Monday, August 18, 2003

Part II

Environmental Protection Agency 40 CFR Parts 141, 142, and 143

National Primary Drinking Water Regulations: Stage 2 Disinfectants and Disinfection Byproducts Rule; National Primary and Secondary Drinking Water Regulations: Approval of Analytical Methods for Chemical Contaminants; Proposed Rule

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00001 Fmt 4717 Sfmt 4717 E:\FR\FM\18AUP2.SGM 18AUP2 49548 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

ENVIRONMENTAL PROTECTION specifies the best available technologies Comments may also be submitted AGENCY (BATs) for the proposed MCLs. EPA is electronically or through hand delivery/ also proposing additional analytical courier by following the detailed 40 CFR Parts 141, 142 and 143 methods for the determination of instructions as provided in section I.C. [FRL–7530–3] disinfectants and disinfection of the SUPPLEMENTARY INFORMATION byproducts (DBPs) in drinking water section. RIN 2040–AD38 and proposing to extend approval of DBP methods for the determination of FOR FURTHER INFORMATION CONTACT: For National Primary Drinking Water additional chemical contaminants. This technical inquiries, contact Tom Regulations: Stage 2 Disinfectants and set of regulations proposed today is Grubbs, Office of Ground Water and Disinfection Byproducts Rule; National known as the Stage 2 Disinfectants and Drinking Water (MC 4607M), U.S. Primary and Secondary Drinking Water Disinfection Byproducts Rule (Stage 2 Environmental Protection Agency, 1200 Regulations: Approval of Analytical DBPR). EPA’s objective for the Stage 2 Pennsylvania Ave., NW., Washington, Methods for Chemical Contaminants DBPR is to reduce the potential risks of DC 20460; telephone (202) 564–5262. For regulatory inquiries, contact Jennifer AGENCY: Environmental Protection reproductive and developmental health Agency. effects and cancer associated with McLain at the same address; telephone disinfection byproducts (DBPs) by (202) 564–5248. For general information ACTION: Proposed rule. reducing peak and average levels of contact the Safe Drinking Water Hotline, SUMMARY: In this document, the DBPs in drinking water supplies. Telephone (800) 426–4791. The Safe Environmental Protection Agency (EPA) The Stage 2 DBPR applies to public Drinking Water Hotline is open Monday is proposing maximum contaminant water systems (PWS) that are through Friday, excluding legal level goals (MCLGs) for chloroform, community water systems (CWSs) or holidays, from 9 a.m. to 5:30 p.m. monochloroacetic acid (MCAA) and nontransient noncommunity water Eastern Time. systems (NTNCWs) that add a primary trichloroacetic acid (TCAA); National SUPPLEMENTARY INFORMATION: Primary Drinking Water Regulations or residual disinfectant other than (NPDWRs) which consist of maximum ultraviolet light or deliver water that has I. General Information been treated with a primary or residual contaminant levels (MCLs) and A. Who Is Regulated by This Action? monitoring, reporting, and public disinfectant other than ultraviolet light. notification requirements for total DATES: The Agency requests comments Entities potentially regulated by the trihalomethanes (TTHM—a sum of on today’s proposal. Comments must be Stage 2 DBPR are community and chloroform, bromodichloromethane, received or post-marked by midnight nontransient noncommunity water dibromochloromethane, and November 17, 2003. systems that add a primary or residual bromoform) and haloacetic acids ADDRESSES: Comments may be disinfectant other than ultraviolet light (HAA5—a sum of mono-, di-, and submitted by mail to: Water Docket, or deliver water that has been treated trichloroacetic acids and mono- and Environmental Protection Agency, Mail with a primary or residual disinfectant dibromoacetic acids); and revisions to Code 4101T, 1200 Pennsylvania Ave., other than ultraviolet light. Regulated the reduced monitoring requirements NW., Washington, DC 20460, Attention categories and entities are identified in for bromate. This document also Docket ID No. OW–2002–0043. the following chart.

Category Examples of regulated entities

Industry ...... Community and nontransient noncommunity water systems that add a primary or residual dis- infectant other than ultraviolet light or deliver water that has been treated with a primary or residual disinfectant other than ultraviolet light. State, Local, Tribal, or Federal Governments .... Community and nontransient noncommunity water systems that add a primary or residual dis- infectant other than ultraviolet light or deliver water that has been treated with a primary or residual disinfectant other than ultraviolet light.

This table is not intended to be preceding section entitled FOR FURTHER for public viewing at the Water Docket exhaustive, but rather provides a guide INFORMATION CONTACT. in the EPA Docket Center, (EPA/DC) for readers regarding entities likely to be B. How Can I Get Copies of This EPA West, Room B102, 1301 regulated by this action. This table lists Document and Other Related Constitution Ave., NW., Washington, the types of entities of which EPA is Information? DC. The EPA Docket Center Public now aware that could potentially be Reading Room is open from 8:30 a.m. to regulated by this action. Other types of 1. Docket. EPA has established an 4:30 p.m., Monday through Friday, entities not listed in this table could official public docket for this action excluding legal holidays. The telephone also be regulated. To determine whether under Docket ID No. OW–2002–0043. number for the Public Reading Room is The official public docket consists of the your facility is regulated by this action, (202) 566–1744, and the telephone documents specifically referenced in you should carefully examine the number for the Water Docket is (202) this action, any public comments definition of ‘‘public water system’’ in 566–2426. For access to docket material, § 141.2 and the section entitled received, and other information related please call (202) 566–2426 to schedule ‘‘coverage’’ (§ 141.3) in Title 40 of the to this action. Although a part of the an appointment. Code of Federal Regulations and official docket, the public docket does applicability criteria in § 141.600 and not include Confidential Business 2. Electronic Access. You may access 141.620 of today’s proposal. If you have Information (CBI) or other information this Federal Register document questions regarding the applicability of whose disclosure is restricted by statute. electronically through the EPA Internet the Stage 2 DBPR to a particular entity, The official public docket is the under the ‘‘Federal Register’’ listings at contact one of the persons listed in the collection of materials that is available http://www.epa.gov/fedrgstr/.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00002 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49549

