect Level I. Background VII. Ralph Flournoy, Drinking Water/ LT1ESTWR: Long-Term 1Enhanced A. Statutory Requirements and Legal Ground Water Management Branch, Surface Water Treatment Rule Authority 726 Minnesota Ave., Kansas City, KS MCL: Maximum contaminant level B. Regulatory History 66101, (913) 551–7374 MCLG: Maximum contaminant level 1. Existing Regulations VIII. Bob Clement, Public Water Supply 2. Public Health Concerns To Be goal Addressed Section (8P2–W–MS), 999 18th Street, M–DBP: Microbial and Disinfectants/ 3. Regulatory Negotiation Process Suite 500, Denver, CO 80202–2466, Disinfection Byproducts 4. Federal Advisory Committee Process (303) 312–6653 mg/L: Milligrams per liter 5. 1997 and 1998 Notices of Data IX. Bruce Macler, Water Supply Section, MRDL: Maximum residual disinfectant Availability (NODA) 75 Hawthorne Street, San Francisco, level II. Summary of Final Stage 1 Disinfection CA 94105, (415) 744–1884 MRDLG: Maximum residual disinfectant Byproduct Rule X. Wendy Marshall, Drinking Water level goal A. Applicability Unit, 1200 Sixth Avenue (OW–136), NDWAC: National Drinking Water B. MRDLGs and MRDLs for Disinfectants Seattle, WA 98101, (206) 553–1890 Advisory Council C. MCLGs and MCLs for TTHMs, HAA5, Chlorite, and Bromate Abbreviations Used in This Document NIST: National Institute of Science and D. Treatment Technique for Disinfection Technology Byproducts Precursors AWWA: American Water Works NOAEL: No Observed Adverse Effect E. BAT for Disinfectants, TTHMs, HAA5, Association Level Chlorite, and Bromate AWWSCo: American Water Works NODA: Notice of Data Availability F. Compliance Monitoring Requirements Service Company NOM: Natural organic matter G. Analytical Methods BAT: Best available technology NPDWR: National Primary Drinking H. Laboratory Certification Criteria BDCM: Bromodichloromethane Water Regulation I. Variances and Exemptions CDC: Centers for Disease Control and NTNCWS: Nontransient noncommunity J. State Recordkeeping, Primacy, Reporting Prevention water system Requirements C.I.: Confidence Intervals K. System Reporting Requirements NTP: National Toxicology Program L. Guidance Manuals CMA: Chemicals Manufacturers NTTAA: National Technology Transfer Association M. Regulation Review and Advancement Act III. Explanation of Final Rule CPE: Comprehensive performance NTU: Nephelometric turbidity unit evaluation A. MCLGs/MRDLGs OMB: Office of Management and Budget 1. MCLG for Chloroform CWS: Community water system PAR: Population attributable risk a. Today’s Rule DBCM: Dibromochloromethane PBMS: Performance based measurement b. Background and Analysis DBP: Disinfection byproducts c. Summary of Comments D/DBP: Disinfectants and disinfection system PE: Performance evaluation 2. MCLG for Bromodichloromethane byproducts (BDCM) DBPR: Disinfection Byproducts Rule PODR: Point of diminishing return PQL: Practical quantitation limit a. Today’s Rule DBPRAM: Disinfection byproducts b. Background and Analysis regulatory analysis model PWS: Public water system c. Summary of Comments DCA: Dichloroacetic acid QC: Quality control 3. MCLG for Dibromochloromethane DOC: Dissolved organic carbon Reg. Neg.: Regulatory Negotiation (DBCM) DWSRF: Drinking Water State Revolving RFA: Regulatory Flexibility Act a. Today’s Rule Fund RfD: Reference dose b. Background and Analysis EC: Enhanced coagulation RIA: Regulatory impact analysis c. Summary of Comments EJ: Environmental justice RSC: Relative source contribution 4. MCLG for Bromoform EPA: United States Environmental SAB: Science Advisory Board a. Today’s Rule b. Background and Analysis Protection Agency SBREFA: Small Business Regulatory Enforcement Fairness Act c. Summary of Comments ESWTR: Enhanced Surface Water 5. MCLG for Dichloroacetic Acid (DCA) Treatment Rule SDWIS: Safe Drinking Water a. Today’s Rule FACA: Federal Advisory Committee Act Information System b. Background and Analysis GAC10: Granular activated carbon with SUVA: Specific ultraviolet absorbance c. Summary of Comments ten minute empty bed contact time SDWA: Safe Drinking Water Act, or the 6. MCLG for Trichloroacetic Acid (TCA) and 180 day reactivation frequency ‘‘Act,’’ as amended 1996 a. Today’s Rule GAC20: Granular activated carbon with SWTR: Surface Water Treatment Rule b. Background and Analysis twenty minute empty bed contact TC: Total coliforms c. Summary of Comments time and 180 day reactivation TCA: Trichloroacetic acid 7. MCLG for Chlorite and MRDLG for frequency TCR: Total Coliform Rule Chlorine Dioxide TOC: Total organic carbon a. Today’s Rule GDP: Gross Domestic Product b. Background and Analysis GWR: Groundwater rule TOX: Total organic halides c. Summary of Comments HAA5: Haloacetic acids TTHM: Total trihalomethanes 8. MCLG for Bromate (five)(chloroacetic acid, dichloroacetic (chloroform, bromdichloromethane, a. Today’s Rule acid, trichloroacetic acid, bromoacetic dibromochloromethane, and b. Background and Analysis acid, and dibromoacetic acid) bromoform) c. Summary of Comments 69392 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

9. MCLG for Chloral Hydrate 2. Background and Analysis G. Executive Order 13045: Protection of a. Today’s Rule 3. Summary of Comments Children from Environmental Health b. Background and Analysis K. System Reporting and Record Keeping Risks and Safety Risks c. Summary of Comments Requirements H. Consultation with the Science Advisory 10. MRDLG for Chlorine 1. Today’s Rule Board, National Drinking Water a. Today’s Rule 2. Summary of Comments Advisory Council, and the Secretary of b. Background and Analysis L. State Recordkeeping, Primacy, and Health and Human Services c. Summary of Comments Reporting Requirements I. Executive Order 12875: Enhancing the 11. MRDLG for Chloramine 1. State Recordkeeping Requirements Intergovernmental Partnership a. Today’s Rule a. Today’s Rule J. Executive Order 13084: Consultation and b. Background and Analysis b. Background and Analysis Coordination with Indian Tribal c. Summary of Comments c. Summary of Comments Governments B. Epidemiology 2. Special Primacy Requirements K. Submission to Congress and the General 1. Cancer Epidemiology a. Today’s Rule Accounting Office a. Today’s Rule b. Background and Analysis L. Likely Effect of Compliance with the b. Background and Analysis c. Summary of Comments Stage 1 DBPR on the Technical, c. Summary of Comments 3. State Reporting Requirements Financial, and Managerial Capacity of 2. Reproductive and Developmental a. Today’s Rule Public Water Systems Epidemiology b. Background and Analysis VI. References a. Today’s Rule c. Summary of Comments b. Background and Analysis M. Variances and Exemptions I. Background c. Summary of Comments 1. Today’s Rule C. MCLs and BAT for TTHM, HAA5, 2. Background and Analysis A. Statutory Requirements and Legal Chlorite, and Bromate; MRDLs and BAT 3. Summary of Comments Authority for Chlorine, Chloramines, and Chlorine N. Laboratory Certification and Approval The Safe Drinking Water Act (SDWA 1. Today’s Rule Dioxide or the Act), as amended in 1986, 1. MCLs for TTHMs and HAA5 2. Background and Analysis 3. Summary of Comments requires USEPA to publish a ‘‘maximum a. Today’s Rule contaminant level goal’’ (MCLG) for b. Background and Analysis IV. Economic Analysis c. Summary of Comments A. Today’s Rule each contaminant which, in the 2. MCL for Bromate B. Background judgement of the USEPA Administrator, 1. Overview of RIA for the Proposed Rule a. Today’s Rule ‘‘may have any adverse effect on the 2. Factors Affecting Changes to the 1994 b. Background and Analysis health of persons and which is known RIA c. Summary of Comments a. Changes in Rule Criteria or anticipated to occur in public water 3. MCL for Chlorite b. New Information Affecting DBP systems’’ (Section 1412(b)(3)(A)). a. Today’s Rule Occurrence and Compliance Forecast MCLGs are to be set at a level at which b. Background and Analysis c. New Epidemiology Information ‘‘no known or anticipated adverse effect c. Summary of Comments C. Cost Analysis on the health of persons occur and 4. MRDL for Chlorine 1. Revised Compliance Forecast a. Today’s Rule which allows an adequate margin of 2. System Level Unit Costs safety’’ (Section 1412(b)(4)). b. Background and Analysis 3. National Costs c. Summary of Comments The Act was amended in August D. Benefits Analysis 1996. As a result of these Amendments, 5. MRDL for Chloramines 1. Exposure Assessment a. Today’s Rule 2. Baseline Risk Assessment Based on several of these provisions were b. Background and Analysis TTHM Toxicological Data renumbered and augmented with c. Summary of Comments 3. Baseline Analysis Based on additional language. Other sections 6. MRDL for Chlorine Dioxide Epidemiology Data were added establishing new drinking a. Today’s Rule 4. Exposure Reduction Analysis water requirements. These b. Background Analysis 5. Monetization of Health Endpoints modifications are outlined below. c. Summary of Comments E. Net Benefits Analysis The Act also requires that at the same D. Treatment Technique Requirement F. Summary of Comments 1. Today’s Rule time USEPA publishes an MCLG, which V. Other Requirements is a non-enforceable health goal, it also 2. Background and Analysis A. Regulatory Flexibility Analysis 3. Summary of Comments 1. Today’s Rule must publish a National Primary E. Predisinfection Disinfection Credit 2. Background and Analysis Drinking Water Regulation (NPDWR) 1. Today’s Rule 3. Summary of Comments that specifies either a maximum 2. Background and Analysis B. Paperwork Reduction Act contaminant level (MCL) or treatment 3. Summary of Comments C. Unfunded Mandates Reform Act technique (Sections 1401(1) and F. Requirements for Systems to Use 1. Summary of UMRA Requirements 1412(a)(3)). USEPA is authorized to Qualified Operators 2. Written Statement for Rules with Federal promulgate a NPDWR ‘‘that requires the G. Analytical Methods Mandates of $100 million or More use of a treatment technique in lieu of 1. Today’s Rule a. Authorizing Legislation 2. Background and Analysis b. Cost Benefit Analysis establishing a MCL,’’ if the Agency finds 3. Summary of Comments c. Estimates of Future Compliance Costs that ‘‘it is not economically or 4. Performance Based Measurement and Disproportionate Budgetary Effects technologically feasible to ascertain the Systems d. Macro-economic Effects level of the contaminant’’. H. Monitoring Requirements e. Summary of State, Local, and Tribal As amended, EPA’s general authority 1. Today’s Rule Government and TheirConcerns to set a maximum contaminant level 2. Background and Analysis f. Regulatory Alternative Considered goal (MCLG) and National Primary 3. Summary of Comments 3. Impacts on Small Governments Drinking Water Regulation (NPDWR) I. Compliance Schedules D. National Technology Transfer and applies to contaminants that may ‘‘have 1. Today’s Rule Advancement Act 2. Background and Analysis E. Executive Order 12866: Regulatory an adverse effect on the health of 3. Summary of Comments Planning and Review persons,’’ that are ‘‘known to occur or J. Public Notice Requirements F. Executive Order 12898: Environmental there is a substantial likelihood that the 1. Today’s Rule Justice contaminant will occur in public water Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69393 systems with a frequency and at levels health effects of exposure to Giardia quarterly averages of all samples. The of public health concern,’’ and for lamblia, viruses, and Legionella, as well value for each sample is the sum of the which ‘‘in the sole judgement of the as many other pathogenic organisms. measured concentrations of chloroform, Administrator, regulation of such Briefly, those requirements include: (1) bromodichloromethane (BDCM), contaminant presents a meaningful requirements for a maintenance of a dibromochloromethane (DBCM) and opportunity for health risk reduction for disinfectant residual in the distribution bromoform. persons served by public water system; (2) removal and/or inactivation The interim TTHM standard only systems’’ (SDWA Section 1412(b)(1)(A)). of 3 logs (99.9%) for Giardia and 4 logs applies to community water systems The amendments, also require EPA, (99.99%) for viruses; (3) combined filter using surface water and/or ground water when proposing a NPDWR that includes effluent performance of 5 nephelometric serving at least 10,000 people that add an MCL or treatment technique, to turbidity unit (NTU) as a maximum and a disinfectant to the drinking water publish and seek public comment on an 0.5 NTU at 95th percentile monthly, during any part of the treatment process. analysis of health risk reduction and based on 4-hour monitoring for At their discretion, States may extend cost impacts. In addition, EPA is treatment plants using conventional coverage to smaller PWSs; however, required to take into consideration the treatment or direct filtration (with most States have not exercised this effects of contaminants upon sensitive separate standards for other filtration option. subpopulations (i.e. infants, children, technologies); and (4) watershed Information Collection Rule. The pregnant women, the elderly, and protection and other requirements for Information Collection Rule (ICR) is a individuals with a history of serious unfiltered systems. monitoring and data reporting rule that illness), and other relevant factors. Total Coliform Rule. The Total was promulgated on May 14, 1996 (61 (Section 1412 (b)(3)(C)). Coliform Rule (TCR) (54 FR 27544; June FR 24354) (EPA, 1996a). The purpose of The amendments established a 29, 1989) applies to all public water the ICR is to collect occurrence and number of regulatory deadlines, systems (EPA, 1989b). This regulation treatment information to help evaluate including schedules for a Stage 1 sets compliance with the MCL for total the need for possible changes to the Disinfection Byproduct Rule (DBPR), an coliforms (TC) as follows. For systems current SWTR and existing microbial Interim Enhanced Surface Water that collect 40 or more samples per treatment practices, and to help evaluate Treatment Rule (IESWTR), a Long-Term month, no more than 5.0% of the the need for future regulation for Final Enhanced Surface Water samples may be TC-positive; for those disinfectants and disinfection Treatment Rule (LTESWTR) affecting that collect fewer than 40 samples, no byproducts (D/DBPs). The ICR will Public Water Systems (PWSs) that serve more than one sample may be TC- provide EPA with additional under 10,000 people, and a Stage 2 positive. In addition, if two consecutive information on the national occurrence DBPR (Section 1412(b)(2)(C)). The Act samples in the system are TC-positive, in drinking water of (1) chemical as amended also requires EPA to and one is also fecal coliform or E. coli- byproducts that form when disinfectants promulgate regulations to address filter positive, then this is defined as an acute used for microbial control react with backwash (Section 1412(b)(14)). Finally, violation of the MCL. If a system naturally occurring compounds already the Act requires EPA to promulgate exceeds the MCL, it must notify the present in source water and (2) disease- regulations specifying criteria for public using mandatory language causing microorganisms, including requiring disinfection ‘‘as necessary’’ for developed by the EPA. The required Cryptosporidium, Giardia, and viruses. ground water systems (Section 1412 monitoring frequency for a system The ICR will also provide engineering (b)(8)). depends on the number of people data on how PWSs currently control for Finally, as part of the 1996 SDWA served and, ranges from 480 samples per such contaminants. This information is Amendments, recordkeeping month for the largest systems to once being collected because the 1992 requirements were modified to apply to annually for certain of the smallest Regulatory Negotiating Committee ‘‘every person who is subject to a systems. All systems must have a (henceforth referred to as the Reg. Neg. requirement of this title or who is a written plan identifying where samples Committee) on microbial pathogens and grantee’’ (Section 1445 (a)(1)(A)). Such are to be collected. disinfectants and DBPs concluded that If a system has a TC-positive sample, additional information was needed to persons are required to ‘‘establish and it must test that sample for the presence assess the potential health problem maintain such records, make such of fecal coliforms or E. coli. The system created by the presence of DBPs and reports, conduct such monitoring, and must also collect a set of repeat samples, pathogens in drinking water and to provide such information as the and analyze for TC (and fecal coliform assess the extent and severity of risk in Administrator may reasonably require or E. coli) within 24 hours of the first order to make sound regulatory and by regulation * * * ’’. TC-positive sample. public health decisions. The ICR will B. Regulatory History The TCR also requires an on-site also provide information to support inspection every 5 years (10 years for regulatory impact analyses for various 1. Existing Regulations non-community systems using only regulatory options, and to help develop Surface Water Treatment Rule. Under protected and disinfected ground water) monitoring strategies for cost-effectively the Surface Water Treatment Rule for each system that collects fewer than implementing regulations. (SWTR) (54 FR 27486, June 29, 1989) five samples per month. This on-site The ICR pertains to large public water (EPA,1989a), EPA set maximum inspection (referred to as a sanitary systems serving populations at least contaminant level goals of zero for survey) must be performed by the State 100,000; a more limited set of ICR Giardia lamblia, viruses, and Legionella; or by an agent approved by the State. requirements pertain to ground water and promulgated NPDWR for all PWSs Total Trihalomethane Rule. In systems serving between 50,000 and using surface water sources or ground November 1979 (44 FR 68624) (EPA, 100,000 people. About 300 PWSs water sources under the direct influence 1979) EPA set an interim MCL for total operating 500 treatment plants are of surface water. The SWTR includes trihalomethanes (TTHM) of 0.10 involved with the extensive ICR data treatment technique requirements for milligrams per liter (mg/L) as an annual collection. Under the ICR, these PWSs filtered and unfiltered systems that are average. Compliance is defined on the monitor for water quality factors intended to protect against the adverse basis of a running annual average of affecting DBP formation and DBPs 69394 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations within the treatment plant and in the are hampered by difficulties of study these studies have suggested an distribution system monthly for 18 design, scope, and sensitivity. On the association, albeit small, between months. In addition, PWSs must other hand, uncertainty is involved in bladder, rectal, and colon cancer and provide operating data and a description the interpretation of results using high exposure to chlorinated surface water. of their treatment plan design and dose animal toxicological studies of a While there are fewer published surface water systems must monitor for few of the numerous byproducts that epidemiology studies that have been bacteria, viruses, and protozoa. Finally, occur in disinfected drinking water to conducted to evaluate the possible a subset of PWSs must perform estimate the risk to humans from relationship between exposure to treatment studies, using either granular chronic exposure to low doses of these chlorinated surface water and activated carbon (GAC) or membrane and other byproducts. Such studies of reproductive and developmental effects, processes, to evaluate DBP precursor individual DBPs is insufficient to a recent study has suggested an removal and control of DBPs. characterize risks from exposure to the association between early term Monitoring for treatment study entire mixture of DBPs in disinfected miscarriage and exposure to drinking applicability began in September 1996. drinking water. While recognizing these water with elevated trihalomethane The remaining occurrence monitoring uncertainties, EPA continues to believe levels. In addition to this study, another began in July 1997. that the Stage 1 DBPR is necessary for new study reported a small increased One initial intent of the ICR was to the protection of public health from risk of neural tube defects associated collect pathogen occurrence data and exposure to potentially harmful DBPs. with consumption of drinking water other information for use in developing A fundamental component in containing high levels of TTHMs. the IESWTR and to estimate national assessing the risk for a contaminant is However, no significant associations costs for various treatment options. the number of people that may be were observed with individual THMs, However, because of delays in exposed to the parameter of concern. In HAAs, and haloacetonitriles (HANs) promulgating the ICR and technical this case, there is a very large and adverse outcomes in this study. As difficulties associated with laboratory population potentially exposed to DBPs with cancer, EPA cannot conclude at approval and review of facility sampling through drinking water in the U.S. Over this time there is a causal link between plans, ICR monitoring did not begin 200 million people are served by PWSs exposure to DBPs and reproductive and until July 1, 1997, which was later than that apply a disinfectant (e.g., chlorine) developmental effects. originally anticipated. As a result of this to water in order to provide protection While EPA recognizes there are data delay and the new statutory deadlines against microbial contaminants. While deficiencies in the information on the for promulgating the Stage 1 DBPR and these disinfectants are effective in health effects from the DBPs and the IESWTR in November of 1998 (resulting controlling many microorganisms, they levels at which they occur, the Agency from the 1996 SDWA amendments), ICR react with natural organic and inorganic believes the weight-of-evidence data were not available in time to matter in the water to form DBPs, some presented by the available support these rules. In place of the ICR of which may pose health risks. One of epidemiological studies on chlorinated data, the Agency worked with the most complex questions facing drinking water and toxicological studies stakeholders to identify other sources of water supply professionals is how to on individual DBPs support a potential data developed since 1994 that could be minimize the risks from DBPs and still hazard concern and warrant regulatory used to support the development of the maintain adequate control over action at this time to reduce DBP levels Stage 1 DBPR and IESWTR. EPA will microbial contaminants. Because of the in drinking water. Recognizing the continue to work with stakeholders in large number of people exposed to deficiencies in the existing data, EPA analyzing and using the comprehensive DBPs, there is a substantial concern for believes the incremental two-stage ICR data and research for developing any risks associated with DBPs that may approach for regulating DBPs, agreed future Enhanced Surface Water impact public health. upon by the regulatory negotiation Treatment requirements and the Stage 2 Since the discovery of chlorination process, is prudent and necessary to DBPR. byproducts in drinking water in 1974, protect public health and meet the numerous toxicological studies have requirements of the SDWA. 2. Public Health Concerns to be been conducted. Results from these In conclusion, because of the large Addressed studies have shown several DBPs (e.g., number of people exposed to DBPs and EPA’s main mission is the protection bromodichloromethane, bromoform, the different potential health risks (e.g., of human health and the environment. chloroform, dichloroacetic acid, and cancer and adverse reproductive and When carrying out this mission, EPA bromate) to be carcinogenic in developmental effects) that may result must often make regulatory decisions laboratory animals . Some DBPs (e.g., from exposure to DBPs, EPA believes with less than complete information and chlorite, BDCM, and certain haloacetic the Stage 1 DBPR is needed to further with uncertainties in the available acids) have also been shown to cause prevent potential health effects from information. EPA believes it is adverse reproductive or developmental DBPs, beyond that controlled for by the appropriate and prudent to err on the effects in laboratory animals. Although 1979 total trihalomethane rule. Both the side of public health protection when many of these studies have been Reg. Neg. Committee for the 1994 there are indications that exposure to a conducted at high doses, EPA believes proposed rule and the Microbial and contaminant may present risks to public the studies provide evidence that DBPs Disinfectants/Disinfection Byproducts health, rather than take no action until present a potential public health risk Advisory Committee (henceforth cited risks are unequivocally proven. that needs to be addressed. as the M-DBP Advisory Committee) In regard to the Stage 1 DBPR, EPA In the area of epidemiology, a number formed in March 1997 under the Federal recognizes that the assessment of public of epidemiology studies have been Advisory Committee Act (FACA), health risks from disinfection of conducted to investigate the agreed with the need for the Stage 1 drinking water currently relies on relationship between exposure to DBPR to reduce potential risks from inherently difficult and preliminary chlorinated surface water and cancer. DBPs in the near term, while empirical analysis. On one hand, While EPA cannot conclude there is a acknowledging additional information epidemiologic studies of the causal link between exposure to is still needed for the Stage 2 DBPR populations in various geographic areas chlorinated surface water and cancer, (especially on health effects), Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69395

3. Regulatory Negotiation Process for both microbial contaminants and D/ major issues that were discussed in DBPs. Notices of Data Availability (NODA) for In 1992 EPA initiated a negotiated The Reg. Neg. Committee agreed that the IESWTR (62 FR 59486, November 3, rulemaking to address public health the schedules for IESWTR and 1997) (EPA, 1997a) and the Stage 1 concerns associated with disinfectants, LTESWTR should be ‘‘linked’’ to the DBPR (62 FR 59388, November 3, 1997) DBPs, and microbial pathogens. The schedule for the Stage 1 DBPR to assure (EPA, 1997b). The major negotiators included representatives of simultaneous compliance and a recommendations addressed by the State and local health and regulatory balanced risk-risk based Committee and in the NODAs were to: agencies, public water systems, elected implementation. The Reg. Neg. (1) Maintain the proposed MCLs for officials, consumer groups and Committee agreed that additional TTHM, HAA5, and bromate; (2) modify environmental groups. The Reg. Neg. information on health risk, occurrence, the enhanced coagulation requirements Committee met from November 1992 treatment technologies, and analytical as part of DBP control; (3) include a through June 1993. methods needed to be developed in microbial benchmarking/profiling to Early in the process, the negotiators order to better understand the risk-risk provide a methodology and process by agreed that large amounts of information tradeoff, and how to accomplish an which a PWS and the State, working necessary to understand how to overall reduction in health risks to both together, assure that there will be no optimize the use of disinfectants to pathogens and D/DBPs. significant reduction in microbial concurrently minimize microbial and Finally the Reg. Neg. Committee protection as the result of modifying DBP risk on a plant-specific basis were agreed that to develop a reasonable set disinfection practices in order to meet unavailable. Nevertheless, the Reg. Neg. of rules and to understand more fully MCLs for TTHM and HAA5; (4) Committee agreed that EPA propose a the limitations of the current SWTR, continue credit for compliance with Stage 1 DBPR to extend coverage to all additional field data were critical. Thus, applicable disinfection requirements for community and nontransient a key component of the regulation disinfection applied at any point prior noncommunity water systems that use negotiation agreement was the to the first customer, consistent with the disinfectants, reduce the current TTHM promulgation of the ICR previously existing SWTR; (5) modify the turbidity MCL, regulate additional DBPs, set described. performance requirements and add limits for the use of disinfectants, and 4. Federal Advisory Committee Process requirements for individual filters; (6) reduce the level of organic precursor establish an MCLG for Cryptosporidium; compounds in the source water that In May 1996, the Agency initiated a (7) add requirements for removal of may react with disinfectants to form series of public informational meetings Cryptosporidium; (8) provide for DBPs. to provide an update on the status of the mandatory sanitary surveys; and (9) 1994 proposal and to review new data EPA’s most significant concern in make a commitment to additional related to microbial and DBP regulations analysis of the role of Cryptosporidium developing regulations for disinfectants that had been developed since July and DBPs was the need to ensure that inactivation as part of a multiple barrier 1994. In August 1996, Congress enacted concept in the context of a subsequent adequate treatment be maintained for the 1996 SDWA Amendments which controlling risks from microbial Federal Register microbial proposal. The contained a number of new new data and analysis supporting the pathogens. One of the major goals requirements, as discussed above, as addressed by the Reg. Neg. Committee technical areas of agreement were well as specifying deadlines for final summarized and explained at length in was to develop an approach that would promulgation of the IESWTR and Stage reduce the level of exposure from EPA’s 1997 NODAs (EPA, 1997a and 1 DBPR. To meet these deadlines and to EPA, 1997b). disinfectants and DBPs without maximize stakeholder participation, the undermining the control of microbial Agency established the M–DBP 5. 1997 and 1998 Notices of Data pathogens. The intention was to ensure Advisory Committee under FACA in Availability that drinking water is microbiologically March 1997, to collect, share, and In November 1997 EPA published a safe at the limits set for disinfectants analyze new information and data, as NODA (USEPA, 1997b) that and DBPs and that these chemicals do well as to build consensus on the summarized the 1994 proposal; not pose an unacceptable health risk at regulatory implications of this new described new data and information that these limits. Thus, the Reg. Neg. information. The Committee consisted the Agency has obtained and analyses Committee also considered a range of of 17 members representing EPA, State that have been developed since the microbial issues and agreed that EPA and local public health and regulatory proposal; provided information should also propose a companion agencies, local elected officials, drinking concerning the July 1997 microbial rule (IESWTR). water suppliers, chemical and recommendations of the M–DBP Following months of intensive equipment manufacturers, and public Advisory Committee on key issues discussions and technical analysis, the interest groups. related to the proposal (described Reg. Neg. Committee recommended the The M–DBP Advisory Committee met above); and requested comment on these development of three sets of rules: a five times in March through July 1997 recommendations, as well as on other two-staged approach for the DBPs to discuss issues related to the IESWTR regulatory implications that flow from (proposal: 59 FR 38668, July 29, 1994) and Stage 1 DBPR. Technical support the new data and information. The (EPA, 1994a), an ‘‘interim’’ ESWTR for these discussions was provided by a Agency solicited additional data and (proposal: 59 FR 38832, July 29, 1994) Technical Work Group (TWG) information that were relevant to the (EPA, 1994b), and an information established by the Committee at its first issues discussed in the DBP NODA. EPA collection rule (proposal: 59 FR 6332, meeting in March 1997. The also requested that any information the February 10, 1994) (EPA, 1994c) Committee’s activities resulted in the Agency should consider as part of the (promulgation: 61FR24354, May 14, collection, development, evaluation, final rule development process 1996) (EPA, 1996a). The approach used and presentation of substantial new data regarding data or views submitted to the in developing these proposals and information related to key elements Agency since the close of the comment considered the constraints of of both proposed rules. The Committee period on the 1994 proposal, be simultaneously treating water to control reached agreement on a number of formally resubmitted during the 90-day 69396 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations comment period unless already in the increasing the Maximum Residual nontransient noncommunity water underlying record in the docket for the Disinfection Level Goal (MRDLG) for systems (NTNCWs) that treat their water NODA. chlorine dioxide from 0.3 mg/L to 0.8 with a chemical disinfectant for either In March 1998, EPA issued a second mg/L. Today’s final rule was developed primary or residual treatment. In DBP NODA (EPA, 1998a) that based on the outcome of the 1992 Reg. addition, certain requirements for summarized new health effects Neg., the 1994 proposed rule, the 1997 chlorine dioxide apply to transient information received and analyzed since FACA process, and both the 1997 and noncommunity water systems the November 1997 NODA and 1998 DBP NODAs, as well as a wide (TNCWSs). requested comments on several issues range of technical comments from related to the simultaneous compliance stakeholders and members of the public. B. MRDLGs and MRDLs for Disinfectants A summary of today’s rule follows. with the Stage 1 DBPR and the Lead and EPA is finalizing the following Copper Rule. The 1998 NODA indicated II. Summary of Final Stage 1 MRDLGs and maximum residual EPA was considering increasing the Disinfection Byproduct Rule disinfectant levels (MRDLs) for chlorine, MCLG for chloroform from zero to 0.3 chloramines, and chlorine dioxide in mg/L and the proposed MCLG for A. Applicability Table II–1. chlorite from 0.08 mg/L to 0.8 mg/L. The final Stage 1 DBPR applies to EPA also requested comment on community water systems (CWSs) and

TABLE II±1.ÐMRDLGS AND MRDLS FOR DISINFECTANTS

Disinfectant residual MRDLG (mg/L) MRDL (mg/L)

Chlorine ...... 4 (as Cl2) 4.0 (as Cl2) Chloramine ...... 4 (as Cl2) 4.0 (as Cl2) Chlorine Dioxide ...... 0.8 (as ClO2) 0.8 (as ClO2)

C. MCLGs and MCLs for TTHMs, HAA5, Chlorite, and Bromate EPA is finalizing the MCLGs and MCLs in Table II–2.

TABLE II±2.ÐMCLGS AND MCLS FOR DISINFECTION BYPRODUCTS

Disinfection byproducts MCLG (mg/L) MCL (mg/L)

Total trihalomethanes (TTHM) 1 ...... N/A 0.080 ÐChloroform ...... 0 ...... ÐBromodichloromethane ...... 0 ...... ÐDibromochloromethane ...... 0.06 ...... ÐBromoform ...... 0 ...... Haloacetic acids (five) (HAA5) 2 ...... N/A 0.060 ÐDichloroacetic acid ...... 0 ...... ÐTrichloroacetic acid ...... 0.3 ...... Chlorite ...... 0.8 1.0 Bromate ...... 0 0.010 N/AÐNot applicable because there are no individual MCLGs for TTHMs or HAAs. 1 Total trihalomethanes is the sum of the concentrations of chloroform, bromodichloromethane, dibromochloromethane, and bromoform. 2 Haloacetic acids (five) is the sum of the concentrations of mono-, di-, and trichloroacetic acids and mono- and dibromoacetic acids.

D. Treatment Technique for Disinfection conventional filtration treatment are Removal will be achieved through a Byproduct Precursors required to remove specified treatment technique (enhanced percentages of organic materials coagulation or enhanced softening) Water systems that use surface water (measured as total organic carbon) that unless a system meets alternative or ground water under the direct may react with disinfectants to form criteria discussed in Section III.D. influence of surface water and use DBPs as indicated in Table II–3.

TABLE II±3.ÐREQUIRED REMOVAL OF TOTAL ORGANIC CARBON BY ENHANCED COAGULATION AND ENHANCED SOFTENING FOR SUBPART H SYSTEMS USING CONVENTIONAL TREATMENT a,b,c

Source Water Alkalinity (mg/L as CaCO Source Water TOC (mg/L) 3) (percent) 0±60 >60±120 >120

>2.0±4.0 ...... 35.0 25.0 15.0 >4.0±8.0 ...... 45.0 35.0 25.0 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69397

TABLE II±3.ÐREQUIRED REMOVAL OF TOTAL ORGANIC CARBON BY ENHANCED COAGULATION AND ENHANCED SOFTENING FOR SUBPART H SYSTEMS USING CONVENTIONAL TREATMENT a,b,cÐContinued

Source Water Alkalinity (mg/L as Source Water TOC (mg/L) CaCO3) (percent) 0±60 >60±120 >120

>8.0 ...... 50.0 40.0 30.0 a Systems meeting at least one of the conditions in Section 141.135(a)(2) (i)±(vi) of the rule are not required to operate the removals in this table. b Softening systems meeting one of the two alternative compliance criteria in Section 141.135(a)(3) of the rule are not required to meet the re- movals in this table. c Systems practicing softening must meet the TOC removal requirements in the last column to the right.

E. BAT for Disinfectants, TTHMs, is set. PWS that are unable to achieve an exemptions). Table II.4 includes the HAA5, Chlorite, and Bromate MCL or MRDL may be granted a BATs for each of the MCLs or MRDLs variance if they use the BAT and meet that EPA is promulgating in today’s Under the SDWA, EPA must specify other requirements (see section III.M for Stage 1 DBPR. the BAT for each MCL (or MRDL) that a discussion of variances and

TABLE II±4.ÐBAT FOR DISINFECTANTS AND DISINFECTION BYPRODUCTS

Disinfectant/DBP Best available technology

Disinfectants

Chlorine residual ...... Control of treatment processes to reduce disinfectant demand and control of disinfection treatment processes to reduce disinfectant levels. Chloramine residual ...... Control of treatment processes to reduce disinfectant demand and control of disinfection treatment processes to reduce disinfectant levels. Chlorine dioxide residual ...... Control of treatment processes to reduce disinfectant demand and control of disinfection treatment processes to reduce disinfectant levels.

Disinfection Byproducts

Total trihalomethanes ...... Enhanced coagulation or enhanced softening or GAC10*, with chlorine as the primary and residual disinfectant. Total haloacetic acids ...... Enhanced coagulation or enhanced softening or GAC10*, with chlorine as the primary and residual disinfectant. Chlorite ...... Control of treatment processes to reduce disinfectant demand and control of disinfection treatment processes to reduce disinfectant levels. Bromate ...... Control of ozone treatment process to reduce production of bromate. * GAC10 means granular activated carbon with an empty bed contact time of 10 minutes and reactivation frequency for GAC of no more than six months.

F. Compliance Monitoring Requirements monitoring of chlorite; two methods for detailed discussion of the lab bromide; one method for the certification requirements, see section Compliance monitoring requirements measurement of bromate; and one III.N. are explained in Section III.H of today’s method for the measurement of UV254. I. Variances and Exemptions rule. EPA has developed routine and Finally, EPA approved all methods reduced compliance monitoring allowed in § 141.89(a) for measuring Variances and exemptions will be schemes for disinfectants and DBPs to alkalinity. These issues are discussed in permitted in accordance with existing be protective from different types of more detail in section III.G. statutory and regulatory authority. For a health concerns, including acute and detailed discussion see section III.M. long-term effects. H. Laboratory Certification Criteria J. State Recordkeeping, Primacy, and G. Analytical Methods Consistent with other drinking water regulations, determinations of Reporting Requirements EPA has approved five methods for compliance with the MCLs may only be The Stage 1 DBPR requires States to measurement of free chlorine, four conducted by certified laboratories. EPA adopt several regulatory requirements, methods for combined chlorine, and six is requiring that analyses can be including public notification for total chlorine. EPA has also conducted by a party acceptable to EPA requirements, MCLs for DBPs, MRDLs approved two methods for the or the State in those situations where for disinfectants, and the requirements measurement of chlorine dioxide the parameter can adequately be in Subpart L. In addition, States are residuals; three methods for the measured by someone other than a required to adopt several special measurement of HAA5; three methods certified laboratory and for which there primacy requirements for the Stage 1 for the measurement of TTHMs; three is a good reason to allow analysis at DPBR. States are also required to keep methods for the measurement of TOC/ other locations (e.g., for samples which specific records in accordance with Dissolved Organic Carbon (DOC); two normally deteriorate before reaching a existing regulations and additional methods for the monthly measurement certified laboratory, especially when records specific to the Stage 1 DBPR. of chlorite and one method for the daily taken at remote locations). For a Finally, the rule does not require any 69398 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

State additional reporting requirements FR 38677 for a detailed discussion of the potential health effects from short- beyond those required under existing the process for establishing MCLGs). term exposure to elevated levels of regulations. These requirements are The final Stage 1 DBPR contains disinfectants can be dismissed. Ozone discussed in more detail in Section III.L. MCLGs for: four THMs (chloroform, does not require an MRDLG because it bromodichloromethane, reacts so completely that it does not K. System Reporting Requirements dibromochloromethane, and occur in water delivered to consumers. System are required to report bromoform); two haloacetic acids Finally, EPA believes the use of the monitoring data to the State as (dichloroacetic acid and trichloroacetic MRDLGs for other disinfectants or discussed in Section III.K. acid); bromate; and chlorite (see table oxidants would not be appropriate since L. Guidance Manuals II–2 for final MCLG levels). These MRDLGs are developed for regulated MCLGs are the same as those proposed compounds controlled by MRDLs or EPA is developing guidance for both in 1994 with the exception of chlorite, treatment techniques and EPA does not systems and States for the which increased from 0.08 mg/L to 0.8 allow these compounds to be used to implementation of the Stage 1 DBPR mg/L. The MCLG for chloral hydrate has demonstrate compliance with and the IESWTR. The guidance manuals been dropped since EPA has concluded disinfection requirements. include: Guidance Manual for Enhanced that it will be controlled by the MCLs The information EPA relied on to Coagulation and Precipitative Softening; for TTHM and HAA5 and the enhanced establish the MCLGs and MRDLGs was Disinfection Benchmark Guidance coagulation treatment technique. described in the 1994 proposal (EPA, Manual; Turbidity Guidance Manual; The final Stage 1 DBPR contains 1994a), the 1997 DBP NODA (EPA, Alternative Disinfectants and Oxidants MRDLGs for chlorine, chloramines and 1997b), and the 1998 NODA (EPA, Guidance Manual; M/DBP Simultaneous chlorine dioxide (see table II–1 for final 1998a). Criteria and assessment Compliance Manual; Sanitary Survey MRDLG levels). The MRDLGs are as the documents to support the MCLGs and Guidance Manual; Unfiltered Systems same as those proposed in 1994, with MRDLGs are included in the docket Guidance Manual; and Uncovered the exception of chlorine dioxide, (EPA, 1993a; EPA, 1994 d–h; EPA, Finished Water Reservoirs. Guidance which increased from 0.3 mg/L to 0.8 1997c; EPA, 1998 b–f; and EPA, 1998p). manuals will be available after the mg/L. A summary of the occurrence and publication of the Stage 1 DBPR. The MRDLG concept was introduced exposure information for this rule are M. Regulation Review in the proposed rule for disinfectants to detailed in ‘‘Occurrence Assessment for reflect the fact that these substances Disinfectants and Disinfection Under the provisions of the SDWA have beneficial disinfection properties. Byproducts in Public Drinking Water (Section 1412(b)(9)), the Agency is As with MCLGs, MRDLGs are Supplies’ (EPA, 1998u). The discussion required to review NPDWRs at least established at the level at which no of the data used to establish the MCLGs once every six years. As mentioned known or anticipated adverse effects on and MRDLGs and a summary of the previously, today’s final rule revises, the health of persons occur and which major public comments for these updates, and supersedes the regulations allows an adequate margin of safety. chemicals are included below. A more for total trihalomethanes, initially MRDLGs are nonenforceable health detailed discussion is included below published in 1979. Since that time, goals based only on health effects and for chloroform, DCA, chlorite, chloride there have been significant changes in exposure information and do not reflect dioxide, and bromate than the other technology, treatment techniques, and the benefit of the addition of the disinfectants and DBPs. This is the case other regulatory controls that provide chemical for control for waterborne because significant new data has for greater protection of human health. microbial contaminants. By using the become available since the 1994 As such, for today’s rule, EPA has term ‘‘residual disinfectant’’ in lieu of proposal for these four DBPs and one analyzed innovations and changes in ‘‘contaminant’’, EPA intends to avoid disinfectant. technology and treatment techniques situations in which treatment plant that have occurred since promulgation operators are reluctant to apply 1. MCLG for Chloroform of the interim TTHM regulations. That disinfectant dosages above the MRDLG a. Today’s Rule. After careful analysis, contained primarily in the cost during short periods of time to control consideration of all public comments, and technology document supporting for microbial risk. EPA has concluded at this time to this rule, supports the changes in the EPA received numerous comments on promulgate an MCLG for chloroform of Stage 1 DBPR from the 1979 TTHM rule. the use of the term MRDLG. The zero as proposed. This conclusion EPA believes that the innovations and majority of commenters agreed that the reflects an interim risk-management changes in technology and treatment term MRDLG was appropriate to use in decision on the part of the Agency. The techniques that result in changes to the place of MCLG for disinfectants. Other Agency recognizes the strength of the 1979 TTHM regulations are feasible commenters agreed, but felt that the science in support of a non-linear within the meaning of SDWA Section language should more strongly reflect approach for estimating carcinogenicity 1412(b). that disinfectants are necessary and that of chloroform. EPA received public III. Explanation of Final Rule short-term exposure to elevated levels of comments that questioned the the disinfectants is not a health concern. underlying basis and approach used to A. MCLGs/MRDLGs Some commenters suggested that reach the science judgment that the MCLGs are set at levels at which no MRDLGs be extended to ozone, mode of chloroform’s carcinogenic known or anticipated adverse health potassium permanganate and iodine. action supports a nonlinear approach. effects occur, allowing for an adequate In response, EPA agrees with the Equally important are the policy and margin of safety. Establishment of an majority of commenters that the use of regulatory issues raised by stakeholders MCLG for each specific contaminant is the term MRDLG is appropriate and that touch on this issue. EPA believes based on the available evidence of therefore the final rule retains this term. that it is essential to pursue a further carcinogenicity or noncancer adverse EPA believes the language on the dialogue with stakeholders on the issues health effects from drinking water importance of disinfectants is adequate raised in the public comments before exposure using EPA’s guidelines for risk in the rule and thus has not changed applying the substantial new data and assessment (see the proposed rule at 59 this language. EPA does not agree that science on the mode of carcinogenic Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69399 action discussed in the 1998 NODA to 1994 proposal on a risk management exposure; and that an MCLG based on the important decision of moving to a decision that a presumptive or low-dose protection against liver toxicity should non-linear cancer extrapolation linear default (i.e, MCLG of zero) was be protective against carcinogenicity approach for drinking water appropriate until more research became given that the putative mode of action contaminants under the SDWA. available and there was an adequate understanding for chloroform involves Moreover, EPA will complete additional opportunity to work with stakeholders cytotoxicity as a key event preceding deliberations with the Agency’s Science and the scientific community to tumor development. The MCLG of 0.3 Advisory Board (SAB) (open to evaluate and assess the technical as well mg/L was calculated using a relative stakeholder presentations to the SAB) as policy and regulatory implications of source contribution (RSC) of 80 percent. on the analytical approach used to such new information. The 1994 The RSC of 80 percent was based on the evaluate and reach conclusions on mode proposal also reflected the Agency’s assumption that most exposure to of action data, and the science basis for 1986 Guidelines for Carcinogen Risk chloroform is likely to come from the mode of carcinogenic action for Assessment (EPA, 1986) which ingestion of drinking water. The 80 chloroform. recommended reliance on the default In evaluating how to proceed in the assumption of low-dose linearity in the percent assumption for the RSC was development of an MCLG for absence of substantial information on consistent with the calculations used to chloroform, the Agency believes two the mechanism of carcinogenicity. derive the MCLGs for D/DBPs in the additional factors must be taken into Since the 1994 proposal, over 30 1994 proposal. Based on information consideration. First, as part of the 1996 toxicological studies have been received during the public comment SDWA amendments, Congress published on chloroform. These studies period for the 1998 NODA, EPA is mandated that the Stage 1 DBPR rule be were discussed in the November 1997 considering revising its estimate of the promulgated by November 1998. EPA Stage 1 DBP NODA (EPA, 1997b). In RSC for chloroform as discussed below. has concluded that it would be addition, EPA published a second DBP Since the 1998 NODA, EPA has impossible to complete the additional NODA in March 1998 (EPA, 1998a) reevaluated elements of the analysis deliberations noted above in time to which discussed recommendations and underlying a revised MCLG of 0.30 mg/ meet this statutory deadline. Second, as findings from a 1997 International Life L. Considering recent information not explained below, the Agency has also Sciences Institute project (ILSI, 1997), fully analyzed as part of the 1998 completed analysis indicating that co-sponsored by EPA, on the cancer NODA, the Agency is considering regardless of whether the MCLG is assessment for chloroform. The ILSI revising the assumption of an 80% RSC based on a low-dose linear or non-linear project included the analysis and from ingestion of drinking water in view extrapolation approach, the MCL conclusions from an expert panel which enforceable standard for TTHMs of 0.08 was convened and charged with of data which indicates that exposure to mg/L will not be affected. In light of reviewing the available database chloroform via inhalation and dermal these issues, EPA believes it is relevant to the carcinogenicity of exposure may potentially contribute a appropriate and consistent with the chloroform, and considering how end substantial percentage of the overall public health goals of the SDWA to points related to mode of action can be exposure to chloroform depending on establish a zero MCLG for chloroform applied in hazard and dose-response the activity patterns of individuals. based on a linear default extrapolation assessment by using guidance provided Also, EPA is in the process of approach until the Agency is able to by the EPA’s 1996 Proposed Guidelines developing a policy for incorporating complete additional deliberations with for Carcinogen Assessment (EPA, inhalation and dermal exposure into the the Agency’s SAB on the analytical 1996b). The panel was made up of 10 derivation of the RSC. Furthermore, approach used to evaluate and reach internationally recognized scientists there is considerable uncertainty conclusions on mode of action data and from academia, industry, government, regarding the potential exposure to the science basis for the mode of and the private sector. Based on a chloroform via the dietary route and carcinogenic action for chloroform, and consideration of the ILSI panel findings there is information which indicates complete the process of further public and an assessment of new data on individuals who are frequent swimmers dialogue on the important question of chloroform since 1994, EPA requested may receive a large amount of moving to a non-linear cancer comment in the 1998 NODA on the chloroform during swimming. There are extrapolation approach. EPA also notes Agency’s science conclusion that additional uncertainties regarding other that its approach is consistent with chloroform is a likely human carcinogen possible highly exposed sub- legislative history of the SDWA (see 56 and that available scientific analysis populations, e.g., from use of FR 3533—EPA, 1991) and the 1996 supports a non-linear mode of action for humidifiers, hot-tubs, and outdoor SDWA Amendments. estimating the carcinogenic risk misters. In conclusion, because there b. Background and Analysis. As part associated with lifetime exposure from may be a potential for exposure to of its 1994 Stage 1 DBP proposal (EPA, ingesting drinking water. chloroform from other routes of 1994a), EPA requested comment on a As part of the 1998 NODA, EPA also exposure than ingestion of drinking zero MCLG for chloroform. This was requested comment on a revised consistent with information provided to chloroform MCLG of 0.30 mg/L. The water, EPA is considering using the 20 the 1992 Reg. Neg. Committee and was revised MCLG was premised on the percent default floor to ensure adequate based on data from a drinking water substantial new science noted above public health protection. The 20 percent study by Jorgensen et al. (1985) that supports a non-linear mode of has been used historically for drinking indicating an increase of kidney tumors action. In calculating the specific water contaminants other than D/DBPs in male rats in a dose-related manner. MCLG, EPA relied upon data relating to when there is uncertainty in the However, at the time of the proposal hepatoxicity in dogs (EPA, 1994a). This available exposure data. The use of the there was insufficient data to determine hepatoxicity endpoint was deemed 20 percent RSC for chloroform would the mode of carcinogenic action for appropriate given that hepatic injury is produce a MCLG of 0.07 mg/L: chloroform. Therefore, EPA based its the primary effect following chloroform 69400 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