An electronic version of the public receipt by EPA, identify the appropriate public docket, and made available in docket is available through EPA’s docket identification number in the EPA’s electronic public docket. electronic public docket and comment subject line on the first page of your c. Disk or CD ROM. You may submit system, EPA Dockets. You may use EPA comment. Please ensure that your comments on a disk or CD ROM that Dockets at http://www.epa.gov/edocket/ comments are submitted within the you mail to the mailing address to submit or view public comments, specified comment period. Comments identified in section I.C.2. These access the index listing of the contents received after the close of the comment electronic submissions will be accepted of the official public docket, and to period will be marked ‘‘late.’’ EPA is not in WordPerfect or ASCII file format. access those documents in the public required to consider these late Avoid the use of special characters and docket that are available electronically. comments. any form of encryption. Once in the system, select ‘‘search,’’ 1. Electronically. If you submit an 2. By Mail. Send three copies of your then key in the appropriate docket electronic comment as prescribed comments and any enclosures to: Water identification number. below, EPA recommends that you Docket, Environmental Protection Certain types of information will not include your name, mailing address, Agency, Mail Code 4101T, 1200 be placed in the EPA Dockets. and an e-mail address or other contact Pennsylvania Ave., NW., Washington, Information claimed as CBI and other information in the body of your DC 20460, Attention Docket ID No. OW– information whose disclosure is comment. Also include this contact 2002–0043. restricted by statute, which is not information on the outside of any disk 3. By Hand Delivery or Courier. included in the official public docket, or CD ROM you submit, and in any Deliver your comments to: Water will not be available for public viewing cover letter accompanying the disk or Docket, EPA Docket Center, in EPA’s electronic public docket. EPA’s CD ROM. This ensures that you can be Environmental Protection Agency, policy is that copyrighted material will identified as the submitter of the Room B102, 1301 Constitution Ave., not be placed in EPA’s electronic public comment and allows EPA to contact you NW., Washington, DC, Attention Docket docket but will be available only in in case EPA cannot read your comment ID No. OW–2002–0043. Such deliveries printed, paper form in the official public due to technical difficulties or needs are only accepted during the Docket’s docket. Although not all docket further information on the substance of normal hours of operation as identified materials may be available your comment. EPA’s policy is that EPA in section I.B.1. electronically, you may still access any will not edit your comment, and any of the publicly available docket D. What Should I Consider as I Prepare identifying or contact information My Comments for EPA? materials through the docket facility provided in the body of a comment will identified in section I.B.1. be included as part of the comment that You may find the following For public commenters, it is is placed in the official public docket, suggestions helpful for preparing your important to note that EPA’s policy is and made available in EPA’s electronic comments: that public comments, whether public docket. If EPA cannot read your 1. Explain your views as clearly as submitted electronically or in paper, possible. will be made available for public comment due to technical difficulties and cannot contact you for clarification, 2. Describe any assumptions that you viewing in EPA’s electronic public used. docket as EPA receives them and EPA may not be able to consider your comment. 3. Provide any technical information without change, unless the comment and/or data you used that support your a. EPA Dockets. Your use of EPA’s contains copyrighted material, CBI, or views. electronic public docket to submit other information whose disclosure is 4. If you estimate potential burden or comments to EPA electronically is restricted by statute. When EPA costs, explain how you arrived at your EPA’s preferred method for receiving identifies a comment containing estimate. comments. Go directly to EPA Dockets copyrighted material, EPA will provide 5. Provide specific examples to at http://www.epa.gov/edocket, and a reference to that material in the illustrate your concerns. follow the online instructions for version of the comment that is placed in 6. Offer alternatives. EPA’s electronic public docket. The submitting comments. Once in the 7. Make sure to submit your entire printed comment, including the system, select ‘‘search,’’ and then key in comments by the comment period copyrighted material, will be available Docket ID No. OW–2002–0043. The identified. in the public docket. system is an ‘‘anonymous access’’ 8. To ensure proper receipt by EPA, Public comments submitted on system, which means EPA will not identify the appropriate docket computer disks that are mailed or know your identity, e-mail address, or identification number in the subject line delivered to the docket will be other contact information unless you on the first page of your response. It transferred to EPA’s electronic public provide it in the body of your comment. would also be helpful if you provided docket. Public comments that are b. E-mail. Comments may be sent by the name, date, and Federal Register mailed or delivered to the Docket will electronic mail (e-mail) to OW- citation related to your comments. be scanned and placed in EPA’s [email protected], Attention Docket ID electronic public docket. Where No. OW–2002–0043. In contrast to Abbreviations Used in This Document practical, physical objects will be EPA’s electronic public docket, EPA’s e- AIPC All Indian Pueblo Council photographed, and the photograph will mail system is not an ‘‘anonymous ALT Alanine aminotransferase be placed in EPA’s electronic public access’’ system. If you send an e-mail AST Aspartate aminotransferase docket along with a brief description comment directly to the Docket without ASTM American Society for Testing written by the docket staff. going through EPA’s electronic public and Materials docket, EPA’s e-mail system AWWA American Water Works C. How and to Whom Do I Submit automatically captures your e-mail Association Comments? address. E-mail addresses that are AwwaRF American Water Works You may submit comments automatically captured by EPA’s e-mail Association Research Foundation electronically, by mail, or through hand system are included as part of the BAT Best available technology delivery/courier. To ensure proper comment that is placed in the official BCAA Bromochloroacetic acid

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00003 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49550 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

BDCM Bromodichloromethane MRDLG Maximum residual Table of Contents CWS Community water system disinfectant level goal I. Summary DBAA Dibromoacetic acid MTBE Methyl tertiary butyl ether A. Why is EPA Proposing the Stage 2 DBCM Dibromochloromethane mWh Megawatt-hours DBPR? DBP Disinfection byproduct NATICH National Air Toxics B. What Does the Stage 2 DBPR Require? DBPR Disinfectants and Disinfection Information Clearinghouse C. What are the Economic Impacts of the Byproducts Rule NDIR Nondispersive infrared detection Stage 2 DBPR? DCAA Dichloroacetic acid NDMA N-nitrosodimethylamine II. Background DOC Dissolved organic carbon NDWAC National Drinking Water A. What is the Statutory Authority for the EA Economic analysis Stage 2 DBPR? Advisory Council B. What is the Regulatory History of the EC Enhanced coagulation NF Nanofiltration EDA Ethylenediamine Stage 2 DBPR? NOAEL No Observed Adverse Effect C. How were Stakeholders Involved in ED10 Maximum likelihood estimate of Level a dose producing effects in 10 Developing the Stage 2 DBPR? NODA Notice of data availability 1. Federal Advisory Committee process percent of animals NPDWR National primary drinking 2. Other outreach processes EPA United States Environmental water regulation III. Public Health Risk Protection Agency NRWA National Rural Water A. Reproductive and Developmental FACA Federal Advisory Committee Association Epidemiology Act NTNCWS Nontransient 1. Reif et al. 2000 FBRR Filter Backwash Recycling Rule noncommunity water system a. Fetal growth GAC Granular activated carbon NTP National Toxicology Program b. Fetal viability GC/ECD Gas chromatography using c. Fetal malformations and other NTTAA National Technology Transfer electron capture detection developmental anomalies GWUDI Ground water under the direct and Advancement Act 2. Bove et al. 2002 ODA o-dianisidine dihydrochloride influence of surface water a. Fetal growth HAA5 Haloacetic acids (five) (sum of OMB Office of Management and b. Fetal viability monochloroacetic acid, Budget c. Fetal malformations OSTP Office of Science and 3. Nieuwenhuijsen et al. 2000 dichloroacetic acid, trichloroacetic Technology Policy 4. Additional epidemiology studies acid, monobromoacetic acid, and PAR Population attributable risk B. Reproductive and Developmental dibromoacetic acid) PE Performance evaluation Toxicology IC Ion chromatography PWS Public water system 1. EPA analysis and research ICR Information Collection Request 2. Tyl, 2000 IC/ICP–MS Ion chromatograph— QC Quality control RAA Running annual average a. Developmental defects coupled to an inductively coupled b. Whole litter resorption RFA Regulatory Flexibility Act plasma mass spectrometer c. Fetal toxicity IDSE Initial distribution system RfD Reference dose d. Male reproductive effects RSC Relative source contribution evaluation 3. World Health Organization review of the ILSI International Life Sciences RSD Relative standard deviation reproductive and developmental SAB Science Advisory Board Institute toxicology literature (2000) IESWTR Interim Enhanced Surface SAC Selective anion concentration 4. New Studies SBAR Small Business Advisory C. Conclusions Drawn from the Water Treatment Rule Reproductive and Developmental Health IPCS International Programme on Review SBREFA Small Business Regulatory Effects Data Chemical Safety D. Cancer Epidemiology IRIS Integrated Risk Information Enforcement Fairness Act SDWA Safe Drinking Water Act, or the 1. Population Attributable Risk analysis System (EPA) 2. New epidemiological cancer studies kWh/yr Kilowatt hours per year ‘‘Act,’’ as amended in 1996 a. New bladder cancer studies SER Small Entity Representative LED10 Lower 95 percent confidence b. New colon cancer studies bound of the maximum likelihood SGA Small for gestational age c. New rectal cancer studies estimate of the dose producing SUVA Specific ultraviolet absorbance d. Other cancers effects in 10 percent of animals SWAT Surface Water Analytical Tool 3. Review of the cancer epidemiology LH Luteinizing hormone SWTR Surface Water Treatment Rule literature (WHO 2000) LOAEL Lowest observed adverse effect TAME Tertiary amyl methyl ether E. Cancer and Other Toxicology 1. EPA criteria documents level TCAA Trichloroacetic acid TCR Total Coliform Rule 2. Other byproducts with carcinogenic LRAA Locational running annual potential average THM Trihalomethane TOC Total organic carbon a. 3-chloro-4-(dichloromethyl)-5-hydroxy- LT1ESWTR Long Term 1 Enhanced 2(5H)-furanone) (MX)—multisite cancer Surface Water Treatment Rule TTHM Total trihalomethanes (sum of b. N-nitrosodimethylamine (NDMA)— LT2ESWTR Long Term 2 Enhanced four THMs: chloroform, multisite cancer Surface Water Treatment Rule bromodichloromethane, 3. Other toxicological effects MBAA Monobromoacetic acid dibromochloromethane, and 4. WHO review of the cancer toxicology MCAA Monochloroacetic acid bromoform) literature (2000) MCL Maximum contaminant level TWG Technical work group F. Conclusions Drawn from the Cancer MCLG Maximum contaminant level UMRA Unfunded Mandates Reform Epidemiology and Toxicology goal Act G. Request for Comment USDOE EIA U.S. Department of IV. DBP Occurrence within Distribution M-DBP Microbial and disinfection Systems byproducts Energy, Energy Information A. Data Sources mg/L Milligram per liter Administration 1. Information Collection Rule Data MRL Minimum reporting level UV 254 Ultraviolet absorption at 254 2. Other Data Sources Used to Support the MRDL Maximum residual disinfectant nm Proposal level WTP Willingness To Pay B. DBPs in Distribution Systems