0.01 mg/kg/d × 70 kg × 0.2 MCLG for chloroform = = 0./L 07 mg 2L/d

In addition to its reassessment of strongly held views associated with appropriate and prudent risk technical assumptions underlying the development of the chloroform MCLG, management decision at this time is to revised MCLG, the Agency has also EPA notes that the one question not establish an MCLG for chloroform at the reviewed and carefully considered in fundamentally at issue is the proposed presumptive default level of detail a number of significant comments establishment of the 0.080 mg/L TTHM zero. As part of this decision, the on the 1998 NODA. These comments MCL. The majority of commenters who Agency will complete additional reflect both substantial scientific addressed the proposed TTHM MCL deliberations with the Agency’s SAB on support as well as significant concerns continue to support it. This is the analytical approach used to evaluate with a possible MCLG of 0.30 mg/L. As particularly important to EPA in light of and reach conclusions on mode of outlined in more detail below, a number congressional action with regard to the action data, and the science basis for the of nationally recognized scientific M–DBP process in the 1996 SDWA mode of carcinogenic action for experts strongly affirmed the data and Amendments. In enacting the chloroform. The SAB’s review will be technical rationale for relying upon a Amendments and particularly in factored into the Agency’s Stage 2 DBP non-linear mode of action for expressing congressional intent in the rulemaking process. EPA will also chloroform. Other commenters, conference Report, Congress was careful include consideration of the regulatory, however, highlighted several scientific to emphasize ‘‘that the new provisions policy, and precedential issues issues they felt were not adequately of this conference agreement not involving chloroform in the Agency’s considered. These commenters also conflict with the parties’ agreement nor Round 2 M–BP stakeholder process. emphasized their concern that the disrupt the implementation of the EPA wishes to make clear that its policy, regulatory, and enforcement regulatory actions,’’ (such as the current interim decision in today’s rule to set an implications related to a revised MCLG agreement on an TTHM MCL of 0.080 MCLG of zero pending SAB review and were not raised by EPA in either the mg/L). Both of these important elements further stakeholder involvement is not 1992 or the 1997 regulatory negotiation of the Congressional intent were intended to prejudge the question of processes leading up to today’s final reflected in the statutory text. Section what the appropriate MCLG should be rule. Thus, these commenters felt that a 1412(b)(2)(C) requires EPA to maintain for purposes of regulatory decisions number of stakeholders who the M–DBP rule staggered promulgation under the Stage 2 DBPR. EPA may recommended support for components strategy agreed to by the negotiated decide to retain the zero MCLG for that of the Stage 1 DBPR rule did so under rulemaking; and Section 1412(b)(6)(C) rule, or to revise it, depending on the one set of conditions and assumptions exempted the future M–DBP rules from outcome of the SAB review, as well as that the Agency subsequently changed the new cost-benefit standard-setting any new scientific evidence that may without providing a sufficient provision (1412(b)(6)(A)) but not from become available. In regard to the opportunity for further debate and the new risk-risk provision (1412(b)(5)), appropriate RSC factor, in case a non- discussion. because the latter was a part of the linear approach should ultimately be EPA believes that an adequate negotiated rulemaking agreement but adopted, the Agency requests that opportunity for notice and comment the former was not. stakeholders provide any data they man was provided as a result of the 1997 and The Agency, itself, also believes that have bearing on this determination. 1998 DBP NODAs on the underlying the underlying logic, data, and rationale The fundamental objective of the scientific data and technical issue of supporting establishment of a TTHM of SDWA is to establish protective public moving to a non-linear extrapolation 0.080 mg/L MCL is compelling, and this health goals (MCLGs) together with approach based on an understanding of is a critical factor in the Agency’s enforceable standards (MCLs or the mode of carcinogenic action for chloroform MCLG decision under treatment techniques) to move the water chloroform and recalculating the today’s rule. Under either a low-dose treatment systems as close to the public chloroform MCLG to a nonzero number. linear or non-linear extrapolation to health goal as is technologically and However, the Agency recognizes that derive the MCLG, the final TTHM MCL economically feasible. In the case of the reliance on a non-linear mode of action remains unaffected. chloroform and TTHMs, this objective is under the SDWA does represent a After thorough review of the data and met with whichever extrapolation significant and precedential, albeit comments, EPA believes the nonlinear approach (low dose linear versus sound, application of new science to the cancer extrapolation approach is the nonlinear) is relied upon. policy development and risk most appropriate means to establish an c. Summary of Comments. EPA management decision making process of MCLG for chloroform based on received numerous comments on both establishing appropriately protective carcinogenic risk. However, in light of the 1994 proposed rule regarding the MCLGs. The Agency also recognizes its own reconsideration of the MCLG of zero for chloroform and the that although, as discussed below, a appropriate RSC for chloroform under MCLG of 0.3 mg/L contained in the revised MCLG for chloroform would not such an approach, considering the range 1998 NODA. Some commenters were affect the TTHM MCL under today’s of policy, regulatory, and enforcement supportive of the MCLG of zero, while rule, the precedential decision to utilize issues raised as part of the public others were supportive of the 0.3 mg/L a non-linear cancer extrapolation comment period, recognizing the MCLG. The major reason raised by approach clearly has important importance of deliberations with SAB commenters for establishing a nonzero implications for the development of before proceeding further and, yet, MCLG (e.g., 0.3 mg/L) was that there future MCLGs where there is also recognizing that this cannot be was convincing scientific evidence to adequate scientific research and data to accomplished within the constraints of conclude that a nonlinear margin of support such a non-linear analysis. meeting the statutory deadline for Stage exposure approach for evaluating the In reviewing the range of scientific, 1 DBPR rule of November 1998, EPA has carcinogenic risk from chloroform is policy, and regulatory analyses and determined that on balance the more warranted. Commenters who were Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69401 against establishing a nonzero MCLG for carcinogens might represent some finite EPA agrees with commenters that chloroform presented policy and level of risk and therefore an MCLG additional review by the FACA of the scientific concerns. Scientific concerns above zero did not meet the statutory regulatory implications of a nonlinear raised by commenters opposed to the requirement that the goal be set where approach is appropriate for policy nonzero MCLG included their no known anticipated adverse effects reasons, and will initiate these perceptions that: there is insufficient occur, allowing for an adequate margin discussions in the context of the Stage scientific evidence of a threshold for of safety (56 FR 3533; EPA, 1991). 2 DBPR FACA deliberations. In light of chloroform; the threshold assumption is However, if there is scientific evidence the November 1998 statutory deadline also invalid because chloroform co- that indicates there is a ‘‘safe threshold’’ to promulgate the Stage 1 DBP rule and occurs with other mutagenic then a non-zero MCLG could be the steps necessary to complete a final carcinogens; EPA ignored human data established with an adequate margin of rule, EPA has concluded that there is in establishing the MCLG for safety (56 FR 3533; EPA, 1991)). Even not enough time to meet with the SAB chloroform; the linkage between though EPA, as an interim matter, is and FACA, provide ample opportunity cytotoxicity and regenerative establishing an MCLG of zero for for debate, resolve differing points of proliferation and kidney tumors is not chloroform in today’s rule, it believes it views, and complete additional analysis supported by the data; and the evidence has the authority to establish nonzero to meet stakeholders policy concerns in for genotoxicity is mixed and it would MCLGs for carcinogens if the scientific the context of the Stage 1 DBP rule. EPA be difficult if not impossible to evidence supports this finding. notes, however, that regardless of the conclude that the evidence demonstrate In response to commenter’s concerns MCLG for chloroform, the MCL for the chloroform has no direct effect on DNA. with EPA using the proposed 1996 THMs remains at 0.08 mg/L. Since the As detailed at greater length in the Guidelines for Carcinogen Risk MCL is the enforceable standard that docket, EPA does not agree with these Assessment (EPA, 1996b) instead of the water systems will be required to meet, comments as a technical matter. The Agency’s 1986 guidelines, EPA believes a nonlinear or low dose linear Agency does agree with the commenters it is important to point out that the 1986 extrapolation to derive the MCLG will view that further discussion of these guidelines provide for departures from not have a direct impact on the compliance obligations of public water issues with both the SAB and as part of default assumptions such as low dose additional public dialogue is systems or on the levels of chloroform linear assessment. For example, the appropriate. allowed in public water systems, 1986 EPA guidelines reflect the position The policy issues raised by although it may be relevant to of the OSTP (1985; Principle 26) ‘‘No commenters included their belief that: a development of enforceable regulatory single mathematical procedure is zero MCLG is required to comply with limits established under future rules. provisions of the SDWA; EPA is recognized as the most appropriate for required to use the 1986 Cancer low-dose extrapolation in 2. MCLG for Bromodichloromethane Guidelines (EPA, 1986) until the 1996 carcinogenesis. When relevant (BDCM) Cancer Guidelines (EPA, 1996b) are biological evidence on mechanisms of a. Today’s Rule. The final MCLG for formally finalized, and under the 1986 action exists (e.g, pharmacokinetics, BDCM is zero. The zero MCLG is based guidelines the MCLG for chloroform target organ dose), the models or on the classification of BDCM as a must be set at zero; EPA did not provide procedure employed should be probable human carcinogen. The MCLG sufficient opportunity for the members consistent with the evidence.’’ The 1986 was determined in a weight-of-evidence of the FACA, established to assist in the guideline goes on to further state ‘‘The evaluation which considered all development of the Stage 1 DBP rule, to Agency will review each assessment as relevant health data including properly consider the potential to the evidence on carcinogenesis carcinogenicity and reproductive and implications of a nonzero MCLG; and mechanisms and other biological or developmental toxicity animal data. setting a MCLG for chloroform (0.3 mg/ statistical evidence that indicates the EPA believes the data are insufficient at L) above the MCL for the TTHMs (0.08 suitability of a particular extrapolation this time to determine the mode of mg/L) is illogical. model.’’ The EPA’s 1996 Proposed carcinogenic action for BDCM, and In response, EPA believes that the Guidelines for Carcinogen Risk therefore a low dose linear extrapolation underlying science for using a nonlinear Assessment allow EPA to use other approach is used to estimate lifetime extrapolation approach to evaluate the default approaches to estimate cancer cancer risk as a default. carcinogenic risk from chloroform is risk than the historic, linearized b. Background and Analysis. In the well founded. As explained above, multistage default when there is an 1994 Stage 1 DBPR proposal, the MCLG because of the issues raised during the understanding of an agent’s mode of of zero for BDCM was based on large public comment period, EPA believes carcinogenic action. EPA believes that intestine and kidney tumor data from a additional review and dialogue with reliance on the 1986 guidance allows National Toxicology Program (NTP) stakeholders is needed prior to EPA to reach the same conclusion on chronic animal study (NTP, 1987). Since departing from a long-held EPA policy the carcinogenic risk from chloroform as the proposal, several new studies have of establishing zero MCLGs for known if the 1996 guidelines were used. The been published on BDCM metabolism or probable carcinogens. EPA will also use of the best available science is a core (EPA, 1997c). In addition, several new complete additional deliberations with EPA principle and is statutorily genotoxicity studies and short-term the Agency’s SAB on the analytical mandated by the SDWA amendments of toxicity studies including reproductive approach used to evaluate and reach 1996. The 1996 Proposed Guidelines for evaluations were found for BDCM (EPA, conclusions on mode of action data, and Carcinogen Risk Assessment reflect new 1997c). These new studies contribute to the science basis for the mode of science and are consistent with the the weight-of-evidence conclusions carcinogenic action for chloroform. existing 1986 Guidelines for Carcinogen reached in the 1994 proposal. Based on In response to the policy issues raised Risk Assessment. EPA considered the this evidence, the final MCLG for BDCM by commenters, EPA, historically, has 1996 proposed guidelines in assessing is zero based on sufficient evidence of established MCLGs of zero for known or the health effects data for chloroform carcinogenicity in animals. probable human carcinogens based on and the other contaminants discussed in c. Summary of Comments. Several the principle that any exposure to the 1998 March NODA. commenters disagreed with the use of a 69402 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations corn oil gavage animal cancer study to commenter agreed with EPA’s decision commenters generally felt that EPA determine the MCLG for BDCM. Some to base the MCLG on noncarcinogenic should use an RfD approach in commenters agreed with the EPA endpoints. Several commenters quantifying the risk for bromoform. decision to use large intestine and disagreed with the use of a corn oil Some commenters encouraged EPA to kidney tumor data from the corn oil gavage study to determine the MCLG for conduct more experiments on gavage study, but not liver tumor data DBCM. bromoform toxicity. Some commenters in the quantitative estimation of In response, because the evidence of were concerned with the use of a corn carcinogenic risk. One commenter carcinogenicity was limited on DBCM oil gavage study to determine agreed that a low-dose linear (i.e., increased tumor response in only carcinogenic risk. extrapolation approach to dose-response one of the two species tested), EPA In response, although the increase in assessment was appropriate at this time classified DBCM as a possible human tumors was small, the increase was and consistent with EPA policy. carcinogen. The additional factor of 10 considered significant because large However, this commenter suggested that to account for possible carcinogenicity intestine tumors in both male and EPA undertake chronic studies that follows EPA’s science policy for female rats are rare and thus provides include a drinking water study of BDCM establishing MCLGs (EPA, 1994a). EPA sufficient evidence to classify and toxicokinetics. One commenter used liver effects from the NTP bromoform as a probable human disagreed with the EPA conclusion that subchronic corn oil gavage study as the carcinogen. EPA does not plan on the evidence on the mutagenicity of basis for the Reference Dose (RfD). EPA conducting additional chronic testing BDCM is adequate. agrees with the comment that this is an for bromoform at this time, but is In response, EPA agrees with appropriate basis for deriving the RfD conducting toxicokinetic studies and commenters that a drinking water study for DBCM. EPA agrees with commenters shorter term drinking water studies to is preferable to a corn oil gavage study that a drinking water study is preferable better understand the potential risk to assess risk from DBPs in drinking to a corn oil gavage study to assess risk associated with exposure to bromoform water. However, the NTP corn oil from DBPs in drinking water. However, in drinking water. EPA agrees with gavage study is the best data available the NTP corn oil gavage study is the best commenters that drinking water studies on BDCM for a quantitative risk data available on DBCM for derivation are preferable to a corn oil gavage study estimation at this time. BDCM is of the MCLG at this time. EPA does not to assess risk from DBPs in drinking currently being tested for toxicokinetics plan to conduct additional chronic water. However, the NTP corn oil and cancer in a chronic BDCM drinking studies for DBCM but is conducting gavage study is the best data available water rodent study by the NTP. When additional toxicokinetics and short term on bromoform for derivation of the these data are available, EPA will drinking water studies on DBCM to MCLG. reassess the cancer risk of BDCM. EPA better understand the potential risk 5. MCLG for Dichloroacetic Acid (DCA) believes that the animal data currently associated with exposure through available on BDCM are consistent with drinking water. a. Today’s Rule. The final MCLG for EPA cancer guidelines on classifying DCA is zero. EPA has developed a 4. MCLG for Bromoform BDCM as a probable human carcinogen weight-of-evidence characterization for given the evidence on mutagenicity and a. Today’s Rule. The final MCLG for DCA in which it evaluated all relevant given there was an increased incidence bromoform is zero. The zero MCLG is health data (both cancer and noncancer of tumors at several sites in the animals. based on a weight-of-evidence effects). The MCLG of zero is based on Additionally, tumors were found in classification that bromoform is a sufficient evidence of carcinogenicity in both sexes of two rodent species. probable human carcinogen based on a animals which indicates that DCA is a Finally, there have been several new consideration of all relevant health data probable human carcinogen (likely studies on the genotoxicity of BDCM including cancer and noncancer effects. under proposed cancer guidelines). EPA that have supported a mutagenic EPA believes the data are insufficient at believes the data are insufficient at this potential for BDCM (EPA, 1997c) this time to determine the mode of time to determine the mode of carcinogenic action for bromoform, and carcinogenic action for DCA and that 3. MCLG for Dibromochloromethane therefore a low dose linear extrapolation the data is insufficient to quantify the (DBCM) approach is used to estimate lifetime potential cancer risk from DCA. a. Today’s Rule. The final MCLG for cancer risk as a default. b. Background and Analysis. EPA DBCM is 0.06 mg/L. This MCLG is b. Background and Analysis. The proposed an MCLG of zero for DCA. based on a weight of evidence proposed MCLG for bromoform was This was based on classifying DCA as a evaluation of the cancer and noncancer zero. This MCLG was based on an NTP probable human carcinogen in data which resulted in the classification chronic animal carcinogenicity study accordance with the 1986 EPA of DBCM as a possible human (NTP, 1989). Since the proposal, new Guidelines for Carcinogen Risk carcinogen. studies on the genotoxicity of Assessment (EPA, 1986). The DCA b. Background and Analysis. In the bromoform were found. However, these categorization was based primarily on 1994 proposal, the MCLG of 0.06 mg/L new studies do not support changing findings of liver tumors in rats and for DBCM was based on observed liver the proposed MCLG of zero for mice, which was regarded as toxicity from a subchronic study and bromoform. The final MCLG for ‘‘sufficient’’ evidence in animals. No possible carcinogenicity (NTP, 1985). bromoform is therefore zero. lifetime risk calculation was conducted EPA is not aware of any new c. Summary of Comments. Several at the time of the proposal because there information that would change its commenters agreed with EPA’s was insufficient data to quantify the risk evaluation of DBCM since the proposal. classification for bromoform as a (EPA, 1994a). The final MCLG is therefore 0.06 mg/L. probable carcinogen. Other commenters As pointed out in the 1997 and 1998 c. Summary of Comments. Several disagreed with this classification stating DBP NODAs, several toxicological commenters disagreed with the that there was insufficient evidence studies have been identified for DCA additional safety factor of 10 to account available because tumors were found in since the 1994 proposal (EPA, 1997c). In for possible carcinogenicity that was only one species and the increased addition, EPA co-sponsored an ILSI used in the MCLG calculation. One number of tumors was small. These project in which an expert panel was Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69403 convened to explore the application of mutational spectra in oncogenes, alter the original assessment of the the EPA’s 1996 Proposed Guidelines for elevated serum glucocorticoid levels, health effects of TCA based on Carcinogen Risk Assessment (EPA, alterations in cell replication and developmental toxicity and limited 1996b) to the available data on the death). EPA considers the mode of evidence of carcinogenicity. Therefore, potential carcinogenicity of chloroform action through which DCA induces liver the MCLG will remain 0.3 mg/L. and DCA. The panel considered data on tumors in both rats and mice to be c. Summary of Comments. Several DCA which included chronic rodent unclear, and thus the likelihood of commenters agreed with the bioassay data and information on human hazard associated with low classification of TCA as a possible mutagenicity, tissue toxicity, levels of DCA usually encountered in human carcinogen. One commenter felt toxicokinetics, and other mode of action the environment or in drinking water is that toxicity data on TCA indicated a information. The panel concluded that not sufficiently understood. EPA threshold. Some commenters disagreed the potential human carcinogenicity of acknowledges that a mutagenic with the study selected for estimating DCA ‘‘cannot be determined’’ primarily mechanism (i.e., direct DNA reactivity) the RfD (Smith et al. 1989). Some because of the lack of adequate rodent may not be an important influence on commenters stated the uncertainty bioassay data (ILSI, 1997). the carcinogenic process at low doses. factors used to establish the RfD were EPA prepared a new hazard EPA believes that the lack of too high. characterization regarding the potential mutagenicity is not a sufficient basis to In response, EPA acknowledges that a carcinogenicity of DCA in humans depart from a low dose linear default DNA reactive mutagenic mechanism (EPA, 1998b). One objective of this extrapolation approach for the cancer may not be involved in TCA’s mode of report was to develop a weight-of- assessment. There must be other carcinogenicity. Because an RfD was evidence characterization using the convincing evidence to explain how the used in lieu of a quantitative cancer principles of the EPA’s 1996 Proposed tumors are caused by the chemical. The assessment for establishing the MCLG, Guidelines for Carcinogen Risk commenters have not presented such however, there was no need to evaluate Assessment (EPA, 1996b) which are evidence. Although DCA tumor effects the mode of carcinogenic action for TCA consistent with the 1986 Guidelines. are associated with high doses used in at this time. EPA believes that the Smith Another objective of the report was to the rodent bioassays, there is et al. (1989) study is appropriate to use consider new data since the 1994 uncertainty regarding whether the mode in quantifying risk from TCA since proposal and to address the issues of tumorgenesis is solely through developmental toxicity was the most raised by the 1997 ILSI panel report. mechanisms that are operative only at critical effect. EPA believes that an EPA agreed with the ILSI panel report high doses. Therefore, as in the 1994 uncertainty factor of 3,000 is that the mode of action through which proposed rule, EPA believes that the appropriate to account for inter and DCA induces liver tumors in both rats MCLG for DCA should remain as zero to intraspecies differences (100), a lowest and mice cannot be reasonably assure public health protection. NTP is observed adverse effects level (LOAEL) determined at this time. EPA disagrees implementing a new two year rodent (10), and lack of a two-generation with the ILSI panel that the potential bioassay that will include full reproductive study (3) (EPA, 1994a). human carcinogenicity cannot be histopathology at lower doses than These uncertainty factors are consistent determined. Based on the those previously studied. Additionally, with current Agency science policy on hepatocarcinogenic effects of DCA in studies on the mode of carcinogenic using uncertainty factors (EPA, 1994a). both rats and mice in multiple studies, action are being done by various 7. MCLG for Chlorite and MRDLG for as well as other date, for example, investigators including the EPA health Chlorine Dioxide showing that DCA alters cell replication research laboratory. and gene expression, EPA concludes a. Today’s Rule. The final MCLG for that DCA should be considered as a 6. MCLG for Trichloroacetic Acid (TCA) chlorite is 0.8 mg/L and the final ‘‘likely’’ (probable) cancer hazard to a. Today’s Rule. The final MCLG for MRDLG for chlorine dioxide is 0.8 mg/ humans (EPA, 1998b). Therefore, as in TCA is 0.3 mg/L, as was proposed in L. The MCLG for chlorite was increased the 1994 proposed rule, EPA believes 1994. This MCLG is based on from the proposed value of 0.08 mg/L to that the MCLG for DCA should remain developmental toxicity and limited 0.8 mg/L based on a weight-of-evidence zero to assure public health protection. evidence of carcinogenicity in animals. evaluation of all health data on chlorite c. Summary of Comments. Some b. Background and Analysis. The including a recent two-generation commenters agreed with the zero MCLG 1994 proposed rule included a MCLG of reproductive rat study sponsored by the for DCA based on positive carcinogenic 0.3 mg/L for TCA based on Chemical Manufactures Association findings in two animal species. Several developmental toxicity and possible (CMA, 1996). The MRDLG for chlorine commenters stated that a zero MCLG carcinogenicity based on limited dioxide was increased from the was inappropriate due to evidence evidence in animal studies (i.e., proposed value of 0.3 mg/L to 0.8 mg/ which indicates a nongenotoxic mode of hepatocarcinogenicity in mice). Since L based on a weight-of-evidence action for DCA. The comment was the proposal, a 2-year carcinogenicity evaluation using all the health data on raised that the animal evidence was study on TCA (DeAngelo et al., 1997) chlorine dioxide including the insufficient to consider DCA a likely found that TCA was not carcinogenic in information on chlorite from the CMA (probable) human carcinogen, and that male rats. As was discussed in the 1997 study. EPA believes that data on chlorite DCA should be considered at most DBP NODA (EPA, 1997b), there have are relevant to assessing the risks of suggestive of carcinogenicity. also been several recent studies chlorine dioxide because chlorine In response, EPA concludes that DCA examining the mode of carcinogenic dioxide is rapidly reduced to chlorite. should be considered as a probable action for TCA. These new studies Therefore, the findings from the CMA (likely under the 1996 proposed suggest that TCA does not operate via study and previously described studies guidelines) cancer hazard to humans mutagenic mechanisms. For a more in in the 1994 proposal were used to assess (EPA, 1998b) based on the depth discussion of this new data refer the risk for both chlorite and chlorine hepatocarcinogenic effects of DCA in to the 1997 DBP NODA (EPA, 1997b) dioxide. both rats and mice in multiple studies, and related support documents (EPA, b. Background and Analysis. The and mode of action related effects (e.g., 1997c). This new information does not 1994 proposal included an MCLG of 69404 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

0.08 mg/L for chlorite. The proposed appropriate. EPA has concluded that the subsequent information submitted by MCLG was based on an RfD of 3 mg/kg/ RfD for chlorine dioxide should be 0.03 the CMA provided clarification on d estimated from a lowest-observed- mg/L (NOAEL of 3 mg/kg/d with an certain aspects of the study design adverse-effect-level (LOAEL) for uncertainty factor of 100) and that a (CMA, 1998). EPA agrees that even with neurodevelopmental effects identified MRDLG of 0.8 mg/L is appropriate. the clarifications that there are some in a rat study by Mobley et al. (1990). c. Summary of Comments. EPA limitations with the This determination was based on a received numerous comments on the neurodevelopmental component of the weight of evidence evaluation of all the 1994 proposal (EPA, 1994a) and 1998 CMA study. EPA believes that the available data at that time (EPA, 1994d). NODA (EPA, 1998a). The major neuropathology components of the CMA An uncertainty factor of 1000 was used comment from the 1994 proposal was study were adequate. The functional to account for inter-and intra-species that reliance on the Mobley et al. (1990) operation battery had some differences in response to toxicity (a study for the MCLG for chlorite and the shortcomings in that forelimb and factor of 100) and to account for use of Orme et al. (1985) study for chlorine hindlimb grip strength and foot splay a LOAEL (a factor of 10). dioxide were inappropriate and that the were not evaluated. EPA believes the The 1994 proposal included an results from the CMA study must be results from the motor activity MRDLG of 0.3 mg/L for chlorine evaluated before any conclusions on the component of the CMA study were dioxide. The proposed MRDLG was MCLG for chlorite or chlorine dioxide difficult to interpret because of the high based on a RfD of 3 mg/kg/d estimated could be drawn. In relation to the 1998 variability in controls. However, in its from a no-observed-adverse-effect-level NODA, several commenters supported evaluation of the MCLG for chlorite and (NOAEL) for developmental changing the MCLG for chlorite and chlorine dioxide, EPA did not rely neurotoxicity identified from a rat study MRDLG for chlorine dioxide while solely on the CMA study, but used a (Orme et al., 1985; EPA, 1994d). This others were concerned that the science weight-of-evidence approach that determination was based on a weight of did not warrant a change in these included consideration of several evidence evaluation of all available values. The major comments submitted studies. Thus, the shortcomings of one health data at that time (EPA, 1994a). against raising the MCLG and MRDLG study are offset by the weight from other An uncertainty factor of 300 was focused on several issues. First, one studies. EPA believes that the CMA applied that was composed of a factor commenter argued that the 1000-fold study contributes to the weight-of the- of 100 to account for inter-and intra- uncertainty factor used for chlorite in evidence. The studies by Orme et al. species differences in response to the proposal should remain in place (1985), Mobley et al. (1990), and CMA toxicity and a factor of 3 for lack of a because the CMA study used to reduce (1996) support a NOAEL of 3 mg/kg/d two-generation reproductive study the uncertainty factor was flawed. based on neurodevelopmental effects necessary to evaluate potential toxicity Second, several commenters indicated (e.g., decreased exploratory, locomotor associated with lifetime exposure. To that the LOAEL should be set at the behavior, decreased brain weight). fill this important data gap, the CMA lowest dose level (35 ppm) because Furthermore, the CMA study was sponsored a two-generation certain effects at the lowest dose tested reviewed by outside scientists as well as reproductive study in rats (CMA, 1996). may have been missed. Finally, some by EPA scientists. EPA’s re-assessment As described in more detail in the commenters argued that an additional for chlorite and chlorine dioxide 1998 NODA (EPA, 1998a), EPA safety factor should be included to presented in the 1998 March NODA was reviewed the CMA study and completed protect children and drinking water reviewed internally and externally in an external peer review of the study consumption relative to the body weight accordance with EPA peer-review (EPA, 1997d). In addition, EPA of children should be used instead of policy. The three outside experts who reassessed the noncancer health risk for the default assumption of 2 L per day reviewed the Agency’s assessment chlorite and chlorine dioxide and 70 kg adult body weight. considering the new CMA study (EPA, EPA agrees with commenters on the agreed with the NOAEL of 3 mg/kg/day 1998d). This reassessment was also peer 1994 proposal that the results from the and the derived RfD. reviewed (EPA, 1998d). Based on this CMA should be factored into any final Finally, EPA disagrees that an reassessment, EPA requested comment decision on the MCLG for chlorite and additional safety factor should be in the 1998 NODA (EPA, 1998a) on chlorine dioxide. As explained in more applied to provide additional protection changing the proposed MCLG for detail in the 1998 DBP NODA (EPA, for children or that drinking water chlorite from 0.08 mg/L to 0.8 mg/L 1998a), EPA considered the findings consumption relative to the body weight based on the NOAEL identified from the from the CMA study along with other of children should be used in new CMA study which reinforced the available data to reach its conclusions developing the MCLG. The MCLG and concern for neurodevelopmental effects regarding the MCLG and MRDLG for MRDLG presented for chlorite and associated with short-term exposures. chlorite and chlorine dioxide. chlorine dioxide are considered to be EPA determined that the NOAEL for EPA disagrees with the commenter protective of susceptible groups, chlorite should be 35 ppm (3 mg/kg/d who suggested that the 1000-fold including children, given that the RfD is chlorite ion, rounded) based on a uncertainty factor for chlorite should based on a NOAEL derived from weight-of-evidence approach. The data remain because the CMA study was developmental testing, which includes a considered to support the NOAEL are flawed. The study design for the two-generation reproductive study. A summarized in EPA (1998d) and neurodevelopmental component of the two-generation reproductive study included the CMA study as well as CMA study was in accordance with evaluates the effects of chemicals on the previous reports on developmental EPA’s testing guidelines at the time the entire developmental and reproductive neurotoxicity and other adverse health study was initiated. EPA had previously life of the organism. Additionally, effects (EPA, 1998d). EPA continues to reviewed the study protocol for the current methods for developing RfDs are believe, as stated in the 1998 NODA CMA neurotoxicity component and had designed to be protective for sensitive (EPA, 1998a), that the RfD for chlorite approved the approach. While EPA populations. In the case of chlorite and should be 0.03 mg/kg/d (NOAEL of 3 initially had some questions regarding chlorine dioxide a factor of 10 was used mg/kg/d with an uncertainty factor of the design of the neurodevelopmental to account for variability between the 100) and that a MCLG of 0.8 mg/L is component of the study (Moser, 1997), average human response and the Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69405 response of more sensitive individuals. uncertainty factors for chlorite and all groups, including fetuses and In addition, the important exposure is chlorine dioxide. EPA believes the CMA children. that of the pregnant and lactating female provided an acceptable two-generation The MCLG for chlorite is based on an and the nursing pup. The 2 liter per day study with which to reduce the RfD of 0.03 mg/kg/d using a NOAEL of water consumption and the 70 kg body uncertainty factors. In addition, EPA 3 mg/kg/d and an uncertainty factor of weight assumptions are viewed as believes potential health concerns in the 100 to account for inter- and intra- adequately protective of all groups. proposal with having a MCLG for species differences. The MCLG for Based on a review of all the data and chlorite significantly below the MCL are chlorite is calculated to be 0.8 mg/L by public comments, EPA believes that the no longer relevant because the MCL for assuming an adult tap water MCLG for chlorite should be 0.8 mg/L chlorite in today’s rule will remain at consumption of 2 L per day for a 70 kg and the MRDLG for chlorine dioxide 1.0 mg/L while the MCLG has been adult and using a relative source should be 0.8 mg/L. EPA believes the revised to 0.8 mg/L. Given the margin of contribution of 80% (because most MCLG and MRDLG are consistent with exposure to chlorite is likely to come safety that is factored into the the discussions during the regulatory from ingestion of drinking water— estimation of the MCLG of 0.8 mg/L, negotiations which recognized the need EPA,1998u). A more detailed discussion for an acceptable two-generation EPA believes that the MCL of 1.0 mg/ of this assessment is included in the reproductive study prior to reducing the L will be protective of public health of public docket for this rule (EPA, 1998d).

0.03 mg/kg/d × 70 kg × 0.8 MCLG for chlorite = = 0./L 84 mg 2L/day MCLG for chlorite = 0.8 mg/L (Rounded)

For chlorine dioxide the MCLG is calculated to be 0.8 mg/L. This MRDLG come from ingestion of drinking water— based on a NOAEL of 3 mg/kg/d and takes into account an adult tap water EPA, 1998u). A more detailed applying an uncertainty factor of 100 to consumption of 2 L per day for a 70 kg discussion of this assessment is account for inter-and intra-species adult and applies a relative source included in the public docket for this differences in response to toxicity, the contribution of 80% (because most rule (EPA, 1998d). revised MRDLG for chlorine dioxide is exposure to chlorine dioxide is likely to

0.03 mg/kg/d × 70 kg × 0.8 MRDLG for chlorine dioxide = = 0./L 84 mg 2L/day MRDLG for chlorine dioxide = 0.8 mg/L (Rounded)

8. MCLG for Bromate associated with bromate exposure (EPA, EPA Guidelines for Carcinogen Risk a. Today’s Rule. The final MCLG for 1998e), had this reassessment peer Assessment (EPA, 1986) and the draft bromate is zero. The zero MCLG is reviewed (EPA, 1998e), and presented 1996 EPA Proposed Guidelines for based on a weight-of-evidence its findings in the March 1998 NODA Carcinogen Risk Assessment (EPA, evaluation of both the cancer and (EPA, 1998a). The new rodent cancer 1996b) weight-of-evidence approach, noncancer effects which indicates there study by DeAngelo et al. (1998) bromate is considered to be a probable is sufficient laboratory animal data to contributes to the weight of the or likely human carcinogen. This weight conclude that bromate is a probable evidence for the potential human of evidence conclusion of potential (likely under the 1996 proposed cancer carcinogenicity of potassium bromate human carcinogenicity is based on guidelines) human carcinogen. EPA and confirms the study by Kurokawa et sufficient experimental findings that believes the data are insufficient at this al. (1986 a,b). include the following: tumors at time to determine the mode of c. Summary of Comments. Several multiple sites in rats; tumor responses carcinogenic action for bromate, and commenters supported the zero MCLG in both sexes; and evidence for therefore a low dose linear extrapolation for bromate. Others believed the MCLG mutagenicity including point mutations approach is used to estimate lifetime of zero was not justified because there and chromosomal aberrations in in vitro cancer risk as a default. is evidence of a carcinogenic threshold. genotoxicity assays. Furthermore, EPA b. Background and Analysis. The This evidence indicates that bromate believes there is insufficient evidence at 1994 proposed rule included a MCLG of causes DNA damage indirectly via lipid this time to draw conclusions regarding zero for bromate based on a peroxidation, which generates oxygen the mode of carcinogenic action for determination that bromate was a radicals which in turn induce DNA bromate. EPA acknowledges there are probable human carcinogen. This damage. Other commenters argued that studies available showing that bromate determination was based on results from even if there is no carcinogenic may generate oxygen radicals which a two species rodent bioassay by threshold, EPA has overstated the increase lipid peroxidation and damage Kurokawa et al. (1986a and 1986b) that potency of bromate by using the DNA. However, no data are available found kidney tumors in rats. Since the linearized multistage model and should that link this proposed mechanism to 1994 proposed rule, EPA has completed instead use the Gaylor-Kodell model. tumor induction. Thus, EPA believes and analyzed a new chronic cancer In response, EPA disagrees with that while there are studies which study in male rats and mice for commenters who believed that the zero provide some evidence to support the potassium bromate (DeAngelo et al., MCLG was inappropriate. At this time, commenters’ claims, these studies are 1998). EPA reassessed the cancer risk under the principles of both the 1986 insufficient at this time to establish 69406 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations lipid peroxidation and free radical are substantially greater than those commenters agreed with the uncertainty production as key events responsible for present in drinking water. Commenters factor of 100 while others felt that it was the induction of the multiple tumor also questioned the need for an MCLG too high. Some commenters encouraged responses seen in the bromate rodent for chloral hydrate because the MCLs for EPA to consider children in estimating bioassays (EPA, 1998e). Given the THMs and HAAs and the treatment risk from chlorine. uncertainty about the mode of technique of enhanced coagulation will In response, EPA believes that an carcinogenic action for bromate, EPA adequately control for chloral hydrate uncertainty factor of 100 is appropriate believes it is appropriate to use the and because there were no monitoring when a NOAEL from a chronic animal default assumption of low dose linearity provisions proposed. Other commenters study is the basis for the RfD. Because to estimate the cancer risk and establish argued that the use of a 10,000 current methods for developing RfDs are the MCLG of zero for bromate. EPA is uncertainty factor and the selection of designed to be protective for sensitive conducting additional studies the Sanders et al. (1982) study as a basis subpopulations, the uncertainty factor investigating the mode of action for for setting the MCLG were of 100 is considered protective of bromate. inappropriate. children. Furthermore, animal studies EPA also disagrees with commenters In response, EPA agrees with indicate that chlorine is not a who suggested that the Gaylor-Kodell commenters that an MCLG for chloral developmental toxicant. model should be used for low-dose hydrate is not needed. This is based on extrapolation of the bromate data. In the the fact that the TTHM and HAA MCLs 11. MRDLG for Chloramine 1998 NODA, a low dose linear and the treatment technique (i.e., a. Today’s Rule. EPA is promulgating extrapolation of the DeAngelo et al. enhanced coagulation/softening) will an MRDLG of 4 mg/L for chloramines (1998) data was conducted using a one- control for chloral hydrate, as well as based on a NOAEL from a chronic stage Weibull time-to-tumor model. The other chlorination byproducts. In rodent study. Weibull model was considered to be the addition, chloral hydrate does not serve b. Background and Analysis. The preferred approach to account for the as an important indicator for other 1994 proposed Stage I DBPR included reduction in animals at risk that may be chlorination byproducts. The final rule, an MRDLG for chloramines at 4 mg/L due to the decreased survival observed therefore, does not contain an MCLG for based on a NOAEL of 9.5 mg/kg/d for in the high dose group toward the end chloral hydrate. In light of this decision, lack of toxicity in chronic rodent of the study. The estimate of cancer risk EPA is not responding to comments on drinking water study and on application from the DeAngelo et al. (1998) study is the uncertainty factor used as the basis of an uncertainty factor of 100 to similar with the risk estimate derived for setting the MCLG. account of inter- and intra-species from the Kurokawa et al. (1986a) study 10. MRDLG for Chlorine differences (EPA, 1994h). New presented in the 1994 proposed rule. information on chloramines has become Based on an evaluation of all the data a. Today’s Rule. EPA is promulgating available since the 1994 proposal and and after review and consideration of an MRDLG of 4 mg/L for chlorine based was included in the 1997 DBP NODA the public comments, EPA believes the on a NOAEL from a chronic study in and is included in the public docket MCLG for bromate should be zero. animals. (EPA, 1997c). This new information did b. Background and Analysis. EPA not contain data that would change the 9. MCLG for Chloral Hydrate proposed an MRDLG of 4 mg/L for MRDLG. EPA has therefore decided to a. Today’s Rule. EPA has decided to chlorine. The MRDLG was based on a finalized the proposed MRDLG of 4 mg/ not include an MCLG for chloral two-year rodent drinking water study in L for chloramines. hydrate in the Stage 1 DBPR. This which chlorine was given to rats at c. Summary of Comments. Several decision is based on an analysis of the doses ranging from 4 to 14 mg/kg/day commenters agreed with the MRDLG of technical comments and on the fact that and mice at doses ranging from 8 to 24 4 mg/L for chloramine (as chlorine). chloral hydrate will be controlled by the mg/kg/day (NTP, 1990). Neither Some commenters felt that the MRDLG MCLs for TTHM and HAAs and by the systemic toxicity, nor effects on body was too low due to conservative treatment technique of enhanced weight and survival were found. Thus, uncertainty factors. Many commenters coagulation. the MRDLG was based on a NOAEL of agreed with EPA’s conclusion that there b. Background and Analysis. The 14 mg/kg/day and application of a 100 is no animal evidence of carcinogenicity 1994 proposed rule included an MCLG fold uncertainty factor to account for for chloramines. Many commenters for chloral hydrate of 0.04 mg/L. This inter- and intra-species differences agreed with the RSC of 80% for was based on a 90-day mice study by (EPA, 1994a). New information on chloramine while other believed that Sanders et al. (1982) which reported chlorine has become available since the the RSC should be higher. liver toxicity. A RfD of 0.0016 mg/kg/d 1994 proposal and was discussed in the In response, EPA believes that the was used (LOAEL of 16 mg/kg/d with an 1997 DBP NODA and is included in the uncertainty factor of 100 in the MRDLG uncertainty factor of 10,000). In the public docket (EPA, 1997c). This new calculation is appropriate to protect 1997 DBP NODA (EPA,1997b) and information did not contain data that public health including that of children supporting documents (EPA, 1997c), would change the MRDLG. EPA has and sensitive subpopulations. EPA additional studies on chloral hydrate therefore decided to finalize the believes that the 80 percent is an were discussed, however, these new proposed MRDLG of 4 mg/L for appropriate ceiling for the RSC due to studies did not indicate a change in the chlorine. lack of exposure data on other sources MCLG for chloral hydrate. c. Summary of Comments. Several c. Summary of Comments. The commenters agreed with EPA’s of exposure. majority of commenters disagreed with conclusion that there is no animal B. Epidemiology the MCLG of 0.04 mg/L for chloral evidence of carcinogenicity for chlorine. hydrate. Several commenters questioned Some commenters also agreed with EPA 1. Cancer Epidemiology the need for an MCLG for chloral that 4 mg/L was the appropriate MCLG. a. Today’s Rule. EPA has evaluated all hydrate. These commenters mentioned Several commenters agreed with the of the cancer epidemiology data and the its low toxic potential and the fact that proposed relative source contribution of corresponding public comments safe concentrations of chloral hydrate 80 percent for chlorine. Some received on the 1994 proposal (EPA, Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69407