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00004 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49551

1. DBPs above the MCL occur at some 3. Request for comment 2. How was this proposal developed? locations in a substantial number of H. Initial Distribution System Evaluation 3. Which new methods are proposed for plants (IDSE) approval? 2. Specific locations in distribution 1. What is EPA proposing today? a. EPA Method 327.0 for chlorine dioxide systems are not protected to MCL levels a. Applicability and chlorite. 3. Stage 1 DBPR maximum residence time b. Data collection b. EPA Method 552.3 for HAA5 and location may not reflect the highest DBP i. Standard monitoring program dalapon occurrence levels ii. System specific study c. ASTM D 6581–00 for bromate, chlorite, C. Request for Comment iii. 40/30 certification and bromide V. Discussion of Proposed Stage 2 DBPR c. Implementation d. EPA Method 317.0 revision 2 for Requirements 2. How was this proposal developed? bromate, chlorite, and bromide A. MCLG for Chloroform a. Applicability e. EPA Method 326.0 for bromate, chlorite, 1. What is EPA proposing today? b. Data collection and bromide 2. How was this proposal developed? c. Implementation f. EPA Method 321.8 for bromate a. Background 3. Request for comment g. EPA 415.3 for TOC and SUVA (DOC and b. Basis of the new chloroform MCLG a. Applicability UV254) i. Mode of action b. Data collection 4. What additional regulated contaminants ii. Metabolism c. Implementation can be monitored by extending approval c. How the MCLG is derived I. Monitoring Requirements and of EPA Method 300.1? i. Reference dose Compliance Determination for Stage 2A 5. Which methods in the 20th edition and ii. Relative source contribution and Stage 2B TTHM and HAA5 MCLs 2003 On-Line Version of Standard iii. Water ingestion and body weight 1. What is EPA proposing today? Methods are proposed for approval? assumptions a. Stage 2A 6. What is the updated citation for EPA iv. MCLG calculation b. IDSE Method 300.1? v. Other considerations c. Stage 2B 7. How is the HAA5 sample holding time d. Feasibility of other options i. Subpart H systems serving 10,000 or being standardized? 3. Request for comment more people 8. How is EPA clarifying which methods B. MCLGs for THMs and HAAs ii. Subpart H systems serving 500 to 9,999 are approved for magnesium 1. What is EPA proposing today? people determinations? 2. How was this proposal developed? iii. Subpart H systems serving fewer than 9. Which methods can be used to a. Trichloroacetic acid 500 people demonstrate eligibility for reduced b. Monochloroacetic acid iv. Ground water systems serving 10,000 or bromate monitoring? 3. Request for comment more people 10. Request for comments C. Consecutive Systems v. Ground water systems serving fewer P. Laboratory Certification and Approval 1. What is EPA proposing today? than 10,000 people 1. What is EPA proposing today? a. Definitions vi. Consecutive systems 2. What changes are proposed for the PE b. Monitoring 2. How was this proposal developed? acceptance criteria? c. Compliance schedules a. Sampling intervals for quarterly 3. What minimum reporting limits are d. Treatment monitoring being proposed? e. Violations b. Reduced monitoring frequency 4. What are the requirements for analyzing f. Public notice and consumer confidence c. Different IDSE sampling locations by IDSE samples? reports disinfectant type 5. Request for comments g. Recordkeeping and reporting d. Population-based monitoring VI. State Implementation h. State special primacy conditions requirements for certain consecutive A. State Primacy Requirements for 2. How was this proposal developed? systems Implementation Flexibility 3. Request for comment 3. Request for comment B. State Recordkeeping Requirements D. MCLs for TTHM and HAA5 a. Proposed IDSE and Stage 2B monitoring C. State Reporting Requirements 1. What is EPA proposing today? requirements D. Interim Primacy 2. How was this proposal developed? b. Plant-based vs. population-based E. IDSE Implementation a. Definition of an LRAA monitoring requirements F. State Burden b. Consideration of regulatory alternatives i. Issues with plant-based monitoring VII. Economic Analysis c. Basis for the LRAA requirements A. Regulatory Alternatives Considered by d. Basis for phasing LRAA compliance ii. Approaches to addressing issues with the Agency e. TTHM and HAA5 as Indicators plant-based monitoring B. Rationale for the Proposed Rule Option 3. Request for comment J. Compliance Schedules 1. Reducing peak exposure E. Requirements for Peak TTHM and HAA5 1. What is EPA proposing? 2. Reducing average exposure Levels 2. How did EPA develop this proposal? C. Benefits of the Proposed Stage 2 DBPR 1. What is EPA proposing today? 3. Request for comments 1. Non-quantifiable health and non-health 2. How was this proposal developed? K. Public Notice Requirements related benefits 3. Request for comment 1. What is EPA proposing? 2. Quantifiable health benefits F. BAT for TTHM and HAA5 2. Request for comments 3. Benefit sensitivity analyses 1. What is EPA proposing today? L. Variances and Exemptions D. Costs of the Proposed Stage 2 DBPR 2. How was this proposal developed? 1. Variances 1. National cost estimates a. Basis for the BAT 2. What are the affordable treatment 2. Water system costs i. BAT analysis using the Information technologies for small systems? 3. State costs Collection Rule treatment studies M. Requirements for Systems to Use 4. Non-quantifiable ii. BAT analysis using the SWAT Qualified Operators E. Expected System Treatment Changes b. Basis for the Consecutive System BAT N. System Reporting and Recordkeeping 1. Pre-Stage 2 DBPR baseline conditions 3. Request for comment Requirements 2. Predicted technology distributions post- G. MCL, BAT, and Monitoring for Bromate 1. Confirmation of applicable existing Stage 2 DBPR 1. What is EPA proposing today? requirements F. Estimated Household Costs of the 2. How was this proposal developed? 2. Summary of additional reporting Proposed Rule a. Bromate MCL requirements G. Incremental Costs and Benefits of the b. Bromate in hypochlorite solutions 3. Request for comment Proposed Stage 2 DBPR c. Criterion for reduced bromate O. Analytical Method Requirements H. Benefits From the Reduction of Co- monitoring 1. What is EPA proposing today? Occurring Contaminants

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00005 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49552 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