1994a), 1997 NODA (EPA, 1997b), and and cancer. Other commenters concern in light of the uncertainties and 1998 NODA (EPA, 1998a). Based on this disagreed with this characterization given the large population (over 200 evaluation, EPA believes that the cancer stating that they believed the evidence million people) potentially exposed. epidemiology data provides important did indicate there was a strong 2. Reproductive and Developmental information that contributes to the association between exposure to Epidemiology weight-of-evidence evaluation on the chlorinated water and cancer. Other potential health risks from exposure to commenters stated that EPA had not a. Today’s Rule. EPA has evaluated all chlorinated drinking water. At this time, clearly articulated the basis for its of the reproductive and developmental however, the cancer epidemiology conclusions on the issue of causality. epidemiology data and the public studies are insufficient to establish a In response, EPA continues to believe comments received on the 1994 causal relationship between exposure to that there is insufficient evidence, based proposal, 1997 NODA, and the 1998 chlorinated drinking water and cancer; on the epidemiology data, to conclude NODA. Based on this evaluation, EPA and are thus considered limited for use there is a causal association between believes that the reproductive and in quantitative risk assessment. EPA’s exposure to chlorinated waters and developmental epidemiology data weight-of-evidence evaluation of the cancer. EPA agrees, however, that the provides important information that potential risk posed by chlorinated basis for its conclusion on causality was contributes to the weight-of-evidence drinking water is further discussed in not clearly articulated. This judgement evaluation on the potential risks from section IV of this preamble. of causality was based on evaluating the exposure to chlorinated drinking water. b. Background and Analysis. The existing cancer epidemiologic database However, the reproductive preamble to the 1994 proposed rule for the following criteria: strength of epidemiology studies are insufficient to discussed numerous cancer association, consistency of the findings, establish a causal relationship between epidemiology studies that had been specificity of the association, as well as exposure to chlorinated drinking water conducted over the past 20 years to other information concerning the and reproductive and developmental examine the relationship between temporal sequence and presence of a effects and are limited for use in the exposure to chlorinated water and dose-response relationship, and quantification of risk. cancer (EPA, 1994a). At the time of the biological plausibility (Federal Focus, b. Background and Analysis. In the regulatory negotiation, there was 1996; EPA, 1986; EPA 1996b). preamble to the 1994 proposed DBPR, disagreement among the members of the EPA applied the criteria stated above EPA discussed several reproductive Reg. Neg. Committee on the conclusions to assess the possible causality of cancer epidemiology studies (EPA, 1994a). At that could be drawn from these studies. using the best available cancer the time of the proposal, EPA concluded Some members of the Committee felt epidemiology studies (Cantor et al., that there was no compelling evidence that the cancer epidemiology data, taken 1985, McGeehin et al., 1993, King and to indicate a reproductive and in conjunction with the results from Marrett, 1996, Cantor et al., 1998, developmental hazard due to exposure toxicological studies, provide ample and Freedman et al., 1997, Hildesheim et al., to chlorinated water because the sufficient weight-of-evidence to 1998, Doyle et al., 1997). These studies epidemiologic evidence was inadequate conclude that exposure to DBPs in found a weak association for bladder and the toxicological data were limited. drinking water could result in increased cancer, although the findings were not In 1993, an expert panel of scientists cancer risk at levels encountered in consistent within and among the was convened by the International Life some public water supplies. Other studies. The specificity of the Sciences Institute to review the members of the Committee concluded association, temporal association, and available human studies for that the cancer epidemiology studies on dose response relationship remain developmental and reproductive the consumption of chlorinated unknown. In addition, the biological outcomes and to provide research drinking water to date were insufficient mode of action has not been recommendations (EPA/ILSI, 1993). The to provide definitive information for the determined. Using the criteria for expert panel concluded that the regulation. causality, the present epidemiologic epidemiologic results should be In the 1998 DBP NODA (EPA, 1998a), data do not support a causal considered preliminary given that the EPA discussed several new relationship between exposure to research was at a very early stage (EPA/ epidemiology studies that had been chlorinated drinking water and ILSI, 1993; Reif et al., 1996). The 1997 published since the 1994 proposal. EPA development of cancer at this time. This NODA and the supporting documents concluded in the 1998 NODA, based on conclusion does not preclude the (EPA, 1997c) presented several new a review of all the cancer epidemiology possibility that a causal link may be studies (Savitz et al., 1995; Kanitz et al. studies (including the more recent established at a later date by future 1996; and Bove et al., 1996) that had studies), that a causal relationship epidemiology and toxicology studies. been published since the 1994 proposed between exposure to chlorinated surface Some commenters argued that the rule and the 1993 ILSI panel review. water and cancer has not yet been epidemiological evidence indicated an Based on the new studies presented in demonstrated. However, several studies increased risk for cancer by exposure to the 1997 NODA, EPA stated that the have suggested a weak association in chlorinated drinking water, while others results were inconclusive with regard to various subgroups. Results from recent argued that the epidemiological the association between exposure to epidemiology studies continue to evidence does not support a health chlorinated waters and adverse support the decision to pursue effects concern. As stated above, EPA reproductive and developmental effects regulations to provide additional DBP believes that, at this time, a causal link (EPA, 1997b). control measures as discussed in section between exposure to chlorinated In the 1998 DBP NODA (EPA, 1998a), IV.D of this preamble. drinking water and development of EPA included the recommendations c. Summary of Comments. Several cancer cannot be determined. However, from an EPA convened expert panel in commenters agreed with EPA’s EPA believes that the epidemiological July 1997 to evaluate epidemiologic characterization that there was evidence suggests a potential increased studies of adverse reproductive or insufficient evidence to conclude that risk for bladder cancer. It is therefore developmental outcomes that may be there was a causal relationship between prudent public health policy to protect associated with the consumption of exposure to chlorinated surface water against this potential public health disinfected drinking water published 69408 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations since the 1993 ILSI panel review. A from exposure to DBPs in drinking also introduced the MRDL for the three report was prepared entitled ‘‘EPA water. disinfectants (chlorine, chloramines, Panel Report and Recommendations for In response, EPA continues to believe and chlorine dioxide). The MRDLs are Conducting Epidemiological Research that the available epidemiology data enforceable standards, analogous to on Possible Reproductive and along with the toxicological findings MCLs, which recognize the benefits of Developmental Effects of Exposure to suggest that exposure to DBPs may have adding a disinfectant to water on a Disinfected Drinking Water’’ (EPA, adverse effects on humans. However, continuous basis and to maintain a 1998f). The 1997 expert panel was also EPA believes the epidemiology evidence residual to control for pathogens in the charged to develop an agenda for further is insufficient at this time to conclude distribution system. As with MCLs, EPA epidemiological research. The 1997 that there is a causal association has set the MRDLs as close to the panel concluded that the results of between exposure to DBPs and adverse MRDLGs as feasible. The Agency has several studies suggest that an increased reproductive and developmental effects. also identified the BAT which is relative risk of certain adverse outcomes As noted in the 1998 NODA, EPA has feasible for meeting the MRDL for each may be associated with the type of water an epidemiology and toxicology disinfectant. source, disinfection practice, or THM research program that is examining the EPA received similar comments on levels. The panel emphasized, however, relationship between exposure to DBPs the use of the term MRDL as with that most relative risks are moderate or and adverse reproductive and MRDLG. The majority of commenters small and were found in studies with developmental effects. In addition, EPA agreed with the use of the term MRDL limitations in design or conduct. The is pursuing appropriate follow-up for the disinfectants and therefore EPA small magnitude of the relative risk studies to see if the observed association is using the term MRDL in the final rule. in the Waller et al. (1998) study can be found may be due to one or more 1. MCLs for TTHMs and HAA5 sources of bias, as well as to residual replicated elsewhere. EPA will also be a. Today’s Rule. In today’s rule, EPA confounding (factors not identified and working with the California Department is promulgating an MCL for TTHMs of controlled). Additional research is of Health Services to improve estimates 0.080 mg/L. TTHM is the sum of needed to assess whether the observed of exposure to DBPs in the existing measured concentrations of chloroform, associations can be confirmed. In Waller et al. study population. EPA will bromodichloromethane, addition, the 1998 DBP NODA included collaborate with the Centers for Disease dibromochloromethane, and a summary of a study by Waller et al. Control and Prevention (CDC) in a series bromoform. EPA is also promulgating an (1998) conducted in California and of studies to evaluate if there is an MCL for HAA5 of 0.060 mg/L. HAA5 is another study by Klotz and Pyrch (1998) association between exposure to DBPs the sum of measured concentrations of conducted in New Jersey. EPA in drinking water and birth defects. EPA is also involved in a collaborative mono-, di-, and trichloroacetic acids, concluded that while the Waller et al. and mono- and dibromoacetic acids. A (1998) study does not prove that testing program with the NTP under which several individual DBPs have system is in compliance with these exposure to THMs in drinking water MCLs when the running annual average causes early term miscarriages, it does been selected for reproductive and developmental laboratory animal of quarterly averages of all samples provide important new information that taken in the distribution system, needs to be explored and that the study studies. This information will be used in developing the Stage 2 DBPR. computed quarterly, is less than or adds to the weight-of-evidence which equal to the MCL. If the running annual suggests that exposure to DBPs may C. MCLs and BAT for TTHM, HAA5, average computed for any quarter have an adverse health effect on Chlorite, and Bromate; MRDLs and BAT exceeds the MCL, the system is out of humans. EPA indicated that the review for Chlorine, Chloramines, and Chlorine compliance. EPA believes that by of the Klotz and Pyrch study (1998) had Dioxide meeting MCLs for TTHMs and HAA5, not been completed in time for the 1998 MCLs are enforceable standards water suppliers will also control the NODA. which are established as close to the formation of other DBPs not currently EPA has completed its review of the MCLG as feasible. Feasible means with regulated that may also adversely affect Klotz and Pyrch (1998) study and the use of the best technology, treatment human health. concluded that the results in the report techniques, and other means which the EPA has identified the best available provide limited evidence to substantiate Administrator finds available (taking (BAT) technology for achieving the hypothesis that DBPs in drinking costs into consideration) after compliance with the MCLs for both water cause adverse reproductive or examining for efficacy under field TTHMs and HAA5 as enhanced developmental effects since the bulk of conditions and not solely under coagulation or treatment with granular the findings are inconclusive. There is, laboratory conditions. activated carbon with a ten minute however, a suggestion in the study that EPA is promulgating MCLs for two empty bed contact time and 180 day total THMs or some other component of groups of DBPs and two inorganic reactivation frequency (GAC10), with surface water is associated with a small byproducts. EPA is also promulgating chlorine as the primary and residual increased risk of neural tube defects; no MRDLs for three disinfectants. EPA is disinfectant, as was proposed in 1994. significant associations, however, were promulgating these MCLs and MRDLs at b. Background and Analysis. The observed with individual THMs, HAAs the levels proposed in 1994. Systems 1994 proposal for the Stage 1 DBPR or other composite measures of will determine compliance with the included MCLs for TTHM and HAA5 at exposure. MCLs and MRDLs in the same manner 0.080 and 0.060 mg/L, respectively c. Summary of Comments. Several as was proposed in 1994, except for (EPA, 1994a). In addition to the commenters agreed with EPA’s chlorite. EPA determined that proposed MCLs, subpart H systems— conclusions on the significance of the additional monitoring requirements for utilities treating either surface water or reproductive and developmental effects chlorite were necessary based on the groundwater under the direct influence from the various studies. Others findings from the CMA two-generation of surface water—that use conventional believed EPA had not accurately reproductive and developmental study. treatment (i.e., coagulation, characterized the potential adverse Along with introducing the concept of sedimentation, and filtration) or reproductive and developmental effects the MRDLG in the proposed rule, EPA precipitative softening would be Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69409 required to remove DBP precursors by 1998g). EPA continues to believe the (although the Agency has an adequate enhanced coagulation or enhanced proposed MCLs are achievable without database of group occurrence). Second, softening. The removal of TOC would be large-scale technology shifts. there is not an adequate understanding used as a performance indicator for DBP c. Summary of Comments. Several of how water quality parameters (such precursor control. commenters questioned whether the as pH, temperature, bromide, and As part of the proposed rule, EPA TTHM MCL of 0.080 mg/L and the alkalinity) affect individual THM and estimated that 17% of PWSs would HAA5 MCL of 0.060 mg/L were set at a HAA formation. Third, EPA does not need to change their treatment process level that would preclude the use of have an adequate understanding of how to alternative disinfectants (ozone or chlorine as an effective disinfectant. treatment technologies control the chlorine dioxide) or advanced precursor EPA does not believe the MCLs will formation of individual THMs and removal (GAC or membranes) in order preclude the use of chlorine. While HAAs to enable specifying appropriate to comply with the Stage 1 there are currently systems that are MCLs for individual TTHMs or HAAs at requirements. This evaluation was exceeding these MCLs, the Agency has this time. Finally, there are inadequate important to assist in determining concluded that most systems will be health data to characterize the potential whether the proposed MCLs were able to achieve compliance by relatively health risks for several of the HAAs and achievable and at what cost. This low cost alternatives such as: improved to then determine the potential benefits evaluation required an understanding of DBP precursor removal through from reduction in exposures. In the baseline occurrence for the DBPs enhanced coagulation or enhanced conclusion, EPA continues to believe and TOC being considered in the Stage softening; moving the point of the most appropriate approach for 1 DBPR, an understanding of the disinfection to reduce the reaction reducing the health risk from all DBPs baseline treatment in-place, and an between chlorine and DBP precursors; is by the combination of TTHM and estimation of what treatment the use of chloramines for residual HAA5 MCLs and DBP precursor technologies systems would use to disinfection instead of chlorine; or a removal. comply with the Stage 1 DBPR combination of these alternatives. Some commenters stated that EPA requirements. Many commenters also questioned the may have underestimated HAA In 1997, at the direction of the M–DBP need for a modified TTHM MCL and a formation, especially in certain areas of Advisory Committee, the TWG reviewed new MCL for HAA5. As discussed in the country. The Agency was aware that MCL compliance predictions developed section I.B.2. of today’s rule, EPA waters in particular regions of the for the 1994 proposal because of believes the potential public health risks country would be more difficult to treat concern by several Committee members do justify a reduction in exposure to in order to control for HAA5 than for that modifications to the rule would DBPs and hence a modification in the TTHM. Based on additional data result in more PWSs not being able to MCL for TTHMs and a new MCL for received since the proposal, EPA meet the new TTHM and HAA5 MCLs HAA5. Also as discussed in section IV continues to believe that the HAA5 MCL without installation of higher cost of this rule, EPA continues to believe can be met by most systems through the technologies such as ozone or GAC. that the potential risks associated with same general low-cost strategies as used Some members were concerned that both TTHM and HAA5 and unregulated for TTHM (e.g., improved DBP allowing disinfection inactivation credit DBPs will be reduced by the precursor removal, moving the point of prior to precursor removal (by enhanced combination of these MCLs and DBP disinfection, use of chloramines for coagulation or enhanced softening) in precursor removal through enhanced residual disinfection) rather than higher order to prevent significant reductions coagulation and enhanced softening. cost alternatives (see section IV.C for in microbial protection would result in While most commenters agreed with cost estimates of technology treatment higher DBP formation and force systems EPA’s definition of GAC10 and GAC20 choices). to install alternative disinfectants or (GAC with a 10 and a 20 minute empty Many commenters also requested that advanced precursor removal to meet the bed contact time, respectively), several States be granted sufficient flexibility in 1994 proposed TTHM and HAA5 MCLs. commenters thought that designating implementing this rule. While the State As discussed later in today’s document GAC as BAT meant that they would must adopt rules that are at least as in Section III.E (Preoxidation CT Credit), have to install GAC at their treatment stringent as those published in today’s most PWSs can achieve significant plant. EPA is required to designate a rule, EPA has given the States and reduction in DBP formation through the BAT for any MCL that the Agency systems much latitude in monitoring combination of enhanced coagulation promulgates; however, a system may plans (frequency and location), (or enhanced softening) while use any technology it wants to comply allowable disinfectants, and other rule maintaining predisinfection. The TWG’s with the MCL. However, a system must elements. Much of this flexibility carries analysis indicated that there would be a install BAT prior to the State issuing a over from the 1979 TTHM Rule (EPA, decrease in the percentage of PWSs that variance to one of these MCLs. 1979). would need to install higher cost Commenters also questioned the use Finally, some commenters stated that technologies. This decrease was of group MCLs for TTHM and HAA5, requirements in this rule are attributed to changes in the proposed instead of MCLs for the individual complicated. EPA acknowledges that IESWTR which altered the constraints DBPs, since a group MCL does not take this rule is complicated, but that this by which systems could comply with into account differing health effects and complexity is necessary in order to the MCLs. The requirements of the potencies of individual DBPs. EPA adequately and economically address IESWTR would also prevent significant continues to believe that regulating the potential DBP risks. EPA was reduction in microbial protection as TTHMs and HAAs as group MCLs is required to consider a host of described in the 1997 NODA (EPA, appropriate at this time for several complicating factors in developing 1997a) and elsewhere in today’s Federal reasons. First, EPA does not have regulatory requirements: different Register. EPA has included a discussion adequate occurrence data for individual disinfectants, different health effects of the prediction of technology choices trihalomethanes and haloacetic acids to (acute and chronic), different DBP in Section IV (Economic Analysis) of develop national occurrence estimates formation kinetics, different source today’s rule and a more detailed which are needed for estimating the water types and qualities, different discussion in the RIA for this rule (EPA, potential costs and benefits of the rule treatment processes, and the need for 69410 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations simultaneous compliance with other that would be achieved by improved monitoring is required when the rules such as the Total Coliform Rule, methods. The PQL of the revised chlorite concentration measured at the Lead and Copper Rule, and Interim method is approximately 0.010 mg/L for entrance to the distribution system Enhanced Surface Water Treatment bromate, as discussed in Section III.G exceeds a chlorite concentration of 1.0 Rule. The Agency chose to evaluate all (Analytical Methods). At the time of the mg/L. Distribution system monitoring these factors by developing November 1997 NODA, EPA was not may be reduced if certain conditions are requirements that minimized impacts aware of any new information that met (described in section III.H of this on various classes of systems while would lower the PQL for bromate and rule). enabling States to implement the rule. thus allow lowering the MCL. As a b. Background and Analysis. For In addition to the further description of result, EPA concluded that the proposed systems using chlorine dioxide, EPA the requirements in today’s rule, EPA bromate MCL was appropriate. proposed a separate MCL for chlorite will publish a State implementation c. Summary of Comments. Several associated with its usage in 1994. The manual, a small system compliance commenters were concerned that the proposed chlorite MCL of 1.0 mg/L was manual, and a series of guidance bromate MCL may have been set at a supported by the Reg. Neg. Committee manuals that will provide additional level that would preclude the use of because 1.0 mg/L was the lowest level information to systems and States in ozone. During the M–DBP Advisory considered practicably achievable by implementing this rule. Committee discussions, the TWG typical systems using chlorine dioxide, EPA has reviewed all comments and evaluated the feasibility of ozone for from both treatment and monitoring determined that the requirements certain systems that were predicted to perspectives. The MCLG was 0.08 mg/ promulgated today are necessary to have problems in complying with the L, due (in part) to data gaps that control the occurrence of TTHM and TTHM and HAA5 MCLs. While ozone required higher uncertainty factors in HAA5 and are feasible to achieve. These was not feasible for all systems, it was the MCLG determination. The CMA requirements take into account the feasible for many that did not have agreed to fund new health effects difficulties in simultaneously elevated source water bromide levels to research on chlorine dioxide and controlling risks from DBPs and react with ozone to form bromate. The chlorite—with EPA approval of the pathogens, while appropriately TWG predicted that most of the systems experimental design—to resolve these addressing implementation and not able to use ozone would be able to data gaps. EPA completed its review of compliance issues. switch to chlorine dioxide for primary the study and published its findings in disinfection. a NODA in March 1998. Those findings 2. MCL for Bromate EPA has reviewed all comments and led to a chlorite MCLG of 0.8 mg/L and a. Today’s Rule. In today’s rule, EPA determined that the requirements support for an MCL of 1.0 mg/L. is promulgating an MCL for bromate of promulgated today are necessary to c. Summary of Comments. Many 0.010 mg/L. Bromate is one of the control the occurrence of bromate and commenters requested that EPA not principal byproducts of ozonation in are feasible to achieve. For additional modify the MCL for chlorite prior to bromide-containing source waters. The discussion on the treatment receipt and evaluation of the CMA proposed MCL for bromate was 0.010 technologies for controlling bromate study, since lowering the MCL could mg/l. A system is in compliance with formation and their costs see the Cost preclude the use of chlorine dioxide for the MCL when the running annual and Technology Document for drinking water disinfection. EPA has average of monthly samples, computed Controlling Disinfectants and evaluated the CMA study and quarterly, is less than or equal to the Disinfection Byproducts (EPA, 1998k). concluded that the MCLG for chlorite MCL. If the running annual average These requirements take into account should be 0.8 mg/L. EPA believes the computed for any quarter exceeds the the difficulties in simultaneously proposed MCL of 1.0 mg/L, based on a MCL, the system is out of compliance. controlling risks from DBPs and three sample average to determine EPA has identified the BAT for pathogens, while appropriately compliance, is appropriate because this achieving compliance with the MCL for addressing compliance and is the lowest level achievable by typical bromate as control of ozone treatment implementation issues. In addition, the systems using chlorine dioxide. In process to reduce formation of bromate, Reg. Neg. Committee and the M–DBP addition, considering the margin of as was proposed in 1994 (EPA, 1994a). Advisory Committee supported these safety that is factored into the estimate b. Background and Analysis. For conclusions. of the MCLG, EPA believes the MCL systems using ozone, a separate MCL will be protective of public health. Once 3. MCL for Chlorite was proposed for the primary inorganic the final MCLG was established, EPA DBP associated with ozone usage: a. Today’s Rule. In today’s rule, EPA decided that the chlorite MCL should be bromate. Although the theoretical 10¥4 is promulgating an MCL for chlorite of finalized at the level proposed which risk level for bromate is 0.005 mg/l, an 1.0 mg/L. EPA has modified the was as close as economically and MCL of 0.010 mg/L was proposed monitoring requirements from the technically feasible to the MCLG, and because available analytical detection proposed rule for the reasons discussed modified the proposed requirements for methods for bromate were reliable only in section III.A.7. The issue of monitoirng and compliance in response to the projected practical quantification monitoring and MCL compliance to the health concerns associated with limit (PQL) of 0.01 mg/L (EPA, 1994a). determinations as they relate to the chlorite. In the preamble to the proposed rule, health effect of concern for chlorite were EPA has reviewed all comments and EPA requested comment on whether discussed in the proposed rule (EPA, determined that the requirements there were ways to set (or achieve) a 1994a). CWSs and NTNCWSs using promulgated today are necessary to lower MCL (i.e., 0.005 mg/L [5 µg/L]) chlorine dioxide for disinfection or control the occurrence of chlorite and and whether the PQL for bromate could oxidation are required to conduct are feasible to achieve. These be lowered to 5 µg/L in order to allow sampling for chlorite both daily at the requirements take into account the compliance determinations for a lower entrance to the distribution system and difficulties in simultaneously MCL in Stage 1 of the proposed rule. monthly (3 samples on the same day) controlling risks from DBPs and The proposed MCL of 0.010 mg/L for within the distribution system. pathogens, while appropriately bromate was based on a projected PQL Additional distribution system addressing compliance and Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69411 implementation issues. In addition, the calculating compliance with the MRDL. EPA has identified the best means Reg. Neg. Committee and the M–DBP This commenter felt that 4.0 mg/L available for achieving compliance with Advisory Committee supported these appropriately allows for disinfection the MRDL for chloramines as control of conclusions. under varying circumstances. One treatment processes to reduce commenter requested that EPA increase disinfectant demand, and control of 4. MRDL for Chlorine the flexibility of utilities to meet the disinfection treatment processes to a. Today’s Rule. Chlorine is a widely MRDL for chlorine during periods when reduce disinfectant levels. used and highly effective water chlorine levels in the distribution b. Background and Analysis. The disinfectant. In today’s rule, EPA is systems may need to be raised to protect 1994 proposed Stage 1 DBPR included promulgating an MRDL for chlorine of public health. an MRDL for chloramines at 4.0 mg/L 4.0 mg/L. As a minimum, CWSs and EPA believes that the MRDL of 4.0 (EPA, 1994a). The MRDL for NTNCWSs must measure the residual mg/L for chlorine is appropriate to chloramines is equal to the MRDLG for disinfectant level at the same points in control for potential health effects chloramines. EPA requested comment the distribution system and at the same (MRDLG is 4.0 mg/L) from chlorine on a number of issues relating to the time as total coliforms, as specified in while high enough to allow for control calculation of the MRDLG for § 141.21. Subpart H systems may use the of pathogens under a variety of chloramines. New information on results of residual disinfectant conditions. EPA also believes that chloramines has become available since concentration sampling done under the compliance based on a running annual the 1994 proposal and was cited in the SWTR (§ 141.74(b)(6) for unfiltered average of monthly averages of all 1997 NODA and is included in the systems, § 141.74(c)(3) for systems that samples, computed quarterly is public docket for this rule (EPA, 1997b). filter) in lieu of taking separate samples. sufficient to allow systems to increase This new information did not contain Monitoring for chlorine may not be residual chlorine levels in the data that would warrant changing the reduced. distribution system to a level and for a MRDL. EPA has therefore decided to A system is in compliance with the promulgate the proposed MRDL of 4.0 MRDL when the running annual average time necessary to protect public health to address specific microbiological mg/L for chloramines. of monthly averages of all samples, c. Summary of Comments. Some contamination problems and still computed quarterly, is less than or commenters remarked that systems with maintain compliance. If a system has equal to the MRDL. Notwithstanding the high concentrations of ammonia would taste and odor problems associated with MRDL, operators may increase residual have difficulty meeting the MRDL for excess chlorine levels it can lower its chlorine levels in the distribution chloramine of 4.0 mg/L and still level of chlorine. Since there may not be system to a level and for a time maintain adequate microbial protection. necessary to protect public health to any health effects associated with taste One commenter felt that there should address specific microbiological and odor problems, EPA does not have not be a limit for chloramine residual contamination problems (e.g., including a statutory requirement to address this due to variations in parameters such as distribution line breaks, storm runoff concern. distribution system configurations and events, source water contamination, or 5. MRDL for Chloramines temperature. One commenter felt that cross-connections). the MRDL for chloramines was too low EPA has identified the best means a. Today’s Rule. Chloramines are and should not be set at the same level available for achieving compliance with formed when ammonia is added during as the chlorine MRDL since chlorine is the MRDL for chlorine as control of chlorination. In today’s rule, EPA is a stronger disinfectant than treatment processes to reduce promulgating an MRDL for chloramines chloramines. This commenter felt that disinfectant demand, and control of of 4.0 mg/L (measured as combined total limiting the chloramine residual would disinfection treatment processes to chlorine). As a minimum, CWSs and reduce the capability to sustain high reduce disinfectant levels. NTNCWSs must measure the residual water quality in the distribution system. b. Background and Analysis. The disinfectant level at the same points in One commenter supported the 1994 proposed Stage I DBPR included the distribution system and at the same chloramine MRDL and the methods of an MRDL for chlorine at 4.0 mg/L (EPA, time as total coliforms, as specified in calculating compliance with the MRDL. 1994a). The MRDL for chlorine is equal § 141.21. Subpart H systems may use the This commenter felt that 4.0 mg/L to the MRDLG for chlorine. EPA results of residual disinfectant adequately allows for disinfection under requested comment on a number of concentration sampling done under the varying circumstances. issues relating to the calculation of the SWTR (§ 141.74(b)(6) for unfiltered EPA believes that compliance based MRDLG for chlorine. New information systems, § 141.74(c)(3) for systems that on a running annual average of monthly on chlorine has become available since filter) in lieu of taking separate samples. averages of all samples, computed the 1994 proposal and was discussed in Monitoring for chloramines may not be quarterly, is sufficient to allow systems the 1997 NODA (EPA, 1997b). EPA reduced. to increase residual chloramine levels in believes that no new information has A PWS is in compliance with the the distribution system to a level and for become available to warrant changing MRDL when the running annual average a time necessary to protect public health the proposed MRDL. EPA has therefore of monthly averages of all samples, to address specific microbiological decided to promulgate the MRDL of 4.0 computed quarterly, is less than or contamination problems and still mg/L for chlorine. equal to the MRDL. Notwithstanding the maintain compliance. The MRDL for c. Summary of Comments. Some MRDL, operators may increase residual chloramine does not limit disinfectant commenters expressed concern that the chloramine levels in the distribution dosage but rather disinfectant residual MRDL for chlorine is too high. These system to a level and for a time in the distribution system. EPA commenters were concerned that 4 mg/ necessary to protect public health to therefore, believes that systems with L levels of chlorine would have a address specific microbiological high levels of ammonia should be able detrimental effect on piping materials contamination problems (e.g., including to comply with the MRDL. Systems that and would cause taste and odor distribution line breaks, storm runoff have difficulty sustaining high water problems. One commenter supported events, source water contamination, or quality in the distribution system the chlorine MRDL and the methods of cross-connections). should consider modifying their 69412 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations treatment or maintenance procedures to If any daily sample taken at the Since this new MRDLG was equal to the reduce demand. Although chlorine is a entrance to the distribution system proposed MRDL for chlorine dioxide, stronger disinfectant than chloramine, exceeds the MRDL and if, on the the MRDL will remain 0.8 mg/L. EPA believes that an MRDL of 4.0 mg/ following day, any sample taken in the c. Summary of Comments. A number L is sufficient to provide adequate distribution system also exceeds the of commenters were concerned that the microbial protection. MRDL, the system will be in acute MRDL for chlorine dioxide not be violation of the MRDL and must take 6. MRDL for Chlorine Dioxide lowered below the proposed level of 0.8 immediate corrective action to lower the mg/L because this would preclude the a. Today’s Rule. Chlorine dioxide is occurrence of chlorine dioxide below use of chlorine dioxide as a water used primarily for the oxidation of taste the MRDL and issue the required acute and odor-causing organic compounds in public notification. Failure to monitor disinfectant. One commenter supported water. It can also be used for the in the distribution system on the day the MRDL for chlorine dioxide based on oxidation of reduced iron and following an exceedance of the chlorine public health protection, adequate manganese and color, and as a dioxide MRDL shall also be considered microbial protection, and technical disinfectant and algicide. Chlorine an acute MRDL violation. feasibility. One commenter agreed that a dioxide reacts with impurities in water If any two consecutive daily samples running annual average of samples for very rapidly, and is dissipated quickly. taken at the entrance to the distribution compliance determination should not be In today’s rule, EPA is promulgating an system exceed the MRDL, but none of allowed for chlorine dioxide. One MRDL of 0.8 mg/L for chlorine dioxide. the samples taken in the distribution commenter was concerned that the Unlike chlorine and chloramines, the system exceed the MRDL, the system chlorine dioxide MRDL was too high MRDL for chlorine dioxide may not be will be in nonacute violation of the and that EPA should consider children exceeded for short periods of time to MRDL and must take immediate and vulnerable populations in address specific microbiological corrective action to lower the establishing drinking water standards. contamination problems because of occurrence of chlorine dioxide below EPA has reassessed the health effects potential health concerns with short- the MRDL. Failure to monitor at the data on chlorine dioxide, including the term exposure to chlorine dioxide above entrance to the distribution system on new CMA two-generation study and the MCL. the day following an exceedance of the determined that the MRDL should CWSs and noncommunity systems chlorine dioxide MRDL shall also be must monitor for chlorine dioxide only remain at 0.8 mg/L as proposed. EPA considered a nonacute MRDL violation. believes that this MRDL is set at a if chlorine dioxide is used by the system EPA has identified the best means technically feasible level for the for disinfection or oxidation. Monitoring available for achieving compliance with majority of chlorine dioxide plants. This for chlorine dioxide may not be the MRDL for chlorine dioxide as reduced. If monitoring is required, control of treatment processes to reduce is the case because EPA considered systems must take daily samples at the disinfectant demand, and control of children and susceptible populations in entrance to the distribution system. If disinfection treatment processes to its MRDLG determination (EPA, 1998h). any daily sample taken at the entrance reduce disinfectant levels. The MRDL is set as close to this MRDLG to the distribution system exceeds the b. Background and Analysis. EPA as is technically and economically MRDL, the system is required to take proposed an MRDL for chlorine dioxide feasible. three additional samples in the of 0.8 mg/L in 1994. The MRDL was D. Treatment Technique Requirement distribution system on the next day. determined considering the tradeoffs Systems using chlorine as a residual between chemical toxicity and the 1. Today’s Rule disinfectant and operating booster beneficial use of chlorine dioxide as a chlorination stations after the first disinfectant. The Reg. Neg. Committee Today’s rule establishes treatment customer must take three samples in the agreed to this MRDL with the technique requirements for removal of distribution system: one as close as reservation that it would be revisited, if TOC to reduce the formation of DBPs by possible to the first customer, one in a necessary, after completion of a two- means of enhanced coagulation or location representative of average generation reproductive study by CMA. enhanced softening. The treatment residence time, and one as close as As discussed above for chlorite, a technique applies to Subpart H systems possible to the end of the distribution two-generation reproductive study on using conventional filtration treatment system (reflecting maximum residence chlorite, which is relevant to health regardless of size. Subpart H systems are time in the distribution system). effects of chlorine dioxide, was systems with conventional treatment Systems using chlorine dioxide or completed by the CMA. EPA completed trains that use surface water or ground chloramines as a residual disinfectant or its review of this study and published water under the influence of surface chlorine as a residual disinfectant and its findings in a NODA in March 1998 water as their source. The treatment not operating booster chlorination (EPA, 1998a). Based on its assessment of technique requirement has two steps of stations after the first customer must the CMA study and a reassessment of application. Step 1 specifies the take three samples in the distribution the noncancer health risk for chlorite percentage of influent TOC a plant must system as close as possible to the first and chlorine dioxide, EPA concluded remove based on the raw water TOC and customer at intervals of not less than six that the MRDLG for chlorine dioxide be alkalinity levels. The matrix in Table hours. changed from 0.3 mg/L to 0.8 mg/L. III–1 specifies the removal percentages.

TABLE III±1.ÐREQUIRED REMOVAL OF TOTAL ORGANIC CARBON BY ENHANCED COAGULATION AND ENHANCED SOFTENING FOR SUBPART H SYSTEMS USING CONVENTIONAL TREATMENT a,b

Source water alkalinity (mg/L as CaCO3) Source water TOC (mg/L) >60±120 (per- >120 c (per- 0±60 (percent) cent) cent)

>2.0±4.0 ...... 35.0 25.0 15.0 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69413

TABLE III±1.ÐREQUIRED REMOVAL OF TOTAL ORGANIC CARBON BY ENHANCED COAGULATION AND ENHANCED SOFTENING FOR SUBPART H SYSTEMS USING CONVENTIONAL TREATMENT a,bÐContinued

Source water alkalinity (mg/L as CaCO3) Source water TOC (mg/L) >60±120 (per- >120 c (per- 0±60 (percent) cent) cent)

>4.0±8.0 ...... 45.0 35.0 25.0 >8.0 ...... 50.0 40.0 30.0 a Systems meeting at least one of the conditions in Section 141.135(a)(2) (i)±(vi) of the rule are not required to meet the removals in this table. b Softening systems meeting one of the two alternative compliance criteria in Section 141.135(a)(3) of the rule are not required to meet the re- movals in this table. c Systems practicing softening must meet the TOC removal requirements in the last column to the right.

Step 2 provides alternate performance indicator of the amount of TOC that is <30µg/L (or the system has made a clear criteria when it is technically infeasible feasible to remove. and irrevocable financial commitment for systems to meet the Step 1 TOC Systems practicing enhanced to technologies that will meet the TTHM removal requirements. For systems softening are not required to perform jar and HAA level); practicing enhanced coagulation, Step 2 testing under today’s treatment (4) the system’s TTHM is <40µg/L, of the treatment technique requirement technique as part of a Step 2 procedure. HAA5 is <30µg/L, and only chlorine is is used to set an alternative TOC Rather, they are required to meet one of used for primary disinfection and removal requirement (i.e. alternative three alternative performance criteria if maintenance of a distribution system percent removal of raw water TOC) for they cannot meet the Step 1 TOC residual; those systems unable to meet the TOC removal requirements. These criteria (5) the system’s source water SUVA removal percentages specified in the are: (1) Produce a finished water with a prior to any treatment is ≤ 2.0 L/mg–m; matrix. The alternative TOC removal SUVA of less than or equal to 2.0 L/mg- and percentage is determined by performing m; (2) remove a minimum of 10 mg/L (6) the system’s treated water SUVA is ≤ jar tests on at least a quarterly basis for magnesium hardness (as CaCO3); or (3) 2.0 L/mg–m. one year. During the jar tests, alum or lower alkalinity to less then 60 mg/L (as Alternative compliance criteria 1, 2, 5, an equivalent dose of ferric coagulant is CaCO3). All three of these alternative and 6 are determined based on monthly added in 10 mg/L increments until the performance criteria are measured monitoring calculated quarterly as a pH is lowered to the target pH value. monthly and can be calculated quarterly running annual average of all The target pH is the value the sample as a running annual average to measurements. Alternative compliance must be at or below before the demonstrate compliance. As discussed criteria 3 is based on monthly incremental addition of coagulant can in the 1997 DBP NODA (EPA 1997b) monitoring for TOC and alkalinity or be discontinued. For the alkalinity EPA has not been able, from a technical quarterly monitoring for TTHMs and ranges 0–60, >60–120, >120–240, and and engineering standpoint, to identify HAA5, calculated quarterly as a running >240 mg/L (as CaCO3), the target pH a Step 2 testing procedure at this time annual average of all measurements. values are 5.5, 6.3, 7.0, and 7.5, that allows softening systems to set an Alternative criteria 4 is determined respectively. Once the Step 2 jar test is alternative TOC removal amount. based on monitoring for TTHMs and complete, the TOC removal (mg/L) is Enhanced softening systems unable to HAA5, calculated quarterly as a running then plotted versus coagulant dose (mg/ meet the Step 1 TOC removal annual average of all measurements. L). The alternative TOC removal requirements or any of the three SUVA, an indicator of DBP precursor percentage is set at the point of alternative performance criteria may removal treatability, is defined as the diminishing returns (PODR) identified apply to the State for a waiver from the UV–254 (measured in m¥1) divided by on the plot. treatment technique requirements. EPA the DOC concentration (measured as Today’s rule defines the PODR as the believes the three alternative mg/L). point on the TOC versus coagulant dose performance criteria listed above 2. Background and Analysis plot where the slope changes from provide assurance that softening greater than 0.3/10 to less than 0.3/10 systems have maximized TOC removal The general structure of the 1994 and remains less than 0.3/10. After to the extent feasible. proposed rule and today’s final rule are identifying the PODR, the alternative Today’s rule also provides alternative similar. The 1994 proposal included an TOC removal percentage can be set. If compliance criteria—which are separate enhanced coagulation and enhanced the TOC removal versus coagulant dose and independent of the Step 2 enhanced softening treatment technique plot does not meet the PODR definition, coagulation procedure and the requirement for Subpart H systems. The the water is considered not amenable to enhanced softening alternative 1994 proposed rule included a TOC enhanced coagulation and TOC removal performance criteria—from the removal matrix for Step 1 TOC removal is not required if the PWS requests, and treatment technique requirements requirements and it also provided for a is granted, a waiver from the enhanced provided certain conditions are met. Step 2 jar test procedure for systems coagulation requirements by the State. These criteria are: practicing enhanced coagulation. The Systems are required to meet the (1) the system’s source water TOC is PODR for the Step 2 procedure was alternative TOC removal requirements <2.0 mg/L; defined as a slope of .3/10 on the TOC during full-scale operation to maintain (2) the system’s treated water TOC is removal versus coagulant dose plot. The compliance with the treatment <2.0 mg/L; Step 2 procedure included a maximum technique. For the technical reasons (3) the system’s source water TOC pH value, now referred to as the ‘‘target outlined in the 1997 DBP NODA (EPA <4.0 mg/L, its source water alkalinity is pH’’ for conducting the jar tests and it 1997b), EPA has concluded that this >60 mg/L (as CaCO3), and the system is also allowed systems to request a waiver definition of the PODR is a reliable achieving TTHM <40µg/L and HAA5 from the State if the PODR was never 69414 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations attained. The target pH values in the enhanced coagulation’s ability to percentages in the bottom two rows may 1994 proposal were the same as those in remove TOC from different source be reasonable and allow 90 percent of today’s final rule. A Step 2 procedure waters (Edwards, 1997; Tseng & plants to avoid the Step 2 procedure. for enhanced softening systems was not Edwards, 1997; Chowdhury, 1997). The Therefore, the TOC removal percentages specified in the proposal. model was subsequently used to analyze in the first row have been lowered 5.0 The proposed rule also provided for a the level or percentage of TOC removal percentage points to enable 90 percent number of exceptions to the enhanced that is operationally feasible to achieve of plants to comply without coagulation and enhanced softening for the boxes in the proposed TOC unreasonable coagulant dosage or requirements, but it did not include use removal matrix. Nine predictive resorting to the Step 2 procedure. of SUVA as an alternative compliance equations for TOC removal were Data Supporting the Use of SUVA as criteria. developed, one for each box of the TOC an Exemption from Treatment A major goal of the TOC removal removal matrix, to select TOC removal Technique Requirements. At the time of treatment technique requirements was percentages that could be ‘‘reasonably’’ the proposal, insufficient data on SUVA to minimize transactional costs to the met by 90 percent of the systems was available to define precise criteria States both in terms of limiting the implementing enhanced coagulation. for when enhanced coagulation would number of systems seeking alternative The equations indicated that many not be effective for removing DBP performance criteria and in providing systems having source waters within the precursors. The M–DBP Advisory relatively simple methodologies for low TOC boxes of the matrix (i.e. 2.0– Committee examined the role of SUVA determining alternative performance 4.0 mg/L, the first row of the matrix) as an indicator of the amount of DBP criteria. In the 1997 DBP NODA (EPA would meet the Step 2 slope criterion precursor material enhanced 1997b), EPA presented new data and before meeting the required TOC coagulation is capable of removing. It analysis and the basis for modifying the removal percentages. In other words, has been well established that proposed criteria to those described in less than 90 percent of the systems in coagulation primarily removes the today’s final rule. The 1997 NODA also this row could achieve the proposed humic fraction of the natural organic solicited public comment on EPA’s TOC removal with reasonable coagulant matter (NOM) in water (Owen et al., intended changes to the proposal and doses. The equations indicated that the 1993). Furthermore, Edzwald and Van the recommendations of the M–DBP TOC removal percentages in the Benschoten (1990) have found SUVA to Advisory Committee to EPA. An medium and high TOC boxes (the be a good indicator of a water’s humic overview of the key points in the 1997 bottom two rows of the matrix) could be content. The humic fraction of a water’s NODA most pertinent to modifying the met by approximately 90 percent of the organic content significantly affects DBP treatment technique requirements are systems in these boxes. The research formation upon chlorination. A study by White et al. (1997) showed presented below. team also examined 90th-percentile Data Supporting Changes in the TOC that waters with high initial SUVA SUVA curves, in conjunction with the Removal Requirements. The proposed values exhibited significant reductions nine TOC removal curves, to predict TOC removal percentages, which were in SUVA as a result of coagulation, what TOC removal percentage is set with the intent that 90% of affected demonstrating a substantial removal of appropriate for each of the nine boxes systems would be able to achieve them, the humic (and other UV-absorbing) of the matrix. were developed with limited data. Since components of the organic matter, the proposal, several jar studies and An analysis of full-scale TOC removal whereas waters with low initial SUVA analyses of full-scale plant TOC removal has also been performed since 1994. values exhibited relatively low performance have been performed. They Data was obtained from 76 treatment reductions in SUVA. For all of the were analyzed by EPA as part of the M– plants of the American Water Works waters examined, the SUVA tended to DBP Advisory Committee process. This Service Company (AWWSCo) system, plateau at high alum doses, reflecting data will not be thoroughly reviewed plants studied by Randtke et al. (1994), that the residual organic matter was here; instead, the major points salient to and plants studied by Singer et al. primarily non-humic and therefore development of the final regulation will (1995). These data represent a one-time unamenable to removal by enhanced be summarized. See the 1997 DBP sampling at each plant under current coagulation. SUVA’s ability to indicate NODA (EPA 1997b) to review EPA’s operating conditions when enhanced the amount of humic matter present, detailed analysis of the new data. coagulation was not being practiced. and enhanced coagulation’s ability to As discussed in greater detail in the This sampling is different from the preferentially remove humic matter, 1997 DBP NODA, research by Singer et proposed compliance requirements logically establishes SUVA as an al. (1995) indicated that a significant which would be based on an annual indicator of enhanced coagulation’s number of waters, especially low-TOC, average of monthly samples. Based on ability to remove humic substances from high-alkalinity waters in the first row of current treatment at the plants in the a given water. The M–DBP Advisory the proposed TOC removal matrix, study, 83 percent of the systems treating Committee therefore recommended that would probably not be able to meet the moderate-TOC, low-alkalinity water a SUVA value ≤ 2.0 L/mg–m be an TOC removal percentages and would removed an amount of TOC greater than exemption from the treatment technique therefore need to use the Step 2 protocol that required by the TOC removal requirement and that this SUVA value to establish alternative performance matrix, whereas only 14 percent of the also be added as a Step 2 procedure. criteria. The Singer et al. (1995) study systems treating water with low TOC Effect of Coagulant Dose on TOC raised concern regarding the number of and high alkalinity met the proposed Removal for Enhanced Softening. At the systems that might need to use the Step TOC removal requirements. The results time of proposal, limited data was 2 procedure to set alternative of the survey, coupled with the available on the effectiveness of TOC performance criteria. A study by information discussed in the preceding removal by enhanced coagulation and Malcolm Pirnie, Inc. and Colorado paragraph, indicate that the proposed enhanced softening and on conditions University addressed this issue by TOC removal percentages in the top row that define feasibility. Several studies developing a nationally representative of the matrix might be too high for 90 examined the relationship between database of 127 source waters and used percent of plants to avoid the Step 2 increased coagulant dose and TOC this data to develop a model to predict procedure, while the removal removal (Shorney et al., 1996; Clark et Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69415 al. 1994). These studies indicate some mentioned earlier in this section, as a curve. Since SUVA is a very good improvement in TOC removal with Step 2 procedure for softening systems. predictor of the humic fraction of TOC, small doses of iron salts (5 mg/L ferric which is the fraction preferentially 3. Summary of Comments sulfate), but no additional TOC removal removed by enhanced coagulation, and during softening occurred with A large number of comments on the the PODR predicted by SUVA and the increased coagulant addition (up to 25 1994 proposal questioned whether the slope criterion agree, EPA believes the mg/L dose). Pilot testing by the City of required TOC removal percentages slope criterion of 0.3 mg/L TOC removal Austin’s softening plant confirmed the could be obtained by 90 percent of per 10 mg/L of coagulant addition study’s jar test results by showing that affected systems. In response, since the accurately predicts the PODR. increasing ferric sulfate doses beyond time of proposal, a large body of additional data and analysis has been The majority of commenters did not the level required for turbidity removal support requiring the use of bench-scale provided no additional TOC removal. developed to help address this question. The analyses discussed above showed filtration as part of the Step 2 enhanced Multiple jar tests on various waters coagulation procedure. The commenters performed by Singer et al. (1996) that the top row of the TOC removal matrix needed to be lowered by 5.0 generally believed that using filtration at examined the relationship between use bench scale is of limited value because of lime and soda ash and TOC removal. percentage points to enable 90 percent of systems within the row to achieve the the great majority of TOC is removed via Only lime and soda ash (no coagulants) sedimentation, not through filtration. were used in the tests. The study required TOC removal without unreasonable coagulant doses. Analysis Additionally, some commentors felt that showed the removal of 10 mg/L of also showed the TOC removal attempting to replicate full-scale magnesium hardness would probably percentages contained in the two lower filtration at bench scale can contain have less of an impact on plant residual rows of the TOC removal matrix inherent inaccuracy. EPA generally generation than using a lime soda-ash accurately reflected the TOC removal 90 agrees that a Step 2 filtration procedure process. However, the amount of percent of these systems could remove. should not be required. The Agency residual material generated under both EPA believes the final TOC removal believes that most of the TOC removed scenarios could be substantial. matrix, which includes the adjustments by conventional treatment plants is Step 2 Requirements for Softening to the top row mentioned above, removed in the sedimentation basin Systems. As stated above, the proposed accurately reflects the TOC removal that rather than in the filters. Therefore, rule did not include a Step 2 procedure 90 percent of the systems affected by the requiring a bench-scale filtration for softening plants because of a lack of rule could practically achieve. procedure as part of Step 2 testing will data. The M-DBP Advisory Committee Commenters questioned why systems not increase the accuracy of the examined new data that had been that meet the DBP Stage 1 MCLs for procedure or its value to the treatment collected since the proposal to TTHM and HAA5 must still practice technique implementation. Accordingly, determine if a Step 2 procedure for enhanced coagulation. The enhanced today’s final rule does not require the softening plants could be identified. coagulation treatment technique is use of a bench scale filtration procedure Data included the current TOC removals designed to remove DBP precursor during Step 2 enhanced coagulation being achieved by softening plants material to help reduce the risks posed testing. Detailed guidance on covered by the ICR (49 plants). The data by DBPs. Also, EPA believes that conducting the Step 2 testing will be were analyzed to find the appropriate enhanced coagulation would reduce the provided in the Guidance Manual for TOC removal levels for softening plants. number of systems switching to Enhanced Coagulation and Enhanced The results of plotting the average TOC alternative disinfectants, which was a Precipatative Softening. percent removals on a percentile basis goal of the Reg. Neg. Committee. EPA indicated that the relative impact of believes that even if systems are meeting Commenters expressed varied meeting the TOC removal requirement the MCLs, an additional risk reduction opinions regarding the frequency of in the proposed rule would be greatest benefit can be achieved through removal Step 2 testing. Several commenters in the low TOC group (>2–4 mg/L). of DBP precursor material at a relatively stated that the rule should not set a However, forcing a plant to increase pH low cost to the system. Therefore, minimum testing frequency, but that it may require it to add soda ash (due to systems that meet the MCLs must still should be left to State discretion based the decrease in alkalinity caused by practice enhanced coagulation to on source water characteristics. Other high lime dose necessary to raise the decrease the risks posed by DBPs in commenters believed a minimum of pH). This would be a significant general. quarterly monitoring should be required treatment change due to the additional The Agency received numerous with a provision for more frequent solids generation and because comments on the 1994 proposal that testing to address source water quality significant amounts of magnesium expressed doubt regarding the definition events. EPA believes that Step 2 testing hydroxide may precipitate at the higher of the PODR. Specifically, the frequency should be related to seasonal pH. Most softening plants are normally commenters stated that the accuracy of and other variations in source water operated without soda ash addition the slope criterion (0.3 mg/L TOC quality as these variations may because of the high cost of soda ash, the removed per 10 mg/L coagulant added) influence the amount of TOC removal additional sludge production, the for determining the PODR was not the treatment plant can achieve. increased chemical addition to stabilize supported with adequate data. The data Accordingly, EPA recommends that the water, and the increased sodium developed since the proposal and the systems utilizing the Step 2 procedure levels in the finished water (Randtke et corresponding analysis demonstrate that for compliance perform Step 2 testing al., 1994 and Shorney et al., 1996). Due the slope criterion accurately predicts quarterly for one year after the effective to these difficulties, EPA does not the PODR. The analyses discussed data of the rule. The system may then currently believe that a lime and soda- above showed that there is a particular apply to the State to reduce testing to a ash softening process would be a viable relationship between SUVA and the minimum of once per year. If the State Step 2 procedure for softening systems. slope criterion, namely, that they both does not approve the request for The final rule instead specifies two predict the PODR at the same point of reduced testing frequency, the system alternative compliance criteria, the TOC removal versus coagulant dose must continue to test quarterly. 69416 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

E. Predisinfection Disinfection Credit assumed that the addition of simultaneously maintaining disinfectant was essentially ‘‘mutually prechlorination, can still result in 1. Today’s Rule exclusive’’ to the goal of reducing DBP decreased DBP formation (especially for Today’s rule does not impose any formation by the removal of TOC. As TOX and TTHM). Greater benefits are constraints on the ability of systems to discussed below, new data developed realized by moving the point of practice predisinfection and take since 1994 suggest this may not be the chlorination to post-rapid mixing or microbial inactivation credit for case. further downstream for HAA5 control, predisinfection to meet the disinfection Reasons for Disinfectant Use. In order and to mid-flocculation or post- requirements of the SWTR. Utilities are to obtain information on the impact that sedimentation for TOX and TTHM free to take disinfection credit for disallowing predisinfection would have control. These data show that the predisinfection, regardless of the on utilities’ disinfection practices, a assumption made in the 1994 proposal, disinfectant used, for disinfection that survey was sent out to ICR utilities to namely that application of any occurs after the last point the source obtain information on their current disinfectant prior to TOC removal water is subject to surface water run-off predisinfection practices. The results of would critically effect DBP formation, and prior to the first customer. the survey of 329 surface water was not accurate. The data indicate that 2. Background and Analysis treatment plants indicated that 80 simultaneous employment of enhanced percent (263) of these plants use coagulation and predisinfection does The 1994 proposed Stage 1 DBPR predisinfection for one or more reasons. not necessarily mean that DBP (EPA,1994a) discouraged the use of The survey indicated that the majority formation cannot be substantially disinfectants prior to precursor of the plants using predisinfection were controlled (see EPA 1997b for detailed (measured as TOC) removal by not doing so for multiple reasons. However, analysis). allowing compliance credit for the the main reason reported for Impact on Softening Plants. In order SWTR’s disinfection requirements to be predisinfection was microbial taken prior to removal of a specified to obtain additional information on the inactivation. Algae control, taste and current TOC removals being achieved percentage of TOC. The proposed odor control, and inorganic oxidation, by softening plants, a survey was sent to IESWTR options were intended to in that order, were the next most all the ICR softening utilities (49 plants) include microbial treatment frequently cited reasons for practicing requesting that they fill out a single page requirements to prevent increases in predisinfection. Seventy-seven percent of information with yearly average, microbial risk due to the loss of of plants that predisinfected reported maximum and minimum values for predisinfection credit. These options that their current levels of Giardia multiple operating parameters for each were to be implemented simultaneously lamblia inactivation would be lowered softening plant. The survey showed that with the Stage 1 DBPR. The purpose of if predisinfection was discontinued and in spite of the fact that 78 percent of not allowing predisinfection credit was no subsequent additional disinfection softening plants are using free chlorine to maximize removal of organic was added to compensate for change in precursors (measured as TOC) prior to practice. Eighty-one percent of plants for at least a portion of their the addition of a disinfectant, thus disinfection, 90 percent of plants are that predisinfected would have to make µ lowering the formation of DBPs. major capital investments to make up currently meeting an 80 g/L MCL level Many drinking water systems use for the lost logs of Giardia lamblia for TTHMS. All the softening plants preoxidation to control a variety of reported average HAA5 levels below 60 inactivation. For example, to maintain µ water quality problems such as iron and the same level of microbial protection g/L. Without predisinfection credit, manganese, sulfides, zebra mussels, currently afforded, construction to these plants may have to provide Asiatic clams, and taste and odor. The provide for additional contact time or disinfection contact time after 1994 proposed rule did not preclude the use of a different disinfectant might be sedimentation, which could mean continuous addition of oxidants to needed if predisinfection credit was significantly increasing the free chlorine control these problems. However, the eliminated. contact time to make up for a shortened proposed regulation, except under a few In addition to the ICR mail survey, detention time. specific conditions, did not allow credit results from EPA’s Comprehensive 3. Summary of Comments for compliance with disinfection Performance Evaluations (CPE) from 307 requirements prior to TOC removal. PWSs (4 to 750 mgd) reported that 71% Most commenters stated that the Analysis supporting the proposed rule of the total number of plants used proposed elimination of predisinfection concluded that many plants would be predisinfection and 93% of those that would result in many plants not being able to comply with the Stage 1 MCLs predisinfected used two or three able to maintain existing levels of for THMs and HAA5 of 0.080 mg/L and disinfectant application points during disinfection or comply with the SWTR 0.060 mg/L, respectively, by reductions treatment. disinfection requirements without in DBP levels as a result of reduced Based on the above information, EPA making significant compensatory disinfection practice in the early stages believes that predisinfection is used by changes in their disinfection practice. of treatment. Also, enhanced a majority of PWSs for microbial Commenters were concerned that coagulation and enhanced softening inactivation, as well as other drinking without predisinfection the level of were thought to lower the formation of water treatment objectives. Therefore, microbial risk their customers were other unidentified DBPs as well. The disallowing predisinfection credit could exposed to could significantly increase, 1994 proposal assumed that addition of influence systems to make changes in and that eliminating microbial disinfectant prior to TOC removal treatment to comply with the inactivation credit for predisinfection to would initiate DBP formation through disinfection requirements of the SWTR comply with the SWTR might influence contact of the chlorine with the TOC, or to maintain current levels of utilities to abandon predisinfection to effectively eliminating the value of microbial inactivation. more easily comply with the TTHM and enhanced coagulation for DBP Impact of Point of Chlorination on HAA5 MCLs. EPA agrees with this reduction. Finally, the analysis DBP Formation. The results of a study concern and therefore the final rule has underlying the 1994 proposed by Summers et al. (1997) indicate that been modified from the proposal to elimination of the preoxidation credit practicing enhanced coagulation, while allow predisinfection credit. Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69417

F. Requirements for Systems to Use MRDL: Standard Methods 4500–ClO2 D 19th Edition Supplement (APHA, 1996). Qualified Operators (APHA, 1995) and 4500–ClO2 E (APHA EPA believes that all of these methods EPA believes that systems that must 1995). EPA did not approve Standard can achieve the precision and detection make treatment changes to comply with Method 4500–ClO2 C (APHA, 1995), level necessary for compliance requirements to reduce the which was included in the 1994 determinations required in today’s rule microbiological risks and risks from proposed rule. The Agency determined, when the quality control (QC) disinfectants and disinfection in concurrence with the majority of procedures contained in the method byproducts should be operated by commenters on this issue, that Standard descriptions and this rule are followed. personnel who are qualified to Method 4500–ClO2 C is outdated and However, while any of these methods recognize and react to problems. inaccurate in comparison to chlorine may be used, EPA advises that a Therefore, in today’s rule, the Agency is dioxide methods approved in today’s consistent method be employed for all requiring that all systems regulated rule and is inadequate for compliance measurements in order to reduce the under this rule be operated by an monitoring. impact of possible instrument bias. individual who meets State specified TTHM. Today’s rule approves three In accordance with the concerns of qualifications, which may differ based methods for determining compliance commenters, today’s rule requires on size and type of the system. Subpart with the TTHM MCL: EPA Methods certain QC procedures for TOC analyses H systems already are required to be 502.2 (EPA, 1995), 524.2 (EPA, 1995), in addition to those contained in the operated by qualified operators under and 551.1 (EPA, 1995). method descriptions. These additional HAA5. Today’s rule approves three the provisions of the SWTR (40 CFR QC steps are designed to increase the methods for determining compliance 141.70(c)). Current qualification or integrity of the analysis and have been with the HAA5 MCL: EPA Methods certification programs developed by the found to be effective in data collection 552.1 (EPA, 1992) and 552.2 (EPA, States should, in many cases, be under the ICR. Filtration of samples 1995) and Standard Method 6251B adequate to meet this requirement for prior to TOC analysis is not permitted, (APHA, 1995). as this could result in removal of Subpart H systems. Also, States must Bromate. Today’s rule approves a maintain a register of qualified organic carbon. Where turbidity method for determining compliance interferes with TOC analysis, samples operators. with the bromate MCL: EPA Method EPA encourages States which do not should be homogenized and, if 300.1 (EPA, 1997e). EPA has already have operator certification necessary, diluted with organic-free demonstrated this method to be capable programs in effect to develop such reagent water. TOC samples must either of quantifying bromate at the MCL of 10 programs. The Reg. Neg. Committee and be analyzed or must be acidified to µg/L under a wide range of solution TWG believed that properly trained achieve pH less than 2.0 by minimal conditions. EPA did not approve EPA personnel are essential to ensure safer addition of phosphoric or sulfuric acid Method 300.0 (EPA, 1993b) for bromate drinking water. States with existing as soon as practical after sampling, not analysis, although this method was to exceed 24 hours. Samples must be operator certification programs may included for analysis of bromate in the wish to update their programs for analyzed within 28 days. 1994 proposed rule. As stated in the SUVA (Specific Ultraviolet qualifying operators under the SWTR. In proposed rule, EPA Method 300.0 is not Absorbance). Today’s rule establishes these cases, States may wish to indicate sensitive enough to measure bromate at SUVA as an alternative criterion for that their operator certification the MCL established in today’s rule. demonstrating compliance with TOC programs are being developed in EPA Method 300.1 was developed removal requirements contained in accordance with EPA’s new guidelines. subsequent to the proposed rule in order today’s rule. SUVA is a calculated G. Analytical Methods to provide a method with adequate parameter defined as the UV absorption ¥1 sensitivity to assess bromate at 254 nm (UV254) (measured as m ) 1. Today’s Rule compliance. divided by the DOC concentration Chlorine (Free, Combined, and Total). Chlorite. Today’s rule approves two (measured as mg/L). If the UV Today’s rule approves four methods for methods for determining compliance absorption is first determined in units of measuring free, combined, and total with the chlorite MCL: EPA Methods cm¥1, the SUVA equation is multiplied chlorine to determine compliance with 300.0 (EPA, 1993b) and 300.1 (EPA, by 100 to convert to m¥1, as shown the chlorine MRDL (using either free or 1997e). As described elsewhere in below: ¥1 total chlorine) and chloramines MRDL today’s rule, chlorite compliance SUVA = 100 (cm/m) [UV254 (cm )/DOC (using either combined or total analyses are made on samples taken in (mg/L)] chlorine): ASTM Method D1253–86 the distribution system during monthly Two separate analytical methods are (ASTM, 1996), Standard Methods 4500– monitoring, or during additional necessary to make this measurement: Cl D (APHA, 1995), 4500–Cl F (APHA, distribution system monitoring as UV254 and DOC. Today’s rule approves 1995), and 4500–Cl G (APHA, 1995). required. Today’s rule establishes the three methods for DOC analysis: Additionally, this rule approves two following method for daily monitoring Standard Methods 5310 B, 5310 C, and methods for measuring total chlorine to of chlorite: Standard Method 4500-ClO2 5310 D, as published in the Standard determine compliance with the chlorine E (APHA, 1995), amperometric titration. Methods 19th Edition Supplement MRDL and chloramines MRDL: As stated elsewhere in today’s rule, (APHA, 1996); and approves Standard Standard Methods 4500–Cl E (APHA, daily monitoring of chlorite is Method 5910 B (APHA, 1995) for UV254 1995) and 4500–Cl I (APHA, 1995). The conducted on samples taken at the analysis. rule also contains an additional method entrance to the distribution system. The final rule contains QC steps for for measuring free chlorine to determine Commenters supported the use of the SUVA analyses that are required in compliance with the chlorine MRDL: amperometric titration as a feasible addition to those mandated in the Standard Method 4500–Cl H (APHA, method for daily monitoring of chlorite. method descriptions. These 1995). TOC. Today’s Rule approves three requirements were developed in Chlorine Dioxide. Today’s rule methods for TOC analysis: Standard response to comments solicited by EPA approves two methods for determining Methods 5310 B, 5310 C, and 5310 D, in the 1997 DBP NODA (EPA, 1997b) compliance with the chlorine dioxide as published in the Standard Methods and are as follows: 69418 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

—sample acquisition (DOC and UV254 soon as practical after sampling, not Bromide. Today’s rule approves the samples used to determine a SUVA to exceed 48 hours.) following two methods for monitoring value must be taken at the same time bromide: EPA Methods 300.0 (EPA, —filtration (Prior to analysis, UV254 and and at the same location. SUVA must DOC samples must be filtered through 1993b) and 300.1 (EPA, 1997e). be determined on water prior to the a 0.45 µm pore-diameter filter. DOC Alkalinity. Today’s rule approves addition of disinfectants/oxidants.) samples must be filtered prior to three methods for measuring alkalinity: —sample preservation (DOC samples acidification.) ASTM Method D1067–92B (ASTM, must either be analyzed or must be 1994), Standard Method 2320 B (APHA, acidified to achieve pH less than 2.0 —background concentrations in the filtered blanks (Water passed through 1995), and Method I–1030–85 (USGS, by minimal addition of phosphoric or 1989). sulfuric acid as soon as practical after the filter prior to filtration of the sampling, not to exceed 48 hours. The sample must serve as the filtered pH. Today’s rule requires the use of methods that have been previously pH of UV254 samples may not be blank. This filtered blank must be adjusted.) analyzed using procedures identical approved in § 141.23(k) for —holding times (DOC samples must be to those used for analysis of the measurement of pH. analyzed within 28 days of sampling. samples and must meet the following Approved analytical methods are UV254 samples must be analyzed as criteria: TOC <0.5 mg/L.) summarized in Table III–2.