I. Are there Increased Risks From Other representative of many other DBPs that information concludes that significant Contaminants? may also be present in the water; public health benefits may be achieved J. Effects on General Population and likewise, a reduction in TTHM and through further cost-effective reduction Subpopulation Groups HAA5 indicates a reduction of total of DBPs in distribution systems. K. Uncertainties in Baseline, Risk, Benefit, Congress required EPA to promulgate and Cost Estimates DBPs. L. Benefit/Cost Determination for the The Stage 2 DBPR is designed to the Stage 2 DBPR as part of the 1996 Proposed Stage 2 DBPR reduce the level of exposure from Safe Drinking Water Act (SDWA) M. Request for Comment disinfectants and DBPs without Amendments (section 1412(b)(2)(C)). VIII. Statutory and Executive Order Reviews undermining the control of microbial Today’s proposal reflects consensus A. Executive Order 12866: Regulatory pathogens. The Long Term 2 Enhanced recommendations from the Stage 2 Planning and Review Surface Water Treatment Rule Microbial/Disinfection Byproducts (M- B. Paperwork Reduction Act (LT2ESWTR) will be finalized and DBP) Federal Advisory Committee (the C. Regulatory Flexibility Act implemented simultaneously with the Advisory Committee). These D. Unfunded Mandates Reform Act Stage 2 DBPR to ensure that drinking recommendations are set forth in the M- E. Executive Order 13132: Federalism F. Executive Order 13175: Consultation water is microbiologically safe at the DBP Agreement in Principle (USEPA and Coordination with Indian Tribal limits set for disinfectants and DBPs. 2000g), which can be accessed on the Governments New information on health effects, edocket Web site (www.epa.gov/ G. Executive Order 13045: Protection of occurrence, and treatment has become edocket). Children from Environmental Health and available since the Stage 1 DBPR, which After considering the new occurrence Safety Risks supports the need for the Stage 2 DBPR. and health effects data and analyses, H. Executive Order 13211: Actions That Several reproductive and developmental EPA has determined that there is an Significantly Affect Energy Supply, studies have recently become available, opportunity to further reduce potential Distribution, or Use and EPA has completed a more risks from DBPs. The Stage 2 DBPR I. National Technology Transfer and Advancement Act extensive analysis of reproductive and being proposed today presents a cost- J. Executive Order 12898: Federal Actions developmental effects associated with effective, risk targeting approach to to Address Environmental Justice in DBPs since the Stage 1 DBPR. Both reduce risks from DBPs. The new Minority Populations or Low Income human epidemiology studies and requirements provide for more Populations animal toxicology studies have shown consistent protection from DBPs across K. Consultations with the Science associations between chlorinated the entire distribution system and the Advisory Board, National Drinking drinking water and reproductive and reduction of DBP peaks. New risk Water Advisory Council, and the developmental endpoints such as targeting provisions require only those Secretary of Health and Human Services spontaneous abortion, stillbirth, neural systems with the greatest risk to make L. Plain Language IX. References tube defects, pre-term delivery, capital improvements. The Stage 2 intrauterine growth retardation, and low DBPR, in conjunction with the I. Summary birth weight. New epidemiology and LT2ESWTR, will help public water toxicology studies evaluating bladder systems deliver safer water to A. Why Is EPA Proposing the Stage 2 and rectal cancers have also increased Americans with the benefits of DBPR? the weight of evidence linking these disinfection to control pathogens but The Environmental Protection Agency health effects to DBP exposure. The with fewer risks from DBPs. is committed to ensuring that all public large number of people (254 million B. What Does the Stage 2 DBPR Require? water systems provide clean and safe Americans) exposed to DBPs and the drinking water. Disinfectants are often identified potential cancer, The Stage 2 DBPR applies to an essential element of drinking water reproductive, and developmental risks community or nontransient treatment because of the barrier they played a significant role in EPA’s noncommunity water systems that add provide against harmful waterborne decision to move forward with a primary or residual disinfectant other microbial pathogens. However, regulatory changes that target lowering than ultraviolet light or deliver water disinfectants react with naturally DBP exposures beyond the requirements that has been treated with a primary or occurring organic and inorganic matter of the Stage 1 DBPR. residual disinfectant other than in source water and distribution systems While the Stage 1 DBPR provided a ultraviolet light. The TTHM and HAA5 to form disinfection byproducts (DBPs) major reduction in DBP exposure, new MCL values will remain the same as in that may pose health risks. The Agency national survey data suggest that some the Stage 1 DBPR, although compliance is proposing the Stage 2 Disinfectants customers are receiving drinking water calculations will be different. The and Disinfection Byproduct Rule with elevated, or peak DBP proposed Stage 2 DBPR includes new (DBPR) to reduce potential cancer, concentrations even when their MCLGs for chloroform, reproductive, and developmental risks distribution systems are in compliance monochloroacetic acid, and from DBPs. with the Stage 1 DBPR. Some of these trichloroacetic acid, but these new The Stage 2 DBPR augments the Stage peak concentrations can be substantially MCLGs do not affect the MCLs for 1 DBPR that was finalized in 1998. The greater than the Stage 1 DBPR maximum TTHM or HAA5. proposed Stage 2 DBPR focuses on contaminant levels (MCLs). The new The risk targeting components of the monitoring and reducing concentrations survey results also showed that Stage 1 Stage 2 DBPR will focus the greatest of two classes of DBPs: total DBPR monitoring sites may not be amount of change where the greatest trihalomethanes (TTHM) and haloacetic representative of peak DBP amount of risk may exist. The acids (HAA5). In part, these two groups concentrations that occur in distribution provisions of the Stage 2 DBPR focus on of DBPs are used as indicators of the systems. In addition, the new identifying and reducing exposure by various byproducts that are present in information indicates that cost-effective reducing DBP peaks in distribution disinfected water. This means that technologies including ultraviolet light systems. The first provision, designed to concentrations of TTHM and HAA5 are (UV) and granular activated carbon address significant variations in monitored for compliance, but their (GAC) may be very effective at lowering exposure, is the Initial Distribution presence in drinking water is DBP levels. EPA’s analysis of this new System Evaluation (IDSE). The purpose

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00006 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49553

of the IDSE is to identify Stage 2 DBPR TTHM and HAA5, respectively. In Stage based monitoring and how it might compliance monitoring sites for 2B, systems must comply with MCLs of affect systems complying with this rule. capturing peaks. Because Stage 2 DBPR 0.080 mg/L and 0.060 mg/L as LRAAs C. What Are the Economic Impacts of compliance will be determined at these for TTHM and HAA5, respectively, the Stage 2 DBPR? new monitoring sites, distribution based on sampling sites identified systems that identify elevated through the IDSE. A more detailed EPA quantified the potential benefits concentrations of TTHM and HAA5 will discussion of the proposed Stage 2 of the Stage 2 DBPR by estimating the need to make treatment or process DBPR MCL requirements can be found reduction in bladder cancer cases that changes to bring the system into in sections V.D., V.I., and V.J. of this may result from the decrease in average compliance with the Stage 2 DBPR. By preamble and in § 141.64(b)(2) and (3), DBP concentrations in disinfected identifying compliance monitoring sites and § 141.136, and subpart V of the rule water. Estimated reductions in DBP- with elevated concentrations of TTHM language. related bladder cancers (including both and HAA5, the IDSE will offer increased The IDSE and LRAA calculation will fatal and non-fatal cases) result in assurance that MCLs are being met lead to overall reductions in DBP annualized benefits ranging from $0 to across the distribution system. Both concentrations and reduce short term $986 million (using a three percent treatment changes and awareness of exposures to high DBP concentrations, discount rate), depending on the risk TTHM and HAA5 levels resulting from but even with this strengthened level assumed. There may also be a number of the IDSE will allow systems to better approach to regulating DBPs it will be important nonquantifiable benefits control for distribution system peaks. possible for individual DBP samples to associated with reducing DBPs in The IDSE is designed to offer exceed the MCLs when systems are in drinking water, the primary ones being flexibility to public water systems. The compliance with the Stage 2 DBPR. The reduced potential risk of adverse IDSE requires TTHM and HAA5 Stage 2 DBPR requires systems that reproductive and developmental effects monitoring for one year on a regular experience significant excursions to schedule that is determined by source including miscarriage, stillbirth, neural evaluate distribution system operational tube defects, heart defects, and cleft water type and system size. Systems practices and identify opportunities to have the option of performing a site- palate. Although a number of studies reduce DBP concentrations in the have found an association between specific study based on historical data, distribution system. This provision will water distribution system models, or reproductive and developmental curtail peaks and reduce exposure to endpoints and short-term exposure to other data; and waivers are available high DBP levels. Significant excursions under certain circumstances. The elevated DBP levels, a causal link has are discussed in greater detail in section not yet been established and proposed IDSE requirements are V.E. discussed in sections V.H., V.I., and V.J. information is not yet available to The Stage 2 DBPR also contains of this preamble and in subpart U of the quantify potential effects. As a result, provisions for regulating consecutive proposed rule. the Agency has not included an estimate The second provision of the Stage 2 systems, defined in the Stage 2 DBPR as of the potential benefits from reducing DBPR, which is designed to address public water systems that buy or reproductive and developmental risks in variations in temporal and spatial otherwise receive some or all of their its primary economic impact analysis of exposure, is the new compliance finished water from another public the Stage 2 DBPR. However, an calculation of the MCLs. The Stage 1 water system on a regular basis. illustrative calculation of potential fetal DBPR running annual average (RAA) Uniform regulation of consecutive loss risk is discussed in Section VII and calculation allows some locations systems provided by the Stage 2 DBPR presented in more detail in the within a distribution system to have will ensure that consecutive systems Economic Analysis (USEPA 2003i) to higher DBP annual averages than others deliver drinking water that meets illustrate the benefits that could be as long as the system-wide average is applicable DBP standards. More associated with this rule. Reduction in below the MCL. The Stage 2 DBPR will information on regulation of other cancers potentially associated base compliance on a locational running consecutive systems can be found in with DBP exposure represent additional annual average (LRAA) calculation sections V.C., V.H., V.I. and V.J. unquantified health benefits. where the annual average at each Today’s document proposes plant- EPA estimates the total annualized sampling location in the distribution based monitoring requirements for non- costs of the Stage 2 DBPR to be $54 to system will be used to determine consecutive systems and certain $64 million. This estimate includes compliance with the MCLs. The LRAA consecutive systems. Plant-based costs associated with treatment changes, will reduce exposures to peak DBP monitoring means that the number of the Initial Distribution System concentrations by ensuring that each compliance monitoring locations within Evaluation, changes in compliance monitoring site is in compliance with a distribution system is based on the monitoring, and rule implementation the MCLs as an annual average, and it number of plants, population served, activities for both public water systems will provide all customers drinking and type of source water used by the and States. EPA estimates that water that more consistently meets the distribution system. EPA is proposing approximately 2.8 percent of all plants MCLs. population-based monitoring for will need to convert to chloramines or EPA is proposing that systems comply consecutive systems that buy all their add advanced treatment to comply with with the Stage 2 DBPR MCLs in two finished water from other public water the Stage 2 DBPR. phases, designated as Stage 2A and systems. EPA is also requesting Table I–1 presents the estimated Stage 2B. In Stage 2A, beginning three comment on whether this approach quantified and unquantified benefits of years after the rule is final, all systems should be extended to all systems the Stage 2 DBPR and the estimated must comply with MCLs of 0.120 mg/L covered by today’s rule. Under a costs. Analyses of unquantified benefits for TTHM and 0.100 mg/L for HAA5 as population-based monitoring structure, suggest that the total benefits associated LRAAs at Stage 1 DBPR sampling sites, the number of compliance monitoring with the Stage 2 DBPR might be much in addition to continuing to comply locations is based only on the greater than these estimates. By with the Stage 1 DBPR MCLs of 0.080 population served and source water targeting risks and building on the solid mg/L and 0.060 mg/L as RAAs for type. Section V.I. describes population- foundation of the Stage 1 DBPR, the