TABLE III±2.ÐAPPROVED ANALYTICAL METHODS

Analyte EPA method Standard method Other

Chlorine (free, combined, total) ...... 4500±Cl D ASTM D1253±8...... 4500±Cl F ...... 4500±Cl G (Total) ...... 4500±Cl E ...... 4500±Cl I (Free) ...... 4500±Cl H Chlorine Dioxide ...... 4500±ClO2 D ...... 4500±ClO2 E TTHM ...... 502.2 524.2 551.1 HAA5 ...... 552.1 625l B 552.2 Bromate ...... 300.1 Chlorite (monthly) ...... 300.0 300.1 (Daily) ...... 4500±ClO2 E TOC/DOC ...... 5310 B ...... 5310 C ...... 5310 D UV254 ...... 5910 B Bromide ...... 300.0 300.1 Alkalinity ...... 2320 B ASTM D1067± 92B. USGS I±1030± 85. pH ...... 150.1 4500±H+B ASTM D1293±84. 150.2

2. Background and Analysis indicator N,N-diethyl-p- impurities in the sodium chloride (Xie, phenylenediamine (DPD). Chlorine (Free, Combined, and Total). 1995). Other changes to EPA Method In the 1994 proposed rule, EPA TTHM. The 1994 proposed rule 551.1 include a buffer addition to included all Standard Methods for included three methods for the analysis stabilize chloral hydrate, elimination of analysis of free, combined, and total of TTHMs. They were EPA Methods the preservative ascorbic acid, and chlorine that were approved in today’s 502.2, 524.2, and 551. In 1995, EPA modification of the extraction procedure rule. Method 551 was revised to EPA Method to minimize the loss of volatile analytes. Chlorine Dioxide. The 1994 proposed 551.1, rev. 1.0 (EPA, 1995), which was The revised method requires the use of rule included the same three methods approved for ICR monitoring under 40 surrogate and other quality control standards to improve the precision and for analyzing chlorine dioxide (ClO2) CFR 141.142. that are approved under the SWTR and EPA Method 551.1 has several accuracy of the method. ICR regulations. Two of these methods, improvements upon EPA Method 551. HAA5. The 1994 proposed rule Standard Methods 4500.ClO2 C (APHA, The use of sodium sulfate is strongly included two methods for the analysis 1992) and 4500.ClO2 E (APHA, 1992), recommended over sodium chloride for of five haloacetic acids—EPA Method are amperometric methods. The third the MTBE extraction of DBPs. This 552.1 (EPA, 1992) and Standard Method proposed method was Standard Method change was in response to a report 6233B (APHA, 1992). Both methods use 4500.ClO2 D (APHA, 1992), a indicating elevated recoveries of some capillary column gas chromatographs colorimetric test using the color brominated DBPs due to bromide equipped with electron capture Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69419 detectors. The two methods differ in the and minimizes the potential for of Standard Methods 5310 B, C, and D sample preparation steps. EPA Method chromatographic interferences from were published in Standard Methods 552.1 uses solid phase extraction disks common anions at concentrations 19th Edition Supplement (APHA, 1996). followed by an acidic methanol 10,000 times greater than bromate ion. The revised version of Standard Method derivitization. Standard Method 6233B For example, quantification of 5.0 µg/L 5310 B recognized the capacity of is a small volume liquid-liquid (micro) bromate is feasible in a matrix certain high temperature instruments to extraction with methyl-t-butyl ether, containing 50 mg/L chloride. achieve detection limits below 1 mg/L followed by a diazomethane Minimizing the interferences permits using this method. derivitization. Following the proposed the introduction of a larger sample SUVA (Specific Ultraviolet rule, Standard Method 6233B was volume to yield method detection limits Absorbance). SUVA analytical methods revised and renumbered 6251B (APHA, in the range of 1–2 µg/L. were not addressed in the 1994 1995) to include bromochloroacetic In the 1997 DBPR NODA (EPA, proposed rule because SUVA had not acid, for which a standard was not 1997b), EPA discussed EPA Method been developed and proposed as a commercially available in 1994. 300.1 and projected that by using it compliance parameter for TOC removal Recognizing these improvements, EPA laboratories would be able to quantify requirements at that time. The analytical approved Standard Method 6251B for bromate with the accuracy and methods and associated QC procedures analysis under the ICR (40 CFR Part 141 precision necessary for compliance for DOC and UV254 approved in today’s or EPA, 1996a). Several commenters determination with an MCL of 10 µg/L. rule are those on which the Agency requested that the revised and Although there would be a limited solicited comment in the 1997 DBPR renumbered method, Standard Method number of laboratories that would be NODA (EPA, 1997b). 6251B, also be approved for the analysis qualified to do such analyses, EPA Bromide. The 1994 proposed rule of haloacetic acids under the Stage 1 determined that there should be included EPA Method 300.0 for analysis DBPR. adequate laboratory capacity for of bromide. EPA believed that the In 1995 EPA published a third bromate ion compliance monitoring by working range of this method method for HAAs, EPA Method 552.2 the time the rule becomes effective. adequately covered the requirements (EPA, 1995), and subsequently approved Chlorite. The proposed rule required proposed for bromide monitoring. As it for HAA analysis under the 1996 ICR systems using chlorine dioxide for described above, EPA developed (40 CFR Part 141 or EPA, 1996a). EPA disinfection or oxidation to perform Method 300.1 for improved bromate Method 552.2 is an improved method, monthly monitoring for chlorite ion in analysis subsequent to the proposed combining the micro extraction the distribution system. EPA designated rule. EPA Method 300.1 can also procedure of Standard Method 6233B EPA Method 300.0 (ion effectively measure bromide at the with the acidic methanol derivitization chromatography) for chlorite analysis. concentration of 50 µg/L, required in procedure of EPA Method 552.1. It is EPA considered other methods using today’s rule for reduced monitoring of capable of analyzing nine HAAs. amperometric and potentiometric bromate. Bromate. The 1994 proposed rule techniques but decided that only the ion Alkalinity. The proposed rule required systems that use ozone to chromatography method (EPA Method included all methods approved by EPA monitor for bromate ion. EPA proposed 300.0) would produce results with the for measuring alkalinity. These methods EPA Method 300.0 (EPA, 1993b) for the accuracy and precision needed for have all been approved in today’s rule. analysis of bromate and chlorite ions. determining compliance. Subsequent to However, at the time of the proposal, the proposed rule, EPA Method 300.0 3. Summary of Comments EPA was aware that EPA Method 300.0 was improved in order to achieve lower Following is a discussion of major was not sensitive enough to measure detection limits for bromate ion and comments on the analytical methods bromate ion concentration at the renumbered as EPA Method 300.1. requirements of the Stage 1 DBPR. proposed MCL of 10 µg/L. EPA TOC. To satisfy requirements of the Chlorine. A commenter to the 1994 recognized that modifications to the Stage 1 DBPR, the 1994 proposed rule proposal recommended approval of method would be necessary to increase directed that a TOC analytical method ASTM method D1253–86. EPA the method sensitivity. Studies at that should have a detection limit of at least determined that this method is time indicated that changes to the 0.5 mg/L and a reproducibility of ± 0.1 equivalent to Standard Method 4500–Cl injection volume and the eluent mg/L over a range of 2 to 5 mg/L TOC. D, and has approved this method in chemistry would decrease the detection The proposed rule included two today’s rule. limit below the MCL. Many commenters methods for analyzing TOC: Standard Chlorine Dioxide. EPA received to the 1994 proposal agreed that EPA Methods 5310 C, which is the comments on the proposed rule Method 300.0 was not sensitive enough persulfate-ultraviolet oxidation method, detailing weaknesses of the methods to determine compliance with a MCL of and 5310 D, the wet-oxidation method selected to calculate ClO2. Commenters 10 µg/L bromate ion, given that MCLs (APHA, 1992). These methods were pointed out that other halogenated are typically set at 5 times the minimum selected because, according to data species, such as free chlorine, detection levels (MDLs). published in Standard Methods (APHA chloramines, and chlorite, as well as Following the proposal, EPA 1992), they could achieve the necessary common metal ions (e.g. copper, improved EPA Method 300.0 and precision and detection limit. Standard manganese, chromate) will interfere renumbered it as EPA Method 300.1 Method 5310 B, the high-temperature with these methods. Additionally, (EPA, 1997b). EPA Method 300.1 combustion method, was considered but where these methods determine specifies a new, high capacity ion not proposed because it was described concentrations by difference, they are chromatography (IC) column that is in Standard Methods (1992, APHA) as potentially inaccurate and subject to used for the analysis of all anions listed having a detection limit of 1 mg/L. The propagation of errors. Commenters in the method, instead of requiring two proposal stated that if planned specifically criticized Standard Method different columns as specified in EPA improvements to the instrumentation 4500–ClO2 C (APHA 1995), Method 300.0. The new column has a used in Standard Method 5310 B were amperometric method I, which was higher ion exchange capacity that successful, the next version would be characterized as outdated and improves chromatographic resolution considered for promulgation. Revisions inaccurate, and stated that Standard 69420 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

Method 4500–ClO2 E (APHA 1995), Moreover, the efficacy of EPA Method A minority of commenters on this amperometric method II, is a 300.1 in a wide range of sample issue suggested that the DPD method substantially better method. matrices is demonstrated by the (Standard Method 4500–ClO2 D (APHA Consequently, in the 1997 DBP NODA, performance validation data contained 1995)) be approved for daily monitoring EPA requested comment on removing in the published method description. of chlorite ion levels. EPA has Standard Method 4500–ClO2 C from the Based on a review of all the public determined that the accuracy and list of approved methods for the comments, EPA is approving EPA precision of the DPD method (Standard analysis of chlorine dioxide for Method 300.1 for bromate analysis in Method 4500–ClO2 D) in the compliance with the MRDL. today’s rule. measurement of chlorite are Comments on the 1997 DBPR NODA Chlorite. EPA solicited comment in substantially worse than with Standard favored eliminating Standard Method the 1997 DBPR NODA on approving Method 4500–ClO2 E, and are 4500.ClO2 C as an approved method for EPA Method 300.1, in addition to EPA insufficient for this method to be used ClO2 compliance analysis. EPA does not Method 300.0, for compliance analysis for daily monitoring of chlorite. As a approve this method in today’s rule. of chlorite. The majority of commenters consequence, EPA has not approved the EPA recognizes that the two methods on this issue favored approval of both DPD method for chlorite monitoring in approved for ClO2 monitoring under methods and today’s rule establishes today’s rule. today’s rule are subject to interferences. both for determining compliance with TOC. EPA received several comments However, EPA believes that these the chlorite MCL. on the 1994 proposal requesting methods can be used effectively to In the 1994 proposed rule, EPA approval of Standard Method 5310 B for indicate compliance with the ClO2 requested comment on changing TOC compliance analysis. Commenters MRDL when the quality control monitoring requirements for chlorite to stated that newer instrumentation could procedures contained in the method reflect concern about potential acute achieve a detection limit of 0.5 mg/L descriptions are followed. Several health effects. Several commenters TOC using this method. Following the commenters also encouraged EPA to stated that daily monitoring of chlorite publication of a revised version of approve a more sensitive and specific would be feasible if an amperometric Method 5310 B in Standard Methods method for ClO2 analysis, and suggested 19th Edition Supplement (APHA 1996) analytical method could be used. alternative methods including Acid which recognized the capacity of some Commenters suggested that daily Chrome Violet K, Lissamine Green B, combustion based TOC analyzers to amperometric analyses for chlorite be and Chlorophenol Red. While EPA achieve detection limits below 1 mg/L, conducted on samples taken from the supports the development of improved EPA requested comment on approving entrance to the distribution system, and analytical methods for chlorine dioxide, Standard Method 5310 B, along with that weekly or monthly analyses using the Agency believes that at this time the Standard Methods 5310 C and 5310 D, ion chromatography still be required as methods suggested by commenters have for the analysis of TOC in the 1997 a check, because ion chromatography is not gone through the necessary DBPR NODA. a more accurate analytical method. performance validation processes to The majority of commenters on TOC warrant their approval for compliance Commenters noted that daily analysis urged EPA to approve all three monitoring. monitoring for chlorite would provide methods. Commenters were concerned, Bromate. In the 1994 proposed rule, improved operational control of plants though, that because these three EPA discussed the fact that the current and reduce the likelihood of systems methods employ different processes to version of EPA Method 300.0 was not incurring compliance violations. oxidize organic carbon to carbon sensitive enough to measure bromate Today’s rule establishes amperometric dioxide, results from different TOC ion concentrations at the proposed MCL titration (Standard Method 4500–ClO2 analyzers could vary to a degree that is and requested comment on E) for daily analyses of chlorite samples of regulatory significance. Specifically, modifications to EPA Method 300.0 to taken at the entrance to the distribution the efficiency of oxidation of large improve its sensitivity. In the 1997 system, along with monthly (or organic particles or very large organic NODA, EPA presented EPA Method quarterly if reduced, or additional as molecules such as tannins, lignins, and 300.1 and requested comment on required), analyses by ion humic acids may be lower with replacing EPA Method 300.0 with EPA chromatography (EPA Methods 300.0 persulfate based instruments (APHA Method 300.1 for the analysis of and 300.1) of chlorite samples taken 1996). Although available data bromate. from within the distribution system. comparing different TOC methods is Commenters agreed that EPA Method EPA believes that the ion limited, one study observed a persulfate 300.1 is a more sensitive method than chromatography method, rather than the catalytic oxidation technique to EPA Method 300.0 for low level bromate amperometric method, should be used underestimate the TOC concentration analysis and the majority suggested that for making chlorite compliance measured by a high temperature EPA Method 300.1 be the approved determinations in the distribution catalytic oxidation technique by 3–6% method for bromate analysis. One system due to its greater accuracy. on stream water and soil water samples commenter requested that However, the amperometric method is (Kaplan, 1992). Standard Methods interlaboratory round-robin testing be sufficient for the purposes of daily recommends checking the oxidation conducted before EPA Method 300.1 is monitoring at the entrance to the efficiency of the instrument with model accepted for Stage 1 DBPR compliance distribution system, which are to compounds representative of the sample monitoring. EPA considers significantly aid in proper operational matrix, because many factors can interlaboratory round-robin testing of control of a treatment plant and to influence conversion of organic carbon EPA Method 300.1 to be unnecessary indicate when distribution system to carbon dioxide (APHA 1996). because this method is essentially an testing is appropriate. For this reason, EPA believes that the potential improvement of EPA Method 300.0 only the ion chromatographic methods regulatory impact of small disparities in which is already approved. EPA Method (EPA Method 300.0 and 300.1), and not oxidation efficiencies between different 300.1 primarily makes use of a superior the amperometric titration methods, are TOC analyzers is minor. Studies using analytical column to achieve increased approved in today’s rule for determining PE samples indicate that for instruments sensitivity for bromate analysis. compliance with the chlorite MCL. calibrated in accordance with the Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69421 procedures specified in Standard especially filtration, in the measurement identical, EPA anticipates that most Methods (APHA, 1996), the magnitude of DOC and observed that sufficient systems will find it economical when of measurement error due to analytical washing of filters prior to filtration of determining SUVA to filter one sample. discrepancies between instruments will samples is critical to preventing The filtrate would then be split into two typically be less than the measurement contamination of the samples by organic portions, one of which would be used uncertainty attributed to a particular carbon from the filters. Several for UV analysis while the other would instrument (EPA, 1994c). In addition, comments on the 1997 DBPR NODA be acid preserved and used for DOC EPA anticipates that most systems will expressed opposition to a requirement analysis. However, EPA has not use a consistent method for TOC that UV254 and DOC analyses be made included a requirement that the DOC analyses and that this will assist in on the same sample filtrate. and UV254 analyses used in the SUVA minimizing the importance of Commenters stated that this is determination be made on the same instrument bias. This practice was impractical because UV analyses are sample filtrate. Instead, EPA requires suggested by several commenters. often conducted at the treatment plant that DOC and UV254 samples used to Commenters also suggested that EPA while DOC analyses are typically run determine a SUVA value must be taken implement a formal certification process off-site. Commenters also noted that at the same time and at the same for laboratories measuring TOC. Some DOC samples should be acid preserved location. commenters recommended that EPA whereas pH adjustment of samples for In the 1997 DBPR NODA, EPA also require a laboratory approval process for UV254 analysis is improper. observed that because disinfectants/ TOC measurements under the Stage 1 Today’s rule establishes that samples oxidants (chlorine, ozone, chlorine DBPR that is similar to what is required for DOC and UV254 analyses must be dioxide, potassium permanganate) under the ICR. EPA requires that TOC filtered through a 0.45 µm-pore- typically reduce UV254 without analyses be conducted by a party diameter filter. EPA does not have substantially impacting DOC, raw water approved by EPA or the State but not specific requirements on the type of SUVA should be determined on water that TOC measurements be subject to filter that is used, provided it has a 0.45 prior to the application of disinfectants/ the same laboratory certification µm pore-diameter, but will provide oxidants. If disinfectants/oxidants are procedures required for the analysis of guidance on this issue in the Guidance applied in raw-water transmission lines DBPs. However, today’s rule contains Manual for Enhanced Coagulation. This upstream of the plant, then raw water QC requirements for TOC analyses manual will be available for public SUVA should be based on a sample which are in addition to those in review after promulgation of the Stage 1 collected upstream of the point of Standard Methods. These additional QC DBPR. Today’s rule addresses filter disinfectant/oxidant addition. For procedures pertain to sample washing prior to analysis by requiring determining settled-water SUVA, if the preservation and holding time, and have that water passed through the filter prior plant applies disinfectants/oxidants been found to be effective for TOC to filtration of the sample serve as the prior to the settled water sample tap, analyses under the ICR. filtered blank. The filtered blank must then settled-water SUVA should be SUVA. In the 1997 DBPR NODA, EPA be analyzed using procedures identical determined in jar testing. No solicited comment on a range of issues to those used for analysis of the samples commenters were opposed to these dealing with the determination of SUVA and must meet the following criteria: provisions and today’s rule requires that including: analytical methods, TOC < 0.5 mg/L. These criteria are the samples used for SUVA determinations sampling, sample preparation, filter maximum allowable background be taken from water prior to the types, pH, interferences to UV, high concentrations specified for these addition of any oxidants/disinfectants. turbidity waters, quality control, and analyses under the ICR. In the Guidance A few commenters stated that SUVA other issues that should be addressed. Manual for Enhanced Coagulation, EPA should not be subject to rigorous The Agency requested comment on will furnish instructions on sample analytical procedures because the approving Standard Method 5910 B for handling and filter washing to assist application of SUVA in this rule is measuring UV254 and Standard Methods systems in achieving acceptable field based on a relationship which is largely 5310 B, C, and D, for measuring DOC. reagent blanks. empirical (i.e. correlations between In requesting comment on filtration, Filtration of samples for DOC analysis SUVA and TOC removal by EPA noted that filtration is necessary must be done prior to acid preservation, coagulation). EPA recognizes the prior to both UV254 and DOC analyses in as stipulated in today’s rule. This is empirical nature of this relationship and order to eliminate particulate matter and necessary because acidification of the the variance it has displayed in studies. separate the operationally defined sample to pH < 2 can cause substantial Regulations, however, must address dissolved organic matter (based on a precipitation of dissolved organic specific SUVA values if SUVA is to 0.45 µm-pore-diameter cut-off). species. Because biological activity will serve as an alternative compliance However, filtration can also corrupt rapidly alter the DOC of a sample that parameter. For compliance with these samples through adsorption of has not been preserved, EPA requires regulations to be meaningful, SUVA carbonaceous material onto the filter or that DOC samples be acidified to pH < must be determined accurately. its desorption from it (APHA 1996). In 2.0 within 48 hours of sampling. Consequently, today’s rule requires addition, EPA requested comment on Consequently, filtration of DOC samples certain QC procedures in the DOC and requiring that UV254 and DOC analyses must be done within 48 hours in order UV254 analyses that are used to calculate be measured from the same sample to allow acid preservation within this SUVA. filtrate. time period. The pH of UV254 samples Today’s rule establishes the removal The majority of commenters on SUVA may not be adjusted. Today’s rule places of 10 mg/L magnesium hardness (as analytical methods recommended that a maximum holding time from sampling CaCO3) as an alternative performance EPA approve Standard Methods 5310 B, to analysis of 2 days for UV254 samples criterion that systems practicing C, and D, for DOC analysis and Standard and 28 days for DOC samples. These enhanced softening can use to Method 5910 B for UV254 analysis. EPA holding times are the same as those demonstrate compliance with the has approved these methods in today’s approved for ICR data collection. treatment technique requirement for rule. In addition, commenters stressed Because the filtration procedures for TOC removal. However, EPA did not the importance of sample preparation, UV254 and DOC samples are largely propose methods for the analysis of 69422 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations magnesium in drinking water and PBMS approach to ensure adequate data H. Monitoring Requirements therefore the final rule does not contain quality. EPA would then specify 1. Today’s Rule any approved methods for magnesium. performance requirements in its EPA expects to propose magnesium regulations to ensure that any method Today’s rule establishes monitoring analytical methods to be used for used for determination of a regulated requirements to support implementation compliance monitoring under the Stage analyte is at least equivalent to the of the enhanced coagulation and 1 DBPR by the end of 1998. performance achieved by other enhanced softening treatment technique, implementation of new 4. Performance Based Measurement currently approved methods. EPA expects to publish its PBMS MCLs for TTHM, HAA5, bromate, and Systems chlorite, and implementation of MRDLs implementation strategy for water On October 6, 1997, EPA published a for chlorine, chloramines, and chlorine programs in the Federal Register by the Document of the Agency’s intent to dioxide. Monitoring for DBPs, end of calendar year 1998. implement a Performance Based disinfectant residuals, and TOC must be Measurement System (PBMS) in all of Once EPA has made its final conducted during normal operating its programs to the extent feasible (EPA, determinations regarding conditions. Failure to monitor in 1997f). The Agency is currently implementation of PBMS in programs accordance with the monitoring plan is determining the specifics steps under the Safe Drinking Water Act, EPA a monitoring violation. Where necessary to implement PBMS in its would incorporate specific provisions of compliance is based on a running programs and preparing an PBMS into its regulations, which may annual average of monthly or quarterly implementation plan. Final decisions include specification of the performance samples or averages and the system’s have not yet been made concerning the characteristics for measurement of failure to monitor makes it impossible to implementation of PBMS in drinking regulated contaminants in the drinking determine compliance with MCLs or water programs. However, EPA is water program regulations. MRDLs, this failure to monitor will be currently evaluating what relevant treated as a violation. performance characteristics should be Tables III–3 and III–4 below specified for monitoring methods used summarize routine and reduced in the drinking water programs under a monitoring requirements of today’s rule.

TABLE III±3.ÐROUTINE MONITORING REQUIREMENTS 1

Requirement Location for sampling Large surface sys- Small surface sys- Large ground water Small ground water (reference) tems 2 tems 2 systems 3 systems 3

TOC and Alkalinity Source Water 4 ...... 1sample/month/plant 3 1 sample/month/ NA ...... NA. (141.132(d)(1)). plant3. Only required for plants with conven- tional filtration treat- ment. TTHMs and HAA5 25% in dist sys at 4/plant/quarter ...... 1/plant/quarter5 ...... 1/plant/quarter 6 ...... 1/plant/year 5,6 (141.132(b)(1)(i)). max res time, 75% at dist sys rep- resentative loca- tions. at maximum resi- at maximum resi- at maximum resi- dence time. dence time. dence time. if pop.<500, then 1/ during warmest plant/yr 8. month. during warmest month. Bromate 7 Dist sys entrance 1/month/trt plant 1/month/trt plant 1/month/trt plant 1/month/trt plant (141.132(b)(3)(i)). point. using O3. using O3. using O3. using O3. Chlorite8 (daily) Dist sys entrance Daily/trt plant using Daily/trt plant using Daily/trt/plant using (141.132(b)(2)(i)(A)). point. CIO2. CIO2. CIO2. Chlorite8 (monthly) Dist sys: 1 near first 3 sample set/month .. 3 sample set/month .. 3 sample set/month .. 3 sample set/month. 141.132(b)(2)(i)(B)). cust, 1 in dist sys middle, 1 at max res time. Chlorine and Same points as total Same times as total Same times as total Same times as total Same times as total chloramines coliform in TCR. coliform in TCR. coliform in TCR. coliform in TCR. coliform in TCR. (141.132(c)(1)(i)). Chlorine dioxide8 Dist sys entrance Daily/trt plant using Daily/trt plant using Daily/trt plant using Daily/trt plant using (141.132(c)(2)(i)). point. CIO2. CIO2. CIO2. CIO. 1 Samples must be taken during representative operating conditions. Provisions for reduced monitoring shown elsewhere. 2 Large surface (subpart H) systems serve 10,000 or more persons. Small surface (subpart H) systems serve fewer than 10,000 persons. 3 Large systems using ground water not under the direct influence of surface water serve 10,000 or more persons. Small systems using ground water not under the direct influence of surface water serve fewer than 10,000 persons. 4 Subpart H systems which use conventional filtration treatment (defined in section 141.2) must monitor 1) source water TOC prior to any treat- ment and 2) treated TOC at the same time; these two samples are called paired samples. Systems must take a source water alkalinity sample at the same time. 5 If the annual monitoring result exceeds the MCL, the system must increase monitoring frequency to 1/plant/quarter. Compliance determina- tions will be based on the running annual average of quarterly monitoring results. Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69423

6 Multiple wells drawing water from a single aquifer may, with State approval, be considered one treatment plant for determining the minimum number of samples. 7 Only required for systems using ozone for oxidation or disinfection. 8 Only required for systems using chlorine dioxide for oxidation or disinfection. Additional chlorite monitoring required if daily sample exceeds MCL. Additional chlorine dioxide monitoring requirements apply if any chlorine dioxide sample exceeds the MRDL.

TABLE III±4.ÐREDUCED MONITORING REQUIREMENTS 1

Requirement Location for reduced 2 (reference) sampling Reduced monitoring frequency and prerequisites

TOC and Alkalinity Paired samples 3 ...... Subpart H systems-reduced to 1 paired sample/plant/quarter if 1) avg TOC < 2.0 mg/l for 2 (141.132(d)(2)). years or 2) avg TOC < 1.0 mg/l for 1 year. TTHMs and HAA5s In dist sys at point with Monitoring cannot be reduced if subpart H system source water TOC > 4.0 mg/l. (141.132(b)(1)(ii)). max res time. Subpart H systems serving 10,000 or more-reduced to 1/plant/qtr if 1) system has completed at least 1 yr of routine monitoring and 2) both TTHM and HAA5 running annual averages are no more than 40 µg/l and 30 µg/l, respectively. Subpart H systems serving <10,000 and ground water systems 4 serving 10,000 or more-re- duced to 1/plant/yr if 1) system has completed at least 1 yr of routine monitoring and 2) both TTHM and HAA5 running annual averages are no more than 40 µg/l and 30 µg/l, re- spectively. Samples must be taken during month of warmest water temperature. Subpart H systems serving <500 may not reduce monitoring to less than 1/plant/yr. Groundwater systems 6 serving<10,000-reduced to 1/plant/3yr if 1) system has completed at least 2 yr of routine monitoring and both TTHM and HAA5 running annual averages are no more than 40 µg/l and 30 µg/l, respectively or 2) system has completed at least 1 yr of routine monitoring and both TTHM and HAA5 annual samples are no more than 20 µg/l and 15 µg/l, respectively. Samples must be taken during month of warmest water tempera- ture. 5 Bromate Dist sys entrance point 1/qtr/trt plant using O3, if system demonstrates 1) avg raw water bromide <0.05 mg/l (based (141.132(b)(3)(ii)). on annual avg of monthly samples). Chlorite 6 Dist sys: 1 near first Systems may reduce routine distribution system monitoring from monthly to quarterly if the (141.132(b)(2)(iii)). cust, 1 in dist sys chlorite concentration in all samples taken in the distribution system is below 1.0 mg/L for a middle, 1 at max res period of one year; 3 samples per quarter. time. Chlorine, chlorine diox- NA ...... Monitoring may not be reduced. ide 6, chloramines (141.132(c)(2)(ii) and (c)(2)(iii). 1 Samples must be taken during representative operating conditions. Provisions for routine monitoring shown elsewhere. 2 Requirements for cancellation of reduced monitoring are found in the regulation. 3 Subpart H systems which use conventional filtration treatment (defined in Section 141.2) must monitor 1) source water TOC prior to any treat- ment and 2) treated TOC before continuous disinfection (except that systems using ozone followed by biological filtration may sample after bio- logical filtration) at the same time; these two samples are called paired samples. 4 Multiple wells drawing water from a single aquifer may, with State approval, be considered one treatment plant for determining the minimum number of samples. 5 Only required for systems using ozone for oxidation or disinfection. 6 Only required for systems using chlorine dioxide for oxidation or disinfection.

The formation rate of DBPs is affected remaining samples must be taken at Reduced Monitoring: To qualify for by type and amount of disinfectant locations representative of at least reduced monitoring, systems must meet used, water temperature, pH, amount average residence time in the certain prerequisites (see Figure III–1). and type of precursor material in the distribution system. Systems eligible for reduced monitoring water, and the length of time that water Subpart H Systems Serving 10,000 or may reduce the monitoring frequency remains in the treatment and More People. Routine Monitoring: CWSs for TTHMs and HAA5 to one sample per distribution systems. For this reason, and NTNCWSs using surface water (or treatment plant per quarter. Systems on today’s rule specifies the points in the ground water under direct influence of a reduced monitoring schedule may distribution system (and, in some cases, surface water) (Subpart H systems) that remain on that reduced schedule as long the time) where samples must be taken. treat their water with a chemical as the average of all samples taken in For purposes of this regulation, multiple disinfectant and serve 10,000 or more the year is no more than 0.060 mg/L for wells drawing raw water from a single people must routinely take four water TTHM and 0.045 mg/L for HAA5. aquifer may, with State approval, be samples each quarter for both TTHMs Systems that do not meet these levels considered one plant for determining and HAA5 for each treatment plant in must revert to routine monitoring in the the minimum number of samples. the system. At least 25 percent of the quarter immediately following the TTHM and HAA5. Any system may samples must be taken at the point of quarter in which the system exceeded take samples in excess of the required maximum residence time in the 0.060 mg/L for TTHM or 0.045 mg/L for frequency. In such cases, at least 25 distribution system. The remaining HAA5. Additionally, the State may percent of all samples collected each samples must be taken at representative return a system to routine monitoring at quarter must be taken at locations points in the distribution system. This the State’s discretion. within the distribution system that monitoring frequency is the same as the represent the maximum residence time frequency required under the current of the water in the system. The TTHM rule (§ 141.30). 69424 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

FIGURE III±1.ÐELIGIBILITY FOR REDUCED TTHM AND HAA5 MONITORING: GROUND WATER SYSTEMS SERVING 10,000 OR MORE PEOPLE AND SUBPART H SYSTEMS SERVING 500 OR MORE PEOPLE

Ground water systems serving 10,000 or more people, and Subpart H systems serving 500 or more people, may reduce monitoring of TTHMs and HAA5 if they meet all of the following conditions: ÐThe annual average for TTHMs is no more than 0.040 mg/L. ÐThe annual average for HAA5 is no more than 0.030 mg/L. ÐAt least one year of routine monitoring has been completed. ÐAnnual average source water TOC level is no more than 4.0 mg/L prior to treatment (applies to Subpart H systems only).

Compliance Determination: A public system must increase monitoring to one Compliance Determination: A PWS is water system (PWS) is in compliance sample per treatment plant per quarter, in compliance with the MCL when the with the MCL when the running annual taken at the point of maximum running arithmetic annual average of arithmetic average of quarterly averages residence time in the distribution quarterly averages of all samples, of all samples, computed quarterly, is system. computed quarterly, is less than or less than or equal to the MCL. If the Reduced Monitoring: These systems equal to the MCL. If the running annual running annual average computed for may not reduce monitoring. Systems on average for any quarter exceeds the any quarter exceeds the MCL, the increased monitoring may return to MCL, the system is out of compliance. system is out of compliance. routine monitoring if the annual average Subpart H Systems Serving 500 to of quarterly samples is no more than Ground Water Systems Serving Fewer 9,999 People. Routine Monitoring: 0.060 mg/L for TTHM and 0.045 mg/L than 10,000 People Routine Monitoring: Systems are required to take one water for HAA5. CWSs and NTNCWSs using only ground sample each quarter for each treatment Compliance Determination: A PWS is water sources not under the direct plant in the system. Samples must be in compliance when the annual sample influence of surface water that treat taken at the point of maximum (or average of annual samples, if their water with a chemical disinfectant residence time in the distribution additional sampling is conducted) is and serve fewer than 10,000 people are system. less than or equal to the MCL. If the required to sample once per year for Reduced Monitoring: To qualify for annual sample exceeds the MCL, the each treatment plant in the system. The reduced monitoring, systems must meet system must increase monitoring to one sample must be taken at the point of certain prerequisites (see Figure III–1). sample per treatment plant per quarter. maximum residence time in the Systems eligible for reduced monitoring If the running annual average of the distribution system during the month of may reduce the monitoring frequency quarterly samples then exceeds the warmest water temperature. If the for TTHMs and HAA5 to one sample per MCL, the system is out of compliance. sample (or the average of annual treatment plant per year. Sample must Ground Water Systems Serving 10,000 samples if more than one sample is be taken at a distribution system or More People. Routine Monitoring: location reflecting maximum residence CWSs and NTNCWSs using only ground taken) exceeds the MCL, the system time and during the month of warmest water sources not under the direct must increase monitoring to one sample water temperature. Systems on a influence of surface water that treat per treatment plant per quarter. reduced monitoring schedule may their water with a chemical disinfectant Reduced Monitoring: To qualify for remain on that reduced schedule as long and serve 10,000 or more people are reduced monitoring, systems must meet as the average of all samples taken in required to take one water sample each certain prerequisites (see Figure III–2). the year is no more than 0.060 mg/L for quarter for each treatment plant in the Systems eligible for reduced monitoring TTHM and 0.045 mg/L for HAA5. system. Samples must be taken at points may reduce the monitoring frequency Systems that do not meet these levels that represent the maximum residence for TTHMs and HAA5 to one sample per must revert to routine monitoring in the time in the distribution system. three-year monitoring cycle. Sample Reduced Monitoring: To qualify for quarter immediately following the must be taken at a distribution system reduced monitoring, systems must meet quarter in which the system exceeded location reflecting maximum residence 0.060 mg/L for TTHM or 0.045 mg/L for certain prerequisites (see Figure III–1). time and during the month of warmest HAA5. Additionally, the State may Systems eligible for reduced monitoring water temperature. Systems on a return a system to routine monitoring at may reduce the monitoring frequency to the State’s discretion. one sample per treatment plant per year. reduced monitoring schedule may Compliance Determination: A PWS is Sample must be taken at a distribution remain on that reduced schedule as long in compliance with the MCL for TTHM system location reflecting maximum as the average of all samples taken in and HAA5 when the annual average of residence time and during the month of the year is no more than 0.060 mg/L for all samples, taken that year, is less than warmest water temperature. Systems on TTHM and 0.045 mg/L for HAA5. or equal to the MCL. If the average for a reduced monitoring schedule may Systems that do not meet these levels these samples exceeds the MCL, the remain on that reduced schedule as long must resume routine monitoring. system is out of compliance. as the average of all samples taken in Systems on increased monitoring may Subpart H Systems Serving Fewer the year is no more than 0.060 mg/L for return to routine monitoring if the than 500 People. Routine Monitoring: TTHM and 0.045 mg/L for HAA5. annual average of quarterly samples is Subpart H systems serving fewer than Systems that do not meet these levels no more than 0.060 mg/L for TTHM and 500 people are required to take one must revert to routine monitoring in the 0.045 mg/L for HAA5. sample per year for each treatment plant quarter immediately following the in the system. The sample must be taken quarter in which the system exceeded Compliance Determination: A PWS is at the point of maximum residence time 0.060 mg/L for TTHM or 0.045 mg/L for in compliance when the annual sample in the distribution system during the HAA5. Additionally, the State may (or average of annual samples) is less month of warmest water temperature. If return a system to routine monitoring at than or equal to the MCL. the annual sample exceeds the MCL, the the State’s discretion. Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69425

FIGURE III±2.ÐELIGIBILITY FOR REDUCED TTHM AND HAA5 MONITORING: GROUND WATER SYSTEMS SERVING FEWER THAN 10,000 PEOPLE

Systems using ground water not under the direct influence of surface water that serve fewer than 10,000 people may reduce monitoring for TTHMs and HAA5 if they meet either of the following conditions: 1. The average of two consecutive annual samples for TTHMs is no more than 0.040 mg/L, the average of two consecutive annual samples for HAA5 is no more than 0.030 mg/L, and at least two years of routine monitoring has been completed. 2. The annual sample for TTHMs is no more than 0.020 mg/L, the annual sample for HAA5 is no more than 0.015 mg/L, and at least one year of routine monitoring has been completed.

Chlorite. Routine Monitoring: CWSs requirement for that month. Further sampling done under the SWTR and NTNCWSs using chlorine dioxide distribution system monitoring will not (§ 141.74(b)(6)(i) for unfiltered systems, for disinfection or oxidation are be required in that month unless the § 141.74(c)(3)(i) for systems that filter) required to conduct sampling for chlorite concentration at the entrance to in lieu of taking separate samples. chlorite both daily at the entrance to the the distribution system again exceeds Reduced Monitoring: Monitoring for distribution system and monthly within 1.0 mg/L. chlorine may not be reduced. the distribution system. Additional Reduced Monitoring: Systems may Compliance Determination: A PWS is distribution system monitoring may be reduce routine distribution system in compliance with the MRDL when the required, and distribution system monitoring for chlorite from monthly to running annual arithmetic average of monitoring may be reduced if certain quarterly if the chlorite concentration in monthly averages of all samples, conditions are met. This monitoring is all samples taken in the distribution computed quarterly, is less than or described below. system is below 1.0 mg/L for a period equal to the MRDL. Notwithstanding the Routine Monthly Monitoring— of one year and the system has not been MRDL, operators may increase residual Systems are required to take a three required to conduct any additional chlorine levels in the distribution sample set each month in the monitoring. Systems that qualify for system to a level and for a time distribution system. One sample must reduced monitoring must continue to necessary to protect public health to be taken at each of the following conduct daily monitoring at the address specific microbiological locations: (1) as close as possible to the entrance to the distribution system. If contamination problems (e.g., including first customer, (2) in a location the chlorite concentration at the distribution line breaks, storm runoff representative of average residence time, entrance to the distribution system events, source water contamination, or and (3) as close as possible to the end exceeds 1.0 mg/L, the system must cross-connections). of the distribution system (reflecting resume routine monthly monitoring. Chloramines. Routine Monitoring: As maximum residence time in the Compliance Determination: A PWS is a minimum, CWSs and NTNCWSs must distribution system). As described out of compliance with the chlorite measure the residual disinfectant level elsewhere in this document, all samples MCL when the arithmetic average (as either total chlorine or combined taken in the distribution system must be concentration of any three sample set chlorine) at the same points in the analyzed by ion chromatography taken in the distribution system is distribution system and at the same time (Methods 300.0 and 300.1). greater than 1.0 mg/L. as total coliforms, as specified in Routine Daily Monitoring—Systems Bromate. Routine Monitoring: CWSs § 141.21. Subpart H systems may use the must take one sample each day at the and NTNCWSs using ozone for results of residual disinfectant entrance to the distribution system. As disinfection or oxidation are required to concentration sampling done under the described elsewhere in this document take at least one sample per month for SWTR (§ 141.74(b)(6) for unfiltered (section III.G), samples taken at the each treatment plant in the system using systems, § 141.74(c)(3) for systems that distribution system entrance may be ozone. The sample must be taken at the filter) in lieu of taking separate samples. analyzed by amperometric titration entrance to the distribution system Reduced Monitoring: Monitoring for (Method 4500–ClO2 E). If the chlorite when the ozonation system is operating chloramines may not be reduced. MCL is exceeded at the entrance to the under normal conditions. Compliance Determination: A PWS is distribution system, the system is not Reduced Monitoring: Systems may in compliance with the MRDL when the out of compliance. However, the system reduce monitoring from monthly to running annual arithmetic average of must carry out addition monitoring as once per quarter if the system monthly averages of all samples, described in the following paragraph. demonstrates that the annual average computed quarterly, is less than or Additional Monitoring: On any day raw water bromide concentration is less equal to the MRDL. Notwithstanding the when the chlorite concentration than 0.05 mg/L, based upon monthly MRDL, operators may increase residual measured at the entrance to the measurements for one year. chloramine levels in the distribution distribution system exceeds the chlorite Compliance Determination: A PWS is system to a level and for a time MCL (1.0 mg/L), the system is required in compliance if the running annual necessary to protect public health to to take a three sample set in the arithmetic average of samples, address specific microbiological distribution system on the following computed quarterly, is less than or contamination problems (e.g., including day, at the locations specified for equal to the MCL. distribution line breaks, storm runoff routine monthly monitoring. If the Chlorine. Routine Monitoring: As a events, source water contamination, or system is required to conduct minimum, CWSs and NTNCWSs must cross-connections). distribution system monitoring as a measure the residual disinfectant level Chlorine Dioxide Routine Monitoring: result of having exceeded the chlorite (as either free chlorine or total chlorine) CWSs, NTNCWSs, and TNCWSs must MCL at the entrance to the distribution at the same points in the distribution monitor for chlorine dioxide only if system, and the average of the three system and at the same time as total chlorine dioxide is used by the system samples taken in the distribution system coliforms, as specified in § 141.21. for disinfection or oxidation. If is below 1.0 mg/L, the system will have Subpart H systems may use the results monitoring is required, systems must satisfied its routine monthly monitoring of residual disinfectant concentration take daily samples at the entrance to the 69426 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations distribution system. If the MRDL (0.8 a monthly basis, with samples taken in (III.A.7), supported EPA’s finding of mg/L) is exceeded, the system must both the source water prior to any neurobehavioral health effects resulting conduct additional monitoring. treatment and in the treated water no from chlorite, along with the rationale Additional Monitoring: If any daily later than the point of combined filter for daily monitoring at the entrance to sample taken at the entrance to the effluent turbidity monitoring. At the the distribution system as a trigger for distribution system exceeds the MRDL, same time, systems must monitor for further compliance monitoring in the the system is required to take three source water alkalinity. distribution system. additional samples in the distribution Reduced Monitoring: Subpart H system on the next day. Samples must systems with an average treated water 3. Summary of Comments be taken at the following locations. TOC of less than 2.0 mg/L for two TOC. Many commenters expressed Systems using chlorine as a residual consecutive years, or less than 1.0 mg/ confusion regarding the raw and disinfectant and operating booster L for one year, may reduce monitoring finished water TOC monitoring scheme chlorination stations after the first for both TOC and alkalinity to one and their relationship to compliance customer—These systems must take paired sample per plant per quarter. calculations. Commenters noted, three samples in the distribution Compliance Determination: correctly, that changes in alkalinity and system: one as close as possible to the Compliance criteria for TOC are TOC level can move the utility to a first customer, one in a location dependent upon a variety of factors and different box of the TOC removal representative of average residence time, is discussed elsewhere in this rule. matrix, and questioned whether this and one as close as possible to the end would affect requisite monitoring. As in 2. Background and Analysis of the distribution system (reflecting the proposal, moving to a different box maximum residence time in the The monitoring requirements in of the matrix will not affect monitoring distribution system). today’s rule are the same as those in the requirements. Utilities are required to Systems using chlorine dioxide or 1994 proposed rule, with the exception take a minimum of one paired (raw and chloramines as a residual disinfectant or of requirements for bromide monitoring finished water) TOC sample per month. chlorine as a residual disinfectant and and chlorite. Commenters were also concerned that not operating booster chlorination Bromide Monitoring for Reduced the TOC monitoring provisions would stations after the first customer—These Bromate Monitoring. The 1994 proposal limit their ability to take additional TOC systems must take three samples in the included a provision for reduced samples for operational control. This distribution system as close as possible bromate monitoring for utilities with concern is unfounded; EPA to the first customer at intervals of not source water bromide concentrations recommends in the Enhanced less than six hours. less than 0.05 mg/L. EPA believes there Coagulation and Enhanced Precipitative Reduced Monitoring: Monitoring for is a very small likelihood that systems Softening Guidance Manual that chlorine dioxide may not be reduced. using ozone will exceed the bromate utilities take as many TOC samples as Compliance Determination: Acute MCL if source water bromide necessary to maintain proper violations—If any daily sample taken at concentrations are below this level. The operational control. EPA also the entrance to the distribution system provision did not specify a bromide recommends that TOC compliance exceeds the MRDL and if, on the monitoring frequency, however. Today’s samples, as opposed to operational following day, one or more of the three rule allows utilities to reduce bromate samples, be taken on a constant samples taken in the distribution system monitoring from monthly to once per schedule or be identified one month exceeds the MRDL, the system will be quarter if the system demonstrates, prior to the samples being taken. This in acute violation of the MRDL and based on representative monthly will allow utilities to take numerous must issue the required acute public samples over the course of a year, that operational samples and still provide for notification. Failure to monitor in the the average raw water bromide unbiased compliance sampling. Systems distribution system on the day following concentration is less than 0.05 mg/L. may use their sampling plans for this an exceedance of the chlorine dioxide Chlorite Monitoring. The proposed purpose. MRDL shall also be considered an acute rule required treatment plants using Chlorite. In the proposal, EPA MRDL violation. chlorine dioxide to monitor for chlorite solicited comment on changing the Nonacute violations—If any two ion by taking a three sample set in the frequency and location of chlorite consecutive daily samples taken at the distribution system, once per month, monitoring in consideration of potential entrance to the distribution system and to analyze these samples using ion acute health effects. Commenters stated exceed the MRDL, but none of the chromatography. However, the proposal that daily monitoring of chlorite would samples taken in the distribution system states that after the Negotiating be feasible if amperometric titration exceed the MRDL, the system will be in Committee had agreed to the above were allowed as an analytical method. nonacute violation of the MRDL. Failure monitoring scheme for chlorite at its last Commenters recommended that daily to monitor at the entrance to the meeting in June, 1993, EPA’s Reference amperometric analyses for chlorite be distribution system on the day following Dose Committee met and determined a conducted on samples taken from the an exceedance of the chlorine dioxide different toxicological endpoint for entrance to the distribution system, and MRDL shall also be considered a chlorite, based on the identification of that weekly or monthly analyses using nonacute MRDL violation. neurobehavioral effects. In light of this ion chromatography still be required as Important Note: Unlike chlorine and finding, EPA asserted that it did not a check since ion chromatography is a chloramines, the MRDL for chlorine believe the proposed monthly more accurate analytical method. dioxide may not be exceeded for short monitoring requirement for chlorite was Several comments stated that daily periods of time to address specific sufficiently protective of public health. monitoring for chlorite would improve microbiological contamination Following the proposed rule, EPA operational control of plants and problems. acquired additional information on decrease the probability of a PWS TOC. Routine Monitoring: CWSs and chlorite toxicity, including the results of exceeding the chlorite MCL in the NTNCWSs which use conventional a two-generation study sponsored by the distribution system. However, filtration treatment must monitor each CMA. This additional information, commenters requested that if daily treatment plant water source for TOC on discussed elsewhere in this document monitoring for chlorite were to be Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69427 required, a provision for reduced I. Compliance Schedules discussed in more detail below, EPA chlorite monitoring be included as well. believes this is both consistent with the 1. Today’s Rule In response to these comments, requirements of section 1412(b)(10) as Today’s action establishes revised today’s rule requires treatment plants well as with legislative history affirming compliance deadlines for States to adopt the Reg. Neg. objectives of simultaneous using chlorine dioxide to conduct daily and for public water systems to monitoring for chlorite by taking one compliance and minimization of risk- implement the requirements in this risk tradeoff. sample at the entrance to the rulemaking. Central to the distribution system. This sample may be determination of these deadlines are the 2. Background and Analysis measured using amperometric titration principles of simultaneous compliance The background, factors, and (Standard Method 4500-ClO 2 E). between the Stage 1 DBPR and the competing concerns that EPA Treatment plants are also required to corresponding rules (Interim Enhanced considered in developing the take a three sample set from the Surface Water Treatment Rule, Long compliance deadlines in today’s rule are distribution system once per month, as Term Enhanced Surface Water explained in detail in both the Agency’s was proposed in 1994. In addition, Treatment Rule, and Ground Water IESWTR and Stage 1 DBPR November today’s rule requires that on any day Rule) to ensure continued microbial 1997 NODAs. As explained in those that the concentration of chlorite protection, and minimization of risk- NODAs, EPA identified four options to measured at the distribution system risk tradeoffs. These deadlines also implement the requirements of the 1996 entrance exceeds the MCL, the reflect new legislative provisions SDWA Amendments. The requirements treatment plant must take a three enacted as part of 1996 SDWA outlined above reflect the fourth option sample set in the distribution system on amendments. Section 1412 (b)(10) of the that EPA requested comment upon in the following day. All samples taken in SDWA as amended provides PWSs must November 1997. the distribution system must be comply with new regulatory By way of background, the SDWA analyzed by ion chromatography requirements 36 months after 1996 Amendments affirmed several key (Method 300.0 or 300.1). promulgation (unless EPA or a State principles underlying the M–DBP EPA recommends that treatment determines that an earlier time is compliance strategy developed by EPA plants keep chlorite levels below 1.0 practicable or that additional time up to and stakeholders as part of the 1992 mg/L and believes that if treatment two years is necessary for capital regulatory negotiation process. First, plants exceed the MCL in finished improvements). In addition, Section under Section 1412(b)(5)(A), Congress water, immediate distribution system 1413(a)(1) provides that States have 24 recognized the critical importance of testing is warranted to ensure that instead of the previous 18 months from addressing risk/risk tradeoffs in chlorite levels are below 1.0 mg/L. EPA promulgation to adopt new drinking establishing drinking water standards has not, however, changed the water standards. and gave EPA the authority to take such Applying the 1996 SDWA risks into consideration in setting MCL compliance determination for chlorite Amendments to today’s action, this or treatment technique requirements. from the 1994 proposed rule. rulemaking provides that States have The technical concerns and policy Compliance is still based on the average two years from promulgation to adopt objectives underlying M/DBP risk/risk of three sample sets taken in the and implement the requirements of this tradeoffs are referred to in the initial distribution system. The results of daily regulation. Simultaneous compliance sections of today’s rule and have monitoring do not serve as a compliance will be achieved as follows. remained a key consideration in EPA’s violation; rather, they can only trigger Subpart H water systems covered by development of appropriate compliance immediate distribution system today’s rule that serve a population of requirements. Second, Congress monitoring. Moreover, if the treatment 10,000 or more generally have three explicitly adopted the phased M–DBP plant is required to take distribution years from promulgation to comply with regulatory development schedule system samples by the results of daily all requirements of this rule. In cases developed by the Negotiating monitoring and the average chlorite where capital improvements are needed Committee. Section 1412(b)(2)(C) concentration in the three distribution to comply with the rule, States may requires that the M/DBP standard system samples is below the MCL, then grant such systems up to an additional setting intervals laid out in EPA’s that sampling will meet the treatment two years to comply. These deadlines proposed ICR rule be maintained even plant’s requirement for routine monthly were consistent with those for the if promulgation of one of the M–DBPRs monitoring in the distribution system IESWTR. is delayed. As explained in the 1997 for that month. Today’s rule also Subpart H systems that serve a NODA, this phased or staggered includes a provision for reduced population of less than 10,000 and all regulatory schedule was specifically chlorite monitoring. Treatment plants ground water systems will be required designed as a tool to minimize risk/risk may reduce routine distribution system to comply with applicable Stage 1 DBPR tradeoff. A central component of this monitoring for chlorite from monthly to requirements within five years from approach was the concept of quarterly if the chlorite concentration in promulgation. Since the Long Term ‘‘simultaneous compliance’’, which all samples both at the entrance to the Enhanced Surface Water Treatment Rule provides that a PWS must comply with distribution system and within the (LT1) requirements that apply to new microbial and DBP requirements at distribution system are below 1.0 mg/L systems under 10,000 and the Ground the same time to assure that in meeting for a period of one year. Water Rule are scheduled to be a set of new requirements in one area, In summary, after review of all public promulgated two years after today’s rule a facility does not inadvertently increase comments and associated data, EPA or in November 2000, the net result of the risk (i.e., the risk ‘‘tradeoff’’) in the believes that these provisions for this staggered deadline is that these other area. chlorite monitoring will be both feasible systems will be required to comply with A complicating factor that EPA took for treatment plants and provide a level both Stage 1 DBPR and LT1/GWR into account in developing today’s of protection to public health requirements three years after deadlines is that the SDWA 1996 commensurate with the toxic effects promulgation of LT1/GWR at the same Amendments changed two statutory associated with chlorite. end date of November 2003. For reasons provisions that elements of the 1992 69428 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

Negotiated Rulemaking Agreement were water systems with the Stage 1 DBPR 3. Summary of Comments based upon. The 1994 Stage 1 DBPR and and the GWR. As in the case of subpart Commenters were in general ICR proposals provided that 18 months H systems serving fewer than 10,000, agreement that the compliance deadline after promulgation large PWS would system operators will not know until strategy contained in the fourth option comply with the rules and States would November 2000 what the final of the 1997 NODA did the best job of adopt and implement the new compliance requirements for both rules complying with the requirements to requirements. As noted above, Section are. EPA thus believes it appropriate to 1996 SDWA Amendments and meeting 1412(b)(10) of the SDWA as amended grant the additional two years for the objectives of the 1993 Reg. Neg. now provides that drinking water rules compliance with the Stage 1 DBPR Agreement that Congress affirmed as shall become effective 36 months after allowed by the statute. part of the 1996 Amendments. promulgation (unless the Administrator EPA has been very successful in Nonetheless, a number of commenters determines that an earlier time is meeting all of the new statutory expressed concern about the ability of practicable or that additional time for large surface water systems that had to capital improvements is necessary—up deadlines and is on track for the LT1 make capital improvements to comply to two years). In addition, Section Rule and GWR. While EPA fully intends with all requirements of the Stage 1 1413(a)(1) now provides that States have to meet the schedule discussed earlier, DBPR and IESWTR. They pointed out 24 instead of the previous 18 months to if those rules are delayed the Agency that capital improvements include more adopt new drinking water standards that will evaluate all available options to than just the construction, but also have been promulgated by EPA. protect against unacceptable risk-risk Today’s compliance deadline trade-offs. Part of this effort is the financing, design, and approval. EPA believes that the provisions of requirements reflect the principle of extensive outreach to systems already Section 1412(b)(10) of the SDWA as simultaneous compliance and the underway to fully inform water supplies amended allow systems the flexibility concern with risk/risk tradeoffs. Subpart of the likely elements in the upcoming H systems serving a population of at rules. In addition, EPA would consider needed to comply. As noted earlier in least 10,000 will be required to comply including provisions for streamlined this section, States may grant up to an with the key provisions of this rule on variance and/or exemption processing additional two years compliance time the same schedule as they will be in these rules if they were delayed, in for an individual system if capital required to comply with the parallel order to enhance State flexibility in improvements are necessary. Moreover, requirements of the accompanying ensuring that compliance with the Stage as both of these rules have been under IESWTR that is also included in today’s 1 DBPR is not required before the negotiation since 1992, proposed in Federal Register. corresponding microbial protection rule. 1994 and further clarified in 1997, EPA believes that most systems have had With regard to subpart H systems Under today’s Stage 1 DBPR, EPA has serving fewer than 10,000, EPA believes substantial time to consider how to already provided small subpart H proceed with implementation and to that providing a five year compliance systems and ground water systems the period under Stage 1 DBPR is initiate preliminary planning. Several two-year extension for capital commenters also supported delaying the appropriate and warranted under improvements since these systems will section 1412(b)(10), which expressly promulgation of the Stage 1 DBPR for not know with certainty until November ground water systems until the GWR is allows five years where necessary for 2000 if capital improvements will be capital improvements. As discussed in promulgated, in order to ensure needed for simultaneous compliance simultaneous compliance with both more detail in the 1997 IESWTR NODA, with the Stage 1 DBPR and LT1/GWR. capital improvements require, of rules. EPA believes that this option States considering whether to grant a would not be consistent with the reg- necessity, preliminary planning and two-year capital improvement extension evaluation. An essential prerequisite of neg agreement, as endorsed by Congress, for compliance with the GWR or LT1 because the agreement specifies that the such planning is a clear understanding will also need to consider the impact of of final compliance requirements that Stage 1 DBPR will apply to all such extensions on compliance with must be met. In the case of the staggered community and nontransient today’s rule, given that a similar M/DBP regulatory schedule established noncommunity water systems. extension for capital improvement has as part of the 1996 SDWA Amendments, Moreover, EPA has committed to the already been provided in the initial LT1 microbial requirements for systems LT1 and GWR promulgation schedule compliance schedule for the Stage 1 under 10,000 are required to be outlined above precisely to address this DBPR. EPA believes, however, that promulgated two years after the final issue. these systems will generally not require Stage 1 DBPR. As a result, small systems In conclusion EPA believes that the extensive capital improvements that will not even know what their final compliance deadlines outlined above take longer than three years to install to combined compliance obligations are for systems covered by this rule are until promulgation of the LT 1 rule. meet Stage 1 DBPR, GWR, and LT1 appropriate and consistent with the Thus, an additional two year period requirements, or will require no capital requirements of the 1996 SDWA reflecting the two year Stage 1 DBPR/LT improvements at all. However if needed, amendments. The Agency notes, 1 regulatory development interval EPA will work with States and utilities however, that some elements of Option established by Congress is required to to address systems that require time 4 outlined in the 1997 NODA apply to allow for the preliminary planning and beyond November 2003 to comply. This systems that may be covered by future design steps which are inherent in any strategy may include exemptions. Long Term Enhanced and Ground Water capital improvement process. In addition, EPA will provide rules. EPA intends to follow the In the case of ground water systems, guidance and technical assistance to deadline strategy outlined in Option 4 the statutory deadline for promulgation States and systems to facilitate timely for these future rules. However, as of the GWR is May 2002. However, EPA compliance with both DBP and today’s action only relates to the Stage intends to promulgate this rule by microbial requirements. EPA will 1 DBPR, the Agency will defer final November 2000, in order to allow three request comment on how best to do this action on deadlines associated with years for compliance and still ensure when the Agency proposes the future rules until those rules, simultaneous compliance by ground LTESWTR and GWR. themselves, are finalized. Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69429

J. Public Notice Requirements 2 violations include monitoring change from that included in the 1994 violations, failure to comply with an Proposed Stage 1 DBPR (EPA, 1994a). 1. Today’s Rule analytical requirement specified by an K. System Reporting and Record Today’s action addresses public NPDWR, and operating under a variance Keeping Requirements notification by promulgating public or exemption. notification language for the regulated Today’s final rule contains specific 1. Today’s Rule compounds in 40 CFR Section 141.32 health effects language for the The Stage 1 DBPR, consistent with the (e). EPA takes this opportunity to note contaminants which are in today’s current system reporting regulations that the 1996 amendments to the SDWA rulemaking. EPA believes that the require the Agency to make certain mandatory health effects language is the under 40 CFR 141.31, requires PWSs to changes to the public notice regulations. most appropriate way to inform the report monitoring data to States within EPA intends to propose changes to the affected public of the potential health ten days after the end of the compliance public notice requirements in the implications of violating a particular period. In addition, systems are required Federal Register shortly after EPA standard. to submit the data required in § 141.134. promulgation of the Stage 1 DBPR. These data are required to be submitted Applicable changes in the public notice 3. Summary of Comments quarterly for any monitoring conducted requirements, when they become EPA received comments on the topic quarterly or more frequently, and within effective, will supersede today’s of the public notification language for 10 days of the end of the monitoring provisions. In general, the public TTHM, HAA5, chlorine, chloramines, period for less frequent monitoring. notification for the Stage 1 DBPR is not chlorine dioxide, and enhanced Systems that are required to do extra substantially changed from that coagulation. Some commenters noted monitoring because of the disinfectant included in the 1994 Proposed Stage 1 that the language in 141.32(e)(79) is used have additional reporting DBPR (EPA, 1994a). satisfactory. One commenter requested requirements specified. This applies to that the language for DBPs be modified systems that use chlorine dioxide (must 2. Background and Analysis to recognize that disinfectants react with report chlorine dioxide and chlorite Under Section 1414(c)(1) of the Act, naturally occurring organic and results) and ozone (must report bromate each owner or operator of a public water inorganic matter to form DBPs. Some results). system must give notice to the persons commenters did not support the use of Subpart H systems that use served by the system of (1) any violation the same public notification language conventional treatment are required to of any MCL, treatment technique for both DBP MCL and enhanced report either compliance/ requirement, or testing provision coagulation treatment technique noncompliance with DBP precursor prescribed by an NPDWR; (2) failure to violations. Several commenters (TOC) removal requirements or report comply with any monitoring suggested that the content of the notices which of the enhanced coagulation/ requirement under section 1445(a) of for chlorine, chloramine, and chlorine enhanced softening exemptions they are the Act; (3) existence of a variance or dioxide should reflect that disinfection meeting. There are additional exemption; (4) failure to comply with is an essential step in surface water requirements for systems that cannot the requirements of a schedule treatment. One commenter suggested meet the required TOC removals and prescribed pursuant to a variance or that the language for chlorine dioxide must apply for an alternate enhanced exemption; and (5) notice of the acute effects should be deleted. Other coagulant level. These requirements are concentration level of any unregulated commenters felt that the notice to included in § 141.134(b). contaminant for which the consumers of chlorine dioxide Calculation of compliance with the Administrator has required public violations at the treatment facility TOC removal requirements is based on notice. which do not result in violations in the normalizing the percent removals over EPA promulgated the current distribution system (nonacute the most recent four quarters, since regulations for public notification on violations) should not require public compliance is based on that period. October 28, 1987 (52 FR 41534—EPA, notification. Normalization, which would prescribe 1987). These regulations specify general In response, EPA has modified the equal weight to the data collected each notification requirements, including public notification language for DBPs to month, is necessary since source water frequency, manner, and content of indicate that disinfectants react with quality changes may change the percent notices, and require the inclusion of naturally occurring organic and TOC removal requirements from one EPA-specified health effects information inorganic matter to form DBPs. EPA month to another. EPA has developed a in each public notice. The public believes it is appropriate to use the same sample reporting and compliance notification requirements divide public notification language for the calculation sheet that will be available violations into two categories (Tier 1 enhanced coagulation treatment in the enhanced coagulation guidance and Tier 2) based on the seriousness of technique violation as for violations for manual to assist utilities in making the violations, with each tier having the TTHM and HAA5 MCLs, since these calculations. different public notification enhanced coagulation is meant to limit requirements. Tier 1 violations include exposure to DBPs. EPA believes the 2. Summary of Comments violations of an MCL, treatment current language in the public There were no significant comments technique, or a variance or exemption notification language is appropriate to on the system reporting and schedule. Tier 1 violations contain reflect that disinfection is an essential recordkeeping requirements and health effects language specified by EPA step in water treatment. EPA believes therefore EPA is finalizing the which concisely and in non-technical that since the potential health effects requirements as proposed. terms conveys to the public the adverse from chlorine dioxide are short-term health effects that may occur as a result that it is appropriate to maintain the L. State Recordkeeping, Primacy, and of the violation. States and water acute effects language to protect the Reporting Requirements utilities remain free to add additional fetus, infants, and children. In general, The SDWA provides that States and information to each notice, as deemed the public notification requirements for eligible Indian Tribes may assume appropriate for specific situations. Tier the Stage 1 DBPR will not substantially primary enforcement responsibilities. 69430 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

Fifty-four out of fifty-six State and interim primacy for future NPDWRs additional recordkeeping requirements. territorial jurisdictions have applied for must obtain primacy for this rule. Other commenters stated that they and received primary enforcement believed the requirements were 1. State Recordkeeping Requirements responsibility (primacy) under the Act. necessary. EPA understands No Tribes have received primacy. To a. Today’s Rule. The current commenters concerns with requiring obtain primacy for the federal drinking regulations in § 142.14 require States recordkeeping requirements that are water regulations, States must adopt with primacy to keep various records, unnecessary, but believes this their own regulations which are at least including analytical results to determine information is important to conduct as stringent as the federal regulations. compliance with MCLs, MRDLs, and effective State program oversight, This section describes the regulations treatment technique requirements; including the review of State decisions and other procedures and policies that system inventories; State approvals; and their basis. After further review, States must adopt to implement the enforcement actions; and the issuance of EPA has decided to eliminate the final Stage 1 DBPR. variances and exemptions. The Stage 1 requirement in the proposal that States To implement the final rule, States are DBPR requires States to keep additional must keep records of systems that apply required to adopt the following records of the following, including all for alternative TOC performance regulatory requirements: supporting information and an criteria. EPA is more concerned with the —Section 141.32, Public Notification; explanation of the technical basis for systems that are required to meet —Section 141.64, MCLs for Disinfection each decision: alternative TOC performance criteria, Byproducts; (1) Records of determinations made not the systems that have applied for the —Section 141.65, MRDLs for by the State when the State has allowed alternative performance criteria. In Disinfectants; systems additional time to install GAC addition, EPA has added three —Subpart L, Disinfectant Residuals, or membrane filtration. These records recordkeeping requirements, two of Disinfectant Byproducts, and must include the date by which the which were originally in the reporting Disinfection Byproduct Precursors. system is required to have completed requirements section and one for which installation; EPA requested comment. In addition to adopting regulations no (2) Records of systems that are The first additional requirement will less stringent than the federal required to meet alternative minimum require States to keep lists of all systems regulations, States must adopt certain TOC removal requirements or for whom required to monitor for various requirements related to this regulation the State has determined that the source disinfectants and DBPs (#8 above). The in order to have their program revision water is not amendable to enhanced second additional requirement will applications approved by EPA. This rule coagulation. These records must include require States to maintain a list of also requires States to keep specific the results of testing to determine laboratories that have completed records and submit specific reports to alternative limits and the rationale for performance sample analyses and EPA. establishing the alternative limits; On April 28, 1998, EPA amended its (3) Records of subpart H systems achieved the quantitative results for State primacy regulations at 40 CFR TOC, TTHMs, HAA5, bromate, and using conventional treatment meeting # 142.12 to incorporate the new process any of the enhanced coagulation or chlorite ( 6 above). EPA believes both of identified in the 1996 SDWA enhanced softening exemption criteria; these recordkeeping requirements are amendments for granting primary (4) Register of qualified operators; necessary to ensure adequate EPA enforcement authority to States while (5) Records of systems with multiple program oversight. As discussed below, their applications to modify their wells considered to be one treatment these two requirements are no longer in primacy programs are under review (63 plant for purposes of determining the State reporting requirements as EPA FR 23362; EPA, 1998i). The new process monitoring frequency; has decided that the requirements in the grants interim primary enforcement (6) Records of the sampling plans for proposal on State reporting authority for a new or revised regulation subpart H systems serving more than requirements are not needed on a during the period in which EPA is 3,300 persons must be keep on file at regular basis, but are needed for making a determination with regard to the State after submission by the system; program oversight. The third additional primacy for that new or revised (7) A list of laboratories that have requirement pertains to the request for regulation. This interim enforcement completed performance sample analyses comment in the proposal on authority begins on the date of the and achieved the quantitative results for maintaining the monitoring plans primacy application submission or the TOC, TTHMs, HAA5, bromate, and submitted by systems (#6 above). effective date of the new or revised State chlorite; and Several commenters supported this regulation, whichever is later, and ends (8) A list of all systems required to additional requirement stating that it when EPA makes a final determination. monitor for disinfectants and DBPs was a necessary element for However, this interim primacy authority under subpart L. implementing the final rule. Others is only available to a State that has b. Background and Analysis. In believed it was not necessary to keep primacy for every existing national addition to requesting comments on the this on file because the public could primary drinking water regulation in requirements (1) through (5), and (7) request this information from the system effect when the new regulation is and (8) listed above, EPA also requested or the State as normal public records. promulgated. comments on whether States should be EPA believes that it is important for As a result, States that have primacy required to keep the monitoring plan States to review, and keep on file the for every existing NPDWR already in submitted by systems serving more than systems monitoring plan to ensure that effect may obtain interim primacy for 3,300 people on file at the State after the PWS is monitoring and calculating this rule, beginning on the date that the submission to make it available for compliance in accordance with the State submits its complete and final public review. plan. This will also enable the public to primacy application for this rule to EPA, c. Summary of Comments. There were view the plan. Thus, EPA is adding this or the effective date of its revised several commenters who suggested that requirement to the final recordkeeping regulations, whichever is later. In EPA should keep in mind State budget requirements. In conclusion, based on a addition, a State which wishes to obtain constraints when requiring specific review of all public comments the final Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69431 rule contains eight State recordkeeping that State programs contain all the installing GAC or membrane technology requirements in addition to those essential elements for an effective and the basis for the additional time; required under current regulations in program. Specifically, EPA believes the (3) A list of laboratories that have § 142.14. special requirements are important to completed performance sample analyses ensure that the process or approach and achieved the quantitative results for 2. Special Primacy Requirements used by the State for evaluating whether TOC, TTHMs, HAA5, bromate, and a. Today’s Rule. To ensure that a State the interim treatment in place for chlorite; program includes all the elements systems granted additional time to (4) A list of all systems using multiple necessary for an effective and install GAC or membranes or alternative ground water wells which draw from enforceable program under today’s rule, enhanced coagulation levels will be the same aquifer and are considered a a State application for program revision protective of public health. The single source for monitoring purposes; approval must include a description of requirement to have qualified operators (5) A list of all Subpart H systems how the State will: is important because the treatment using conventional treatment which are (1) Determine the interim treatment technologies used to comply with the not required to operate with enhanced requirements for systems granted Stage 1 DBPR and the IESWTR coagulation, and the reason why additional time to install GAC and simultaneously are complex and will enhanced coagulation is not required for membrane filtration under 141.64(b)(2). require a certain level of expertise. The each system; and (2) Qualify operators of community requirement to approve parties for (6) A list of all systems with State- and nontransient noncommunity water conducting analyses of specific water approved alternate performance systems subject to this regulation under quality parameters is important because standards (alternate enhanced 141.130(c). Qualification requirements each of the parameters required to be coagulation levels). established for operators of systems tested is critical to a specific component c. Summary of Comments. Several subject to 40 CFR Part 141 Subpart H of the final rule (e.g., bromide ion is commenters stated that the reporting (Filtration and Disinfection) may be important because for bromate it is requirements were not necessary to used in whole or in part to establish possible to reduce monitoring from operate an oversight program and that operator qualification requirements for monthly to once per quarter, if a system these reports could be made available meeting subpart L requirements if the demonstrates that the average raw water for EPA review during annual audits. State determines that the subpart H bromide concentration is less than 0.05 EPA agrees with commenters that the requirements are appropriate and mg/L based upon representative reports are not necessary to operate an applicable for meeting subpart L monthly measurements for one year). oversight program, and that if needed requirements. Finally, it is important to define the EPA could request this information from (3) Approve DPD colorimetric tests criteria used to determine if multiple the States. However, EPA does believe kits for free and total chlorine wells are to be considered a single it is important that States maintain this measurements under 141.131(c)(2). source as this could have significant information in their records. In State approval granted under subpart H implications for monitoring. conclusion, based on commenters (§ 141.74(a)(2)) for the use of DPD c. Summary of Comments. There were concerns and for the reasons cited colorimetric test kits for free chlorine no significant comments on the primacy above, the final rule contains no testing would be considered acceptable requirements. The only change from the additional State reporting requirements approval for the use of DPD test kits in proposal was to delete the requirement other than those required by 142.15. measuring free chlorine residuals as that States must have approved parties required in subpart L. M. Variances and Exemptions (4) Approve parties to conduct to perform temperature evaluations. 1. Today’s Rule analyses of water quality parameters This requirement was included in the under 141.132(a)(2) (pH, alkalinity, proposed rule because of the need to Variances may be granted in bromide, and residual disinfectant have accurate measurements as a part of accordance with section 1415(a)(1)(A) of concentration measurements). The the process for not allowing the SDWA and in accordance with State’s process for approving parties predisinfection credit. Since the final 1415(e) and EPA’s regulations. performing water quality measurements rule allows credit for compliance with Exemptions may be granted in for systems subject to subpart H applicable disinfection requirements accordance with section 1416(a) of the requirements may be used for approving consistent with the SWTR, the SDWA and EPA’s regulations. temperature requirement was removed. parties measuring water quality 2. Background and Analysis parameters for systems subject to 3. State Reporting Requirements subpart L requirements, if the State Variances. The SDWA provides for determines the process is appropriate a. Today’s Rule. EPA currently two types of variances—general and applicable. requires in § 142.15 that States report to variances and small system variances. (5) Define criteria to use in EPA information such as violations, Under section 1415(a)(1)(A) of the determining if multiple wells are being variance and exemption status, and SDWA, a State which has primary drawn from a single aquifer and enforcement actions. The Stage 1 DBPR enforcement responsibility (primacy), or therefore can be considered as a single does not add any additional reporting EPA as the primacy agency, may grant source under 141.132(a)(2). Such requirements. variances from MCLs to those public criteria will be used in determining the b. Background and Analysis. The water systems of any size that cannot monitoring frequency for systems using preamble to the proposed rule included comply with the MCLs because of only ground water not under the direct six State reporting requirements. These characteristics of the water sources. The influence of surface water. included: primacy agency may grant general (6) Approve alternative TOC removal (1) A list of all systems required to variances to a system on condition that levels as allowed under 141.135(b). monitor for various disinfectants and the system install the best available b. Background and Analysis. As disinfection byproducts; technology, treatment techniques, or discussed above, EPA included several (2) A list of all systems for which the other means, and provided that special primacy requirements to ensure State has granted additional time for alternative sources of water are not 69432 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations reasonably available to the system. At these technologies often were not the Act or improve the quality of the time this type of variance is granted, applicable to small water systems. drinking water. the State must prescribe a compliance While EPA will continue to use If EPA or the State grants an schedule and may require the system to feasibility for large systems in setting exemption to a public water system, it implement additional control measures. NPDWRs, the 1996 amendments to the must at the same time prescribe a Furthermore, before EPA or the State SDWA specifically require EPA to make schedule for compliance (including may grant a general variance, it must small system technology assessments for increments of progress or measures to find that the variance will not result in both existing and future regulations. develop an alternative source of water an unreasonable risk to health (URTH) The 1996 amendments to the SDWA supply) and implementation of to the public served by the public water identifies three categories of small appropriate control measures that the system. public water systems that need to be State requires the system to meet while Under section 1413(a)(4), States that addressed: (1) those serving a the exemption is in effect. Under section choose to issue general variances must population between 3301 to 10,000; (2) 1416(b)(2)(A), the schedule prescribed do so under conditions, and in a those serving a population of 501— shall require compliance as manner, that are no less stringent than 3300; and (3) those serving a population expeditiously as practicable (to be section 1415. Of course, a State may of 26—500. The SDWA requires EPA to determined by the State), but no later adopt standards that are more stringent make determinations of available than 3 years after the effective date for than the EPA standards. EPA specifies compliance technologies and, if needed, the regulations established pursuant to BATs for general variance purposes. variance technologies for each size section 1412(b)(10). For public water EPA may identify as BAT different category. A compliance technology is a systems which do not serve more than treatments under section 1415 for technology that is affordable and that a population of 3,300 and which need variances other than the BAT under achieves compliance with the MCL and/ financial assistance for the necessary section 1412 for MCLs. EPA’s section or treatment technique. Compliance improvements, EPA or the State may 1415 BAT findings may vary depending technologies can include point-of-entry renew an exemption for one or more on a number of factors, including the or point-of-use treatment units. Variance additional two-year periods, but not to number of persons served by the public technologies are only specified for those exceed a total of 6 years, if the system water system, physical conditions system size/source water quality establishes that it is taking all related to engineering feasibility, and combinations for which there are no practicable steps to meet the the costs of compliance with MCLs. In listed compliance technologies. requirements above. this final rule, EPA is not specifying EPA has completed an analysis of the A public water system shall not be different BAT for variances under affordability of DBP control granted an exemption unless it can section 1415(a). Section 1415(e) technologies for each of the three size establish that either: (1) the system authorizes the primacy Agency (EPA or categories included above. Based on this cannot meet the standard without the State) to issue variances to small analysis, multiple affordable capital improvements that cannot be public water systems (those serving less compliance technologies were found for completed prior to the date established than 10,000 persons) where the system each of the three system sizes (EPA, pursuant to section 1412(b)(10); (2) in cannot afford to comply with an MCL 1998q and EPA, 1998r) and therefore the case of a system that needs financial and where the primacy agency variance technologies were not assistance for the necessary determines that the terms of the identified for any of the three size implementation, the system has entered variances ensure adequate protection of categories. The analysis was consistent into an agreement to obtain financial public health (63 FR 1943–57; EPA, with the methodology used in the assistance pursuant to section 1452 or 1998j). These variances also may only document ‘‘National-Level Affordability any other Federal or state program; or be granted where EPA has identified a Criteria Under the 1996 Amendments to (3) the system has entered into an variance technology under Section the Safe Drinking Water Act’’ (EPA, enforceable agreement to become part of 1412(b)(15) for the contaminant, system 1998s) and the ‘‘Variance Technology a regional public water system. size and source water quality in Findings for Contaminants Regulated 3. Summary of Comments on Variance question. Before 1996’’ (EPA, 1998t). Prior to the 1996 SDWA amendments, Exemptions. Under section 1416(a), and Exemptions EPA was required to set the MCL for a EPA or a State may exempt a public In the 1994 proposal, EPA requested contaminant as close to the MCLG as is water system from any requirements comment on whether exemptions to the feasible. Section 1412(b)(4)(D) of the related to an MCL or treatment rule should be granted if a system could SDWA states that ‘‘the term ‘‘feasible’’ technique of an NPDWR, if it finds that demonstrate to the State that due to means with the use of the best (1) due to compelling factors (which unique water quality characteristics it technology, treatment techniques and may include economic factors such as could not avoid, through the use of other means which the Administrator qualification of the PWS as serving a BAT, the possibility of increasing total finds, after examination for efficacy disadvantaged community), the PWS is health risk to its consumers by under field conditions and not solely unable to comply with the requirement complying with the Stage 1 regulations. under laboratory conditions, are or implement measure to develop an The Agency requested information available (taking cost into alternative source of water supply; (2) under which such a scenario may consideration).’’ the exemption will not result in an unfold. Several commenters supported The cost assessment for the feasibility unreasonable risk to health; and; (3) the granting exemptions provided a system determinations have historically been PWS was in operation on the effective could demonstrate that installation of based upon impacts to regional and date of the NPWDR, or for a system that BAT will increase the total health risk. large metropolitan water systems was not in operation by that date, only After additional consideration, EPA serving populations greater than 50,000 if no reasonable alternative source of believes it is not appropriate, for several people. Since large systems served as drinking water is available to the new reasons, to grant exemptions based on a the basis for the feasibility system; and (4) management or demonstration that the use of BAT determinations, the technical and/or restructuring changes (or both) cannot could increase the total health risk by cost considerations associated with reasonably result in compliance with complying with the Stage 1 DBPR. First, Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69433

EPA does not believe the analytical conducted by approved parties. [40 CFR bound) without artificially imposing tools and methodologies are currently 141.89 and 141.74]. EPA’s ‘‘Manual for unnecessarily strict criteria (the lower available that would allow a the Certification of Laboratories bound). Therefore, EPA is establishing determination of whether the total Analyzing Drinking Water’’, EPA 815– the following acceptance limits for health risk from the installation of BAT B–97–001—(EPA, 1997g), specifies the measurement of bromate, chlorite, each would increase. Second, at the time of criteria for implementation of the haloacetic acid, and each proposal there was concern that in drinking water laboratory certification trihalomethane in a PE sample. waters with high bromide program. EPA is defining acceptable concentrations it may be possible to In today’s rule, EPA is promulgating performance for each chemical increase the concentrations of certain MCLs for TTHMs, HAA5, bromate, and measured in a PE sample from estimates brominated DBPs when using precursor chlorite. Today’s rule requires that only derived at a 95% confidence interval removal processes even though the certified laboratories be allowed to from the data generated by a statistically concentrations of the TTHMs and HAA5 analyze samples for compliance with significant number of laboratories may decrease. Also, at the time of the proposed MCLs. For the participating in the PE study. However, proposal, the health risks associated disinfectants and certain other EPA requires that these acceptance with many of the brominated DBPs was parameters in today’s rule, which have criteria not exceed ±50% nor be less unknown, and it was unclear whether MRDLs or monitoring requirements, than ±15% of the study mean. If the benefits of lowering the EPA is requiring that analyses be insufficient PE study data are available concentrations of chlorinated DBPs conducted by a party acceptable to the to derive the estimates required for any outweigh the possible downside risks of State. of these compounds, the acceptance increasing certain brominated DBPs. Performance evaluation (PE) samples, limit for that compound will be set at Since the proposal, some additional which are an important tool in the ±50% of the study true value. The true health effects research has been SDWA laboratory certification program value is the concentration of the completed evaluating the toxicity of (laboratories seeking certification) may chemical that EPA has determined was brominated DBPs. However, this be obtained from a PE provider in the PE sample. approved by the National Institute of research is still preliminary and no EPA recognizes that when using Science and Technology (NIST). To conclusions can be drawn on the multianalyte methods, the data receive and maintain certification, a potential for increased risks from the generated by laboratories that are laboratory must use a promulgated brominated DBPs. In addition, it is performing well will occasionally method and, at least once per year, unclear to what extent the use of exceed the acceptance limits. Therefore, successfully analyze an appropriate PE precursor removal processes will change to be certified to perform compliance sample. In the drinking water PE the concentrations of certain brominated monitoring using a multianalyte DBPs. The ICR data should provide studies, NIST-approved providers will provide samples for bromate, chlorite, method, laboratories are required to some additional information that may generate acceptable data for at least 80% be helpful in this area along with five haloacetic acids, four trihalomethanes, free chlorine, and of the regulated chemicals in the PE additional ongoing research. This sample that are analyzed with the information will be available for alkalinity. The NIST-approved PE providers will provide total chlorine method. If fewer than five compounds consideration in the Stage 2 rule are included in the PE sample, data for deliberations. Based on the reasons and TOC samples in the wastewater PE studies and have the potential to each of the analytes in that sample must stated above, EPA does not believe it is provide these samples for drinking meet the minimum acceptance criteria appropriate to allow exemptions to the water studies. Due to the lability of in order for the laboratory to be rule based on a finding that the chlorine dioxide, EPA does not expect certified. installation of BAT would increase the a suitable PE sample can be designed for Approval Criteria for Disinfectants total risk from DBPs. chlorine dioxide measurements. and Other Parameters. Today’s rule N. Laboratory Certification and PE Sample Acceptance Limits for establishes MRDLs for the three Approval Laboratory Certification. Historically, disinfectants—chlorine, chloramines, EPA has set minimum PE acceptance and chlorine dioxide. In addition, EPA 1. Today’s Rule limits based on one of two criteria: has established monitoring EPA recognizes that the effectiveness statistically derived estimates or fixed requirements for TOC, alkalinity, and of today’s regulations depends on the acceptance limits. Statistical estimates bromide; there are no MCLs for these ability of laboratories to reliably analyze are based on laboratory performance in parameters. In previous rules [40 CFR the regulated disinfectants and DBPs at the PE study. Fixed acceptance limits 141.28, .74, and .89], EPA has required the MRDL or MCL, respectively. are ranges around the true concentration that measurements of alkalinity, Laboratories must also be able to of the analyte in the PE sample. Today’s disinfectant residuals, pH, temperature, measure the trihalomethanes and rule combines the advantages of these and turbidity be made with an approved haloacetic acids at the reduced approaches by specifying statistically- method and conducted by a party monitoring trigger levels, which are derived acceptance limits around the approved (not certified) by the State. In between 25 and 50 percent of the MCLs study mean, within specified minimum today’s rule, EPA requires that samples for these compound classes. EPA has and maximum fixed criteria. collected for compliance with today’s established State primacy requirements EPA believes that specifying requirements for alkalinity, bromide, for a drinking water laboratory statistically-derived PE acceptance residual disinfectant, and TOC be certification program for the analysis of limits with upper and lower bounds on conducted with approved methods and DBPs. States must adopt a laboratory acceptable performance provides the by a party approved by the State. certification program as part of primacy. flexibility necessary to reflect Other Laboratory Performance [40 CFR 142.10(b)]. EPA has also improvement in laboratory performance Criteria. For all contaminants and specified laboratory requirements for and analytical technologies. The parameters required to be monitored in analyses of DBP precursors and acceptance criteria maintain minimum today’s rule, the States may impose disinfectant residuals which must be data quality standards (the upper other requirements for a laboratory to be 69434 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations certified or a party to be approved to percent cost of capital ($230 million). reductions in the cancer risks were conduct compliance analyses. The basis for these estimates, and proportional to reductions in TTHM, alternate cost estimates using different HAA5, or TOC levels (predicted from 2. Background and Analysis cost of capital assumptions are compliance forecasts). Negotiators The laboratory certification and described later in this section. While the agreed that the range of possible risks approval requirements that today’s rule benefits of this rule are difficult to attributed to chlorinated water should establishes are unchanged from those quantify because of the uncertainty consider both toxicological data and proposed by EPA in 1994. associated with risks from exposure to epidemiological data, including the 3. Summary of Comments DBPs (and the resultant reductions in Morris et al. (1992) estimates. No risk due to the decreased exposure from consensus, however, could be reached EPA received few comments on DBPs), EPA believes that there is a on a single likely risk estimate. laboratory certification and approval. reasonable likelihood that the benefits Therefore, the predicted benefits for the Commenters requested clarification of will exceed the costs. Various proposal ranged from one to several the use of the ±50% upper bound and ± approaches for assessing the benefits are thousands cases of cancer being avoided 15% lower bound, along with the use considered and described in the benefits per year after implementation of the of statistically derived limits. EPA and net benefits sections of this Stage 1 DBPR. Despite, the uncertainty believes that statistically derived limits preamble. in quantifying the benefits from the provide flexibility to allow laboratory Stage 1 DBPR, the Reg. Neg. Committee B. Background certification standards to reflect recognized that risks from chlorinated improvement in laboratory performance 1. Overview of RIA for the Proposed water could be large, and therefore and analytical technologies. As Rule should be reduced. The Reg. Neg. laboratories become more proficient in Committee also recommended that the conducting these analyses, statistically In the RIA for the 1994 proposed Stage 1 DBPR (EPA, 1994i) EPA proposed Stage 1 DBPR provided the derived acceptance limits may drop. best means for reducing risks from DBPs However, to prevent the exclusion of estimated the national capital and annualized utility costs (sum of until better information become laboratories capable of producing data available. of sufficient quality for compliance amortized capital and annual operating For a more detailed discussion of the purposes, EPA has established a lower costs, assuming 10% cost of capital) for cost and benefit analysis of the 1994 bound for acceptance limits of ±15%. all systems at $4.4 billion and $1.04 proposed DBPR refer to the preamble of EPA is imposing an upper bound on billion, respectively. The cost and the proposed rule (EPA, 1994a) and the acceptable performance to establish reduction in DBP exposure estimates of RIA for the proposed rule (EPA, 1994i). minimum data quality standards. the 1994 RIA were derived using a Results outside of this range have Disinfection Byproduct Regulatory 2. Factors Affecting Changes to the 1994 unacceptable accuracy for compliance Analysis Model (DBPRAM). The RIA determinations. These upper and lower DBPRAM consisted of a collection of a. Changes in Rule Criteria. Based on bounds were not determined analytical models which used Monte the new data reflecting the feasibility of statistically; they are the data quality Carlo simulation techniques to produce enhanced coagulation, as discussed objectives the Agency has determined as national forecasts of compliance and previously, the enhanced coagulation acceptable. exposure reductions for different regulatory scenarios. The TWG, requirements were modified by IV. Economic Analysis representing members of the Reg. Neg. decreasing the percent TOC removal Under Executive Order 12866, Committee, used the best available requirements by 5 percent for systems Regulatory Planning and Review, EPA information at the time as inputs to the with low TOC level waters (i.e., 2–4 mg/ must estimate the costs and benefits of DBPRAM, and for making further L TOC). These new percent TOC the Stage 1 DBPR in a Regulatory Impact adjustments to the model predictions. removal requirements were used with Analysis (RIA) and submit the analysis The Stage 1 DBPR compliance and new source and finished water TOC to Office of Management and Budget exposure forecasts were affected by occurrence data to revise the estimates (OMB) in conjunction with publishing constraints imposed by the 1994 for the number of systems requiring the final rule. EPA has prepared an RIA proposed IESWTR option which would enhanced coagulation. to comply with the requirements of this have required systems to provide The IESWTR was revised from the Order. This section provides a summary enough disinfection, while not allowing proposal to allow inactivation credit for of the information from the RIA for the for disinfection credit prior to TOC disinfection prior to and during stages Stage 1 DBPR (USEPA 1998g). removal by enhanced coagulation, to of treatment for precursor removal. achieve a 10¥4 annual risk of infection Also, the proposed IESWTR was revised A. Today’s Rule from Giardia (EPA, 1994a). The to include disinfection benchmark EPA has estimated that the total compliance forecast assumed that a criteria, in lieu of requiring treatment to annualized cost, for implementing the substantial number of systems would an acceptable risk level, to prevent Stage 1 DBPR is $701 million in 1998 need to install advanced technologies to increases in microbial risk while dollars (assuming a 7 percent cost of meet the Stage 1 DBPR because of systems complied with the Stage 1 capital). This estimate includes needing to achieve the 10¥4 annual risk DBPR. These two rule changes were annualized treatment costs to utilities level from Giardia while no longer being considered in revising the forecasts of ($593 million), start-up and annualized allowed disinfection credit prior to TOC compliance and changes in exposure monitoring costs to utilities ($91.7 removal. resulting from the Stage 1 DBPR. million), and startup and annualized Predicted benefits for the proposed b. New Information Affecting DBP monitoring costs to states ($17.3 Stage 1 DBPR were derived assuming a Occurrence and Compliance Forecasts. million). Annualized treatment costs to baseline risk ranging from 1 to 10,000 Since the rule was proposed, new utilities includes annual operation and cancer cases per year (based on analysis sources of data have become available maintenance costs ($362 million) and of available toxicological and that were used to update the 1994 RIA. annualized capital costs assuming 7 epidemiological data) and assuming The new data includes: Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69435

• Updated costs for different an reassessment of the Morris et al. C. Cost Analysis treatment technologies (e.g., (1992) meta-analysis (Poole, 1997). membranes) used in the DBP Cost and Review of the meta-analysis indicated National cost estimates of compliance Technology Document, (EPA, 1998k); that the estimate of cancer cases had with the Stage 1 DBPR were derived • 1996 data from the AWWA Water limited utility for risk assessment from estimates of utility treatment costs, Industry Data Base on TOC, TTHM and purposes for methodological reasons monitoring and reporting costs, and HAA5 occurrence, and disinfection (EPA, 1998l and EPA, 1998m). EPA has start-up costs. Utility treatment costs practices; decided not to use the Morris et al. were derived using compliance forecasts • Plant schematics of treatment (1992) meta-analysis to estimate the of technologies to be used and unit costs processes for ICR utilities; for the different technologies. • potential benefits from the Stage 1 Research data from numerous DBPR. EPA has considered new sources regarding the efficacy of 1. Revised Compliance Forecast epidemiology studies conducted since enhanced coagulation for precursor the time of proposal and completed an The TWG, supporting the M–DBP removal and resultant DBP formation assessment of the potential number of Advisory Committee, used the 1996 (Krasner, 1997; and EPA, 1997b); AWWA Water Industry Data Base • New research results produced in bladder cancer cases that could be attributed to exposure from chlorinated (WIDB) to reevaluate the compliance jar tests by TWG members documenting decision tree used in the RIA for the the effect of moving the point of surface waters. Based on this assessment 1994 proposal. The WIDB provided predisinfection under varying of epidemiological studies, EPA occurrence data on TOC level in raw conditions (Krasner, 1997 and EPA, estimates that between 1100–9300 water and finished water, TTHM and 1997b). bladder cancer cases per year could be HAA5 levels within distribution This new information has been attributed to exposure to chlorinated systems, and information on described in the 1997 DBP NODA (EPA, surface waters (EPA, 1998c). Due to the 1997b). Public comments received in wide uncertainty in these estimates, the predisinfection practices. 1997, supported using the above true number of attributable cases could The above information was used to information in revising the decision tree also be zero. The basis for these bladder predict treatment compliance choices analysis. Discussion on the decision tree cancer case estimates and potential that plants would likely make under the changes are in section IV.C of this reductions in risk resulting from the Stage 1 DBPR. Table IV–1 illustrates preamble. Stage 1 DBPR is discussed further in the how the compliance forecast changed c. New Epidemiology Information. benefits and net benefits sections that for large systems using surface water Since the proposal, EPA has completed follow. since the time of proposal.

TABLE IV±1.ÐCOMPARISONS OF COMPLIANCE FORECASTS FOR SURFACE WATER SYSTEMS SERVING ≥10,000 POPULATION FROM THE 1994 PROPOSAL AND FINAL RULE

1994 1998 Treatment # systems % systems # systems % systems

(A) No Further Treatment ...... 386 27.7 544 39.0 (B) Chlorine/Chloramines ...... 41 2.9 231 16.6 (C) Enhanced Coagulation + Chloramines ...... 136 9.7 265 19.0 (D) Enhanced Coagulation + Chlorine ...... 600 43.0 265 19.0 (E) Ozone, Chlorine Dioxide, Granular Activated Carbon, Membranes ...... 232 16.6 90 6.5

Total * ...... 1,395 100 1,395 100 * May not add to total due to independent rounding.

Notable is that the percentage of systems DBPR. It also appears that a substantial membrane technologies) is attributed to predicted to use advanced technologies number of systems may have already the change in the IESWTR (as described (ozone, chlorine dioxide, GAC, or made treatment changes to comply with above) from the time of proposal. membrane) dropped from 17 percent to the 1994 proposed rule. However, unlike for large systems, the 6.5 percent since proposal, and the Table IV–2 illustrates how the overall percentage of systems predicted percentage of systems not affected by compliance forecast changed for small to require treatment modifications did the rule increased from 28 percent to 39 systems using surface water since the not change. A higher percentage of percent. This shift in predicted time of proposal. As for large systems, small systems (70 percent) are predicted compliance choices is mainly attributed the percentage of systems predicted to to be affected than large systems (61 to less stringent disinfection use advanced technologies dropped percent) because previously smaller requirements under the IESWTR which substantially, from 17 percent to 6.5 systems did not have to comply with a would reduce the formation of DBPs percent. This drop in use of advanced TTHM standard. and reduce the number of systems technology (i.e., ozone/chloramines and requiring treatment to meet the Stage 1 69436 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

TABLE IV±2.ÐCOMPARISON OF COMPLIANCE DECISION TREE FOR SURFACE WATER SYSTEMS SERVING <10,000 POPULATION FROM THE 1994 PROPOSAL AND FINAL RULE

1994 1998 # systems % systems # systems % systems

No Further Treatment ...... 1,549 30 1,549 30 Number of Affected Systems ...... 3,615 70 3,615 70 Treatment: Chlorine/Chloramine ...... 155 3.0 826 16.0 Enhanced Coagulation ...... 2,169 42.0 1,983 38.4 Enhanced Coagulation/Chloramine ...... 465 9.0 465 9.0 Ozone/Chloramine ...... 258 5.0 184 3.6 Enhanced Coagulation+Ozone, Chloramine ...... 258 5.0 0 0 Membranes ...... 310 6.0 157 3.0

Table IV–3 illustrates the compliance ground water systems are anticipated to is more prevalent in large versus small forecast for ground water systems. This need treatment changes (12 percent) ground water systems. forecast did not change from the time of than large ground water systems (15 proposal. A smaller percentage of small percent) because the use of disinfectants

TABLE IV±3.ÐCOMPLIANCE DECISION TREE FOR ALL GROUND WATER SYSTEMS

Systems <10,000 Systems ≥10,000 # systems % systems # systems % systems

No Further Treatment ...... 59,847 88 1,122 85 Percentage of Affected Systems ...... 8,324 12 198 15 Treatment: Chlorine/Chloramine ...... 5,403 8 119 9 Ozone/Chloramine ...... 0 0 26 2 Membranes ...... 2,921 4 53 4

2. System Level Unit Costs Tables IV–4 and IV–5 present the unit cost estimates in 1998 dollars that were utilized for each of the different treatment technologies in each system size category. Unit costs are presented in $ per 1000 gallons which includes operation and maintenance costs and amortized capital costs (using a 7% discount rate and a 20 year amortization period). One dollar per thousand gallons equates to approximately $100 per household per year as an average for communities in the U.S. More detailed information on these unit costs is available from the EPA’s Cost and Technology Document (EPA, 1998k).

TABLE IV±4.ÐSURFACE WATER SYSTEMS COSTS FOR DBP CONTROL TECHNOLOGIES ($/KGAL) AT 7% COST OF CAPITAL

Population size category 100K± 25±100 100±500 500±1K 1±3.3K 3.3±10K 10±25K 25±50K 50±75K 75±100K 500K 500K±1M >1M

Chlorine/Chloramine ...... 0.71 0.19 0.06 0.03 0.03 0.02 0.01 0.01 0.01 0.01 0.01 0.01 Enhanced Coagulation (EC) 0.15 0.13 0.12 0.11 0.09 0.08 0.07 0.07 0.07 0,07 0.06 0.06 EC/Chloramine ...... 0.87 0.32 0.18 0.14 0.12 0.09 0.08 0.08 0.08 0.07 0.07 0.07 Ozone/Chloramine ...... 12.67 3.21 1.05 0.52 0.38 0.23 0.13 0.10 0.08 0.06 0.04 0.04 EC+Ozone, Chloramine ...... 12.82 3.34 1.17 0.63 0.47 0.30 0.20 0.17 0.15 0.13 0.11 0.10 EC+GAC10 ...... 6.24 2.43 1.21 0.81 0.59 0.46 0.37 0.35 0.29 0.24 0.19 0.16 EC+GAC20 ...... 14.11 5.87 3.45 2.45 1.87 1.48 1.05 1.00 0.90 0.64 0.48 0.41 Chlorine Dioxide ...... 24.33 5.73 1.65 0.64 0.24 0.11 0.07 0.07 0.06 0.05 0.04 0.04 Membranes ...... 3.40 3.47 3.39 2.65 1.72 0.96 0.96 0.87 0.87 0.87 0.87 0.87

TABLE IV±5.ÐGROUND WATER SYSTEMS COSTS FOR DBP CONTROL TECHNOLOGIES ($/KGAL) AT 7% COST OF CAPITAL

Population size category 100K± 25±100 100±500 500±1K 1±3.3K 3.3±10K 10±25K 25±50K 50±75K 75±100K 500K 500K±1M >1M

Chlorine/Chloramine ...... 0.72 0.19 0.06 0.03 0.03 0.02 0.01 0.01 0.01 0.01 0.01 0.01 Ozone/Chloramine ...... 12.67 3.21 1.05 0.52 0.38 0.23 0.13 0.10 0.08 0.06 0.04 0.04 Membranes ...... 3.41 3.47 3.39 2.65 1.72 0.96 0.96 0.87 0.87 0.87 0.87 0.87

3. National Costs Table IV–6 provides a detailed summary of national costs in 1998 dollars under the Stage 1 DBPR for different cost of capital assumptions under a 20 year amortization period. A cost of capital rate of 7 percent was used to calculate the unit costs for the national compliance cost model. This rate represents the standard discount rate preferred by OMB for benefit-cost analyses of government programs and regulations. The 3 percent and 10 percent rates are provided as a sensitivity analysis to show different assumptions about the cost of capital that would affect estimated Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69437 costs. The 10 percent rate also provides a link to the 1994 Stage 1 DBPR cost analysis which was based on a 10 percent rate. EPA believes that the cost estimates presented in Table IV–6 are probably within +/¥30 percent. Uncertainty around the cost estimates pertain to compliance forecast estimates, unit cost estimates for the different technologies as they may pertain to individual sites, and estimated costs associated with monitoring.

TABLE IV±6.ÐSUMMARY OF COSTS UNDER THE STAGE 1 DBPR ($000)

Surface water systems Ground water systems Utilities Costs All systems Small Large Total Small Large Total

Summary of Costs at 3 Percent Cost of Capital

Treatment Costs Total Capital Costs ...... 242,652 554,564 797,216 997,537 528,539 1,526,076 2,323,292 Annual O&M ...... 23,068 201,308 224,376 83,910 54,243 137,153 362,530 Annualized Capital Costs ...... 16,326 37,161 53,487 67,287 35,618 102,905 156,392 Annual Utility Treatment Costs ...... 39,394 238,469 277,863 151,197 89,861 240,058 518,922 Monitoring and Reporting Cost: Start-Up Costs ...... 59 28 87 674 26 700 787 Annual Monitoring ...... 10,867 14,619 25,486 38,803 26,326 65,129 90,615 State Costs: Start-Up Costs ...... 2,919 Annual Monitoring ...... 13,243

Total Annual Costs at 3 Percent Cost of Capital ...... 626,486

Summary of Costs at 7 Percent Cost of Capital

Total Capital Costs ...... 242,652 554,564 797,216 997,537 528,539 1,526,076 2,323,292 Annual O&M ...... 23,068 201,308 224,376 83,910 54,243 137,153 362,530 Annualized Capital Costs ...... 22,786 62,355 85,141 94,403 50,046 144,499 229,590 Annual Utility Treatment Costs ...... 45,855 263,663 309,518 178,313 104,289 282,602 592,120 Monitoring and Reporting Cost: Start-Up Costs ...... 82 39 121 946 36 982 1,103 Annual Monitoring ...... 10,867 14,619 25,486 38,803 26,326 65,129 90,615

State Costs: Start-Up Costs ...... 4,099 Annual Monitoring ...... 13,243

Total Annual Costs at 7 Percent Cost of Capital ...... 701,180

Summary of Costs at 10 Percent Cost of Capital

Total Capital costs ...... 242,652 554,564 797,216 997,537 528,539 1,526,076 2,323,292 Annual O&M ...... 23,068 201,308 224,376 83,910 54,243 137,153 362,530 Annualized Capital Costs ...... 28,423 74,639 103,062 117,328 62,522 179,850 282,912 Annual Utility Treatment Costs ...... 51,491 275,947 327,438 201,238 116,765 317,003 645,442 Monitoring and Reporting Cost: Start-Up Costs ...... 102 48 150 1,177 45 1,222 1,372 Annual Monitoring ...... 10,867 14,619 25,486 38,803 26,326 65,129 90,615 State Costs: Start-Up Costs ...... 5,100 Annual Monitoring ...... 13,243

Total Annual Costs at 10 Percent Cost of Capital ...... 755,772

The total national costs of the final anticipated to need advanced order to provide protection against Stage 1 DBPR are less than estimated in technologies and lower membrane microbial contaminants. Because of the the RIA for the proposed rule in 1994. technology costs as described above. large number of people potentially The estimated capital costs of the 1994 D. Benefits Analysis exposed to DBPs, there is a substantial proposal in 1998 dollars is $4.97 billion concern for any health risks which may and the total annual cost (assuming a 10 1. Exposure Assessment be associated with exposure to DBPs. percent cost of capital as was assumed A large portion of the U.S. population Several factors are necessary to assess in 1994) is $1.3 billion. The drop in is exposed to DBPs via drinking water. the exposure to DBPs: the size of the national costs from the 1994 proposal is Over 200 million people in the U.S. are population potentially at risk; the mainly attributed to the lowering of the served by PWSs which apply a method and rate of ingestion; and the number of surface water systems disinfectant (e.g., chlorine) to water in concentration of DBPs in drinking 69438 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations water. Because DBPs are formed in System data (SDWIS) is estimated in estimated to be served by water systems drinking water by the reaction of Table IV–7. Based on recent information that disinfect and are potentially disinfectants with natural organic and from SDWIS, it was assumed that all exposed to DBPs. This widespread inorganic matter, the population at risk surface water systems disinfect and a exposure represents over 88 percent of is identified as the population served by portion of ground water systems the total U.S. population (270 million). drinking water systems that disinfect. disinfect (95 percent by population The route of exposure is through The population served by each of four among large systems and 83 percent by drinking disinfected tap water. system categories, taken from recent population among small systems). Safe Drinking Water Act Information Approximately 239 million persons are

TABLE IV±7.ÐPOPULATION POTENTIALLY EXPOSED TO DBPS

% of popu- Population Population lation receiv- served by sys- served ing disinfected tems that dis- water infect

Large Surface Water: >10,000 persons ...... 141,297,000 100 141,297,000 Small Surface Water: <10,000 persons ...... 17,232,000 100 17,232,000 Large Ground Water: >10,000 persons ...... 56,074,000 95 53,270,300 Small Ground Water: < 10,000 persons ...... 32,937,000 83 27,337,710

Total ...... 239,137,010

In general, little data are available on categories described in Table IV–7. of TTHMs in drinking water, the annual the occurrence of DBPs on a national TTHM levels among systems serving drinking water unit risk is multiplied by basis. Although there is sufficient greater than 10,000 people were the number of units, in this case the occurrence data available for THMs in estimated based on average concentration of TTHMs in µg/L, broken large water systems to develop a concentrations among systems in out into individual THMs based on the national occurrence distribution for that AWWA’s WIDB. TTHM levels in proportions presented above. Based on subset of systems, data are limited for systems serving less than 10,000 people these cancer risk estimates derived from small water systems. Similarly, some were estimated through modeling. laboratory animal studies, the annual occurrence data for HAA5 are available Modeling consisted of applying TTHM 95th percentile upper bound number of for large surface water systems, but not predictive equations to estimates of DBP cancer cases attributable to TTHMs is small surface water and groundwater precursor levels and treatment approximately 100. This means that systems. conditions. The mean weighted average there is a 95 percent chance that the baseline TTHM concentrations among annual number of cases are less than or 2. Baseline Risk Assessment Based on all the system type categories was 44 µg/ equal to 100. Using the maximum TTHM Toxicological Data L. likelihood or ‘‘best’’ estimates, the EPA performed a quantitative risk Occurrence data from an EPA DBP annual number of cancer cases is about assessment using the dose-response field study indicate that chloroform is 2. information on THMs. This assessment, the most common THM (in general, however, captures only a portion of the about 70 percent of total THMs), with 3. Baseline Analysis Based on potential risk associated with DBPs in bromoform being the least common (1 Epidemiology Data drinking water. It is not possible, given percent). Bromodichloromethane has an Epidemiological studies can be used existing toxicological and exposure occurrence of approximately 20 percent to assess the overall population risk data, to gauge how much of the total of the total THMs, with associated with a particular exposure. cancer risk associated with the dibromochloromethane comprising the Since the late 1970s, epidemiological consumption of chlorinated drinking final 8 percent of the total THMs. In the investigations have attempted to assess water is posed by TTHMs alone. An absence of more detailed occurrence whether chlorinated drinking water assessment of THMs, however, provides data, these proportions are used to contributes to the incidence of bladder, some estimation of the potential human divide the average TTHM concentration colon, rectal, and other cancers. Several risk, albeit limited. into the concentration for the four studies have reported a weak Performing the risk assessment based individual compounds. association between bladder cancer and on TTHM toxicological data requires Two estimates of risk factors were exposure to chlorinated drinking water, making several assumptions and used to estimate the cancer incidence. but a causal relationship has not been extrapolations (from a nonhuman The first set of lifetime unit risk factors confirmed (Freedman, et al., 1997). species to humans, from high doses in represent the upper 95 percent Several cancer epidemiological the laboratory study to lower confidence limit of the dose-response studies examining the association environmental exposures, and from a function. The second estimate of between exposure to chlorinated surface nondrinking water route to the relevant lifetime unit risk is the maximum water and cancer were published route of human exposure). Assumptions likelihood estimate used in the 1994 subsequent to the 1994 proposed rule are also made about the occurrence of analysis that represents the central and the 1992 meta-analysis. In general, TTHMs and the individual DBPs. EPA tendency of the dose-response function these new studies are better designed estimated the pre-Stage 1 DBPR TTHM (Bull, 1991). The annual unit risk is than the studies published prior to the concentration levels by calculating a calculated by dividing the lifetime risk 1994 proposal. The new studies include weighted average (based on populations by a standard assumption of 70 years incidence of disease, interviews with receiving disinfected waters) of TTHM per lifetime. To calculate the annual the study subjects, and better exposure levels among the different system type incidence of cancer due to consumption assessments. More evidence is available Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69439 on bladder cancer for a possible to evaluate the relationship between light of these concerns, the Agency association to exposure to chlorinated exposure to chlorinated drinking water agrees that causality between exposure surface water than other cancer sites. and incidence of cancer cases, based on to chlorinated water and bladder cancer Because of the limited data available for personal interviews; (all finally selected has not been established and that the other cancer sites such as colon and studies were population-based, case- number of cases attributable to such rectal cancer, the RIA focuses on control studies) exposures could be zero. 2. The study was of high quality and bladder cancer. Based on the estimate of 54,500 new well designed (e.g., adequate sample Based on the best studies, a range of bladder cancer cases per year nationally, size, high response rate, adjusted for potential risks was developed through as projected by the National Cancer the use of the population attributable known confounding factors); and, 3. The study had adequate exposure Institute for 1997, the numbers of risk (PAR) concept. Epidemiologists use possible bladder cancer cases per year PAR to quantify the fraction of disease assessments (e.g., residential histories, actual THM data). potentially associated with exposures to burden in a population (e.g., bladder DBPs in chlorinated drinking water cancer) that could be eliminated if the Using the above criteria, five bladder cancer studies were selected for estimated from the five studies range exposure (e.g., chlorinated drinking × estimating the range of PARs. from 1,100 (0.02 54,500) to 9,300 (.17 water) was absent. PAR (also referred to • × 54,500) cases. As noted above, due to as attributable risk, attributable portion, Cantor, et al., 1985; • McGeehin, et al., 1993; the uncertainty in these estimates, the or etiologic fraction) provides a • King and Marrett, 1996; number of cases could also be zero. In perspective on the potential magnitude • Freedman, et al., 1997; and making these estimates it is necessary to of risks associated with various • Cantor, et al., 1998. assume that these bladder cancer cases exposures under the assumption of The PARs from the five bladder are attributed to DBPs in chlorinated causality. For example, the National cancer studies ranged from 2 percent to surface water, even though the studies Cancer Institute estimates that there will 17 percent. These values were derived examined the relationship between be 54,500 new cases of bladder cancer from measured risks (Odds Ratio and chlorinated surface water and bladder in 1997. If data from an epidemiological Relative Risk) based on the number of cancer. This derived range is not study analyzing the impact of years exposed to chlorinated surface accompanied by confidence intervals consuming chlorinated drinking water water. Because of the uncertainty in (C.Is), but the C.Is. are likely to be very reports a PAR of 1 percent, it can be these estimates, it is possible that the wide. EPA believes that the mean risk estimated that 545 (54,500 × .01) PAR could also be zero. The estimates from each of the five studies bladder cancer cases in 1997 may be uncertainties associated with these PAR provides a reasonable estimate of the attributable to chlorinated drinking estimates are large due to the common water. potential range of risk suggested by the prevalence of both the disease (bladder different epidemiological studies. Table Under the Executive Order #12866 cancer) and exposure (chlorinated that requires EPA to conduct a RIA, EPA IV–8 contains a summary of the risk drinking water). estimates from the 1994 draft RIA and has chosen to estimate an upper bound In order to apply these PAR estimates the estimates derived from the more bladder cancer risk range for chlorinated to the U.S. population to estimate the recent analysis. drinking water using the PAR. EPA number of bladder cancer cases suggested this approach in the 1998 attributable to DBPs in drinking water, A related analysis based on odds NODA (EPA, 1998a). While EPA a number of assumptions must be made. ratios was conducted to derive a range recognizes the limitations of the current These include: (1) that the study of plausible estimates for cancer epidemiologic data base for making populations selected for each of the epidemiologic studies (EPA, 1998n). these estimates, the Agency considers cancer epidemiology studies are This analysis was also based on bladder the data base reasonable for use in reflective of the entire population that cancer studies (the five studies cited developing an upper bound estimate of develops bladder cancer; (2) that the above in addition to Doyle et al. 1997). bladder cancer risk for use in the RIA. percentage of those cancer cases in the For the purpose of this exercise, the In light of the toxicological evidence, studies exposed to chlorinated drinking annual U.S. expected number of 47,000 EPA recognizes that the risks from water are reflective of the bladder bladder cancers cited by Morris et chlorinated drinking water may be cancer cases in the U.S.; (3) that DBPs al.(1992) was used to calculate estimates considerably lower than those derived were the only carcinogens in these of the cancers prevented. The number of from the currently available chlorinated surface waters; and (4) that cancers attributable to DBP exposure epidemiological studies. EPA selected the relationship between DBPs in was estimated not to exceed 2,200– studies for inclusion in the quantitative chlorinated drinking water exposure 9,900 per year and could include zero. analysis if they contained the pertinent and bladder cancer is causal. As would be expected from related data to perform a PAR calculation and The last of these assumptions is analysis performed in the same data, met all three of the following criteria: perhaps the most open to question. As this range is similar to the 1,100–9300 1. The study was a population-based, noted in the March 1998 NODA, the PAR range. EPA has used the 1100–9300 case-control, or cohort study conducted results of the studies are inconsistent. In PAR range for the RIA.