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00007 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49554 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

Stage 2 DBPR will deliver cost-effective reductions in DBP levels and associated potential public health risks.

TABLE I–1.—COSTS AND BENEFITS OF THE STAGE 2 DBPR BASED ON ANNUALIZATION DISCOUNT RATE OF 3%

Costs Benefits Unquantified benefits

$54–64 M ...... $0–986 M Reduction in potential reproductive and developmental health effects, potential reduction in colon and rectal cancer, improved taste and odor of drinking water, control of contaminants that may be regulated in the future.

II. Background water by—(i) increasing the surface water. The SWTR includes A combination of factors have concentration of other contaminants in treatment technique requirements for influenced the development of the drinking water; or (ii) interfering with filtered and unfiltered systems that are proposed Stage 2 DBPR. These include the efficacy of drinking water treatment intended to protect against the adverse the initial 1992–1994 Microbial and techniques or processes that are used to health effects of exposure to Giardia Disinfection Byproduct (M–DBP) comply with other national primary lamblia, viruses, and Legionella, as well stakeholder deliberations and EPA’s drinking water regulations’’ (section as other pathogenic organisms. Stage 1 DBPR proposal; the 1996 Safe 1412(b)(5)(A)). When establishing an EPA also promulgated the Total Drinking Water Act (SDWA) MCL or treatment technique under this Coliform Rule (TCR) on June 29, 1989 Amendments; the 1996 Information authority, ‘‘the level or levels of (54 FR 27544)(USEPA 1989b) to provide Collection Rule; the 1998 Stage 1 DBPR; treatment techniques shall minimize the protection from microbial other new data, research, and analysis overall risk of adverse health effects by contamination in distribution systems of on disinfection byproduct (DBP) balancing the risk from the contaminant all types of public water supplies. The occurrence, treatment, and health effects and the risk from other contaminants TCR established an MCLG of zero for since the Stage 1 DBPR; and the Stage the concentrations of which may be total and fecal coliform bacteria, and an 2 DBPR Microbial and Disinfection affected by the use of a treatment MCL based on the percentage of positive Byproducts Federal Advisory technique or process that would be samples collected during a compliance Committee. The following shows how employed to attain the MCL or levels’’ period. Under the TCR, no more than 5 EPA arrived at this proposal for (section 1412(b)(5)(B)). percent of distribution system samples regulating disinfection byproducts. Finally, section 1412(b)(2)(C) of the collected in any month may contain Act requires EPA to promulgate a Stage coliform bacteria. A. What Is the Statutory Authority for 2 DBPR 18 months after promulgation of Together, the SWTR and the TCR the Stage 2 DBPR? the Long Term 1 Enhanced Surface were intended to address risks The SDWA, as amended in 1996, Water Treatment Rule (LT1ESWTR). associated with microbial pathogens authorizes EPA to promulgate a national Consistent with statutory requirements that might be found in source waters or primary drinking water regulation for risk balancing (section associated with distribution systems. (NPDWR) and publish a maximum 1412(b)(5)(B)), EPA will finalize the However, while reducing exposure to contaminant level goal (MCLG) for LT2ESWTR concurrently with the Stage pathogenic organisms, the SWTR also contaminants the Administrator 2 DBPR to ensure simultaneous increased the use of disinfectants in determines ‘‘may have an adverse effect protection from microbial and DBP some public water systems and, as a on the health of persons,’’ is ‘‘known to risks. result, exposure to DBPs in those occur or there is a substantial likelihood systems. that the contaminant will occur in B. What Is the Regulatory History of the In 1992, prompted by concerns about public water systems with a frequency Stage 2 DBPR? health risk tradeoffs between and at levels of public health concern,’’ The first rule to regulate DBPs was disinfection byproducts and microbial and for which ‘‘in the sole judgement of promulgated on November 29, 1979. pathogens, EPA initiated a negotiated the Administrator, regulation of such The Total Trihalomethanes Rule (44 FR rulemaking with a wide range of contaminant presents a meaningful 68624) (USEPA 1979) set an MCL of stakeholders. The negotiators included opportunity for health risk reduction for 0.10 mg/L for total trihalomethanes representatives of State and local health persons served by public water (TTHMs). Compliance was based on the and regulatory agencies, public water systems’’ (SDWA section 1412(b)(1)(A)). running annual average (RAA) of systems, elected officials, consumer MCLGs are non-enforceable health goals quarterly averages of all samples groups, and environmental groups. The set at a level at which ‘‘no known or collected throughout the distribution Regulatory Negotiating Committee met anticipated adverse effects on the health system. This TTHM standard applied from November 1992 through June 1993. of persons occur and which allows an only to community water systems using Following months of intensive adequate margin of safety’’. These surface water and/or ground water that discussions and technical analyses, the health goals are published at the same served at least 10,000 people and added Regulatory Negotiating Committee time as the NPDWR (sections 1412(b)(4) a disinfectant to the drinking water recommended the development of three and 1412(a)(3)). during any part of the treatment process. sets of rules: an Information Collection The Agency may also consider Under the Surface Water Treatment Rule, a two-staged approach for additional health risks from other Rule (SWTR) (54 FR 27486, June 29, regulating DBPs, and an ‘‘interim’’ contaminants and establish an MCL ‘‘at 1989) (USEPA 1989a), EPA set MCLGs Enhanced Surface Water Treatment Rule a level other than the feasible level, if of zero for Giardia lamblia, viruses, and (IESWTR) to be followed by a ‘‘final’’ the technology, treatment techniques, Legionella; and promulgated NPDWRs Enhanced Surface Water Treatment Rule and other means used to determine the for all public water systems using (USEPA 1996a, USEPA 1998c, USEPA feasible level would result in an surface water sources or ground water 1998d). EPA took the first step towards increase in the health risk from drinking sources under the direct influence of implementing this strategy by proposing