TABLE IV±8.ÐNUMBER OF CANCER CASES ATTRIBUTABLE TO DBPS: COMPARISON OF ESTIMATES IN 1994 AND 1998

1994 estimates 1998 estimates

Number of New Bladder Cancer Cases/Year ...... Approx. 50,000 ...... 54,500. Number of Estimated Deaths Due to Bladder Cancer/Year ...... Did not state ...... 12,500. Attributable to DBPs in Drinking Water Data Source ...... >15 studies ...... 5 studies that meet specific criteria. Causality ...... No ...... No. Percent Attributable to DBPs ...... Did not state ...... 2% to 17%. 69440 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

TABLE IV±8.ÐNUMBER OF CANCER CASES ATTRIBUTABLE TO DBPS: COMPARISON OF ESTIMATES IN 1994 AND 1998Ð Continued

1994 estimates 1998 estimates

Number of Cancer Cases Attributable to DBPs: Estimated Using Toxicological Data ...... Less than 1* ...... Zero to 100.** Estimated Using Epidemiological Data ...... Over 10,000*** ...... Zero to 9,300.**** * Based on maximum likelihood estimates of risk from THMs. ** Based on IRIS 95th percent C.I. estimates of risk from THMs. *** Indicates rectal and bladder cancer cases. **** Indicates only bladder cancer cases.

The current benefits analysis is and nonfatal bladder cancer cases with underestimate of potential benefits that structured in roughly the same manner the estimate of the number of bladder will be realized as a consequence of as that presented in the 1994 RIA. The cancer cases reduced by the rule. The today’s action. While the low end of the baseline cancer risks could lie anywhere following assumptions are used to range cannot extend below $0, it is from zero to 100 cases per year based on estimate the range of potential benefits: possible that the high end of the range toxicological data; and zero to 9,300 • An estimate of the number of could extend beyond $4 billion if the cases per year based on epidemiological bladder cancer cases attributable to other reductions in risk could be data. Consequently, the task is to assess DPBs in drinking water ranging from 0 quantified and monetized. No discount the economic benefit of the final Stage to 9,300 annually. factor has been applied to these 1 DBPR in the face of this broad range • A 24 percent reduction in exposure valuations, although there is likely to be of possible risk. to TTHMs due to the Stage 1 DBPR (75 a time lag between compliance with the percent CI of 19 to 30 percent) will rule and the realization of benefits. 4. Exposure Reduction Analysis result in an equivalent reduction in Given this wide range of potential EPA predicted exposure reductions bladder cancer cases benefits and the uncertainty involved in due to the current Stage 1 DBPR relative • A value per statistical life saved for estimating the risk attributable to DBPs, to the present baseline. EPA used the fatal bladder cancers represented by a EPA undertook five different concentration of TTHMs as a marker to distribution with a mean of $5.6 million approaches to assessing the net benefits measure the exposure to the range of • A willingness to pay to avoid a of the Stage 1 DBPR. These approaches DBPs because data are available on the nonfatal case of bladder cancer are described in the net benefits section baseline occurrence and formation of represented by a distribution with a and should be considered both TTHMs. There are limited data on the mean of $587,500 individually and in the aggregate. total mix of byproducts in drinking Using the low end of the risk range of water. Therefore, the reduction in 0 bladder cancer cases attributable to E. Net Benefits Analysis TTHMs is assumed to reflect the DBPs results in a benefits estimate of $0. The potential economic benefits of the reduction in exposure to all DBPs. To To calculate the high end of the range, Stage 1 DBPR derive from the increased determine the change in exposure, it is the 9,300 estimate of attributable cases level of public health protection and necessary to estimate the pre-Stage 1 is multiplied by the percent reduction associated decreased level of risk. The baseline average TTHM concentration in exposure to derive the number of quantification of the benefits resulting and the post Stage 1 average TTHM bladder cancer cases reduced (9,300 × from DBP control is complicated by the concentration. The difference in the pre- .24 = 2,232 bladder cancer cases uncertainty in the understanding of the and post-Stage 1 TTHM concentrations reduced). This assumes a linear health risks. Epidemiological studies, reflect the potential reduction in relationship between reduction in referred to previously, suggest an TTHMs and thus in DBPs. TTHMs concentrations and reduction in association between bladder cancer and As described previously, the cancer risk (e.g., 24 percent reduction in exposure to chlorinated surface water; estimated pre-Stage 1 TTHM weighted TTHMs concentration is associated with however, these risks are uncertain. The average concentration is 44 µg/L for all a 24 percent reduction in cancer risk). lowest estimate in the selected system sizes and types of systems. The Assuming 23 percent of the bladder epidemiological studies of the number post Stage 1 TTHM concentrations for cancer cases end in fatality and 77 of new bladder cancer cases per year each system category were estimated percent are nonfatal, the number of fatal attributable to chlorinated surface water based on the technology compliance bladder cancer cases reduced is 513 is 1,100 cases, while the highest is 9,300 forecasts previously discussed and (2,232 × .23) and the number of nonfatal cases. EPA recognizes that while these estimated reductions in TTHM levels bladder cancer cases is 1,719 (2,232 × risks may be real, they also could be depending upon technology. The post- .77). Based on the valuation zero. Assessment of risks based only on Stage 1 TTHM weighted average distributions described above, the toxicological data for THMs, indicate a concentration is estimated at 33 µg/L. estimate of benefits at the mean much lower risk (2 cancer cases per year This represents a 24 percent reduction associated with reducing these bladder at the most likely estimate, to about 100 in TTHM levels resulting from the Stage cancer cases is approximately $4 billion. cases per year using the 95 percent 1 DBPR. Further details of the above It should be noted that these estimates confidence level upper bound), but analysis is described in the RIA for the do not include potential benefits from THMs represent only a few of the many Stage 1 DBPR (USEPA, 1998g). reducing other health effects (e.g, colon/ DBPs in drinking water. rectal cancer and reproductive EPA explored several alternative 5. Monetization of Health Endpoints endpoints) that cannot be quantified at approaches for assessing the benefits of The range of potential benefits from this time. As a result, EPA believes that the Stage 1 DBPR: Overlap of Benefit the Stage 1 DBPR can be estimated by the potential benefits discussed in and Cost Estimates; Minimizing Total applying the monetary values for fatal today’s rule may be a substantial Social Losses; Breakeven Analysis; Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69441

Household Costs; and Decision-Analytic show that there is substantial overlap in $1 billion (each avoided non-fatal case Model. A summary of the analysis of the estimates of benefits and costs. The results in a cost savings of $0.6 million). each approach is presented below. More range of quantified benefits extends The sum of the cost savings is detailed descriptions are described in from zero to over $4 billion. The zero approximately $4 billion. The high end the RIA (USEPA, 1998g). end of the range of estimated benefits of the benefits range could potentially Overlap of Benefit and Cost Estimates. represents the possibility that there is be higher if other health damages are One method to characterize net benefits essentially no health benefit from avoided. The range of cost estimates is is to compare the relative ranges of reducing exposure to DBPs. The other significantly smaller, ranging from $500 benefits and costs. Conceptually, an end of the range assumes there are 9,300 million to $900 million annually. overlap analysis tests whether there is bladder cancer cases per year Although these cost estimates have enough of an overlap between the range attributable to DBPs and there is a 24 uncertainty, the degree of uncertainty is of benefits and the range of costs for percent annual reduction in exposure of little consequence to the decisions there to be a reasonable likelihood that with the promulgation of the rule, being made given the scale of the benefits will exceed costs. In a resulting in avoidance of 2,232 cases. uncertainty for the benefits. theoretical case where the high end of Assuming that number of avoided cases, the range of benefits estimates does not approximately 513 would have been Figure IV–1b, on the other hand, overlap the low end of the range of cost fatalities and would result in a cost indicates that while the quantified estimates, a rule would be difficult to savings of approximately $3 billion benefits could exceed the costs, there is justify based on traditional benefit-cost (each avoided fatality results in a cost the possibility that there could be rationale. savings of $5.6 million). Additionally, negative net benefits if there were no For the Stage 1 DBPR, the overlap 1,719 non-fatal cases avoided would health benefits. analysis (Figures IV–1a and IV–1b) result in a cost savings of approximately BILLING CODE 6560±50±U 69442 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

Figure IV–1a Overlap of Estimated Benefits and Costs of the Stage 1 DBPR Figure IV–1b Overlap of the Ranges of the Estimated Benefits and Costs of the Stage 1 DBPR

BILLING CODE 6560±50±C Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69443

Minimizing Total Social Losses risk level after implementation of the Total ‘‘social loss’’ for each risk Analysis. Minimizing Total Social policy option. scenario are also indicated in Table IV– Losses analysis, sometimes called Under this analysis the ‘‘total social 9. The ‘‘social loss’’ is the cost to society ‘‘minimizing regrets’’ analysis, is a costs’’ (water treatment costs plus costs of making a wrong choice among the decision-aiding tool that is suited for of health damages still remaining after regulatory alternatives. It is computed as use in situations where it is impossible treatment) are calculated for three the difference between the total social to pin down the exact nature and extent regulatory alternatives (No Action, Stage cost (water treatment cost plus of a risk. The basic premise of 1, and Strong Intervention—otherwise remaining health damages) of an Minimizing Total Social Losses analysis known as the proposed Stage 2 alternative at a given risk scenario and requirements of the 1994 proposal) is to estimate total social costs for policy the total social cost of the best across a range of risk scenarios (< 1; 100; alternatives over a range of plausible alternative (least total social cost 1,000; 2,500; 5,000; 7,500; and 10,000 alternative for that risk scenario). The risk scenarios. The actual, or ‘‘true’’ risk attributable bladder cancer cases regulatory alternatives across the is unknowable, so instead this analysis annually). Total social costs for each different risk levels can also be asks what range and level of risks could regulatory alternative for different risk compared to see which alternative be true, and then evaluates the total assumptions are presented in Table IV– minimizes the maximum potential loss. costs to society if particular risk levels 9. The results indicate that the Stage 1 The best alternative, by this ‘‘mini-max’’ within that range turned out to be the DBPR has the least social cost among ‘‘true’’ value. Total social costs include the three alternatives analyzed across criteria, would be the one in which the both the cost to implement the policy the range of risks from 2,500 through upper bound of potential losses is option, plus costs related to residual 7,500 attributable bladder cancer cases smallest. (i.e., remaining) health damages at each annually. BILLING CODE 6560±50±U 69444 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

BILLING CODE 6560±50±C Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69445

Under the Stage 1 DBPR alternative, estimates are needed in order for the Household costs provide a common the worst loss that could happen would benefits, as measured in avoided health sense test of benefit/cost relationships occur if the lowest end of the risk range damages associated with bladder cancer, and are another useful benchmark for is true. This would result in total social to equal the costs. comparing the willingness-to-pay to losses of $0.7 billion per year. It is Two important concepts for this reduce the possible risk posed by DBPs concluded that the maximum potential analysis are the cost of illness measure in drinking water. It is essentially a loss of the Stage 1 alternative is smaller and the willingness-to-pay measure. The household level breakeven analysis. It than that of No Action ($4.1 billion) by cost of illness measure includes medical works backwards from the cost to ask a factor of 6 and smaller than that of costs and lost wages associated with whether the implied amount of benefits Strong Intervention ($2.9 billion) by a being unable to work as a result of (willingness-to-pay) needed to cover factor of 4. Thus, the Stage 1 DBPR is illness. In comparison, willingness-to- costs is a plausible amount. the best of the 3 alternatives at pay measures how much one would pay minimizing the maximum social loss. to reduce the risk of having all the About 115 million households are The 1994 Reg. Neg. and 1997 M–DBP discomfort and costs associated with located in service areas of systems Advisory Committees implicitly applied nonfatal cancer if such an option affected by the Stage 1 DBPR. Of these this type of ‘‘minimizing maximum existed. The main difference between households, 71 million (62 percent) are loss’’ framework when developing and these two methods is that willingness- served by large surface water systems. evaluating the DBP regulatory options. to-pay incorporates pain and suffering, Approximately 4.2 million (4 percent) In the face of large uncertainty regarding as well as changes in behavior into the are served by small surface water risk from DBPs, they decided that a valuation, while cost of illness does not. systems. Large ground water systems moderate response, relying on the more EPA has estimated the cost of a non- served 24 million households (21 cost-effective of the available treatment fatal case of bladder cancer at $121,000 percent) and small ground water methods was appropriate as an interim using the cost of illness method, and at systems serve 15.7 million households step until more information on risk $587,500 using the willingness-to-pay (14 percent). becomes available. approach. Break Even Analysis. Breakeven Assuming an annual cost of $701 All of the households served by analysis represents another approach to million and assumptions about the systems affected by the Stage 1 DBPR assessing the benefits of the Stage 1 monetary value of preventing both fatal will incur some additional costs (e.g., DBPR given the scientific uncertainties. and nonfatal bladder cancer cases, the monitoring costs), even if the system Breakeven is a standard benchmark of Stage 1 DBPR would need to reduce 438 does not have to change treatment to cost effectiveness and economic bladder cancer cases per year using the comply with the proposed rule. The efficiency, and is essentially the point willingness-to-pay measure for nonfatal costs calculated below include both where the benefits of the Stage 1 DBPR cancers or 574 cases per year using the monitoring and treatment costs. are equal to the costs. Normally, the cost of illness measure. If exposure is The cumulative distribution of benefits and costs of an option are reduced by 24 percent, the baseline household costs for all systems and by calculated separately and then number of bladder cancer cases each system type is displayed in Figures compared to assess whether and by attributable to DBPs in chlorinated IV–2a, IV–2b, IV–2c. The distributions what amount benefits exceed costs. In drinking water required to break even show that the large percentage of the case of the Stage 1 DBPR, would need to range from 1,820 to 2,390 households will incur small additional independently estimating benefits is new cases annually. Although these costs, with a small portion of systems difficult, if not impossible, because of values are well above the range facing higher costs. At the highest end the 10,000-fold uncertainty surrounding indicated by existing toxicological data of the distribution, approximately 1,400 the risk. Instead, the breakeven analysis for THMs alone, they fall within the households served by surface water works backwards from those variables attributable risk range suggested by the systems in the 25–100 size range that are less uncertain. In this case, epidemiological studies. implementation costs for the rule and Household Cost Analysis. A fourth switching to membrane technology will the monetary value associated with the approach for assessing the net benefits face an average annual cost increase of health endpoints are used to calculate of the Stage 1 DBPR is to calculate the $400 per year ($33 per month). what baseline risk and risk reduction costs per household for the rule. BILLING CODE 6560.50-U 69446 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

BILLING CODE 6560-50-C Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69447

The households have been sorted into million) are in the highest category For households served by large three cost categories for the ease of ($10–$33.40 per month). ground water systems (Figure IV–2b), 95 comparison (Table IV–10). The first For households served by large percent will incur less than $1 per category includes households with a surface water systems (Figure IV–2b), 98 month, 4 percent will incur between $1 cost increase of less than $12 per year, percent will incur less than $1 per and $10 per month, and 1 percent will less than $1 per month. The second month, 2 percent will incur between $1 incur greater than $10 per month. The category contains households with costs and $10 per month, and 0.03 percent highest cost ($125 annually, $10.40 greater than $12 per year, but less than will incur greater than $10 per month. monthly) is faced by households served $120 per year ($10 per month). The The highest cost ($125 annually, $10.40 by systems in the 10,000 to 25,000 size monthly) is faced by households served third category includes households with range implementing membrane by systems in the 10,000 to 25,000 size cost increases greater than $120 per year technology. range implementing membrane For households served by small to $400 per year ($33 per month). technology. ground water systems (Figure IV–2c), 91 Across all system categories (see For households served by small percent will incur less than $1 per Figure IV–2a), 95 percent of the surface water systems (Figure IV–2c), 71 month, 5 percent will incur between $1 households (110.1 million) fall within percent will incur less than $1 per and $10 per month, and 4 percent will the first category and will incur less month, 28 percent will incur between incur greater than $10 per month. The than $1 per month additional costs due $1 and $10 per month, and 1 percent will incur greater than $10 per month. highest cost ($357 annually, $29.75 to the Stage 1 DBPR. An additional 4 monthly) is faced by households served percent (4.4 million) are in the second The highest cost ($400 annually, $33 monthly) is faced by households served by systems in the 25–100 size range category at between $1 and $10 per implementing membrane technology. month cost increase and 1 percent (1.0 by systems in the 25–100 size range implementing membrane technology. BILLING CODE 6560±50±U 69448 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69449 69450 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

BILLING CODE 6560-50-C Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69451

In the small proportion of systems customer survey research implying a the resources used to complete the where household costs are shown to be demand for this benefit. Poole report would have been to much greater—up to several hundreds Decision Analytical model. The RIA complete a new meta-analysis using the of dollars per year—these results are also discusses a fifth type of analysis in more recent studies that have come out driven by the assumption that which probability functions are used to since the Morris et al. (1992) meta- membrane technologies will be the model the uncertainty surrounding analysis and that the Poole evaluation selected treatment, as noted above. three variables (rule cost, exposure did not advance the science in this area. Additionally, two points must be made: reduction, and attributable bladder Several commenters were critical of the (1) a number of these systems may find cancer risk) in order to derive a PAR analysis (described in EPA, 1998a) less expensive means of compliance probability distribution function for used to characterize the potential (e.g., selection of alternative source annual net benefit of the Stage I rule. baseline bladder cancer cases per year water, purchased water, or Because there is little actual data on that could be attributable to exposure to consolidation with other systems); and these probability functions, this chlorinated drinking water. They (2) if these systems do install approach should be considered present several arguments including: membranes, they may receive additional illustrative only. It is not discussed questioning whether such an analysis is water quality and/or compliance further here, but is discussed in Chapter warranted given the inconsistencies in benefits beyond those associated with 6 of the RIA for the Stage 1 DBPR (EPA, the studies used to complete the DBPs. For example, because membranes 1998g). analysis; stating that the use of the term are so effective, systems that install While any one of the above analytical upper bound of any suggested risk of membranes are likely to incur lower approaches by itself may not make a cancer is inappropriate because this compliance costs for future definitive case for the benefit-cost does not include the potential risks from rulemakings. effectiveness for the Stage 1 DBPR, other cancer sites such as colon and Given the uncertain nature of the risks taken collectively EPA believes they rectal; using the assumption of causality associated with DBPs, household costs indicate that the Stage 1 DBPR benefits is not warranted given the to society will exceed the costs. The provide a common sense estimate of inconsistencies in the studies used to monetized benefits in the five willingness-to-pay to reduce the risks: complete the PAR analysis; and the PAR alternatives represent only a portion of Would the average household (95 analysis should include a lower bound total potential benefits. Benefits percent of households) be willing to pay estimate of zero. associated with other cancer sites (rectal less than $1 per month ($12 per year) to EPA agrees that the use of the Morris and colon) and other health endpoints reduce the potential risks posed by et al. (1992) meta-analysis for estimating (such as developmental and DBPs? benefits is not appropriate for the reproductive effects) could not be reasons cited by commenters (e.g., Willingness to pay studies are not quantified at this time, and while they studies of different designs and available to directly answer this could be nil, they also could be quite discussed in more detail in the 1998 question. Taking the $1 per month large. Based on a careful weighing of the DBP NODA). EPA is currently figure as a measure of implied public projected costs against the potential considering whether a new meta- health benefit at the household level, it quantified and non-quantified benefits, analysis that uses the most recent is useful to ask what benefits can be EPA has determined that the benefits of epidemiology studies would be useful identified that could balance a $1 per the rule justify its costs. for the Stage 2 rulemaking. The Poole month expenditure. First, it is entirely F. Summary of Comments (1997) report considered a meta analysis possible that there is much more than a of the available data. Poole used several dollar-a-month’s worth of tangible Many commenters expressed concern techniques to evaluate the data and health benefit based on reduced risk of about the wide range of benefits given included several new studies that were bladder cancer alone. Second, the broad the high national cost of the rule. EPA available at the time of his analysis. exposure to DBPs and the possible has revised the benefits analysis; and Poole concluded that the cancer health effects involved offer the while the associated uncertainties epidemiology data considered in his possibility that there are significant remain large, EPA believes the benefits evaluation should not be combined into additional health benefits of a tangible of the Stage 1 DBPR justify its costs. a single summary estimated and that the nature. However, the agency recognizes Other commenters expressed concern data had limited utility for risk that in the small percentage of situations with using the data from Morris et al. assessment purposes. More recent where the costs per household is (1992) for quantifying benefits. They studies by Cantor et al. (1998), Doyle et between $120 to $400 per year, this may believed that the studies used in the al. (1997) and Freedman et al. (1997) indeed be a difficult financial burden to meta-analysis were different in design were not available at the time of his meet (e.g., may exceed household and thus not appropriate to use in meta- evaluation. willingness-to-pay). analysis. In addition the commenters EPA understands commenters Finally, the preventive weighing and believed that potential confounding concerns with the PAR analysis, balancing of public health protection factors or bias may not have been especially concerns with assuming also provides a margin of safety—a adequately controlled in the selected ‘‘causality’’ in the PAR evaluation when hedge against uncertainties. Recent studies. Others believed there was it is stated in other sections of the survey research conducted in the utility in using the meta-analysis to preamble that EPA does not believe drinking water field provides provide a perspective on the potential causality has been established. Even compelling empirical evidence that the cancer risks. Several commenters were though causality has not been number one priority of water system supportive of the Poole (1997) established, EPA is required to estimate customers is the safety of their water. evaluation of the Morris et al. (1992) the potential impacts of major Although definitive economic research meta-analysis stating that they regulations such as the DBP Stage 1 has not been performed to investigate concurred that the Morris analysis rule. The Agency believes it is the extent of household willingness-to- should not be used for estimating appropriate to conduct the PAR analysis pay for such a margin of safety, there is benefits for the Stage 1 DBPR. Other as described in the 1998 DBP NODA strong evidence from conventional commenters suggested a better use of (EPA, 1998a), to provide estimates of the 69452 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations potential risk that may need to be on a substantial number of small comments, and any changes to the rule reduced. EPA agrees that the use of the entities, EPA is not required to prepare in response to the comments; (3) term ‘‘upper bound of any suggested a regulatory flexibility analysis. describe, and where feasible, estimate risk’’ is not appropriate because there Throughout the 1992–93 negotiated the number of small entities to which are other potential risks that have not rulemaking process for the Stage 1 the final rule will apply; (4) describe the been quantified that may contribute to DBPR and IESWTR and in the July 1994 projected reporting, record keeping, and the overall risk estimates. In addition, proposals for these rules, a small PWS other compliance requirements of the EPA agrees that the estimates of the was defined as a system serving fewer rule, including an estimate of the classes potential cancer cases should include than 10,000 persons. This definition of small entities that will be subject to zero as this is a possibility given the reflects the fact that the original 1979 the requirements and the type of uncertainties in the data. EPA agrees standard for total trihalomethanes professional skills necessary for that several assumptions are made in applied only to systems serving at least preparation of reports or records; and (5) the analysis regarding the national 10,000 people. The definition thus describe the steps the Agency has taken extrapolation of the results and that recognizes that baseline conditions from to minimize the impact on small there is insufficient information at this which systems serving fewer than entities, including a statement of the time to validate these assumptions. 10,000 people will approach reasons for selecting the chosen option However, given the need to develop disinfection byproduct control and and for rejecting other options which national estimates of risk, EPA believes simultaneous control of microbial would alter the impact on small entities. it is appropriate to make these pathogens is different than that for EPA has considered and addressed all assumptions in order to provide a systems serving 10,000 or more persons. the above requirements in the perspective on the potential risks from EPA again discussed this approach to Regulatory Impact Analysis (RIA) for the exposure to chlorinated surface waters. the definition of a small system for these Stage 1 DBPR (EPA 1998g). The Commenters expressed concerns with rules in the 1998 DBP NODA (EPA, following is a summary of the RFA. the high costs associated with systems 1998a). EPA is continuing to define The first requirement is discussed in that must adopt alternative advanced ‘‘small system’’ for purposes of this rule section I of today’s rule. The second, technologies, especially for small and the IESWTR as a system which third and fifth requirements are systems. Since the 1994 proposal, the serves fewer than 10,000 people. summarized below. The fourth The Agency has since proposed and projected national costs for the Stage 1 requirement is discussed in V.B taken comment on its intent to define DBPR have dropped significantly (as (Paperwork Reduction Act) and the ‘‘small entity’’ as a public water system discussed above). This is mainly due to Information Collection Requirement. the revised compliance forecast and that serves 10,000 or fewer persons for Number of Small Entities Affected. lower membrane technology costs. In purposes of its regulatory flexibility EPA estimates that 69,491 groundwater the revised compliance forecast, fewer assessments under the RFA for all future systems will be affected by the Stage 1 systems using surface water will need drinking water regulations. (See DBPR, with 68,171 (98%) of these advanced technologies to comply. This Consumer Confidence Reports Rule, 63 systems serving less than 10,000 shift to lesser use of advanced FR 7620, Feb. 13, 1998.) In that persons. Of the 68,171 small systems technologies to comply with the Stage 1 proposal, the Agency discussed the affected, EPA estimates that 8,323 (12%) DBPR also pertains to small systems basis for its decision to use this will have to modify treatment to comply (those serving less than 10,000 people). definition and to use a single definition Commenters expressed concern for of small public water system whether with the Stage 1 DBPR. Of these, 5,403 the high costs associated with the Stage the system was a ‘‘small business’’, systems (8%) will use chloramines to 2 DBPR and whether EPA would obtain ‘‘small nonprofit organization’’, or comply and 2,921 systems (4.3%) will enough information to adequately ‘‘small governmental jurisdiction.’’ EPA use membranes to comply. Use of these understand the risks that might be also consulted with the Small Business technologies by small groundwater avoided to justify such a rule. EPA Administration on the use of this systems will result in total capital costs agrees that additional health effects definition as it relates to small of $998 million and an annualized information is needed before businesses. Subsequently, the Agency treatment cost of $180 million. reproposing the Stage 2 DBPR and will has used this definition in developing EPA estimates that 6,560 surface address this issue in the next round of its regulations under the Safe Drinking water systems will be affected by the FACA deliberations. Based on new data Water Act. This approach is virtually Stage 1 DBPR, with 5,165 (79%) of these generated through research, EPA will identical to the approach used in the systems serving less than 10,000 reevaluate the Stage 2 regulations and Stage 1 DBPR and IESWTR. Since, EPA persons. It is estimated that 3,616 (70%) re-propose, as appropriate. is not able to certify that the final Stage of these small systems will have to 1 DBPR will not have a significant modify treatment to comply with the V. Other Requirements economic impact on a substantial Stage 1 DBPR and 3,459 (67%) of these A. Regulatory Flexibility Act number of small entities, EPA has systems will use a combination of completed a final RFA and will publish enhanced coagulation, chloramines, and 1. Today’s Rule a small entity compliance guidance to ozone, while another 157 systems (3%) Under the Regulatory Flexibility Act, help small entities comply with this will use membranes. Use of these 5 U.S.C. 601 et seq. (RFA), as amended regulation. technologies by small surface water by the Small Business Regulatory systems will result in total capital costs Enforcement Fairness Act, EPA 2. Background and Analysis of $243 million and an annualized generally is required to conduct a The Regulatory Flexibility Act treatment cost of $46 million. regulatory flexibility analysis describing requires EPA to address the following EPA has included several provisions the impact of the regulatory action on when completing a final RFA: (1) state which will reduce the economic burden small entities as part of rulemaking. succinctly the objectives of, and legal of compliance for these small systems. However, under section 605(b) of the basis for, the final rule; (2) summarize These requirements, discussed in RFA, if EPA certifies that the rule will public comments on the initial RFA, the greater detail in the RIA (EPA, 1998g), not have a significant economic impact Agency’s assessment of those include: Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69453

—Less routine monitoring. Small when needed. EPA considered an This is based on an estimate that there systems are required to monitor less option in the development of the final will be 4,631 respondents on average frequently for such contaminants as rule for large systems to have MCLs of per year who will need to provide about TTHMs and HAA5. Also, ground 80 ug/L for TTHMs and 60 ug/L for 9,449 responses and that the average water systems (the large majority of HAAs and for small systems to have a response will take 33 hours. The annual small systems) are required to monitor simple TTHM standard of 100 ug/L. labor cost is estimated to be about $12 less frequently than Subpart H This option was rejected because million. In the first 3 years after systems (surface water systems and allowing small systems to comply with promulgation of the rule, only labor groundwater under the direct a different MCL level would not costs are incurred. The costs are influence of surface water) of the adequately protect the health of the incurred for the following activities: same size. population served by these systems. reading and understanding the rule; —Extended compliance dates. Systems EPA did not consider excluding small planning; and training. that use only ground water not under systems from the Stage 1 DBPR, because An Agency may not conduct or the direct influence of surface water these systems do not currently have any sponsor, and a person is not required to serving fewer than 10,000 people have standards for DBPs and the Agency respond to a collection of information 60 months from promulgation of this believed there was a public health unless it displays a currently valid OMB rule to comply. This is in contrast to concern that needed to be addressed. control number. The OMB control large Subpart H systems which have For a more detailed description of the numbers for EPA’s regulations are listed 36 months to comply. These extended alternatives considered in the in 40 CFR Part 9 and 48 CFR Chapter compliance dates will allow smaller development of the final rule see the 15. EPA is amending the table in 40 CFR systems to learn from the experience final RIA (EPA, 1998g) or the final Part 9 of currently approved ICR control of larger systems on how to most cost Unfunded Mandates Reform Act numbers issued by OMB for various effectively comply with the Stage 1 Analysis for the Stage 1 DBPR (EPA, regulations to list the information DBPR. In addition, larger systems will 1998o). generate a significant amount of requirements contained in this final treatment and cost data from the ICR B. Paperwork Reduction Act rule. This ICR was previously subject to and in their efforts to achieve The Office of Management and Budget public notice and comment prior to compliance with the Stage 1 (OMB) has approved the information OMB approval. As a result, EPA finds requirements. EPA intends to collection requirements contained in that there is ‘‘good cause’’ under section summarize this information and make this rule under the provisions of the 553 (b)(B) of the Administrative it available through guidance manuals Paperwork Reduction Act, 44 U.S.C. Procedures Act (5 U.S.C. 553 (b) (B)) to (i.e., the Small Entities Guidance 3501 et seq. and has assigned OMB amend this table without prior notice Manual). EPA believes this control number 2040–0204. and comment. Due to the technical information will assist smaller The information collected as a result nature of the table, further notice and systems in achieving compliance with of this rule will allow the States and the comment would be unnecessary. EPA to evaluate PWS compliance with the Stage 1 DBPR. C. Unfunded Mandates Reform Act the rule. For the first three years after 3. Summary of Comments promulgation of the Stage 1 DBPR, the 1. Summary of UMRA Requirements Several commenters expressed major information requirements pertain concern with the significant economic to preparation for monitoring activities, Title II of the Unfunded Mandates burden that the Stage 1 DBPR would and for compliance tracking. Responses Reform Act of 1995 (UMRA), Public place on small systems. Other to the request for information are Law 104–4, establishes requirements for commenters suggested more flexibility mandatory (Part 141). The information Federal agencies to assess the effects of be given for small systems and that a collected is not confidential. their regulatory actions on State, local, longer compliance period for small EPA is required to estimate the and tribal governments and the private systems should be included in the final burden on PWS for complying with the sector. Under UMRA section 202, EPA Stage 1 DBPR. Several commenters final rule. Burden means the total time, generally must prepare a written suggested small systems should not be effort, or financial resources expended statement, including a cost-benefit included in the final Stage 1 DBPR by persons to generate, maintain, retain, analysis, for proposed and final rules because the costs for implementing the or disclose or provide information to or with ‘‘Federal mandates’’ that may rule would exceed the potential benefits for a Federal agency. This includes the result in expenditures to State, local, for these systems. time needed to review instructions; and tribal governments, in the aggregate, EPA understands commenters’ develop, acquire, install, and utilize or to the private sector, of $100 million concerns with the potential significant technology and systems for the purposes or more in any one year. Before economic burden on small systems. of collecting, validating, and verifying promulgating an EPA rule, for which a Because of this potential significant information, processing and written statement is needed, section 205 impact, EPA has provided several maintaining information, and disclosing of the UMRA generally requires EPA to requirements which will reduce the and providing information; adjust the identify and consider a reasonable burden on these systems. These existing ways to comply with any number of regulatory alternatives and requirements which are discussed above previously applicable instructions and adopt the least costly, most cost- and also in greater detail in the RIA requirements; train personnel to be able effective or least burdensome alternative (EPA, 1998g) include: (1) less routine to respond to a collection of that achieves the objectives of the rule. monitoring; and (2) extended information; search data sources; The provisions of section 205 do not compliance dates. EPA also believes complete and review the collection of apply when they are inconsistent with small systems can reduce their information; and transmit or otherwise applicable law. Moreover, section 205 economic burden by; (1) consolidation disclose the information. allows EPA to adopt an alternative other with larger systems; (2) using money EPA estimates that the annual burden than the least costly, most cost effective from the State revolving fund loans; and on PWS and States for reporting and or least burdensome alternative if the (3) using variances and exemptions recordkeeping will be 314,471 hours. Administrator publishes with the final 69454 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations rule an explanation on why that b. Cost Benefit Analysis. Section IV each fiscal year to assist in running the alternative was not adopted. discusses the cost and benefits State drinking water program. Before EPA establishes any regulatory associated with the Stage 1 DBP rule. c. Estimates of Future Compliance requirements that may significantly or Also, the EPA’s Regulatory Impact Costs and Disproportionate Budgetary uniquely affect small governments, Analysis of the Stage 1 Disinfectants/ Effects. To meet the UMRA requirement including tribal governments, it must Disinfection Byproducts Rule (EPA, in section 202, EPA analyzed future have developed, under section 203 of 1998g) contains a detailed cost benefit compliance costs and possible the UMRA, a small government agency analysis. Today’s rule is expected to disproportionate budgetary effects. The plan. The plan must provide for have a total annualized cost of Agency believes that the cost estimates, notification to potentially affected small approximately $701 million using a 7 indicated above and discussed in more governments, enabling officials of percent cost of capital. The analysis detail in Section IV of this rule, affected small governments to have includes both qualitative and monetized accurately characterize future meaningful and timely input in the benefits for improvements to health and compliance costs of the rule. development of EPA regulatory safety. Because of scientific uncertainty In regard to the disproportionate proposals with significant Federal regarding the exposure assessment and impacts, EPA considered available data intergovernmental mandates; and the risk assessment for DBPs, the sources in analyzing the informing, educating, and advising Agency has used five analytical disproportionate impacts upon small governments on compliance with geographic or social segments of the the regulatory requirements. approaches to assess the benefits of the Stage 1 DBP. These analyses were based nation or industry. This analysis was 2. Written Statement for Rules With on the quantification of bladder cancer difficult because impacts will most Federal Mandates of $100 Million or health damages avoided. However, this likely depend on a system’s source More rule may also reduce colon and rectal water characteristics and this data is not EPA has determined that this rule cancers, as well as decrease adverse available for all systems. However, it contains a Federal mandate that may reproductive and developmental effects. should be noted that the rule uniformly result in expenditures of $100 million or This would further increase the benefits protects the health of all drinking water more for State, local, and tribal of this rule. system users regardless of the size or governments, in the aggregate, and the type of system. Further analysis Various Federal programs exist to revealed that no geographic or social private sector in any one year. provide financial assistance to State, Accordingly, EPA has prepared, under segment patterns were likely for this local, and Tribal governments in rule. One observation is that the section 202 of the UMRA, a written complying with this rule. The Federal statement addressing the following historical pattern of development in this government provides funding to States country led most large cities to be areas: (1) authorizing legislation; (2) that have primary enforcement cost-benefit analysis including an developed near rivers and other bodies responsibility for their drinking water analysis of the extent to which the costs of water useful for power, programs through the Public Water to State, local and Tribal governments transportation, and drinking water. To Systems Supervision Grants program. will be paid for by the federal the extent that this rule affects surface Additional funding is available from government; (3) estimates of future water, it in most ways reflects the compliance costs and disproportionate other programs administered either by distribution of population and budgetary effects; (4) macro-economic EPA or other Federal agencies. These geography of the nation. No rationale for effects; and (5) a summary of EPA’s include the Drinking Water State disproportionate impacts by geography consultation with State, local, and Revolving Fund (DWSRF) and Housing or social segment was identified. This Tribal governments, and a summary of and Urban Development’s Community analysis, therefore, developed three their concerns, and a summary of EPA’s Development Block Grant Program. For other measures: reviewing the impacts evaluation of their concerns. A more example, SDWA authorizes the on small systems versus large systems; detailed description of this analysis is Administrator of the EPA to award reviewing the costs to public versus presented in EPA’s Unfunded Mandates capitalization grants to States, which in private water systems; and reviewing Reform Act Analysis for the Stage 1 DBP turn can provide low cost loans and the household costs of the final rule. Rule (EPA, 1998o) which is included in other types of assistance to eligible First, the national impacts on small the docket for this rule. public water systems. The DWSRF systems (those serving fewer than a. Authorizing Legislation. Today’s assists public water systems with 10,000 people) versus large systems rule is promulgated pursuant to Section financing the costs of infrastructure (those serving 10,000 people or more) is 1412(b)(2) of the 1996 amendments to needed to achieve or maintain indicated in Table V–1. The higher cost the SDWA; paragraph C of this section compliance with SDWA requirements. to the small ground water systems is establishes a statutory deadline of Each State will have considerable mostly attributable to the large number November 1998 to promulgate this rule. flexibility to determine the design of its of these types of systems (i.e. there are This rule supersedes the TTHM Rule program and to direct funding toward 68,171 small ground water systems, (EPA, 1979). In addition, the Stage 1 its most pressing compliance and public 1,320 large ground water systems, 5,165 DBP rule is closely integrated with the health protection needs. States may small surface water systems, and 1,395 IESWTR, which also has a statutory also, on a matching basis, use up to ten large surface water surface water deadline of November 1998. percent of their DWSRF allotments for systems).

TABLE V±1.ÐANNUAL COST OF COMPLIANCE FOR SMALL AND LARGE SYSTEMS ($000)*

Small systems Large systems (population (population < 10,000) ≥ 10,000)

Surface Water Systems (All) ...... $56,804 $278,321 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69455

TABLE V±1.ÐANNUAL COST OF COMPLIANCE FOR SMALL AND LARGE SYSTEMS ($000)*ÐContinued

Small systems Large systems (population (population < 10,000) ≥ 10,000)

Ground Water System (All) ...... 218,062 130,651

Total ...... 274,866 408,972 * Costs calculated at a 7 percent cost of capital and include one time start-up costs.

The second measure of compliance with existing regulations. governmental entities and the private disproportionate impact evaluated is the EPA believes that smaller systems sector affected by this rule through relative total costs to public versus incurring the highest household costs various means. This included private water systems, by size. EPA may also incur the highest reduction in participation on a Regulatory believes the implementation of the rule risk. This is because smaller systems Negotiation Committee chartered under affects both public and private water have not had to previously comply with the Federal Advisory Committee Act systems equally, with the variance in a TTHMs standard of 100 ug/L. In the (FACA) in 1992–93 that included total cost by system size merely a RIA, EPA estimates that on average, stakeholders representing State and function of the number of affected small systems will achieve about twice local governments, public health systems. as much reduction in risk as achieved organizations, public water systems, The third measure, household costs, by larger systems (EPA,1998g). elected officials, consumer groups, and can also be used to gauge the impact of Based on the analysis above, EPA environmental groups. a regulation and to determine whether does not believe there will be After the amendments to SDWA in there are disproportionately high disproportionate impacts on small 1996, the Agency initiated a second impacts in particular segments of the systems, public versus private systems, FACA process, similarly involving a population. A detailed analysis of or generally by household. A more broad range of stakeholders, and held household cost impacts by system size detailed description of this analysis is meetings during 1997 to address the and system type are presented in presented in the EPA’s Unfunded expedited deadline for promulgation of Section IV.E. In summary, for large Mandates Reform Act Analysis for the the Stage 1 DBPR in November 1998. surface water systems EPA estimates Stage 1 DBP Rule (EPA,1998o). EPA established the M–DBP Advisory that 98 percent of households will incur d. Macro-economic Effects. As Committee to collect, share, and analyze costs of less than $1 per month while required under UMRA Section 202, EPA new data reviewed since the earlier Reg. 0.3 percent of households will incur is required to estimate the potential Neg. process and also to build a costs greater than $10 per month. For macro-economic effects of the consensus on the regulatory large groundwater systems, EPA regulation. Macro-economic effects tend implications of this new information. estimates that 95 percent of households to be measurable in nationwide The M–DBP Advisory Committee will incur costs of less than $1 per econometric models only if the established a technical working group to month while 1.0 percent of households economic impact of the regulation assist them with the many scientific will incur costs greater than $10 per reaches 0.25 percent to 0.5 percent of issues surrounding this rule. The month. For small surface water systems Gross Domestic Product (GDP). In 1997, Committee included representatives EPA estimates the 71 percent of real GDP was $7,188 billion so a rule from organizations such as the National households will incur costs of less than would have to cost at least $18 billion League of Cities, the National $1 per month while 1 percent of to have a measurable effect. A regulation Association of City and County Health households will incur costs of greater with a smaller aggregate effect is Officials, the Association of than $10 per month. For small unlikely to have any measurable impact Metropolitan Water Agencies, the groundwater systems EPA estimates that unless it is highly focused on a Association of State Drinking Water 91 percent of households will incur particular geographic region or Administrators, and the National costs of less than $1 per month while 4 economic sector. The macro-economic Association of Water Companies. In percent of households will incur costs effects on the national economy from addition, the Agency invited the Native of greater than $10 per month. the Stage 1 DBPR should be negligible American Water Association to The household analysis tends to based on the fact that the total annual participate in the FACA process to overestimate the costs per household costs are about $701 million per year (at develop this rule. Although they because of the structure and a 7 percent cost of capital) and the costs eventually decided not to take part, the assumptions of the methodology. For are not expected to be highly focused on Association continued to be informed of example, the highest per-household cost a particular geographic region or sector. meetings and developments through a would be incurred in a system using e. Summary of EPA’s Consultation stakeholders mailing list. membrane technology. These systems, with State, Local, and Tribal Stakeholders who participated in the conversely, might seek less costly Governments and Their Concerns. FACA processes, as well as all other alternatives such as point-of-use Under UMRA section 202, EPA is to interested members of the public, were devices, selection of alternative water provide a summary of its consultation invited to comment on the proposed sources, or connecting into a larger with elected representatives (or their rule and NODAs. Also, as part of the regional water system. The overall effect designated authorized employees) of Agency’s Communication Strategy, EPA is that costs are higher in smaller affected State, local and Tribal sent copies of the proposed rule and systems, and a higher percentage of governments in this rulemaking. NODAs to many stakeholders, including those systems are publicly owned. Although this rule was proposed before six tribal associations. Smaller systems, however, represent a UMRA became a statutory requirement, In addition, the Agency notified larger portion of systems that are not in EPA initiated consultations with governmental entities and the private 69456 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations sector of opportunities to provide input commenters preferred the option which effective enhanced coagulation (i.e., on this Stage 1 DBPR in the Federal provides the maximum flexibility obtaining efficiencies of TOC removal Register on July 29, 1994 (59 FR allowed under the SDWA for systems to without excessive sludge production 38668—EPA, 1994A), November 3, 1997 comply with the Stage 1 DBPR, and this and associated costs). (62 FR 59485—EPA, 1997b), and on is the option EPA selected for the final In conclusion, EPA believes that the March 31, 1998 (63 FR 15974—EPA, rule. alternative selected for the Stage 1 DBPR 1998a). Additionally, EPA extended the f. Regulatory Alternatives Considered. is the most cost-effective option that comment period for the March 31, 1998 As required under Section 205 of the achieves the objectives of the rule. For NODA and announced a public meeting UMRA, EPA considered several a complete discussion of this issue see to address new information. EPA regulatory alternatives developed by the EPA’s Regulatory Impact Analysis of the received approximately 213 written Reg Neg Committee and M–DBP Stage 1 Disinfectants/Disinfection comments on the July 29, 1994 notice, Advisory Committee and suggested by Byproducts Rule (EPA,1998g). stakeholders. approximately 57 written comments on 3. Impacts on Small Governments the November 3, 1997 notice, and The Reg Neg Committee considered approximately 41 written comments on several options including a proposed The 1994 Stage 1 DBPR proposal was the March 31, 1998 notice. Of the 213 TTHMs MCL of 80 µg/L and HAA5 MCL done without the benefit of the UMRA comments received concerning the 1994 of 60 µg/L for large systems (and a requirements. However, in preparation proposed rule, 11% were from States simple standard of 100 µg/l for small for the final rule, EPA conducted and 41% were from local governments. systems). Another option called for the analysis on small government impacts Also, one comment on the 1994 use of precursor removal technology to and included small government officials proposal was from a tribal group that reduce the level of total organic carbon or their designated representatives in represented 43 tribes. Of the 57 with alternative levels ranging from 4.0 the rule making process. The FACA comments received concerning the 1997 to 0.5. Other options evaluated included processes gave a variety of stakeholders, Notice of Data Availability, 18% were a 80 µg/L for TTHMs, 60 µg/L for HAA5, including small governments, the from States and 37% were from local and 4.0 for TOC. Finally, an option was opportunity for timely and meaningful governments. Of the 41 comments evaluated of a 80 µg/L for TTHMs, 60 participation in the regulatory received on the 1998 Notice of Data µg/L for HAA5, and 5.0 for TOC. The development process. Representatives of Availability prior to the close of the final consensus included a combination small government organizations were on comment period, 5% were from States of MCLs which would be equal for all both the Reg. Neg. Committee and the and 15% were from local governments. system size categories and a target TOC M–DBP Advisory Committee and their The public docket for this rulemaking level. Allowing small systems to comply representatives attended public contains all comments received by the with a different MCL levels was rejected stakeholder meetings. Groups such as Agency and provides details about the because the rule would not adequately the National Association of City and nature of State, local, and tribal protect the health of the population County Health Officials and the government’s concerns. State and local served by these systems. A more National League of Cities participated in governments raised several concerns detailed description of these alternatives the rulemaking process. Through such including: the need for the Stage 1 is discussed in the document Unfunded participation and exchange, EPA DBPR; the high costs of the rule in Mandates Reform Act Analysis for the notified potentially affected small relation to the uncertain benefits; the Stage 1 DBPR Rule which can be found governments of requirements under belief that not allowing predisinfection in the docket (EPA, 1998o). consideration and provided officials of credit would increase the microbial risk; Other regulatory alternatives were affected small governments with an and the need for flexibility in considered by the M–DBP Advisory opportunity to have meaningful and implementing the Stage 1 DPBR and Committee and these alternatives had timely input into the development of IESWTR to insure the rules are the overall effect of reducing the cost of regulatory proposals. implemented simultaneously. The one the final rule. For example, the M–DBP In addition, EPA will educate, inform, tribal comment noted that compliance Advisory Committee recommended and advise small systems including would come at a cost of diverting funds maintaining the predisinfection credit those run by small government about away from other important drinking after reviewing data which suggested DBPR requirements. One of the most water needs such as maintaining that many systems could probably meet important components of this process is drinking water infrastructure. the proposed MCLs for DBPs while the Small Entity Compliance Guide, as EPA understands the State, local, and maintaining current disinfection required by the Small Business tribal governments concerns with the practices. This decision was important Regulatory Enforcement Fairness Act of costs of the rule and the need to provide because systems would have had to 1996. This plain-English guide will additional public health protection for incur large capital costs to remain in explain what actions a small entity must the expenditure. The Agency believes compliance with disinfection take to comply with the rule. Also, the the final Stage 1 DPBR will provide requirements if predisinfection credits Agency is developing fact sheets that public health benefits to individuals by were disallowed. Thus by allowing concisely describe various aspects and reducing their exposures to DBPs, while predisinfection, the overall cost of the requirements of the DBPR. not requiring excessive capital rule was lowered. expenditures. As discussed above, the Also, the Committee recommended D. National Technology Transfer and majority of households will incur exempting systems for the enhanced Advancement Act additional costs of less than $1 per coagulation requirements based on their Under section 12(d) of the National month. As discussed in section III.E, the raw water quality. For example, systems Technology Transfer and Advancement final rule maintains the existing with raw-water TOC of less than or Act (NTTAA), the Agency is required to predisinfection credit. Finally, in the equal to 2.0 mg/L and raw-water SUVA use voluntary consensus standards in its 1997 DBP NODA (EPA, 1997b), EPA of less than or equal to 2.0 L/mg-m regulatory activities unless to do so requested comment on four alternative would be exempt from the enhanced would be inconsistent with applicable schedules for complying with the Stage coagulation requirements. This law or otherwise impractical. Voluntary 1 DBPR. Most State and local exclusion was intended to promote cost- consensus standards are technical Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69457 standards (e.g., materials specifications, ‘‘significant regulatory action’’ as one Finally, as part of EPA’s test methods, sampling procedures, that is likely to result in a rule that may: responsibilities to comply with E.O. business practices, etc.) that are 1. Have an annual effect on the 12898, the Agency held a stakeholder developed or adopted by voluntary economy of $100 million or more or meeting on March 12, 1998 to address consensus standards bodies. Where adversely affect in a material way the various components of pending available and potentially applicable economy, a sector of the economy, drinking water regulations; and how voluntary consensus standards are not productivity, competition, jobs, the they may impact sensitive sub- used by EPA, the Act requires the environment, public health or safety, or populations, minority populations, and Agency to provide Congress, through State, local, or tribal governments or low-income populations. Topics OMB, an explanation of the reasons for communities; discussed included treatment not using such standards. 2. Create a serious inconsistency or techniques, costs and benefits, data EPA’s process for selecting the otherwise interfere with an action taken quality, health effects, and the analytical test methods is consistent or planned by another agency; regulatory process. Participants with section 12(d) of the NTTAA. EPA 3. Materially alter the budgetary included national, state, tribal, performed literature searches to identify impact of entitlement, grants, user fees, municipal, and individual stakeholders. analytical methods from industry, or loan programs or the rights and EPA conducted the meetings by video academia, voluntary consensus obligations of recipients thereof; or conference call between eleven cities. standards bodies, and other parties that 4. Raise novel legal or policy issues This meeting was a continuation of could be used to measure disinfectants, arising out of legal mandates, the stakeholder meetings that started in DBPs, and other parameters. In addition, President’s priorities, or the principles 1995 to obtain input on the Agency’s EPA’s selection of the methods set forth in the Executive Order. Drinking Water Programs. The major benefited from the recommendations of Pursuant to the terms of Executive objectives for the March 12, 1998 an Advisory Committee established Order 12866, it has been determined meeting were: under the FACA Act to assist the • that this rule is a ‘‘significant regulatory Solicit ideas from EJ stakeholders Agency with the Stage 1 DBPR. The action’’ because it will have an annual on known issues concerning current Committee made available additional effect on the economy of $100 million drinking water regulatory efforts; technical experts who were well-versed • Identify key issues of concern to EJ or more. As such, this action was in both existing analytical methods and stakeholders; and new developments in the field. submitted to OMB for review. Changes • Receive suggestions from EJ The results of these efforts form the made in response to OMB suggestions or stakeholders concerning ways to basis for the analytical methods in recommendations are documented in increase representation of EJ today’s rule which includes: eight the public record. communities in OGWDW regulatory methods for measuring different DBPs, F. Executive Order 12898: efforts. of which five are EPA methods and Environmental Justice In addition, EPA developed a plain- three are voluntary consensus English guide specifically for this standards; nine methods for measuring Executive Order 12898 establishes a meeting to assist stakeholders in disinfectants, all of which are voluntary Federal policy for incorporating understanding the multiple and consensus standards; three voluntary environmental justice into Federal sometimes complex issues surrounding consensus methods for measuring TOC; agency missions by directing agencies to drinking water regulation. two EPA methods for measuring identify and address disproportionately Overall, EPA believes this rule will bromide; one voluntary consensus high and adverse human health or equally protect the health of all minority environmental effects of its programs, method for measuring UV254, and both and low-income populations served by governmental and voluntary consensus policies, and activities on minority and systems regulated under this rule from methods for measuring alkalinity. low-income populations. The Agency exposure to DBPs. Where applicable voluntary consensus has considered environmental justice G. Executive Order 13045: Protection of standards were not approved, this was related issues concerning the potential Children From Environmental Health due to their inability to meet the data impacts of this action and has consulted Risks and Safety Risks quality objectives (e.g. accuracy, with minority and low-income sensitivity, quality control procedures) stakeholders. Executive Order 13045 applies to any necessary for demonstration of Two aspects of today’s rule comply rule initiated after April 21, 1997, or compliance with the relevant with the Environmental Justice proposed after April 21, 1998, that (1) is requirement. Executive Order which requires the determined to be ‘‘economically In the 1997 NODA, EPA requested Agency to consider environmental significant’’ as defined under E.O. 12866 comment on voluntary consensus justice issues in the rulemaking and to and (2) concerns an environmental standards that had not been addressed consult with Environmental Justice (EJ) health or safety risk that EPA has reason and which should be considered for stakeholders. They can be classified as to believe may have a disproportionate addition to the list of approved follows: (1) the overall nature of the effect on children. If the regulatory analytical methods in the final rule. No rule, and (2) the convening of a action meets both criteria, the Agency additional consensus methods were stakeholder meeting specifically to must evaluate the environmental health suggested by commenters. address environmental justice issues. or safety effects of the planned rule on The Stage 1 DBPR applies to community children, and explain why the planned E. Executive Order 12866: Regulatory water systems and nontransient regulation is preferable to other Planning and Review noncommunity water systems that treat potentially effective and reasonably Under Executive Order 12866, (58 FR their water with a chemical disinfectant feasible alternatives considered by the 41344—EPA, 1993c) the Agency must for either primary or residual treatment. Agency. determine whether the regulatory action Consequently, the health protection The final Stage 1 DBPR is not subject is ‘‘significant’’ and therefore subject to benefits this rule provides are equal to the Executive Order because EPA OMB review and the requirements of across all income and minority groups published a notice of proposed the Executive Order. The Order defines within these communities. rulemaking before April 21, 1998. 69458 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