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00008 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49555

the Stage 1 DBPR and IESWTR in 1994. DBPs. Removal is achieved through After analyzing the data, the Advisory Congress affirmed the phased microbial enhanced coagulation or enhanced Committee reached three significant and disinfection byproduct rulemaking softening, unless a system meets conclusions that led the Advisory strategy in the 1996 SDWA alternative compliance criteria. Committee to recommending further Amendments by requiring that EPA EPA finalized the IESWTR at the same control of DBPs in public water systems. develop these three sets of rules on a time as the Stage 1 DBPR to ensure The data from the Information specific schedule that stipulates simultaneous compliance and address Collection Rule show that the RAA simultaneous promulgation of risk tradeoff issues. The IESWTR compliance calculation allows elevated requirements governing microbial applies to all water systems that use DBP levels to regularly occur at some protection and DBPs. surface water or ground water under the locations in the system when the overall In March 1997, the Agency direct influence of surface water that average at all locations is below the established the Microbial and serve at least 10,000 people. The MCL. Customers served at those Disinfection Byproduct (M–DBP) purpose of the IESWTR is to improve sampling locations that regularly exceed Advisory Committee under the Federal control of microbial pathogens in the MCLs are experiencing higher Advisory Committee Act (FACA) to drinking water, specifically the exposure compared to customers served collect, share, and analyze new protozoan Cryptosporidium. at locations that consistently meet the information and data available since the The Filter Backwash Recycle Rule MCLs. 1994 proposals of the Stage 1 DBPR and (FBRR) and the Long Term 1 Enhanced Second, the new data demonstrated the IESWTR, as well as to build Surface Water Treatment Rule how single samples can be substantially consensus on the regulatory (LT1ESWTR) round out the first group above the MCLs. The new information implications of the new information. of regulations balancing microbial and showed that it is possible for customers The Advisory Committee consisted of DBP risks. EPA promulgated the FBRR to receive drinking water with 17 members representing EPA, State and in 2001 (USEPA 2001c) and the concentrations of DBPs up to 75% above local public health and regulatory LT1ESWTR in 2002 (USEPA 2002b) to the MCLs even when their water system agencies, local elected officials, drinking increase protection of finished drinking is in compliance with the Stage 1 DBPR. water suppliers, chemical and water supplies from contamination by Studies have shown that DBP exposure equipment manufacturers, and public Cryptosporidium and other microbial during short, critical time windows may interest groups. The Advisory pathogens. The LT1ESWTR extends adversely impact reproductive and Committee met five times in March protection against Cryptosporidium and developmental health. through July 1997 to discuss issues other disease-causing microbes to water Third, data from the Information related to the IESWTR and the Stage 1 systems that use surface water or ground Collection Rule revealed that the highest DBPR. The Advisory Committee reached water under the direct influence of TTHM and HAA5 levels are not always consensus on a number of major issues surface water that serve fewer than located at the maximum residence time that were incorporated into the Stage 1 10,000 people. While the Ground Water monitoring sites specified by the Stage DBPR and the IESWTR. Rule, proposed in May 2000, (USEPA 1 DBPR. These sites were required for The Stage 1 DBPR and IESWTR, 2000h) will add significant protection monitoring by the Stage 1 DBPR because finalized in December 1998, were the from pathogens in vulnerable ground previous data suggested that water in first rules to be promulgated under the water systems, it does not pose as many the distribution system for the 1996 SDWA Amendments (USEPA risk-risk tradeoff considerations as the maximum residence time would have 1998c and 1998d). The Stage 1 DBPR surface water rules because only a small the highest TTHM levels. The fact that applies to all community and percentage of ground water systems the locations with the highest DBP nontransient noncommunity water subject to the Stage 2 DBPR have high levels varied in different public water systems that add a chemical disinfectant DBP levels. systems indicates that the Stage 1 DBPR to water. The rule established maximum EPA reconvened the Advisory monitoring sites may not be residual disinfectant level goals Committee in March 1999 to develop representative of the high DBP (MRDLGs) and enforceable maximum recommendations on issues pertaining concentrations that actually exist in residual disinfectant level (MRDL) to the Stage 2 DBPR and LT2ESWTR. distribution systems, and additional standards for three chemical The Advisory Committee collected, monitoring is needed to identify disinfectants—chlorine, chloramine, developed, and evaluated new distribution system locations with and chlorine dioxide; maximum information that became available after elevated DBP levels. This information contaminant level goals (MCLGs) for the Stage 1 DBPR was published. The encouraged the Advisory Committee to three THMs, two haloacetic acids Information Collection Rule provided recommend additional measures to (HAAs), bromate, and chlorite; and new data on DBP exposure, and control; identify locations with high LRAAs. enforceable maximum contaminant it also included new data on occurrence Section IV provides a complete level (MCL) standards for TTHM, five and treatment of pathogens. The discussion of the new occurrence data. haloacetic acids (HAA5), chlorite, and unprecedented amount of information The analysis of the new data also bromate calculated as running annual collected under the Information indicates that certain technologies are averages (RAAs). The Stage 1 DBPR uses Collection Rule was supplemented by a effective at reducing DBP TTHMs and HAA5 as indicators of the survey conducted by the National Rural concentrations. Bench- and pilot-scale various DBPs that are present in Water Association, data provided by studies for granular activated carbon disinfected water. Under the Stage 1 various States, the Water Utility (GAC) and membrane technologies DBPR, water systems that use surface Database (which contains data collected required by the Information Collection water or ground water under the direct by the American Water Works Rule provided information on the influence of surface water and use Association), and Information effectiveness of the two technologies. conventional filtration treatment are Collection Rule Supplemental Surveys. Other studies found UV light to be required to remove specified This large body of data allowed the highly effective for inactivating percentages of organic materials, Advisory Committee to reach new Cryptosporidium and Giardia at low measured as total organic carbon (TOC), conclusions regarding DBP exposure doses without promoting the formation that may react with disinfectants to form and new treatment options. of DBPs (Malley et al. 1996; Zheng et al.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00009 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49556 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