However, EPA’s policy since November the proposed Stage 1 DBP rule to the consultation with State, local, and tribal 1, 1995, is to consistently and explicitly Science Advisory Board, National governments; the nature of the consider risks to infants and children in Drinking Water Advisory Council government concerns; and EPA’s all risk assessments generated during its (NDWAC), and the Secretary of Health position supporting the need to issue decision making process including the and Human Services for their review. this rule. setting of standards to protect public EPA has evaluated comments received J. Executive Order 13084: Consultation health and the environment. from these organizations and considered and Coordination With Indian Tribal EPA’s Office of Water has historically them in developing the final Stage 1 Governments considered risks to sensitive DBP rule. populations (including fetuses, infants, Under Executive Order 13084, EPA and children) in establishing drinking I. Executive Order 12875: Enhancing the may not issue a regulation that is not water assessments, advisories or other Intergovernmental Partnership required by statute, that significantly or guidance, and standards (EPA, 1989c Under Executive Order 12875, EPA uniquely affects the communities of and EPA, 1991). The disinfection of may not issue a regulation that is not Indian tribal governments, and that public drinking water supplies to required by statute and that creates a imposes substantial direct compliance prevent waterborne disease is the most mandate upon a State, local or tribal costs on those communities, unless the successful public health program in U.S. government, unless the Federal Federal government provides the funds history. However, numerous chemical government provides the funds necessary to pay the direct compliance byproducts (DBPs) result from the necessary to pay the direct compliance costs incurred by the tribal reaction of chlorine and other costs incurred by those governments, or governments, or EPA consults with disinfectants with naturally occurring EPA consults with those governments. If those governments. If EPA complies by organic and inorganic material in source EPA complies by consulting, Executive consulting, Executive Order 13084 water, and these may have potential Order 12875 requires EPA to provide to requires EPA to provide to the Office of health risks. Thus, maximizing health the Office of Management and Budget a Management and Budget, in a separately protection for sensitive subpopulations description of the extent of EPA’s prior identified section of the preamble to the requires balancing risks to achieve the consultation with representatives of rule, a description of the extent of EPA’s recognized benefits of controlling affected State, local and tribal prior consultation with representatives waterborne pathogens while minimizing governments, the nature of their of affected tribal governments, a risk of potential DBP toxicity. Human concerns, copies of any written summary of the nature of their concerns, experience shows that waterborne communications from the governments, and a statement supporting the need to disease from pathogens in drinking and a statement supporting the need to issue the regulation. In addition, water is a major concern for children issue the regulation. In addition, Executive Order 13084 requires EPA to and other subgroups (elderly, immune Executive Order 12875 requires EPA to develop an effective process permitting compromised, pregnant women) develop an effective process permitting elected officials and other because of their greater vulnerabilities elected officials and other representatives of Indian tribal (Gerba et al., 1996). Based on animal representatives of State, local and tribal governments ‘‘to provide meaningful studies, there is also a concern for governments ‘‘to provide meaningful and timely input in the development of potential risks posed by DBPs to and timely input in the development of regulatory policies on matters that children and pregnant women (EPA, regulatory proposals containing significantly or uniquely affect their 1994a; EPA, 1998a). significant unfunded mandates.’’ communities.’’ In developing this regulation, risks to EPA has concluded that this rule will EPA has concluded that this rule will sensitive subpopulations (including create a mandate on State, local, and significantly affect communities of fetuses and children) were taken into tribal governments and that the Federal Indian tribal governments. It will also account in the assessments of government will not provide all of the impose substantial direct compliance disinfectants and disinfection funds necessary to pay the direct costs costs on such communities, and the byproducts. A description of the data incurred by the State, local, and tribal Federal government will not provide all available for evaluating risks to children governments in complying with the the funds necessary to pay the direct and the conclusions drawn can be found mandate. In developing this rule, EPA costs incurred by the tribal governments in the public docket for this rulemaking consulted with State and local in complying with the rule. In (EPA, 1998h). In addition, the Agency governments to enable them to provide developing this rule, EPA consulted has evaluated alternative regulatory meaningful and timely input in the with representatives of tribal options and selected the option that will development of this rule. EPA also governments pursuant to both Executive provide the greatest benefits for all invited the Native American Water Order 12875 and Executive Order people including children. See the Association to participate in the FACA 13084. EPA’s consultation, the nature of regulatory impact analysis for a process to develop this rule, but they the governments’ concerns, and EPA’s complete discussion of the different decided not to take part in the position supporting the need for this options considered. It should also be deliberations. rule are discussed above in the noted that the IESTWR, which As described in Section V.C.2.e, EPA preamble section that addresses accompanies this final rule, provides held extensive meetings with a variety compliance with Executive Order better controls of pathogens and of State and local representatives, who 12875. Specifically in developing this achieves the goal of increasing the provided meaningful and timely input rule, the Agency invited the Native protection of children. in the development of the proposed American Water Association to rule. State and local representatives participate in the FACA process to H. Consultations With the Science were also part of the FACA committees develop this rule. Although they Advisory Board, National Drinking involved in the development of this eventually decided not to take part, the Water Advisory Council, and the rule. Summaries of the meetings have Association continued to be informed of Secretary of Health and Human Services been included in the public docket for meetings and developments through a In accordance with section 1412 (d) this rulemaking. See section V.C.2.e for stakeholders mailing list. As described and (e) of the Act, the Agency submitted summaries of the extent of EPA’s in Section V.C.2.e of the discussion on Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69459

UMRA, EPA held extensive meetings drinking water? Is the source of • Fiscal management and controls. that provided the opportunity for generally good quality and adequately Are adequate books and records meaningful and timely input in the protected? maintained? Are appropriate budgeting, development of the proposed rule. • Infrastructure adequacy. Can the accounting, and financial planning Summaries of the meetings have been system provide water that meets SDWA methods used? Does the system manage included in the public docket for this standards? What is the condition of its its revenues effectively? rulemaking. infrastructure, including well(s) or There are 76,051 systems affected by source water intakes, treatment, storage, this rule. Of these, 12,998 will have to K. Submission to Congress and the and distribution? What is the modify their treatment process and General Accounting Office infrastructure’s life expectancy? Does undertake disinfectant and DBP The Congressional Review Act, 5 the system have a capital improvement monitoring and reporting. Some of this U.S.C. 801 et seq., as added by the Small plan? smaller group may also be required to Business Regulatory Enforcement • Technical knowledge and do DBP precursor monitoring and Fairness Act of 1996, generally provides implementation. Is the system’s operator reporting. The other 63,063 systems will that before a rule may take effect, the certified? Does the operator have need to do disinfectant and DBP agency promulgating the rule must sufficient technical knowledge of monitoring and reporting, but will not submit a rule report, which includes a applicable standards? Can the operator need to modify their treatment process. copy of the rule, to each House of the effectively implement this technical Some of this larger group may also be Congress and to the Comptroller General knowledge? Does the operator required to do DBP precursor of the United States. EPA will submit a understand the system’s technical and monitoring and reporting. report containing this rule and other operational characteristics? Does the Systems not modifying treatment are required information to the U.S. Senate, system have an effective operation and not generally expected to require the U.S. House of Representatives, and maintenance program? significantly increased technical, the Comptroller General of the United Managerial capacity is the ability of a financial, or managerial capacity to States prior to publication of the rule in water system to conduct its affairs in a comply with these new requirements. the Federal Register. A major rule manner enabling the system to achieve Certainly some individual facilities may cannot take effect until 60 days after it and maintain compliance with SDWA have weaknesses in one or more of these is published in the Federal Register. requirements. Managerial capacity refers areas but overall, systems should have This rule is a ‘‘major rule’’ as defined by to the system’s institutional and or be able to obtain the capacity needed 5 U.S.C. 804(2). This rule will be administrative capabilities. for these activities. effective February 16, 1999. Systems needing to modify treatment Managerial capacity can be assessed will employ one or more of a variety of L. Likely Effect of Compliance With the through key issues and questions, steps. The steps expected to be Stage 1 DBPR on the Technical, including: employed by 50% or more of subpart H Financial, and Managerial Capacity of • Ownership accountability. Are the systems and by eight percent or more of Public Water Systems system owner(s) clearly identified? Can ground water systems covered by the Section 1420(d)(3) of the SDWA as they be held accountable for the system? rule include a combination of low cost amended requires that, in promulgating • Staffing and organization. Are the alternatives, including switching to a NPDWR, the Administrator shall system operator(s) and manager(s) chloramines for residual disinfection, include an analysis of the likely effect clearly identified? Is the system moving the point of disinfectant of compliance with the regulation on properly organized and staffed? Do application, and improving precursor the technical, financial, and managerial personnel understand the management removal. EPA estimates that less than capacity of public water systems. The aspects of regulatory requirements and seven percent of systems in any category following analysis has been performed system operations? Do they have will resort to higher cost alternatives, to fulfill this statutory obligation. adequate expertise to manage water such as switching to ozone or Overall water system capacity is system operations? Do personnel have chloramines for primary disinfection or defined in EPA guidance (EPA 816–R– the necessary licenses and using GAC or membranes for precursor 98–006) as the ability to plan for, certifications? removal. These higher cost alternatives achieve, and maintain compliance with • Effective external linkages. Does the may also provide other treatment applicable drinking water standards. system interact well with customers, benefits, so the cost may be somewhat Capacity has three components: regulators, and other entities? Is the offset by eliminating the need for technical, managerial, and financial. system aware of available external technologies to remove other Technical capacity is the physical and resources, such as technical and contaminants. Some of these systems operational ability of a water system to financial assistance? may choose nontreatment alternatives meet SDWA requirements. Technical Financial capacity is a water system’s such as consolidation with another capacity refers to the physical ability to acquire and manage sufficient system or changing to a higher quality infrastructure of the water system, financial resources to allow the system water source. including the adequacy of source water to achieve and maintain compliance Furthermore, there are a number of and the adequacy of treatment, storage, with SDWA requirements. actions that are expected to be taken and distribution infrastructure. It also Financial capacity can be assessed disproportionately by smaller sized refers to the ability of system personnel through key issues and questions, systems (that is to say, a greater to adequately operate and maintain the including: percentage of smaller sized systems will system and to otherwise implement • Revenue sufficiency. Do revenues undertake than will larger sized requisite technical knowledge. A water cover costs? Are water rates and charges systems). These steps include increased system’s technical capacity can be adequate to cover the cost of water? plant staffing and additional staff determined by examining key issues • Credit worthiness. Is the system training to understand process control and questions, including: financially healthy? Does it have access strategy. Small systems will be required • Source water adequacy. Does the to capital through public or private to do this since larger systems have system have a reliable source of sources? already undertaken these changes to 69460 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations some extent for compliance with the 9. Cantor K. P., Hoover R., Hartge P., et al. 21. Federal Focus. 1996. Principles for 1979 TTHM rule. 1985. Drinking Water Source and Evaluating Epidemiological Data in Bladder Cancer: A Case-Control Study. Regulatory Risk Assessment. Developed For many systems serving less than In Jolley R.L., Bull R.J., Davis W.P., et al. By An Expert Panel at a Conference in 10,000 persons which need to make (eds), Water Chlorination: Chemistry, London, England October 1995. treatment modifications, an Environmental Impact and Health Appendix B: The ‘‘Bradford Hill’’ or enhancement of technical, financial, Effects, Vol. 5. Lewis Publishers, Inc., ‘‘Surgeon General’s’’ Criteria for Judging and managerial capacity may likely be Chelsea, MI pp 145–152. the Causal Significance of an Exposure- needed. As the preceding paragraph 10. Cantor K.P., Hoover R., Hartge P. et al. Effect Association Indicated by an makes clear, these systems will be 1987. Bladder Cancer, Drinking Water Epidemiological Study or Studies. August 1996. Federal Focus, Inc. making structural improvements and Source and Tap Water Consumption: A Case Control Study. JNCI; 79:1269–79. Washington, DC. enhancing laboratory and staff capacity. 11. Cantor K.P., Lunch C.F., Hildesheim M., 22. Freedman M., Cantor K. P., Lee N. L., Larger sized systems have typically Dosemeci M., Lubin J., Alavanja M., Chen L. S., Lei H. H., Ruhl C. E., and already made these improvements as Craun G.F.. 1998. Drinking Water Source Wang S. S. 1997. Bladder cancer and part of normal operations. Meeting the and Chlorination Byproducts. I. Risk of drinking water: a population-based case- requirements of the Stage 1 DBPR will Bladder Cancer. Epidemiology; 9:21–28. control study in Washington County, require operating at a higher level of 12. Chowdhury, Z. 1997. Presentation to Maryland (United States). Cancer Causes sophistication and in a better state of Technical Work Group January, 1997. and Control. 8, pp 738–744. repair than some plants serving less Cincinnati, OH. 23. Gerba, CP., J.B Rose, and C.N Haas. 1996. than 10,000 people have considered 13. Clark, S.C., J. Wiginton, and J.T. Sensitive Populations: Who is at the Greatest Risk. Int. J. Food and acceptable in the past. Musgrove. 1994. Enhanced Lime Softening: Is Your TOC Removal Maxed Microbiology. 30:113–123. Certainly there will be exceptions in Out? AWWA Enhanced Coagulation 24. Heywood R., Sortwell R.J., Noel PRB, systems serving both below 10,000 Workshop, December 1994. Street AE, Prentice DE, Roe FJC, persons and above. Some larger plants 14. CMA. 1996. Sodium Chlorite: Drinking Wadsworth PF, Worden AN, Van Abbe will doubtless find their technical, Water Rat Two-Generation Reproductive NJ. 1979. Safety Evaluation of managerial, and financial capacity taxed Toxicity Study. Chemical Manufacturers Toothpaste Containing Chloroform. III. by the new requirements. Likewise, Association. Quintiles Report Ref. CMA/ Long-term Study in Beagle Dogs. J. Environ. Pathol. Toxicol. 2:835–851. some plants serving less than 10,000 17/96. 25. Hildesheim M.E., Cantor K.P., Lynch C.F., persons will already have more than 15. CMA. 1998. Letter to Michael Cox, U.S. EPA, OGWDW. From Marian K. Stanley, Dosemeci M., Lubin J., Alavanja M., and adequate technical, financial, and Chemical Manufacturers Association. Craun G.F. 1998. Drinking Water Source managerial capacity to meet these Regarding the two-generation and Chlorination Byproducts: Risk of requirements. However, in general, the reproduction/developmental Colon and Rectal Cancers. Epidemiology. systems serving less than 10,000 neurotoxicity study with sodium 9:1, pp: 29–35. persons needing to make treatment chlorite. January 26, 1998. 20 pp. 26. ILSI. 1997. An Evaluation of EPA’s modifications will be the ones most 16. DeAngelo, A. B., Daniel, F. B., Most, B. Proposed Guidelines for Carcinogen Risk needing to enhance their capacity. M. and G. R. Olsen. 1997. The Failure of Assessment Using Chloroform and Monochloroacetic Acid and Dichloroacetate as Case Studies: Report VI. References Trichloroacetic Acid Administered in of an Expert Panel. International Life the Drinking Water to Produce Liver Sciences Institute, Health and 1. APHA. 1992. Standard Methods for the Cancer in Male F344/N rats. J. Toxicol. Environmental Sciences Institute Examination of Water and Wastewater, Environ. Health (in press). November, 1997. 18th Edition. American Public Health 27. Jorgenson T.A., Meierhenry E.F., Association, Washington, DC. 17. DeAngelo A. B., George M. H., Kilburn S. R., Moore T. M., Wolf D. C. 1998. Rushbrook C.J. Bull RJ, Robinson M. 2. APHA. 1995. Standard Methods for the Carcinogenicity of Potassium Bromate 1985. Carcinogenicity of Chloroform in Examination of Water and Wastewater, Administered in the Drinking Water to Drinking Water to Male Osborne-Mendel 19th Edition. American Public Health Make B6C3F1 Mice and F344/N Rats, Rats and Female B6C3F1 Mice. Fundam. Association, Washington DC. Toxicologic Pathology vol.26, No.4 (in Appl. Toxicol. 5:760–769. 3. APHA. 1996. Standard Methods for the press). 28. Kanitz, S. et al. 1996. Association Examination of Water and Wastewater, 18. Doyle T. J., Sheng W., Cerhan J. R., Hong Between Drinking Water Disinfection 19th Edition, Supplement. American C. P., Sellers T. A., Kushi, L. H., Folsom and Somatic Parameters at Birth. Public Health Association, Washington A. R. 1997. The Association of Drinking Environ. Health Perspectives, 104(5), DC. Water Source and Chlorination By- 516–520. 4. ASTM. 1993. Methods D–1067–88B, D– products with Cancer Incidence Among 29. Kaplan, L.A. 1992. Comparison of High 2035–80. Annual Book of ASTM Postmenopausal Women in Iowa: A Temperature and Persulfate Oxidation Standards. Vol. 11.01, American Society Prospective Cohort Study. American Methods for Determination of Dissolved for Testing and Materials. Journal of Public Health. 87:7. Organic Carbon in Freshwaters. Limnol. 5. ASTM. 1994. Methods D–1067–92B. 19. Edwards, M. 1997. Predicting DOC Oceanogr. 37 (5): 1119–25. Annual Book of ASTM Standards. Vol. Removal During Enhanced Coagulation. 30. King, W. D. and L. D. Marrett. 1996. Case- 11.01, American Society for Testing and Jour. AWWA (89:5:78). Control Study of Water Source and Materials. 20. Edzwald, J. K., and J. E. Van Benschoten. Bladder Cancer. Cancer Causes and 6. ASTM. 1996. Methods D–1253–86. Annual 1990. Aluminum Coagulation of Natural Control, 7:596–604. Book of ASTM Standards. Vol. 11.01, Organic Matter. Proc. Fourth Int’l 31. Klotz, J. B. and Pyrch, L. A. 1998. A Case- American Society for Testing and Gothenburg Symposium on Chemical Control Study of Neural Tube Defects Materials. Treatment, Madrid, Spain (Oct. 1990). and Drinking Water Contaminants. New 7. Bove, F.J., et al. 1995. Public Drinking Jersey Department of Health and Senior Water Contamination and Birth Services. Outcomes. Amer. J. Epidemiol., 141(9), 32. Krasner, Stuart. April 1997. Issue Paper 850–862. on Enhanced Coagulation to the M-DBP 8. Bull, R.J. and Kopfler, F.C. 1991. Health Advisory Committee. Effects of Disinfectants and Disinfection Byproducts. Prepared for the American Water Works Research Foundation. Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69461

33. Kurokawa et al. 1986a. Dose-response 46. Poole, C. 1997. Analytical Meta-Analysis 57. U.S. EPA. 1979. National Interim Primary Studies on the Carcinogenicity of of Epidemiological Studies of Drinking Water Regulations; Control of Potassium Bromate in F344 Rats after Chlorinated Drinking Water and Cancer: Trihalomethanes in Drinking Water. Fed. Long-term Oral Administration. J. Natl. Quantitative Review and Reanalysis of Reg., 44:231:68624. (November 29, 1979) Cancer Inst. 77:977–982. the Work Published by Morris et al., Am 58. U.S. EPA. 1983. EPA Method 310.1. 34. Kurokawa et al. 1986b. Long-term in vitro J Public Health 1992:82:955–963. Methods of Chemical Analysis of Water Carcinogenicity Tests of Potassium National Center for Environmental and Wastes. Envir. Monitoring Systems Bromate, Sodium Hypochlorite and Assessment, Office of Research and Laboratory, Cincinnati, OH. EPA 600/4– Sodium Chlorite Conducted in Japan. Development, September 30, 1997. 79–020. 460 pp. Environ. Health Perspect. 69:221–236. 47. Randtke, S. J.; Hoehn, R. C.; Knocke, W. 59. U.S. EPA. 1986. Guidelines for 35. McGeehin, M.A. et al. 1993. Case-Control R.; Dietrich, A. M.; Long, B. W.; and N. Carcinogen Risk Assessment, Fed. Reg. Study of Bladder Cancer and Water A. Wang. 1994. Comprehensive 51(185):33992–34003. EPA/600/8–87/ Disinfection Methods in Colorado. Am. J. Assessment of DBP Precursor Removal 045. NTIS PB88–123997. Epidemiology, 138:492–501. by Enhanced Coagulation and Softening. 60. U.S. EPA. 1987. Drinking Water 36. Mobley, S. A., D. H. Taylor, R. D. Laurie, Proc. AWWA Ann. Conf. (Water Regulations; Public Notification; Final and R. J. Pfohl. 1990. Chlorine dioxide Quality), New York, NY, pp. 737–777. Rule. Federal Register. Vol. 52, No. 208, depresses T3 uptake and delays 48. Reif, J. S. et al. 1996. Reproductive and Wednesday, Oct. 28, 1987—Part II. pp. development of locomotor activity in Developmental Effects of Disinfection 41534–41550. young rats. In: Water Chlorination: By-products in Drinking Water. 61. U.S. EPA. 1988. EPA Method 502.2. Chemistry, Environmental Impact and Environmental Health Prospectives. Methods for the Determination of Health Effects. Vol 6. Lolley, Condie, 104(10):1056–1061. Organic Compounds in Drinking Water. Johnson, Katz, Mattice and Jacobs, ed. 49. Sanders, V.M., B.M. Kauffman, K.L. EPA 600/4–88–039. PB91–231480. lewis Publ., Inc. Chelsea MI., pp 347– White, K.A. Douglas, D.W. Barnes, L.E. Revised July 1991. 360. Sain, T.J. Bradshaw, J.F. Borzelleca and 62. U.S. EPA. 1989a. National Primary 37. Morris, R. D. et al. 1992. Chlorination, A.E. Munson. 1982. Toxicology of Drinking Water Regulations; Filtration, Chlorination By-products, and Cancer: A chloral hydrate in the mouse. Environ. Disinfection; Turbidity, Giardia lamblia, Meta-Analysis. American Journal of Health Perspect. 44:137–146. Viruses, Legionella, and Heterotrophic Public Health, 82(7): 955–963. 50. Savitz, D. A., Andrews, K. W. and L. M. Bacteria; Final Rule. Part II. Fed. Reg., 38. Moser, G. September 9, 1997. Letter to Pastore. 1995. Drinking Water and 54:124:27486. (June 29, 1989) Yogi Patel, U.S. EPA, OW. Regarding the Pregnancy Outcome in Central North 63. U.S. EPA 1989b. National Primary Neurodevelopment Section of the CMA Carolina: Source, Amount, and Drinking Water Regulations; Total Study of Chlorite. Trihalomethane levels. Environ. Health Coliforms (Including Fecal Coliform and 39. NTP. 1985. National Toxicology Program. Perspectives. 103(6), 592–596. E. Coli); Final Rule. Fed. Reg., Toxicology and carcinogenesis studies of 51. Shorney, H. L., Randtke, S. J., Hargette, 54:124:27544. (June 29, 1989) chlorodibromomethane in F344/N rats P. H., Mann, P. D., Hoehn, R.C., Knocke, 64. U.S. EPA. 1989c. Review of and B6C3F1 mice (gavage studies). Tech. W. R., Dietrich, A. M. and B. W. Long. Environmental Contaminants and Rep. Ser. No. 282. 1996. The Influence of Raw Water Toxicology. USEPA. Office of Drinking 40. NTP. 1987. National Toxicology Program. Quality on Enhanced Coagulation and Water Health Advisories, Volume 106. Toxicity and carcinogenesis studies of Softening for the Removal of NOM and 225pp. bromodichloromethane in F344/N rats DBP Formation Potential, Proceedings 65. U.S. EPA. 1990. EPA Methods 551, 552. and B6C3F1 mice (gavage studies). 1996 AWWA Annual Conference, Methods for the Determination of Technical Report Series No. 321. Toronto, Ontario, Canada. Organic Compounds in Drinking Water— 41. NTP. 1989. National Toxicology Program. 52. Singer, P. C., Harrington, G. W., Supplement I. EPA 600/4–90–020. Toxicology and carcinogenesis studies of Thompson, J. D. and M. C. White. 1995. PB91–146027. bromoform in F344/N rats and B6C3F1 Enhanced Coagulation and Enhanced 66 U.S. EPA. 1991. National Primary mice (gavage studies). Tech. Rep. Ser. Softening for the Removal of Disinfection Drinking Water Regulations: Final Rule. No. 350. By-Product Precursors: An Evaluation. Fed, reg., 56:20, January 30, 1991 3526– 42. NTP. 1990. National Toxicology Program. Report prepared for the AWWA 3597. NTP Technical Report on the Toxicology Government Affairs Office, Washington, 67. U.S. EPA. 1992. EPA Methods 524.2, and Carcinogenesis studies of DC, by the Dept. of Environmental 552.1. Methods for the Determination of chlorinated and chloraminated water in Sciences and Engineering, UNC, Chapel Organic Compounds in Drinking Water— Hill, NC. F344/N rats and B6C3F1 mice (drinking Supplement II. EPA 600/R–92/129. water studies). NTP TR 392, National 53. Singer, P. C., Harrington, G. W., PB92–207703. Institutes of Health, 474pp. Thompson, J. and M. White. 1996. 68. U.S. EPA. 1993a. Draft Drinking Water 43. Orme, J. D.H. Taylor, R.D. Laurie, and R.J. Enhanced Coagulation and Enhanced Health Criteria Document for Bromate. Bull. 1985. Effects of Chlorine Dioxide Softening for the Removal of Disinfection Office of Science and Technology, Office on Thyroid Function in Neonatal Rats. J. By-Product Precursors: An Evaluation, of Water. Sep. 30, 1993. Tox. and Environ. Health. 15:315–322. Report to AWWA Disinfectants/ 69. U.S. EPA. 1993b. EPA Method 300.0. The 44. OSTP. 1985. Chemical Carcinogens; A Disinfection By-Products Technical Determination of Inorganic Anions by Review of the Science and Its Associated Advisory Workgroup of the Water Utility Ion Chromatography in the Manual Principles, February 1985. Presented in Council, December 1996. ‘‘Methods for the Determination of Risk Analysis: A guide to Principles and 54. Smith, M.K., Randall, J.L., Read, E.J., and Inorganic Substances in Environmental Methods for Analyzing Health and Stober, J.A. 1989. Teratogenic activity of Samples,’’ EPA/600/R/93/100. NTIS, Environmental Risks. Appendix G. trichloroacetic acid in the rat. Teratology PB94120821. Federal Register, March 14, 1985. Pages 40:445–451. 70. U.S. EPA. 1993c. Executive Order 12866: 10371–10442. 55. Summers, R.S., G. Solarik, V.A. Hatcher, Regulatory Planning and Review. 45. Owen, D. M.; Amy, G. L. and Z. K. R.S. Isabel, and J.F. Stile. 1997. Federal Register. Vol. 58, No. 190. Chowdhury. 1993. Characterization of Analyzing the Impacts of Predisinfection October 4, 1993. 51735–51744. Natural Organic Matter and Its Through Jar Testing, Proceedings, 71. U.S. EPA/ILSI. 1993. A Review of Relationship to Treatability. AWWA AWWA Water Quality Technology Evidence on Reproductive and Research Foundation & AWWA, Denver, Conference, Denver, CO. Developmental Effects of Disinfection CO. 56. Tseng, T. and M. Edwards. 1997. By-Products in Drinking Water. Considerations in Optimizing Washington: U.S. Environmental Coagulation. Proc. 1996 AWWA Water Protection Agency and International Life Qual. Technol. Conf., Boston, Mass. Sciences Institute. 69462 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

72. U.S. EPA. 1994a. National Primary 88. U.S. EPA. 1997e. Method 300.1. 101. U.S. EPA. 1998k. Cost and Technology Drinking Water Regulations; Determination of Inorganic Anions in Document for Controlling Disinfectants Disinfectants and Disinfection Drinking Water by Ion Chromatography. and Disinfection Byproducts. Office of Byproducts; Proposed Rule. Fed. Reg., Revision 1.0. USEPA National Exposure Ground Water and Drinking Water. 59:145:38668. (July 29, 1994). Research Laboratory, Cincinnati OH. Washington, DC. EPA 815–R–98–014. PB 73. U.S. EPA. 1994b. National Primary 89. U.S. EPA. 1997f. Performance Based 99–111486. Drinking Water Regulations; Enhanced Measurement System. Notice of Intent. 102. U.S. EPA. 1998l. Synthesis of the Peer- Surface Water Treatment Requirements; Federal Register, October 6, 1997. Vol. Review of Meta-analysis of Proposed Rule. Fed. Reg., 59:145:38832. 62, No. 193., 52098–52100. Epidemiologic Data on Risks of Cancer (July 29, 1994). 90. U.S. EPA. 1997g. Manual for the from Chlorinated Drinking Water. 74. U.S. EPA. 1994c. National Primary Certification of Laboratories Analyzing National Center for Environmental Drinking Water Regulations; Monitoring Drinking Water, Fourth Edition, Office of Assessment, Office of Research and Requirements for Public Drinking Water Water Resource Center (RC–4100), EPA Development, February 16, 1998. Supplies; Proposed Rule. Fed. Reg., 815–B–97–001. March 1997. 103. U.S. EPA. 1998m. NCEA Position Paper 59:28:6332. (February 10, 1994). 91. U.S. EPA. 1998a. National Primary Regarding Risk Assessment Use of the 75. U.S. EPA. 1994d. Final Draft Drinking Drinking Water Regulations; Results from the Published Study: Morris Water Health Criteria Document for Disinfectants and Disinfection et al. Am J Public Health 1992;82:955– Chlorine Dioxide, Chlorite and Chlorate. Byproducts; Notice of Data Availability; 963. National Center for Environmental Office of Science and Technology, Office Proposed Rule. Fed. Reg., 63 (No. 61): Assessment, Office of Research and of Water. March 31, 1994. 15606–15692. (March 31, 1998). Development, October 7, 1997. 76. U.S. EPA. 1994e. Draft Drinking Water 92. U.S. EPA. 1998b. Dichloroacetic acid: 104. U.S. EPA. 1998n. A Suggested Approach Health Criteria Document for Chlorine, Carcinogenicity Identification for Using the Current Epidemiologic Hypochlorous Acid and Hypochlorite Characterization Summary. National Literature to Estimate the Possible Ion. Office of Science and Technology, Center for Environmental Assessment— Cancer Risk from Water Chlorination, for Office of Water. Washington Office. Office of Research the Purposes of the Regulatory Impact 77. U.S. EPA. 1994f. Health and Ecological and Development. March 1998. EPA Analysis. ORD, National Center for Criteria Div., OST. Final Draft for the 815–B–98–010. PB 99–111387. Environmental Assessment. August 27, Drinking Water Criteria Document on 93. U.S. EPA. 1998c. Quantification of 1998. Trihalomethanes. Apr. 8. 1994. Bladder Cancer Risk from Exposure to 105. U.S. EPA. 1998o. Unfunded Mandates 78. U.S. EPA. 1994g. Draft Drinking Water Chlorinated Surface Water. Office of Reform Act Analysis for the Stage 1 Health Criteria Document for Science and Technology, Office of Water. Disinfectant and Disinfection Byproduct Chlorinated Acetic Acids/Alcohols/ November 9, 1998. Rule. Office of Groundwater and Aldehydes and Ketones. Office of 94. U.S. EPA. 1998d. Health Risk Drinking Water. Science and Technology, Office of Water. Assessment/Characterization of the 106. U.S. EPA. 1998p. Health Risk 79. U.S. EPA. 1994h. Draft Drinking Water Drinking Water Disinfection Byproduct Assessment/Characterization of the Health Criteria Document for Chlorine Dioxide and the Degradation Drinking Water Disinfection Byproduct Chloramines. Office of Science and Byproduct Chlorite. Office of Science Chloroform. Office of Science and Technology, Office of Water. and Technology, Office of Water. Technology, Office of Water. November 80. U.S. Environmental Protection Agency. October 15, 1998. EPA 815–B–98–008. 4, 1998. EPA 815–B–98–006. PB 99– 1994i. Regulatory Impact Analysis of PB 99–111361. 111346. Proposed Disinfectant/Disinfection 95. U.S. EPA. 1998e. Health Risk 107. U.S. EPA. 1998q. Small System Byproduct Regulations. Washington, DC. Assessment/Characterization of the Compliance Technology List for the EPA–68–C3–0368. Drinking Water Disinfection Byproduct Stage 1 DBP Rule. Office of Groundwater 81. U.S. EPA. 1995. Methods for the Bromate. Office of Science and and Drinking Water. EPA 815–R–98–017. Determination of Organic Compounds in Technology, Office of Water. September PB 99–111510. Drinking Water. Supplement III. EPA– 30, 1998. EPA 815–B–98–007. PB 99– 108. U.S. EPA. 1998r. Technologies and Costs 600/R–95/131. NTIS, PB95261616. 111353. for Point-of-Entry (POE) and Point-of-Use 82. U.S. EPA. 1996a. National Primary 96. U.S. EPA. 1998f. Panel Report and (POU) Devices for Control of Disinfection Drinking Water Regulations: Monitoring Recommendation for Conducting Byproducts. Office of Groundwater and Requirements for Public Drinking Water Epidemiological Research on Possible Drinking Water. EPA 815–R–98–016. PB Supplies; Final Rule. Fed. Reg., Reproductive and Developmental Effects 99–111502. 61:94:24354. (May 14, 1996) of Exposure to Disinfected Drinking 109. U.S. EPA. 1998s. National-Level 83. U.S. EPA. 1996b. Proposed Guidelines for Water. Office of Research and Affordability Criteria Under the 1996 Carcinogen Risk Assessment. U.S. EPA, Development. February 12, 1998. Amendments to the Safe Drinking Water April 23, 1996. 97. U.S. EPA. 1998g. Regulatory Impact Act. Office of Groundwater and Drinking 84. U.S. EPA. 1997a. National Primary Analysis of Final Disinfectant/ Water. August 19, 1998. Drinking Water Regulations; Interim Disinfection By-Products Regulations. 110. U.S. EPA. 1998t. Variance Technology Enhanced Surface Water Treatment Rule; Washington, D.C. EPA Number 815–B– Findings for Contaminants Regulated Notice of Data Availability; Proposed 98–002. PB 99–111304. Before 1996. Office of Water. September Rule. Fed. Reg., 62 (No. 212): 59486– 98. U.S. EPA. 1998h. Health Risks to Fetuses, 1998. EPA 815–R–98–003. 59557. (November 3, 1997). Infants, and Children (final Stage 1 DBP 111. U.S. EPA. 1998u. Occurrence 85. U.S. EPA. 1997b. National Primary Rule). Office of Science and Technology. Assessment for Disinfectants and Drinking Water Regulations; Office of Water. November 19, 1998. EPA Disinfection Byproducts in Public Disinfectants and Disinfection 815–B–98–009. PB 99–111379. Drinking Water Supplies. Office of Byproducts; Notice of Data Availability; 99. U.S. EPA. 1998i. Revisions to State Groundwater and Drinking Water. EPA Proposed Rule. Fed. Reg., 62 (No. 212): Primacy Requirements To Implement 815–B–98–004. November 13, 1998. PB 59388–59484. (November 3, 1997). Safe Drinking Water Act Amendments: 99–111320. 86. U.S. EPA. 1997c. Summaries of New Final Rule. Federal Register, Tuesday, 112. USGS. 1989. Method I–1030–85. Health Effects Data. Office of Science April 28, 1998, Vol. 63, No.81, 23362– Techniques of Water Resources and Technology, Office of Water. 23368. Investigations of the U.S. Geological October 1997. 100. U.S. EPA. 1998j. Revision of Existing Survey. Book 5, Chapter A–1, 3rd ed., 87. U.S. EPA. 1997d. External Peer Review of Variance and Exemption Regulations to U.S. Government Printing Office. CMA Study –2– Generation, EPA Comply with Requirements of the Safe 113. Waller K., Swan S. H., DeLorenze G., Contract No. 68–C7–0002, Work Drinking Water Act; Final Rule. Federal Hopkins B., 1998. Trihalomethanes in Assignment B–14, The Cadmus Group, Register, Vol 63, No. 157. Friday, Aug. drinking water and spontaneous Inc., October 9, 1997. 14, 1998. pp. 43833–43851. abortion. Epidemiology. 9(2):134–140. Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69463

114. White, M. C., Thompson, D., Harrington, PART 141ÐNATIONAL PRIMARY chlorine dioxide) in the distribution G. W., and P.S. Singer. 1997. Evaluating DRINKING WATER REGULATIONS system to a level and for a time Criteria for Enhanced Coagulation necessary to protect public health to Compliance. AWWA, 89:5:64. 3. The authority citation for part 141 address specific microbiological 115. Xie, Yuefeng. 1995. Effects of Sodium continues to read as follows: Chloride on DBP Analytical Results, contamination problems caused by Extended Abstract, Division of Authority: 42 U.S.C. 300f, 300g–1, 300g–2, circumstances such as distribution line Environmental Chemistry, American 300g–3, 300g–4, 300g–5, 300g–6, 300j–4, breaks, storm runoff events, source 300j–9, and 300j–11. Chemical Society Annual Conference, water contamination, or cross- Chicago, IL, Aug. 21–26, 1995. 4. Section 141.2 is amended by connections. adding the following definitions in List of Subjects * * * * * alphabetical order to read as follows: Maximum residual disinfectant level 40 CFR Part 9 goal (MRDLG) means the maximum § 141.2 Definitions. level of a disinfectant added for water Environmental protection, Reporting * * * * * treatment at which no known or and recordkeeping requirements. Enhanced coagulation means the anticipated adverse effect on the health 40 CFR Parts 141 and 142 addition of sufficient coagulant for of persons would occur, and which improved removal of disinfection allows an adequate margin of safety. Analytical methods, Drinking water, byproduct precursors by conventional MRDLGs are nonenforceable health Environmental protection, Incorporation filtration treatment. goals and do not reflect the benefit of by reference, Intergovernmental * * * * * the addition of the chemical for control relations, Public utilities, Reporting and Enhanced softening means the of waterborne microbial contaminants. recordkeeping requirements, Utilities, improved removal of disinfection * * * * * Water supply. byproduct precursors by precipitative Subpart H systems means public Dated: November 30, 1998. softening. water systems using surface water or Carol M. Browner, * * * * * ground water under the direct influence Administrator. GAC10 means granular activated of surface water as a source that are carbon filter beds with an empty-bed subject to the requirements of subpart H For the reasons set out in the contact time of 10 minutes based on of this part. preamble, title 40, chapter I of the Code average daily flow and a carbon * * * * * of Federal Regulations is amended as reactivation frequency of every 180 SUVA means Specific Ultraviolet follows: days. Absorption at 254 nanometers (nm), an * * * * * PART 9Ð[AMENDED] indicator of the humic content of water. Haloacetic acids (five) (HAA5) mean It is a calculated parameter obtained by the sum of the concentrations in 1. The authority citation for part 9 dividing a sample’s ultraviolet milligrams per liter of the haloacetic continues to read as follows: absorption at a wavelength of 254 nm =1 acid compounds (monochloroacetic (UV 254) (in m ) by its concentration of Authority: 7 U.S.C. 135 et seq., 136–136y; acid, dichloroacetic acid, trichloroacetic 15 U.S.C. 2001, 2003, 2005, 2006, 2601–2671; dissolved organic carbon (DOC) (in mg/ acid, monobromoacetic acid, and L). 21 U.S.C. 331j, 346a, 348; 31 U.S.C. 9701; 33 dibromoacetic acid), rounded to two U.S.C. 1251 et seq., 1311, 1313d, 1314, 1318, * * * * * 1321, 1326, 1330, 1342, 1344, 1345 (d) and significant figures after addition. Total Organic Carbon (TOC) means (e), 1361; E.O. 11735, 38 FR 21243, 3 CFR, * * * * * total organic carbon in mg/L measured 1971–1975 Comp. p. 973; 42 U.S.C. 241, Maximum residual disinfectant level using heat, oxygen, ultraviolet 242b, 243, 246, 300f, 300g, 300g–1, 300g–2, (MRDL) means a level of a disinfectant irradiation, chemical oxidants, or 300g–3, 300g–4, 300g–5, 300g–6, 300j–1, added for water treatment that may not combinations of these oxidants that 300j–2, 300j–3, 300j–4, 300j–9, 1857 et seq., be exceeded at the consumer’s tap convert organic carbon to carbon 6901–6992k, 7401–7671q, 7542, 9601–9657, without an unacceptable possibility of 11023, 11048. dioxide, rounded to two significant adverse health effects. For chlorine and figures. chloramines, a PWS is in compliance 2. In § 9.1 the table is amended by * * * * * adding under the indicated heading: the with the MRDL when the running 5. Section 141.12 is revised to read as new entries in numerical order to read annual average of monthly averages of follows: as follows: samples taken in the distribution system, computed quarterly, is less than § 141.12 Maximum contaminant levels for § 9.1 OMB approvals under the Paperwork or equal to the MRDL. For chlorine total trihalomethanes. Reduction Act. dioxide, a PWS is in compliance with The maximum contaminant level of * * * * * the MRDL when daily samples are taken 0.10 mg/L for total trihalomethanes (the at the entrance to the distribution sum of the concentrations of OMB con- system and no two consecutive daily bromodichloromethane, 40 CFR citation trol No. samples exceed the MRDL. MRDLs are dibromochloromethane, enforceable in the same manner as tribromomethane (bromoform), and ***** maximum contaminant levels under trichloromethane (chloroform)) applies National Primary Drinking Section 1412 of the Safe Drinking Water to subpart H community water systems Water Regulations Act. There is convincing evidence that which serve a population of 10,000 addition of a disinfectant is necessary people or more until December 16, ***** for control of waterborne microbial 2001. This level applies to community 141.130±141.132 ...... 2040±0204 contaminants. Notwithstanding the water systems that use only ground 141.134±141.135 ...... 2040±0204 MRDLs listed in § 141.65, operators may water not under the direct influence of ***** increase residual disinfectant levels of surface water and serve a population of chlorine or chloramines (but not 10,000 people or more until December 69464 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

16, 2003. Compliance with the public water system as soon as possible has determined that chlorine dioxide is maximum contaminant level for total but in no case later than 72 hours after a health concern at certain levels of trihalomethanes is calculated pursuant the violation. *** exposure. Chlorine dioxide is used in to § 141.30. After December 16, 2003, * * * * * water treatment to kill bacteria and this section is no longer applicable. (E) Violation of the MRDL for chlorine other disease-causing microorganisms 6. Section 141.30 is amended by dioxide as defined in § 141.65 and and can be used to control tastes and revising the the first sentences in determined according to § 141.133(c)(2). odors. Disinfection is required for paragraphs (d) and (f) and adding * * * * * surface water systems. However, at high paragraph (h) to read as follows: (c) * * * The owner or operator of a doses, chlorine dioxide-treated drinking water has been shown to affect blood in § 141.30 Total trihalomethanes sampling, community water system must give a analytical and other requirements. copy of the most recent public notice for laboratory animals. Also, high levels of any outstanding violation of any chlorine dioxide given to laboratory * * * * * animals in drinking water have been (d) Compliance with § 141.12 shall be maximum contaminant level, or any shown to cause neurological effects on determined based on a running annual maximum residual disinfectant level, or the developing nervous system. These average of quarterly samples collected any treatment technique requirement, or neurodevelopmental effects may occur by the system as prescribed in any variance or exemption schedule to as a result of a short-term excessive paragraph (b)(1) or (2) of this all new billing units or new hookups chlorine dioxide exposure. To protect section. * * * prior to or at the time service begins. against such potentially harmful * * * * * * * * * * (e) * * * exposures, EPA requires chlorine (f) Before a community water system dioxide monitoring at the treatment makes any significant modifications to (76) Chlorine. The United States Environmental Protection Agency (EPA) plant, where disinfection occurs, and at its existing treatment process for the representative points in the distribution purposes of achieving compliance with sets drinking water standards and has determined that chlorine is a health system serving water users. EPA has set § 141.12, such system must submit and a drinking water standard for chlorine obtain State approval of a detailed plan concern at certain levels of exposure. Chlorine is added to drinking water as dioxide to protect against the risk of setting forth its proposed modification these adverse effects. and those safeguards that it will a disinfectant to kill bacteria and other implement to ensure that the disease-causing microorganisms and is Note: In addition to the language in this bacteriological quality of the drinking also added to provide continuous introductory text of paragraph (e)(78), systems must include either the language in water served by such system will not be disinfection throughout the distribution system. Disinfection is required for paragraph (e)(78)(i) or (e)(78)(ii) of this adversely affected by such section. Systems with a violation at the modification. * ** surface water systems. However, at high treatment plant, but not in the distribution doses for extended periods of time, * * * * * system, are required to use the language in (h) The requirements in paragraphs (a) chlorine has been shown to affect blood paragraph (e)(78)(i) of this section and treat through (g) of this section apply to and the liver in laboratory animals. EPA the violation as a nonacute violation. subpart H community water systems has set a drinking water standard for Systems with a violation in the distribution system are required to use the language in which serve a population of 10,000 or chlorine to protect against the risk of these adverse effects. Drinking water paragraph (e)(78)(ii) of this section and treat more until December 16, 2001. The the violation as an acute violation. requirements in paragraphs (a) through which meets this EPA standard is (i) The chlorine dioxide violations (g) of this section apply to community associated with little to none of this risk reported today are the result of water systems which use only ground and should be considered safe with exceedances at the treatment facility water not under the direct influence of respect to chlorine. (77) Chloramines. The United States only, and do not include violations surface water that add a disinfectant Environmental Protection Agency (EPA) within the distribution system serving (oxidant) in any part of the treatment sets drinking water standards and has users of this water supply. Continued process and serve a population of determined that chloramines are a compliance with chlorine dioxide levels 10,000 or more until December 16, 2003. health concern at certain levels of within the distribution system After December 16, 2003, this section is exposure. Chloramines are added to minimizes the potential risk of these no longer applicable. 7. Section 141.32 is amended by drinking water as a disinfectant to kill violations to present consumers. (ii) The chlorine dioxide violations revising the heading in paragraph (a) bacteria and other disease-causing reported today include exceedances of introductory text, the first sentence of microorganisms and are also added to the EPA standard within the paragraph (a)(1)(iii) introductory text, provide continuous disinfection distribution system serving water users. and the first sentence of paragraph (c), throughout the distribution system. Violations of the chlorine dioxide and adding paragraphs (a)(1)(iii)(E) and Disinfection is required for surface standard within the distribution system (e) (76) through (81), to read as follows: water systems. However, at high doses for extended periods of time, may harm human health based on short- § 141.32 Public notification. chloramines have been shown to affect term exposures. Certain groups, * * * * * blood and the liver in laboratory including pregnant women, infants, and (a) Maximum contaminant levels animals. EPA has set a drinking water young children, may be especially (MCLs), maximum residual disinfectant standard for chloramines to protect susceptible to adverse effects of levels (MRDLs). * ** against the risk of these adverse effects. excessive exposure to chlorine dioxide- (1) * * * Drinking water which meets this EPA treated water. The purpose of this notice (iii) For violations of the MCLs of standard is associated with little to none is to advise that such persons should contaminants or MRDLs of disinfectants of this risk and should be considered consider reducing their risk of adverse that may pose an acute risk to human safe with respect to chloramines. effects from these chlorine dioxide health, by furnishing a copy of the (78) Chlorine dioxide. The United violations by seeking alternate sources notice to the radio and television States Environmental Protection Agency of water for human consumption until stations serving the area served by the (EPA) sets drinking water standards and such exceedances are rectified. Local Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69465 and State health authorities are the best § 141.53ÐMaximum contaminant level goals system must take. Failure to meet the sources for information concerning for disinfection byproducts. schedule or interim treatment alternate drinking water. MCLGs for the following disinfection requirements constitutes a violation of a (79) Disinfection byproducts and byproducts are as indicated: National Primary Drinking Water Regulation. treatment technique for DBPs. The MCLG United States Environmental Protection Disinfection byproduct (mg/L) (c) The Administrator, pursuant to Agency (EPA) sets drinking water Section 1412 of the Act, hereby standards and requires the disinfection Chloroform ...... Zero identifies the following as the best of drinking water. However, when used Bromodichloromethane ...... Zero technology, treatment techniques, or in the treatment of drinking water, Bromoform ...... Zero other means available for achieving disinfectants react with naturally- Bromate ...... Zero compliance with the maximum Dichloroacetic acid ...... Zero contaminant levels for disinfection occurring organic and inorganic matter Trichloroacetic acid ...... 0.3 present in water to form chemicals byproducts identified in paragraph (a) of Chlorite ...... 0.8 this section: called disinfection byproducts (DBPs). Dibromochloromethane ...... 0.06 EPA has determined that a number of Disinfec- DBPs are a health concern at certain § 141.54 Maximum residual disinfectant tion by- Best available technology levels of exposure. Certain DBPs, level goals for disinfectants. product including some trihalomethanes (THMs) MRDLGs for disinfectants are as TTHM ... Enhanced coagulation or en- and some haloacetic acids (HAAs), have follows: been shown to cause cancer in hanced softening or GAC10, with chlorine as the primary and laboratory animals. Other DBPs have Disinfectant residual MRDLG(mg/L) been shown to affect the liver and the residual disinfectant HAA5 .... Enhanced coagulation or en- nervous system, and cause reproductive Chlorine ...... 4 (as Cl ). 2 hanced softening or GAC10, Chloramines ...... 4 (as Cl ). or developmental effects in laboratory 2 with chlorine as the primary and Chlorine dioxide ...... 0.8 (as ClO ) animals. Exposure to certain DBPs may 2 residual disinfectant. produce similar effects in people. EPA Bromate Control of ozone treatment proc- has set standards to limit exposure to 9. Subpart G is amended by revising ess to reduce production of bro- THMs, HAAs, and other DBPs. the subpart heading and adding mate. §§ 141.64 and 141.65 to read as follows: Chlorite Control of treatment processes to (80) Bromate. The United States reduce disinfectant demand and Environmental Protection Agency (EPA) Subpart GÐNational Revised Primary control of disinfection treatment sets drinking water standards and has Drinking Water Regulations: Maximum processes to reduce disinfectant determined that bromate is a health Contaminant Levels and Maximum levels. concern at certain levels of exposure. Residual Disinfectant Levels Bromate is formed as a byproduct of § 141.65 Maximum residual disinfectant ozone disinfection of drinking water. * * * * * levels. Ozone reacts with naturally occurring § 141.64 Maximum contaminant levels for bromide in the water to form bromate. (a) Maximum residual disinfectant disinfection byproducts. levels (MRDLs) are as follows: Bromate has been shown to produce (a) The maximum contaminant levels cancer in rats. EPA has set a drinking (MCLs) for disinfection byproducts are Disinfectant residual MRDL (mg/L) water standard to limit exposure to as follows: bromate. Chlorine ...... 4.0 (as Cl2). Chloramines ...... 4.0 (as Cl2). (81) Chlorite. The United States Disinfection byproduct MCL Environmental Protection Agency (EPA) (mg/L) Chlorine dioxide ...... 0.8 (as ClO2). sets drinking water standards and has Total trihalomethanes (TTHM) ...... 0.080 (b) Compliance dates. determined that chlorite is a health Haloacetic acids (five) (HAA5) ...... 0.060 concern at certain levels of exposure. Bromate ...... 0.010 (1) CWSs and NTNCWSs. Subpart H Chlorite is formed from the breakdown Chlorite ...... 1.0 systems serving 10,000 or more persons of chlorine dioxide, a drinking water must comply with this section disinfectant. Chlorite in drinking water (b) Compliance dates. (1) CWSs and beginning December 16, 2001. Subpart has been shown to affect blood and the NTNCWSs. Subpart H systems serving H systems serving fewer than 10,000 developing nervous system. EPA has set 10,000 or more persons must comply persons and systems using only ground a drinking water standard for chlorite to with this section beginning December water not under the direct influence of protect against these effects. Drinking 16, 2001. Subpart H systems serving surface water must comply with this water which meets this standard is fewer than 10,000 persons and systems subpart beginning December 16, 2003. associated with little to none of these using only ground water not under the (2) Transient NCWSs. Subpart H risks and should be considered safe direct influence of surface water must systems serving 10,000 or more persons with respect to chlorite. comply with this section beginning and using chlorine dioxide as a * * * * * December 16, 2003. disinfectant or oxidant must comply (2) A system that is installing GAC or with the chlorine dioxide MRDL 8. Subpart F is amended by revising membrane technology to comply with beginning December 16, 2001. Subpart the subpart heading and adding this section may apply to the State for H systems serving fewer than 10,000 §§ 141.53 and 141.54 to read as follows: an extension of up to 24 months past the persons and using chlorine dioxide as a Subpart FÐMaximum Contaminant dates in paragraphs (b)(1) of this section, disinfectant or oxidant and systems Level Goals and Maximum Residual but not beyond December 16, 2003. In using only ground water not under the Disinfectant Level Goals granting the extension, States must set direct influence of surface water and a schedule for compliance and may using chlorine dioxide as a disinfectant * * * * * specify any interim measures that the or oxidant must comply with the 69466 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations chlorine dioxide MRDL beginning requirements of this subpart as follows. Methods for the Determination of December 16, 2003. Subpart H systems serving 10,000 or Organic Compounds in Drinking Water- (c) The Administrator, pursuant to more persons must comply with this Supplement II, USEPA, August 1992, Section 1412 of the Act, hereby subpart beginning December 16, 2001. EPA/600/R–92/129 (available through identifies the following as the best Subpart H systems serving fewer than National Information Technical Service technology, treatment techniques, or 10,000 persons and systems using only (NTIS), PB92–207703). EPA Methods other means available for achieving ground water not under the direct 502.2, 524.2, 551.1, and 552.2 are in compliance with the maximum residual influence of surface water must comply Methods for the Determination of disinfectant levels identified in with this subpart beginning December Organic Compounds in Drinking Water- 16, 2003. paragraph (a) of this section: control of Supplement III, USEPA, August 1995, (2) Transient NCWSs. Subpart H treatment processes to reduce EPA/600/R–95/131. (available through systems serving 10,000 or more persons disinfectant demand and control of NTIS, PB95–261616). EPA Method disinfection treatment processes to and using chlorine dioxide as a 300.0 is in Methods for the reduce disinfectant levels. disinfectant or oxidant must comply Determination of Inorganic Substances 10. A new subpart L is added to read with any requirements for chlorine as follows: dioxide and chlorite in this subpart in Environmental Samples, USEPA, beginning December 16, 2001. Subpart August 1993, EPA/600/R–93/100. Subpart LÐDisinfectant Residuals, H systems serving fewer than 10,000 (available through NTIS, PB94–121811). Disinfection Byproducts, and persons and using chlorine dioxide as a EPA Method 300.1 is titled USEPA Disinfection Byproduct Precursors disinfectant or oxidant and systems Method 300.1, Determination of using only ground water not under the Inorganic Anions in Drinking Water by Sec. direct influence of surface water and Ion Chromatography, Revision 1.0, 141.130 General requirements. using chlorine dioxide as a disinfectant 141.131 Analytical requirements. USEPA, 1997, EPA/600/R–98/118 141.132 Monitoring requirements. or oxidant must comply with any (available through NTIS, PB98-169196); 141.133 Compliance requirements. requirements for chlorine dioxide and also available from: Chemical Exposure 141.134 Reporting and recordkeeping chlorite in this subpart beginning Research Branch, Microbiological & requirements. December 16, 2003. Chemical Exposure Assessment 141.135 Treatment technique for control of (c) Each CWS and NTNCWS regulated Research Division, National Exposure disinfection byproduct (DBP) precursors. under paragraph (a) of this section must Research Laboratory, U.S. be operated by qualified personnel who § 141.130 General requirements. Environmental Protection Agency, meet the requirements specified by the Cincinnati, OH 45268, Fax Number: (a) The requirements of this subpart L State and are included in a State register constitute national primary drinking 513–569–7757, Phone number: 513– of qualified operators. 569–7586. Standard Methods 4500-Cl D, water regulations. (d) Control of disinfectant residuals. 4500-Cl E, 4500-Cl F, 4500-Cl G, 4500- (1) The regulations in this subpart Notwithstanding the MRDLs in § 141.65, Cl H, 4500-Cl I, 4500-ClO2 D, 4500-ClO2 establish criteria under which systems may increase residual E, 6251 B, and 5910 B shall be followed community water systems (CWSs) and disinfectant levels in the distribution in accordance with Standard Methods nontransient, noncommunity water system of chlorine or chloramines (but systems (NTNCWSs) which add a not chlorine dioxide) to a level and for for the Examination of Water and chemical disinfectant to the water in a time necessary to protect public Wastewater, 19th Edition, American any part of the drinking water treatment health, to address specific Public Health Association, 1995; copies process must modify their practices to microbiological contamination problems may be obtained from the American meet MCLs and MRDLs in §§ 141.64 and caused by circumstances such as, but Public Health Association, 1015 141.65, respectively, and must meet the not limited to, distribution line breaks, Fifteenth Street, NW, Washington, DC treatment technique requirements for storm run-off events, source water 20005. Standard Methods 5310 B, 5310 disinfection byproduct precursors in contamination events, or cross- C, and 5310 D shall be followed in § 141.135. connection events. accordance with the Supplement to the (2) The regulations in this subpart 19th Edition of Standard Methods for establish criteria under which transient § 141.131 Analytical requirements. the Examination of Water and NCWSs that use chlorine dioxide as a (a) General. (1) Systems must use only Wastewater, American Public Health disinfectant or oxidant must modify the analytical method(s) specified in Association, 1996; copies may be their practices to meet the MRDL for this section, or otherwise approved by obtained from the American Public chlorine dioxide in § 141.65. EPA for monitoring under this subpart, Health Association, 1015 Fifteenth (3) EPA has established MCLs for to demonstrate compliance with the Street, NW, Washington, DC 20005. TTHM and HAA5 and treatment requirements of this subpart. These ASTM Method D 1253–86 shall be technique requirements for disinfection methods are effective for compliance followed in accordance with the Annual byproduct precursors to limit the levels monitoring February 16, 1999. Book of ASTM Standards, Volume of known and unknown disinfection (2) The following documents are 11.01, American Society for Testing and byproducts which may have adverse incorporated by reference. The Director Materials, 1996 edition; copies may be health effects. These disinfection of the Federal Register approves this obtained from the American Society for byproducts may include chloroform; incorporation by reference in Testing and Materials, 100 Barr Harbor bromodichloromethane; accordance with 5 U.S.C. 552(a) and 1 Drive, West Conshohoken, PA 19428. dibromochloromethane; bromoform; CFR part 51. Copies may be inspected dichloroacetic acid; and trichloroacetic at EPA’s Drinking Water Docket, 401 M (b) Disinfection byproducts. (1) acid. Street, SW, Washington, DC 20460, or at Systems must measure disinfection (b) Compliance dates. (1) CWSs and the Office of the Federal Register, 800 byproducts by the methods (as modified NTNCWSs. Unless otherwise noted, North Capitol Street, NW, Suite 700, by the footnotes) listed in the following systems must comply with the Washington DC. EPA Method 552.1 is in table: Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69467

APPROVED METHODS FOR DISINFECTION BYPRODUCT COMPLIANCE MONITORING

Byproduct measured 1 2 EPA meth- Methodology od Standard method TTHM HAA5 Chlorite 4 Bromate

P&T/GC/ElCD & PID ...... 3502.2 X P&T/GC/MS ...... 524.2 X LLE/GC/ECD ...... 551.1 X LLE/GC/ECD ...... 6251 B X SPE/GC/ECD ...... 552.1 X LLE/GC/ECD ...... 552.2 X Amperometric Titration ...... 4500-ClO2 EX IC ...... 300.0 X IC ...... 300.1 XX 1 X indicates method is approved for measuring specified disinfection byproduct. 2 P&T = purge and trap; GC = gas chromatography; ElCD = electrolytic conductivity detector; PID = photoionization detector; MS = mass spec- trometer; LLE = liquid/liquid extraction; ECD = electron capture detector; SPE = solid phase extractor; IC = ion chromatography. 3 If TTHMs are the only analytes being measured in the sample, then a PID is not required. 4 Amperometric titration may be used for routine daily monitoring of chlorite at the entrance to the distribution system, as prescribed in § 141.132(b)(2)(i)(A). Ion chromatography must be used for routine monthly monitoring of chlorite and additional monitoring of chlorite in the dis- tribution system, as prescribed in § 141.132(b)(2)(i)(B) and (b)(2)(ii).

(2) Analysis under this section for EPA or the State. In these analyses of PE minimum acceptance limit of +/¥50% disinfection byproducts must be samples, the laboratory must achieve and +/¥15% of the study mean. conducted by laboratories that have quantitative results within the (c) Disinfectant residuals. (1) Systems received certification by EPA or the acceptance limit on a minimum of 80% must measure residual disinfectant State. To receive certification to conduct of the analytes included in each PE concentrations for free chlorine, analyses for the contaminants in sample. The acceptance limit is defined combined chlorine (chloramines), and § 141.64(a), the laboratory must carry as the 95% confidence interval out annual analyses of performance calculated around the mean of the PE chlorine dioxide by the methods listed evaluation (PE) samples approved by study data between a maximum and in the following table:

APPROVED METHODS FOR DISINFECTANT RESIDUAL COMPLIANCE MONITORING

Residual Measured 1 Standard Methodology method ASTM method Free Combined Total Chlorine chlorine chlorine chlorine dioxide

Amperometric Titration ...... 4500-Cl D D 1253±86 XXX Low Level Amperometric Titration ...... 4500-Cl E X DPD Ferrous Titrimetric ...... 4500-Cl F XXX DPD Colorimetric ...... 4500-Cl G XXX Syringaldazin e (FACTS) ...... 4500-Cl H X Iodometric Electrode ...... 4500-Cl I X DPD ...... 4500-ClO2 D X Amperometric Method II ...... 4500-ClO2 E X 1 X indicates method is approved for measuring specified disinfectant residual.

(2) If approved by the State, systems (3) Total Organic Carbon (TOC). as mg/L). In order to determine SUVA, may also measure residual disinfectant Standard Method 5310 B (High- it is necessary to separately measure concentrations for chlorine, Temperature Combustion Method) or UV254 and DOC. When determining chloramines, and chlorine dioxide by Standard Method 5310 C (Persulfate- SUVA, systems must use the methods using DPD colorimetric test kits. Ultraviolet or Heated-Persulfate stipulated in paragraph (d)(4)(i) of this (3) A party approved by EPA or the Oxidation Method) or Standard Method section to measure DOC and the method State must measure residual disinfectant 5310 D (Wet-Oxidation Method). TOC stipulated in paragraph (d)(4)(ii) of this concentration. samples may not be filtered prior to section to measure UV254. SUVA must analysis. TOC samples must either be be determined on water prior to the (d) Additional analytical methods. analyzed or must be acidified to achieve addition of disinfectants/oxidants by the Systems required to analyze parameters pH less than 2.0 by minimal addition of system. DOC and UV254 samples used to not included in paragraphs (b) and (c) phosphoric or sulfuric acid as soon as determine a SUVA value must be taken of this section must use the following practical after sampling, not to exceed at the same time and at the same methods. A party approved by EPA or 24 hours. Acidified TOC samples must location. the State must measure these be analyzed within 28 days. parameters. (i) Dissolved Organic Carbon (DOC). (4) Specific Ultraviolet Absorbance Standard Method 5310 B (High- (1) Alkalinity. All methods allowed in (SUVA). SUVA is equal to the UV Temperature Combustion Method) or § 141.89(a) for measuring alkalinity. absorption at 254nm (UV254) (measured Standard Method 5310 C (Persulfate- (2) Bromide. EPA Method 300.0 or in m-1 divided by the dissolved organic Ultraviolet or Heated-Persulfate EPA Method 300.1. carbon (DOC) concentration (measured Oxidation Method) or Standard Method 69468 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

5310 D (Wet-Oxidation Method). Prior must be measured at 253.7 nm (may be (3) Failure to monitor in accordance to analysis, DOC samples must be rounded off to 254 nm). Prior to with the monitoring plan required filtered through a 0.45 µm pore-diameter analysis, UV254 samples must be filtered under paragraph (f) of this section is a filter. Water passed through the filter through a 0.45 µm pore-diameter filter. monitoring violation. prior to filtration of the sample must The pH of UV254 samples may not be (4) Failure to monitor will be treated serve as the filtered blank. This filtered adjusted. Samples must be analyzed as as a violation for the entire period blank must be analyzed using soon as practical after sampling, not to covered by the annual average where procedures identical to those used for exceed 48 hours. compliance is based on a running analysis of the samples and must meet (5) pH. All methods allowed in annual average of monthly or quarterly the following criteria: DOC < 0.5 mg/L. § 141.23(k)(1) for measuring pH. samples or averages and the system’s DOC samples must be filtered through µ § 141.132 Monitoring requirements. failure to monitor makes it impossible to the 0.45 m pore-diameter filter prior to determine compliance with MCLs or (a) General requirements. (1) Systems acidification. DOC samples must either MRDLs. be analyzed or must be acidified to must take all samples during normal achieve pH less than 2.0 by minimal operating conditions. (5) Systems may use only data addition of phosphoric or sulfuric acid (2) Systems may consider multiple collected under the provisions of this as soon as practical after sampling, not wells drawing water from a single subpart or subpart M of this part to to exceed 48 hours. Acidified DOC aquifer as one treatment plant for qualify for reduced monitoring. samples must be analyzed within 28 determining the minimum number of (b) Monitoring requirements for days. TTHM and HAA5 samples required, disinfection byproducts. (1) TTHMs and (ii) Ultraviolet Absorption at 254 nm with State approval in accordance with HAA5. (i) Routine monitoring. Systems (UV254). Method 5910 B (Ultraviolet criteria developed under § 142.16(f)(5) must monitor at the frequency indicated Absorption Method). UV absorption of this chapter. in the following table:

ROUTINE MONITORING FREQUENCY FOR TTHM AND HAA5

Type of system Minimum monitoring frequency Sample location in the distribution system

Subpart H system serving at least Four water samples per quarter At least 25 percent of all samples collected each quarter at locations 10,000 persons. per treatment plant. representing maximum residence time. Remaining samples taken at locations representative of at least average residence time in the distribution system and representing the entire distribution system, taking into account number of persons served, different sources of water, and different treatment methods.1 Subpart H system serving from One water sample per quarter per Locations representing maximum residence time.1 500 to 9,999 persons. treatment plant. Subpart H system serving fewer One sample per year per treat- Locations representing maximum residence time.1 If the sample (or than 500 persons. ment plant during month of average of annual samples, if more than one sample is taken) ex- warmest water temperature. ceeds MCL, system must increase monitoring to one sample per treatment plant per quarter, taken at a point reflecting the maximum residence time in the distribution system, until system meets re- duced monitoring criteria in paragraph (c) of this section. System using only ground water One water sample per quarter per Locations representing maximum residence time.1 not under direct influence of sur- treatment plant 2. face water using chemical dis- infectant and serving at least 10,000 persons. System using only ground water One sample per year per treat- Locations representing maximum residence time.1 If the sample (or not under direct influence of sur- ment plant 2 during month of average of annual samples, if more than one sample is taken) ex- face water using chemical dis- warmest water temperature. ceeds MCL, system must increase monitoring to one sample per infectant and serving fewer than treatment plant per quarter, taken at a point reflecting the maximum 10,000 persons. residence time in the distribution system, until system meets criteria in paragraph (c) of this section for reduced monitoring. 1 If a system elects to sample more frequently than the minimum required, at least 25 percent of all samples collected each quarter (including those taken in excess of the required frequency) must be taken at locations that represent the maximum residence time of the water in the dis- tribution system. The remaining samples must be taken at locations representative of at least average residence time in the distribution system. 2 Multiple wells drawing water from a single aquifer may be considered one treatment plant for determining the minimum number of samples required, with State approval in accordance with criteria developed under § 142.16(f)(5) of this chapter.

(ii) Systems may reduce monitoring, except as otherwise provided, in accordance with the following table: Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69469

Reduced Monitoring Frequency for TTHM and HAA5

You may reduce monitoring if you If you are a ... have monitored at least one year To this level and your ...

Subpart H system serving at least TTHM annual average ≤0.040 mg/ One sample per treatment plant per quarter at distribution system lo- 10,000 persons which has a L and HAA5 annual average cation reflecting maximum residence time. source water annual average ≤0.030 mg/L. TOC level, before any treatment, ≤4.0 mg/L. Subpart H system serving from TTHM annual average ≤0.040 mg/ One sample per treatment plant per year at distribution system loca- 500 to 9,999 persons which has L and HAA5 annual average tion reflecting maximum residence time during month of warmest a source water annual average ≤0.030 mg/L. water temperature. NOTE: Any Subpart H system serving fewer TOC level, before any treatment, than 500 persons may not reduce its monitoring to less than one ≤4.0 mg/L. sample per treatment plant per year. System using only ground water TTHM annual average ≤0.040 mg/ One sample per treatment plant per year at distribution system loca- not under direct influence of sur- L and HAA5 annual average tion reflecting maximum residence time during month of warmest face water using chemical dis- ≤0.030 mg/L. water temperature infectant and serving at least 10,000 persons. System using only ground water TTHM annual average ≤0.040 mg/ One sample per treatment plant per three year monitoring cycle at not under direct influence of sur- L and HAA5 annual average distribution system location reflecting maximum residence time dur- face water using chemical dis- ≤0.030 mg/L for two consecutive ing month of warmest water temperature, with the three-year cycle infectant and serving fewer than years OR TTHM annual average beginning on January 1 following quarter in which system qualifies 10,000 persons. ≤0.020 mg/L and HAA5 annual for reduced monitoring. average ≤0.015 mg/L for one year.

(iii) Systems on a reduced monitoring customer, at a location representative of monitoring under paragraph (b)(2)(ii) of schedule may remain on that reduced average residence time, and at a location this section. The system may remain on schedule as long as the average of all reflecting maximum residence time in the reduced monitoring schedule until samples taken in the year (for systems the distribution system. Any additional either any of the three individual which must monitor quarterly) or the routine sampling must be conducted in chlorite samples taken quarterly in the result of the sample (for systems which the same manner (as three-sample sets, distribution system under paragraph must monitor no more frequently than at the specified locations). The system (b)(2)(i)(B) of this section exceeds the annually) is no more than 0.060 mg/L may use the results of additional chlorite MCL or the system is required and 0.045 mg/L for TTHMs and HAA5, monitoring conducted under paragraph to conduct monitoring under paragraph respectively. Systems that do not meet (b)(2)(ii) of this section to meet the (b)(2)(ii) of this section, at which time these levels must resume monitoring at requirement for monitoring in this the system must revert to routine the frequency identified in paragraph paragraph. monitoring. (ii) Additional monitoring. On each (b)(1)(i) of this section in the quarter (3) Bromate. (i) Routine monitoring. day following a routine sample immediately following the quarter in Community and nontransient monitoring result that exceeds the which the system exceeds 0.060 mg/L noncommunity systems using ozone, for chlorite MCL at the entrance to the and 0.045 mg/L for TTHMs and HAA5, disinfection or oxidation, must take one distribution system, the system is respectively. sample per month for each treatment required to take three chlorite (iv) The State may return a system to plant in the system using ozone. distribution system samples at the routine monitoring at the State’s Systems must take samples monthly at following locations: as close to the first discretion. the entrance to the distribution system (2) Chlorite. Community and customer as possible, in a location while the ozonation system is operating nontransient noncommunity water representative of average residence time, under normal conditions. systems using chlorine dioxide, for and as close to the end of the disinfection or oxidation, must conduct distribution system as possible (ii) Reduced monitoring. Systems monitoring for chlorite. (reflecting maximum residence time in required to analyze for bromate may (i) Routine monitoring. (A) Daily the distribution system). reduce monitoring from monthly to monitoring. Systems must take daily (iii) Reduced monitoring. (A) Chlorite once per quarter, if the system samples at the entrance to the monitoring at the entrance to the demonstrates that the average source distribution system. For any daily distribution system required by water bromide concentration is less than sample that exceeds the chlorite MCL, paragraph (b)(2)(i)(A) of this section 0.05 mg/L based upon representative the system must take additional samples may not be reduced. monthly bromide measurements for one in the distribution system the following (B) Chlorite monitoring in the year. The system may remain on day at the locations required by distribution system required by reduced bromate monitoring until the paragraph (b)(2)(ii) of this section, in paragraph (b)(2)(i)(B) of this section may running annual average source water addition to the sample required at the be reduced to one three-sample set per bromide concentration, computed entrance to the distribution system. quarter after one year of monitoring quarterly, is equal to or greater than 0.05 (B) Monthly monitoring. Systems must where no individual chlorite sample mg/L based upon representative take a three-sample set each month in taken in the distribution system under monthly measurements. If the running the distribution system. The system paragraph (b)(2)(i)(B) of this section has annual average source water bromide must take one sample at each of the exceeded the chlorite MCL and the concentration is ≥0.05 mg/L, the system following locations: near the first system has not been required to conduct must resume routine monitoring 69470 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations required by paragraph (b)(3)(i) of this § 141.2) must monitor each treatment water to a consecutive system, under the section. plant for TOC no later than the point of provisions of § 141.29, the sampling (c) Monitoring requirements for combined filter effluent turbidity plan must reflect the entire distribution disinfectant residuals. (1) Chlorine and monitoring and representative of the system. chloramines. (i) Routine monitoring. treated water. All systems required to Systems must measure the residual monitor under this paragraph (d)(1) § 141.133 Compliance requirements. disinfectant level at the same points in must also monitor for TOC in the source (a) General requirements. (1) Where the distribution system and at the same water prior to any treatment at the same compliance is based on a running time as total coliforms are sampled, as time as monitoring for TOC in the annual average of monthly or quarterly specified in § 141.21. Subpart H systems treated water. These samples (source samples or averages and the system’s may use the results of residual water and treated water) are referred to failure to monitor for TTHM, HAA5, or disinfectant concentration sampling as paired samples. At the same time as bromate, this failure to monitor will be conducted under § 141.74(b)(6)(i) for the source water sample is taken, all treated as a monitoring violation for the unfiltered systems or § 141.74(c)(3)(i) for systems must monitor for alkalinity in entire period covered by the annual systems which filter, in lieu of taking the source water prior to any treatment. average. Where compliance is based on separate samples. Systems must take one paired sample a running annual average of monthly or (ii) Reduced monitoring. Monitoring and one source water alkalinity sample quarterly samples or averages and the may not be reduced. per month per plant at a time system’s failure to monitor makes it (2) Chlorine dioxide. (i) Routine representative of normal operating impossible to determine compliance monitoring. Community, nontransient conditions and influent water quality. with MRDLs for chlorine and noncommunity, and transient (2) Reduced monitoring. Subpart H chloramines, this failure to monitor will noncommunity water systems that use systems with an average treated water be treated as a monitoring violation for chlorine dioxide for disinfection or TOC of less than 2.0 mg/L for two the entire period covered by the annual oxidation must take daily samples at the consecutive years, or less than 1.0 mg/ average. entrance to the distribution system. For L for one year, may reduce monitoring (2) All samples taken and analyzed any daily sample that exceeds the for both TOC and alkalinity to one under the provisions of this subpart MRDL, the system must take samples in paired sample and one source water must be included in determining the distribution system the following alkalinity sample per plant per quarter. compliance, even if that number is day at the locations required by The system must revert to routine greater than the minimum required. paragraph (c)(2)(ii) of this section, in monitoring in the month following the (3) If, during the first year of addition to the sample required at the quarter when the annual average treated monitoring under § 141.132, any entrance to the distribution system. water TOC ≥2.0 mg/L. individual quarter’s average will cause (ii) Additional monitoring. On each (e) Bromide. Systems required to the running annual average of that day following a routine sample analyze for bromate may reduce bromate system to exceed the MCL, the system monitoring result that exceeds the monitoring from monthly to once per is out of compliance at the end of that MRDL, the system is required to take quarter, if the system demonstrates that quarter. three chlorine dioxide distribution the average source water bromide (b) Disinfection byproducts. (1) system samples. If chlorine dioxide or concentration is less than 0.05 mg/L TTHMs and HAA5. (i) For systems chloramines are used to maintain a based upon representative monthly monitoring quarterly, compliance with disinfectant residual in the distribution measurements for one year. The system MCLs in § 141.64 must be based on a system, or if chlorine is used to must continue bromide monitoring to running annual arithmetic average, maintain a disinfectant residual in the remain on reduced bromate monitoring. computed quarterly, of quarterly distribution system and there are no (f) Monitoring plans. Each system arithmetic averages of all samples disinfection addition points after the required to monitor under this subpart collected by the system as prescribed by entrance to the distribution system (i.e., must develop and implement a § 141.132(b)(1). If the running annual no booster chlorination), the system monitoring plan. The system must arithmetic average of quarterly averages must take three samples as close to the maintain the plan and make it available covering any consecutive four-quarter first customer as possible, at intervals of for inspection by the State and the period exceeds the MCL, the system is at least six hours. If chlorine is used to general public no later than 30 days in violation of the MCL and must notify maintain a disinfectant residual in the following the applicable compliance the public pursuant to § 141.32, in distribution system and there are one or dates in § 141.130(b). All Subpart H addition to reporting to the State more disinfection addition points after systems serving more than 3300 people pursuant to § 141.134. If a PWS fails to the entrance to the distribution system must submit a copy of the monitoring complete four consecutive quarters’ (i.e., booster chlorination), the system plan to the State no later than the date monitoring, compliance with the MCL must take one sample at each of the of the first report required under for the last four-quarter compliance following locations: as close to the first § 141.134. The State may also require period must be based on an average of customer as possible, in a location the plan to be submitted by any other the available data. representative of average residence time, system. After review, the State may (ii) For systems monitoring less and as close to the end of the require changes in any plan elements. frequently than quarterly, compliance distribution system as possible The plan must include at least the must be based on an average of samples (reflecting maximum residence time in following elements. taken that year under the provisions of the distribution system). (1) Specific locations and schedules § 141.132(b)(1). If the average of these (iii) Reduced monitoring. Chlorine for collecting samples for any samples exceeds the MCL, the system dioxide monitoring may not be reduced. parameters included in this subpart. must increase monitoring to once per (d) Monitoring requirements for (2) How the system will calculate quarter per treatment plant. disinfection byproduct precursors compliance with MCLs, MRDLs, and (iii) Systems on a reduced monitoring (DBPP). (1) Routine monitoring. Subpart treatment techniques. schedule whose annual average exceeds H systems which use conventional (3) If approved for monitoring as a the MCL will revert to routine filtration treatment (as defined in consecutive system, or if providing monitoring immediately. These systems Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69471 will not be considered in violation of chloramines for residual disinfection day following an exceedance of the the MCL until they have completed one during the year, compliance must be chlorine dioxide MRDL at the entrance year of routine monitoring. determined by including together all to the distribution system is also an (2). Bromate. Compliance must be monitoring results of both chlorine and MRDL violation and the system must based on a running annual arithmetic chloramines in calculating compliance. notify the public of the violation in average, computed quarterly, of monthly Reports submitted pursuant to § 141.134 accordance with the provisions for samples (or, for months in which the must clearly indicate which residual nonacute violations under system takes more than one sample, the disinfectant was analyzed for each § 141.32(e)(78). average of all samples taken during the sample. (d) Disinfection byproduct precursors month) collected by the system as (2) Chlorine dioxide. (i) Acute (DBPP). Compliance must be prescribed by § 141.132(b)(3). If the violations. Compliance must be based determined as specified by § 141.135(b). average of samples covering any on consecutive daily samples collected Systems may begin monitoring to consecutive four-quarter period exceeds by the system under § 141.132(c)(2). If determine whether Step 1 TOC the MCL, the system is in violation of any daily sample taken at the entrance removals can be met 12 months prior to the MCL and must notify the public to the distribution system exceeds the the compliance date for the system. This pursuant to § 141.32, in addition to MRDL, and on the following day one (or monitoring is not required and failure to reporting to the State pursuant to more) of the three samples taken in the monitor during this period is not a § 141.134. If a PWS fails to complete 12 distribution system exceed the MRDL, violation. However, any system that consecutive months’ monitoring, the system is in violation of the MRDL does not monitor during this period, compliance with the MCL for the last and must take immediate corrective and then determines in the first 12 four-quarter compliance period must be action to lower the level of chlorine months after the compliance date that it based on an average of the available dioxide below the MRDL and must is not able to meet the Step 1 data. notify the public pursuant to the requirements in § 141.135(b)(2) and (3) Chlorite. Compliance must be procedures for acute health risks in must therefore apply for alternate based on an arithmetic average of each § 141.32(a)(1)(iii)(E). Failure to take minimum TOC removal (Step 2) three sample set taken in the samples in the distribution system the requirements, is not eligible for distribution system as prescribed by day following an exceedance of the retroactive approval of alternate § 141.132(b)(2)(i)(B) and chlorine dioxide MRDL at the entrance minimum TOC removal (Step 2) § 141.132(b)(2)(ii). If the arithmetic to the distribution system will also be requirements as allowed pursuant to average of any three sample set exceeds considered an MRDL violation and the § 141.135(b)(3) and is in violation. the MCL, the system is in violation of system must notify the public of the Systems may apply for alternate the MCL and must notify the public violation in accordance with the minimum TOC removal (Step 2) pursuant to § 141.32, in addition to provisions for acute violations under requirements any time after the reporting to the State pursuant to § 141.32(a)(1)(iii)(E). compliance date. § 141.134. (ii) Nonacute violations. Compliance (c) Disinfectant residuals. (1) Chlorine must be based on consecutive daily § 141.134 Reporting and recordkeeping and chloramines. (i) Compliance must samples collected by the system under requirements. be based on a running annual arithmetic § 141.132(c)(2). If any two consecutive (a) Systems required to sample average, computed quarterly, of monthly daily samples taken at the entrance to quarterly or more frequently must report averages of all samples collected by the the distribution system exceed the to the State within 10 days after the end system under § 141.132(c)(1). If the MRDL and all distribution system of each quarter in which samples were average of quarterly averages covering samples taken are below the MRDL, the collected, notwithstanding the any consecutive four-quarter period system is in violation of the MRDL and provisions of § 141.31. Systems required exceeds the MRDL, the system is in must take corrective action to lower the to sample less frequently than quarterly violation of the MRDL and must notify level of chlorine dioxide below the must report to the State within 10 days the public pursuant to § 141.32, in MRDL at the point of sampling and will after the end of each monitoring period addition to reporting to the State notify the public pursuant to the in which samples were collected. pursuant to § 141.134. procedures for nonacute health risks in (b) Disinfection byproducts. Systems (ii) In cases where systems switch § 141.32(e)(78). Failure to monitor at the must report the information specified in between the use of chlorine and entrance to the distribution system the the following table:

If you are a... You must report...1

System monitoring for TTHM and HAA5 under the requirements of (1) The number of samples taken during the last quarter. §§ 141.132(b) on a quarterly or more frequent basis. (2) The location, date, and result of each sample taken during the last quarter. (3) The arithmetic average of all samples taken in the last quarter. (4) The annual arithmetic average of the quarterly arithmetic averages of this section for the last four quarters. (5) Whether the MCL was exceeded. System monitoring for TTHMs and HAA5 under the requirements of (1) The number of samples taken during the last year. §§ 141.132(b) less frequently than quarterly (but at least annually). (2) The location, date, and result of each sample taken during the last quarter. (3) The arithmetic average of all samples taken over the last year. (4) Whether the MCL was exceeded. System monitoring for TTHMs and HAA5 under the requirements of (1) The location, date, and result of the last sample taken. § 141.132(b) less frequently than annually. (2) Whether the MCL was exceeded. 69472 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

If you are a... You must report...1

System monitoring for chlorite under the requirements of § 141.132(b) .. (1) The number of samples taken each month for the last 3 months. (2) The location, date, and result of each sample taken during the last quarter. (3) For each month in the reporting period, the arithmetic average of all samples taken in the month. (4) Whether the MCL was exceeded, and in which month it was ex- ceeded. System monitoring for bromate under the requirements of § 141.132(b) (1) The number of samples taken during the last quarter. (2) The location, date, and result of each sample taken during the last quarter. (3) The arithmetic average of the monthly arithmetic averages of all samples taken in the last year. (4) Whether the MCL was exceeded.

(c) Disinfectants. Systems must report the information specified in the following table:

If you are a... You must report...1

System monitoring for chlorine or chloramines under the requirements (1) The number of samples taken during each month of the last quar- of § 141.132(c). ter. (2) The monthly arithmetic average of all samples taken in each month for the last 12 months. (3) The arithmetic average of all monthly averages for the last 12 months. (4) Whether the MRDL was exceeded. System monitoring for chlorine dioxide under the requirements of (1) The dates, results, and locations of samples taken during the last § 141.132(c). quarter. (2) Whether the MRDL was exceeded. (3) Whether the MRDL was exceeded in any two consecutive daily samples and whether the resulting violation was acute or nonacute. 1 The State may choose to perform calculations and determine whether the MRDL was exceeded, in lieu of having the system report that infor- mation.

(d) Disinfection byproduct precursors information specified in the following and enhanced coagulation or enhanced table: softening. Systems must report the

If you are a . .. You must report . ..1

System monitoring monthly or quarterly for TOC under the require- (1) The number of paired (source water and treated water, prior to con- ments of § 141.132(d) and required to meet the enhanced coagula- tinuous disinfection) samples taken during the last quarter. tion or enhanced softening requirements in § 141.135(b)(2) or (3). (2) The location, date, and result of each paired sample and associ- ated alkalinity taken during the last quarter. (3) For each month in the reporting period that paired samples were taken, the arithmetic average of the percent reduction of TOC for each paired sample and the required TOC percent removal. (4) Calculations for determining compliance with the TOC percent re- moval requirements, as provided in § 141.135(c)(1). (5) Whether the system is in compliance with the enhanced coagula- tion or enhanced softening percent removal requirements in § 141.135(b) for the last four quarters. System monitoring monthly or quarterly for TOC under the require- (1) The alternative compliance criterion that the system is using. ments of § 141.132(d) and meeting one or more of the alternative compliance criteria in § 141.135(a)(2) or (3). (2) The number of paired samples taken during the last quarter. (3) The location, date, and result of each paired sample and associ- ated alkalinity taken during the last quarter. (4) The running annual arithmetic average based on monthly averages (or quarterly samples) of source water TOC for systems meeting a criterion in §§ 141.135(a)(2)(i) or (iii) or of treated water TOC for sys- tems meeting the criterion in § 141.135(a)(2)(ii). Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69473

If you are a . .. You must report . ..1

(5) The running annual arithmetic average based on monthly averages (or quarterly samples) of source water SUVA for systems meeting the criterion in § 141.135(a)(2)(v) or of treated water SUVA for sys- tems meeting the criterion in § 141.135(a)(2)(vi). (6) The running annual average of source water alkalinity for systems meeting the criterion in § 141.135(a)(2)(iii) and of treated water alka- linity for systems meeting the criterion in § 141.135(a)(3)(i). (7) The running annual average for both TTHM and HAA5 for systems meeting the criterion in § 141.135(a)(2)(iii) or (iv). (8) The running annual average of the amount of magnesium hardness removal (as CaCO3, in mg/L) for systems meeting the criterion in § 141.135(a)(3)(ii). (9) Whether the system is in compliance with the particular alternative compliance criterion in § 141.135(a)(2) or (3). 1 The State may choose to perform calculations and determine whether the treatment technique was met, in lieu of having the system report that information.

§ 141.135 Treatment technique for control L, respectively; or prior to the effective practicing enhanced softening that of disinfection byproduct (DBP) precursors. date for compliance in § 141.130(b), the cannot achieve the TOC removals (a) Applicability. (1) Subpart H system has made a clear and irrevocable required by paragraph (b)(2) of this systems using conventional filtration financial commitment not later than the section may use the alternative treatment (as defined in § 141.2 ) must effective date for compliance in compliance criteria in paragraphs operate with enhanced coagulation or § 141.130(b) to use of technologies that (a)(3)(i) and (ii) of this section in lieu of enhanced softening to achieve the TOC will limit the levels of TTHMs and complying with paragraph (b) of this percent removal levels specified in HAA5 to no more than 0.040 mg/L and section. Systems must still comply with paragraph (b) of this section unless the 0.030 mg/L, respectively. Systems must monitoring requirements in system meets at least one of the submit evidence of a clear and § 141.132(d). alternative compliance criteria listed in irrevocable financial commitment, in paragraph (a)(2) or (a)(3) of this section. addition to a schedule containing (i) Softening that results in lowering (2) Alternative compliance criteria for milestones and periodic progress reports the treated water alkalinity to less than enhanced coagulation and enhanced for installation and operation of 60 mg/L (as CaCO3), measured monthly softening systems. Subpart H systems appropriate technologies, to the State for according to § 141.131(d)(1) and using conventional filtration treatment approval not later than the effective date calculated quarterly as a running annual may use the alternative compliance for compliance in § 141.130(b). These average. criteria in paragraphs (a)(2)(i) through technologies must be installed and (ii) Softening that results in removing (vi) of this section to comply with this operating not later than June 16, 2005. at least 10 mg/L of magnesium hardness section in lieu of complying with Failure to install and operate these (as CaCO3), measured monthly and paragraph (b) of this section. Systems technologies by the date in the approved calculated quarterly as an annual must still comply with monitoring schedule will constitute a violation of running average. requirements in § 141.132(d). National Primary Drinking Water (b) Enhanced coagulation and (i) The system’s source water TOC Regulations. enhanced softening performance level, measured according to (iv) The TTHM and HAA5 running requirements. (1) Systems must achieve § 141.131(d)(3), is less than 2.0 mg/L, annual averages are no greater than the percent reduction of TOC specified calculated quarterly as a running annual 0.040 mg/L and 0.030 mg/L, in paragraph (b)(2) of this section average. respectively, and the system uses only between the source water and the (ii) The system’s treated water TOC chlorine for primary disinfection and combined filter effluent, unless the State level, measured according to maintenance of a residual in the § 141.131(d)(3), is less than 2.0 mg/L, distribution system. approves a system’s request for alternate calculated quarterly as a running annual (v) The system’s source water SUVA, minimum TOC removal (Step 2) average. prior to any treatment and measured requirements under paragraph (b)(3) of (iii) The system’s source water TOC monthly according to § 141.131(d)(4), is this section. level, measured as required by less than or equal to 2.0 L/mg-m, (2) Required Step 1 TOC reductions, § 141.131(d)(3), is less than 4.0 mg/L, calculated quarterly as a running annual indicated in the following table, are calculated quarterly as a running annual average. based upon specified source water average; the source water alkalinity, (vi) The system’s finished water parameters measured in accordance measured according to § 141.131(d)(1), SUVA, measured monthly according to with § 141.131(d). Systems practicing is greater than 60 mg/L (as CaCO3), § 141.131(d)(4), is less than or equal to softening are required to meet the Step calculated quarterly as a running annual 2.0 L/mg-m, calculated quarterly as a 1 TOC reductions in the far-right average; and either the TTHM and running annual average. column (Source water alkalinity >120 HAA5 running annual averages are no (3) Additional alternative compliance mg/L) for the specified source water greater than 0.040 mg/L and 0.030 mg/ criteria for softening systems. Systems TOC: 69474 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

STEP 1 REQUIRED REMOVAL OF TOC BY ENHANCED COAGULATION AND ENHANCED SOFTENING FOR SUBPART H SYSTEMS USING CONVENTIONAL TREATMENT 1, 2

Source-water alkalinity, mg/L as CaCO3 Source-water TOC, mg/L ≤60±120 (per- >120 3 (per- 0±60 (percent) cent) cent)

>2.0±4.0 ...... 35.0 25.0 15.0 >4.0±8.0 ...... 45.0 35.0 25.0 >8.0 ...... 50.0 40.0 30.0 1 Systems meeting at least one of the conditions in paragraph (a)(2)(i)±(vi) of this section are not required to operate with enhanced coagula- tion. 2 Softening systems meeting one of the alternative compliance criteria in paragraph (a)(3) of this section are not required to operate with en- hanced softening. 3 Systems practicing softening must meet the TOC removal requirements in this column.

(3) Subpart H conventional treatment value based on the results of a new a waiver of enhanced coagulation systems that cannot achieve the Step 1 bench- and pilot-scale test. Failure to requirements. TOC removals required by paragraph achieve State-set alternative minimum (c) Compliance calculations. (1) (b)(2) of this section due to water quality TOC removal levels is a violation of Subpart H systems other than those parameters or operational constraints National Primary Drinking Water identified in paragraph (a)(2) or (a)(3) of must apply to the State, within three Regulations. this section must comply with months of failure to achieve the TOC (ii) Bench- or pilot-scale testing of requirements contained in paragraph removals required by paragraph (b)(2) of enhanced coagulation must be (b)(2) of this section. Systems must this section, for approval of alternative conducted by using representative water calculate compliance quarterly, minimum TOC (Step 2) removal samples and adding 10 mg/L increments beginning after the system has collected requirements submitted by the system. of alum (as aluminum) (or equivalent 12 months of data, by determining an If the State approves the alternative amounts of ferric salt) until the pH is annual average using the following minimum TOC removal (Step 2) reduced to a level less than or equal to method: requirements, the State may make those the enhanced coagulation Step 2 target (i) Determine actual monthly TOC requirements retroactive for the pH shown in the following table: percent removal, equal to: purposes of determining compliance. (1—(treated water TOC/source water Until the State approves the alternate ENHANCED COAGULATION STEP 2 TOC)) × 100 minimum TOC removal (Step 2) TARGET PH (ii) Determine the required monthly requirements, the system must meet the TOC percent removal (from either the Step 1 TOC removals contained in Alkalinity (mg/L as CaCO3) Target pH table in paragraph (b)(2) of this section paragraph (b)(2) of this section. or from paragraph (b)(3) of this section). (4) Alternate minimum TOC removal 0±60 ...... 5.5 (iii) Divide the value in paragraph (Step 2) requirements. Applications >60±120 ...... 6.3 (c)(1)(i) of this section by the value in made to the State by enhanced >120±240 ...... 7.0 >240 ...... 7.5 paragraph (c)(1)(ii) of this section. coagulation systems for approval of (iv) Add together the results of alternative minimum TOC removal paragraph (c)(1)(iii) of this section for (Step 2) requirements under paragraph (iii) For waters with alkalinities of the last 12 months and divide by 12. (b)(3) of this section must include, as a less than 60 mg/L for which addition of (v) If the value calculated in minimum, results of bench- or pilot- small amounts of alum or equivalent paragraph (c)(1)(iv) of this section is less scale testing conducted under paragraph addition of iron coagulant drives the pH than 1.00, the system is not in (b)(4)(i) of this section and used to below 5.5 before significant TOC compliance with the TOC percent determine the alternate enhanced removal occurs, the system must add removal requirements. coagulation level. necessary chemicals to maintain the pH (2) Systems may use the provisions in (i) Alternate enhanced coagulation between 5.3 and 5.7 in samples until the paragraphs (c)(2)(i) through (v) of this level is defined as coagulation at a TOC removal of 0.3 mg/L per 10 mg/L section in lieu of the calculations in coagulant dose and pH as determined by alum added (as aluminum) (or paragraph (c)(1)(i) through (v) of this the method described in paragraphs equivalant addition of iron coagulant) is section to determine compliance with (b)(4)(i) through (v) of this section such reached. TOC percent removal requirements. that an incremental addition of 10 mg/ (iv) The system may operate at any (i) In any month that the system’s L of alum (as aluminum) (or equivalent coagulant dose or pH necessary treated or source water TOC level, amount of ferric salt) results in a TOC (consistent with other NPDWRs) to measured according to § 141.131(d)(3), removal of ≤ 0.3 mg/L. The percent achieve the minimum TOC percent is less than 2.0 mg/L, the system may removal of TOC at this point on the removal approved under paragraph assign a monthly value of 1.0 (in lieu of ‘‘TOC removal versus coagulant dose’’ (b)(3) of this section. the value calculated in paragraph curve is then defined as the minimum (v) If the TOC removal is consistently (c)(1)(iii) of this section) when TOC removal required for the system. less than 0.3 mg/L of TOC per 10 mg/ calculating compliance under the Once approved by the State, this L of incremental alum dose (as provisions of paragraph (c)(1) of this minimum requirement supersedes the aluminum) at all dosages of alum (or section. minimum TOC removal required by the equivalant addition of iron coagulant), (ii) In any month that a system table in paragraph (b)(2) of this section. the water is deemed to contain TOC not practicing softening removes at least 10 This requirement will be effective until amenable to enhanced coagulation. The mg/L of magnesium hardness (as such time as the State approves a new system may then apply to the State for CaCO3), the system may assign a Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations 69475 monthly value of 1.0 (in lieu of the Authority: 42 U.S.C. 300f, 300g-1, 300g-2 chlorine, TTHM, and HAA5, if any, are value calculated in paragraph (c)(1)(iii) 300g-3, 300g-4, 300g-5, 300g-6, 300j-4, 300j- measured. of this section) when calculating 9, and 300j-11. * * * * * compliance under the provisions of 13. Section 142.14 is amended by 14. Section 142.16 is amended by paragraph (c)(1) of this section. adding new paragraphs (d)(12), (d)(13), adding paragraph (h) to read as follows. (iii) In any month that the system’s (d)(14), (d)(15), and (d)(16) to read as § 142.16 Special primacy requirements. source water SUVA, prior to any follows. treatment and measured according to * * * * * § 141.131(d)(4), is ≤2.0 L/mg-m, the § 142.14 Records kept by States. (h) Requirements for States to adopt system may assign a monthly value of * * * * * 40 CFR part 141, subpart L. In addition 1.0 (in lieu of the value calculated in (d) * * * to the general primacy requirements elsewhere in this part, including the paragraph (c)(1)(iii) of this section) (12) Records of the currently requirement that State regulations be at when calculating compliance under the applicable or most recent State provisions of paragraph (c)(1) of this least as stringent as federal determinations, including all supporting requirements, an application for section. information and an explanation of the (iv) In any month that the system’s approval of a State program revision technical basis for each decision, made that adopts 40 CFR part 141, subpart L, finished water SUVA, measured under the following provisions of 40 according to § 141.131(d)(4), is ≤2.0 L/ must contain a description of how the CFR part 141, subpart L for the control State will accomplish the following mg-m, the system may assign a monthly of disinfectants and disinfection value of 1.0 (in lieu of the value program requirements: byproducts. These records must also (1) Section 141.64(b)(2) of this chapter calculated in paragraph (c)(1)(iii) of this include interim measures toward (interim treatment requirements). section) when calculating compliance installation. Determine any interim treatment under the provisions of paragraph (c)(1) (i) States must keep records of requirements for those systems electing of this section. systems that are installing GAC or to install GAC or membrane filtration (v) In any month that a system membrane technology in accordance and granted additional time to comply practicing enhanced softening lowers with § 141.64(b)(2) of this chapter. with § 141.64 of this chapter. alkalinity below 60 mg/L (as CaCO3), the These records must include the date by (2) Section 141.130(c) of this chapter system may assign a monthly value of which the system is required to have (qualification of operators). Qualify 1.0 (in lieu of the value calculated in completed installation. operators of public water systems paragraph (c)(1)(iii) of this section) (ii) States must keep records of subject to 40 CFR part 141, subpart L. when calculating compliance under the systems that are required, by the State, Qualification requirements established provisions of paragraph (c)(1) of this to meet alternative minimum TOC for operators of systems subject to 40 section. removal requirements or for whom the CFR part 141, subpart H—Filtration and (3) Subpart H systems using State has determined that the source Disinfection may be used in whole or in conventional treatment may also water is not amenable to enhanced part to establish operator qualification comply with the requirements of this coagulation in accordance with requirements for meeting 40 CFR part section by meeting the criteria in § 141.135(b)(3) and (4) of this chapter, 141, subpart L requirements if the State paragraph (a)(2) or (3) of this section. respectively. These records must determines that the 40 CFR part 141, (d) Treatment technique requirements include the alternative limits and subpart H requirements are appropriate for DBP precursors. The Administrator rationale for establishing the alternative and applicable for meeting subpart L identifies the following as treatment limits. requirements. techniques to control the level of (iii) States must keep records of (3) Section 141.131(c)(2) of this disinfection byproduct precursors in subpart H systems using conventional chapter (DPD colorimetric test kits). drinking water treatment and treatment meeting any of the alternative Approve DPD colorimetric test kits for distribution systems: For Subpart H compliance criteria in § 141.135(a)(2) or free and total chlorine measurements. systems using conventional treatment, (3) of this chapter. State approval granted under enhanced coagulation or enhanced § 141.74(a)(2) of this chapter for the use (iv) States must keep a register of softening. of DPD colorimetric test kits for free qualified operators that have met the 11. Section 141.154 is amended by chlorine testing is acceptable for the use State requirements developed under adding paragraph (e) to read as follows: of DPD test kits in measuring free § 142.16(f)(2). chlorine residuals as required in 40 CFR § 141.154 Required additional health (13) Records of systems with multiple information. part 141, subpart L. wells considered to be one treatment (4) Sections 141.131(c)(3) and (d) of * * * * * plant in accordance with § 141.132(a)(2) this chapter (State approval of parties to (e) Community water systems that of this chapter and § 142.16(f)(5). conduct analyses). Approve parties to detect TTHM above 0.080 mg/l, but (14) Monitoring plans for subpart H conduct pH, bromide, alkalinity, and below the MCL in § 141.12, as an annual systems serving more than 3,300 residual disinfectant concentration average, monitored and calculated persons in accordance with § 141.132(f) measurements. The State’s process for under the provisions of § 141.30, must of this chapter. approving parties performing water include health effects language (15) List of laboratories approved for quality measurements for systems prescribed by paragraph (73) of analyses in accordance with subject to 40 CFR part 141, subpart H appendix C to subpart O. § 141.131(b) of this chapter. requirements in paragraph (b)(2)(i)(D) of PART 142ÐNATIONAL PRIMARY (16) List of systems required to this section may be used for approving DRINKING WATER REGULATIONS monitor for disinfectants and parties measuring water quality IMPLEMENTATION disinfection byproducts in accordance parameters for systems subject to with part 141, subpart L of this chapter. subpart L requirements, if the State 12. The authority citation for part 142 The list must indicate what determines the process is appropriate continues to read as follows: disinfectants and DBPs, other than and applicable. 69476 Federal Register / Vol. 63, No. 241 / Wednesday, December 16, 1998 / Rules and Regulations

(5) Section 141.132(a)(2) of this therefore be considered a single source allowed under the provisions of chapter (multiple wells as a single for compliance with monitoring § 141.135(b) of this chapter. source). Define the criteria to use to requirements. [FR Doc. 98–32887 Filed 12–15–98; 8:45 am] determine if multiple wells are being (6) Approve alternate minimum TOC drawn from a single aquifer and removal (Step 2) requirements, as BILLING CODE 6560±50±U