1999). This new treatment information LRAAs. RAA calculations average all water systems as they transition to added to the treatment options available samples collected within a distribution compliance with the Stage 2 DBPR to utilities for controlling DBPs beyond system over a one-year period, but while expanding public health the requirements of the Stage 1 DBPR. LRAA calculations average all samples protection. The proposed requirements New data on the health effects of taken at each individual sampling of the Stage 2 DBPR are described in DBPs also influenced the Advisory location in a distribution system during detail in section V of this preamble. Committee’s recommendation to further a one-year period. Systems would also regulate DBPs. Although bladder cancer carry out an Initial Distribution System C. How Were Stakeholders Involved in risks were the focus of the Stage 1 M– Evaluation (IDSE) to select new Developing the Stage 2 DBPR? DBP negotiations, potential compliance monitoring sites that more 1. Federal Advisory Committee Process reproductive and developmental health accurately reflect higher TTHM and The Stage 2 M–DBP Advisory effects were central to the Stage 2 M– HAA5 levels occurring in the Committee consisted of 21 DBP Advisory Committee discussions. distribution system. The second phase organizational members representing Recent human epidemiology studies of compliance would require MCLs of EPA, State and local public health and and animal toxicology studies have both 0.080 mg/L for TTHM and 0.060 mg/L regulatory agencies, local elected shown associations between chlorinated for HAA5 calculated as LRAAs at officials, Native American Tribes, large drinking water and reproductive and individual monitoring sites identified and small drinking water suppliers, developmental health effects such as through the IDSE. spontaneous abortion, stillbirth, neural The Agreement in Principle also chemical and equipment manufacturers, tube defects, pre-term delivery, provided recommendations for environmental groups, and other intrauterine growth retardation, and low simultaneous compliance with the stakeholders. Technical support for the birth weight. A critical review of the LT2ESWTR so that the reduction of Advisory Committee’s discussions was epidemiology literature pertaining to potential health hazards of DBPs does provided by a technical working group reproductive and developmental effects not compromise microbial protection. established by the Advisory Committee. of exposure to DBPs completed in 2000 The recommendations for the The Advisory Committee held ten (Reif et al. 2000) concluded that ‘‘the LT2ESWTR included treatment meetings to discuss issues pertaining to weight of evidence from the requirements for Cryptosporidium based the Stage 2 DBPR and LT2ESWTR from epidemiological studies also suggests on the results of source water September 1999 to July 2000 which that they [DBPs] are likely to be monitoring, a toolbox of options for were open to the public. There was also reproductive toxicants in humans under providing additional treatment at high an opportunity for public comment at appropriate exposure conditions * * * risk facilities, use of microbial each meeting. and that measures aimed at reducing the indicators to reduce Cryptosporidium In September 2000, the Advisory concentrations of byproducts could monitoring burden on small systems, Committee signed the Agreement in have a positive impact on public and future monitoring to determine if Principle, a full statement of the health.’’ source water quality remains constant consensus recommendations of the While there has been substantial after completion of initial monitoring. group. The agreement was published by research to date, the Advisory The Agreement also encouraged EPA to EPA in a December 29, 2000 Federal Committee recognized that significant develop guidance and criteria to Register notice (65 FR 83015), together uncertainty remains regarding the risk facilitate the use of UV light for with the list of committee members and associated with DBPs in drinking water. compliance with drinking water their organizations. The Agreement is The Advisory Committee carefully disinfection requirements. The complete divided into Parts A and B. The considered the analyses described text of the Agreement in Principle recommendations in each part stand previously, as well as costs and (USEPA 2000g) can be found at the alone and are independent of one potential impacts on public water edocket Web site (http://www.epa.gov/ another. The entire Advisory Committee systems, and concluded that a targeted edocket). reached consensus on Part A, which protective public health approach After extensive analysis and contains provisions that directly apply should be taken to address exposure to investigation of available data and rule to the proposed Stage 2 DBPR and DBPs beyond the requirements of the options considered by the Advisory LT2ESWTR. The full Advisory Stage 1 DBPR. After reaching this Committee, EPA is proposing a Stage 2 Committee, with the exception of the conclusion, the Advisory Committee DBPR control strategy that is consistent National Rural Water Association developed an Agreement in Principle with the key elements of the Agreement (NRWA), also agreed to Part B, which (USEPA 2000g) that laid out their in Principle signed in September 2000 has recommendations for future recommendations on how to further by the participants in the Stage 2 M– activities by EPA in the areas of control DBPs in public water systems. DBP Advisory Committee. EPA distribution systems and microbial In the Agreement in Principle, the determined that the risk-targeting water quality criteria. Advisory Committee recommended measures recommended in the 2. Other Outreach Processes maintaining the MCLs for TTHM and Agreement in Principle will require HAA5 at 0.080 mg/L and 0.060 mg/L only those systems with the greatest risk EPA received valuable input from respectively, but changing the to make treatment and operational small system operators as part of an compliance calculation in two phases to changes and will maintain simultaneous Agency outreach initiative under the facilitate systems moving from the protection from the potential health Regulatory Flexibility Act (RFA). EPA running annual average (RAA) hazards of DBPs and microbial also conducted outreach conference calculation to a locational running contaminants. EPA has carefully calls to solicit feedback and information annual average (LRAA) calculation. In evaluated and expanded upon the from Small Entity Representatives the first phase, systems would continue recommendations of the Advisory (SERs) on issues related to Stage 2 DBPR to comply with the Stage 1 DBPR MCLs Committee to more fully develop impacts on small systems. The Agency as RAAs and, at the same time, comply today’s proposal. EPA also made consulted with State, local, and Tribal with MCLs of 0.120 mg/L for TTHM and simplifications where possible to governments on the proposed Stage 2 0.100 mg/L for HAA5 calculated as minimize complications for public DBPR. Section VIII includes a complete

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00010 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49557

description of the many stakeholder TTHM Rule in 1979 and the Stage 1 information EPA analyzed, some activities which contributed to the DBPR in 1998. The Stage 1 DBPR conclusions of these studies and reports, development of the Stage 2 DBPR. primarily addressed possible and implications for the Stage 2 DBPR. The Agency held two meetings to carcinogenic effects (e.g., bladder, colon Further discussion of the implications discuss consecutive system issues and rectal cancers) reported in both and EPA’s conclusions can be found in relevant to the proposal (February 22– human epidemiology and laboratory the Stage 2 Economic Analysis (USEPA 23, 2001 in Denver, CO and March 28, animal studies. Since the Stage 1 DBPR, 2003i). 2001 in Washington, DC). new health studies continue to support EPA has evaluated recently published Representatives from States, EPA an association between bladder, colon epidemiological studies examining the Regions, and public water systems and rectal cancers from long-term relationship between exposure to participated in the discussions. EPA exposure to chlorinated surface water. contaminants in chlorinated surface also briefed the National Drinking Water In addition to cancer effects, recent water and adverse reproductive and Advisory Committee at their November studies have reported associations developmental outcomes. EPA also 2001 meeting on consecutive system between use of chlorinated drinking considered critical reviews of the issues associated with the rule to water and a number of reproductive and epidemiological literature by Reif et al. receive input on the implementation developmental endpoints including (2000), Bove et al. (2002), and strategy selected. This Advisory spontaneous abortion, still birth, neural Nieuwenhuijsen et al. (2000). Based on Committee generally supported EPA’s tube defect, pre-term delivery, low birth these evaluations, EPA believes that the approach. Section V describes EPA’s weight and intrauterine growth reproductive and developmental analysis of consecutive system issues, retardation (small for gestational age). epidemiology data contribute to the comments and input received during Short-term, high-dose animal screening weight of evidence on the potential these sessions, and how the proposed studies on individual byproducts (e.g., health risks from exposure to requirements will apply to consecutive bromodichloromethane (BDCM), and chlorinated drinking water. Although systems. EPA also consulted with the certain haloacetic acids) have also the data are not suitable for a Science Advisory Board in December reported adverse reproductive and quantitative risk assessment at this time, 2001 on the requirements of the Stage 2 developmental effects (e.g., whole litter due in part to inconsistencies in the DBPR. resorption, reduced fetal body weight) findings, they do suggest that exposure Finally, EPA posted a pre-proposal that are similar to those reported in the to DBPs is a potential reproductive and draft of the Stage 2 DBPR preamble and human epidemiology studies. This developmental health hazard. regulatory language on an EPA Internet section discusses the new studies that 1. Reif et al. 2000 site (http://www.epa.gov/safewater/ have become available since mdbp/st2dis.html) on October 17, 2001. promulgation of the Stage 1 DBPR and Reif et al. (2000) completed a critical This public review period allowed how they contribute to the weight of review of the epidemiology literature readers to comment on the Stage 2 evidence for an association between pertaining to reproductive and DBPR’s consistency with the Agreement health effects and exposure to developmental effects of exposure to in Principle of the Stage 2 M–DBP chlorinated surface water. disinfection byproducts in drinking Advisory Committee. EPA received While the Stage 1 DBPR was targeted water as a report to Health Canada. The important suggestions on this pre- primarily at reducing long-term review focused on 16 peer-reviewed proposal draft from 14 commenters exposures to elevated levels of DBPs to scientific manuscripts and published which included public water systems, address chronic health risks from reports and evaluated associations State governments, laboratories, and cancer, the Stage 2 DBPR targets between DBP exposure and outcomes other stakeholders. While EPA will not reducing short-term exposures to grouped as effects on: (1) Fetal growth— formally respond to these comments, address potential reproductive and low birth weight (<2500g); very low EPA has carefully considered them in developmental health risks and cancer birth weight (<1500g); preterm delivery developing today’s proposal. risks. (<37 weeks of gestation) and Based on the weight of evidence from intrauterine growth retardation (or small III. Public Health Risk both the human epidemiology and for gestational age); (2) fetal viability Chlorine has been widely used as a animal toxicology data on cancer and (spontaneous abortion and stillbirth) chemical disinfectant, serving as a reproductive and developmental health and (3) fetal malformations (all principal barrier to microbial effects and consideration of the large malformations, oral cleft defects, major contaminants in drinking water. number of people exposed to cardiac defects, neural tube defects, and However, the microbial risk reduction chlorinated byproducts in drinking chromosomal abnormalities). attributes of chlorination have been water (approximately 254 million), EPA a. Fetal growth. Reif et al. (2000) increasingly scrutinized due to concerns concludes that: (1) Current reproductive found inconsistent epidemiological about potential increased health risks and developmental health effects data evidence for an association between from exposure to disinfection support a hazard concern, (2) new DBPs and fetal growth. Some studies byproducts, which are formed when cancer data strengthens the evidence of found weak but statistically significant certain disinfectants interact with an association of chlorinated water with associations (Gallagher et al. 1998; Bove organic and inorganic material in source bladder cancer and suggests an et al. 1992 and 1995), while two studies waters. Since the discovery of association for colon and rectal cancers, found no association (Dodds et al. 1999; chlorination byproducts in drinking and (3) the combined health data and Savitz et al. 1995) with fetal growth. water in 1974, numerous toxicological warrant regulatory action beyond the b. Fetal viability. Reif et al. 2000’s studies have shown several DBPs (e.g., Stage 1 DBPR. review of the literature found bromodichloromethane, bromoform, inconsistencies in the epidemiological chloroform, dichloroacetic acid, A. Reproductive and Developmental evidence for the association between trichloroacetic acid and bromate) to be Epidemiology DBP exposure and fetal viability. For carcinogenic in laboratory animals. The following section briefly instance, the study by Waller et al. 1998 These findings of carcinogenicity discusses reproductive and found an apparent dose-dependent influenced EPA to promulgate the developmental epidemiology increase in rates of spontaneous

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00011 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49558 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

abortions associated with TTHMs in effects on (1) fetal growth—small for found an association between cardiac California. On the other hand, Savitz et gestational age (SGA) as defined in each defects and TTHM, but Dodds et al. al. (1995) found little evidence of an study (usually defined as the fifth or 1999, 2001 and Shaw et al. 1991 did association using either the tenth percentile weight by gestational not. An association between concentration of TTHM ≥81 µg/L or a week of birth); (2) fetal viability— chlorination and urinary tract defects dose estimate based on the amount of spontaneous abortion and stillbirth; and was found in the three studies that tap water consumed. An increased risk (3) fetal malformations (neural tube evaluated that endpoint (Ka¨lle´n et al. of stillbirth was reported for women in defects, oral clefts, and cardiac defects). 2000; Magnus et al. 1999; Aschengrau et Nova Scotia by Dodds et al. 1999, but a. Fetal growth. Bove et al. found that, al. 1993). in New Jersey, Bove et al. (1992, 1995) although the studies that evaluated SGA Bove et al. (2002) concluded that the found little evidence of an association had several limitations, three studies current reproductive and developmental with TTHM at 80 µg/L, but did report out of eight (Kramer et al. 1992, Bove et epidemiological database for exposure a weak association between stillbirth al. 1995, and Gallagher et al. 1998) to chlorinated byproducts in drinking and use of surface water systems. ‘‘provided moderate evidence for a water presents moderate evidence for Aschengrau et al. (1993) found an causal relationship between a narrow associations between DBP exposure and association between stillbirth and the definition of SGA * * * and TTHM SGA, neural tube defects and use of a chlorinated vs. chloraminated levels that could be found currently in spontaneous abortion. The authors surface water supply, but not for some U.S. public water systems.’’ They acknowledged the difficulties in exposure to surface water. also concluded that the study with the assessing exposure with any precision c. Fetal malformations and other best exposure assessment found the in the studies reviewed, but held the developmental anomalies. Reif et al. strongest association between SGA and opinion that misclassification of (2000) considered the data for TTHM exposure (Gallagher et al. 1998). exposure would tend to underestimate congenital anomalies to be inconsistent One study found a very weak rather than overestimate the risk. across the six studies that have explored association (Dodds et al. 1999) and the 3. Nieuwenhuijsen et al. 2000 these outcomes. For example, two of the other four did not observe an four studies on neural tube defects association (Yang et al. 2000, Kanitz et Nieuwenhuijsen et al. (2000) (Bove et al. 1995; Magnus et al. 1999) al. 1996, Kallen et al. 2000, and Jaakkola reviewed the toxicological and reported significant excess risks, but the et al. 2001). epidemiological literature and evaluated remaining two studies (Dodds et al. b. Fetal viability. Bove et al. evaluated the potential risk of chlorination DBPs 1999; Klotz and Pyrch et al. 1999) did three studies on spontaneous abortion on human reproductive health. The not. These studies found lower risks or and three studies on stillbirth. Again, authors state that ‘‘some studies have no evidence of an association with Bove et al. found that the study shown associations for DBPs and other TTHM. However, those studies were employing the best methods found the outcomes such as spontaneous conducted in locations with either very strongest association between TTHM abortions, stillbirths and birth defects, low or high concentrations of DBPs exposure and spontaneous abortions and although the evidence for these which may have limited the contrast in (Waller et al. 1998). The other two associations is weaker it is gaining exposures, thereby reducing the ability studies (Savitz et al. 1995 and weight.’’ Nieuwenhuijsen et al. also to detect increased risks. An assessment Aschengrau et al. 1989) found weak concluded that, ‘‘although studies report of congenital anomalies is also difficult associations. Two of the studies small risks that are difficult to interpret, due to the relatively small number of investigating stillbirths found an the large number of people exposed to cases available for evaluation. association between stillbirths and chlorinated water supplies constitutes a Overall, Reif et al. (2000) conclude chlorinated surface water (Dodds et al. public health concern.’’ that the weight of evidence from the 2001 and Aschengrau et al. 1993). The epidemiological studies suggest that third study (Bove et al. 1995) found no 4. Additional Epidemiology Studies ‘‘DBPs are likely to be reproductive association, however this study did not Three new reproductive and toxicants in humans under appropriate evaluate individual THM levels or cause developmental epidemiological studies exposure conditions.’’ Reif et al. of death information. were completed that were not included comment that data from animal studies c. Fetal malformations. Bove et al. in the Reif et al. 2000, Bove et al. 2002, of individual DBPs provide biological evaluated seven studies that or Nieuwenhuijsen et al. 2000 literature plausibility for the effects observed in investigated the relationship between reviews. epidemiological studies. Although the birth defects and DBP exposure. This Waller et al. 2001, recalculated the authors recognize that the ‘‘data are evaluation found ‘‘consistency among total trihalomethane exposures from primarily at the stage of hazard these studies in the findings for neural their original publication (Waller et al. identification,’’ they conclude that tube defects and oral cleft defects, but 1998) to evaluate two exposure ‘‘measures aimed at reducing the not for cardiac defects. Associations assessment methods (closest site and concentrations of byproducts could were found for neural tube defects in all utility-wide average). The new have a positive impact on public three studies that examined neural tube calculations were intended to reduce health.’’ defects. These studies also evaluated exposure misclassification by levels of THM exposure (Bove et al. employing weighting factors and subset 2. Bove et al. 2002 1995; Dodds et al. 1999; Klotz et al. analyses. As in the 1998 publication, the Bove et al. (2002) conducted a 1999).’’ Two studies evaluated oral cleft new methods found a relationship qualitative review of 14 epidemiological defects and levels of THMs; one found between spontaneous abortion and THM studies that evaluated possible an association with TTHM (Bove et al. exposure, although the unweighted developmental and reproductive 1995) and the other found an utility-wide point estimate was lower endpoints associated with exposure to association with chloroform (Dodds et than reported in the original chlorination byproducts in drinking al. 2001). A third study that did not manuscript. water. Similar to Reif et al., Bove et al. evaluate THM levels did not identify an Hwang et al. 2002, assessed the effect evaluated associations between DBP association with oral cleft defects of water chlorination byproducts on exposure and outcomes grouped as (Jaakkola et al. 2001). Bove et al. 1995 specific birth defects in Norway by

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00012 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49559

classifying exposure on the basis of The Agency is supporting several published after the 1998 Stage 1 DBPR, chlorination (yes/no) and amount of studies using improved study designs to are summarized in the updated natural organic matter in the water. provide better information for children’s health document, ‘‘Health Statistically significant associations characterizing potential risks. Details on Risks to Fetuses, Infants, and Children: with exposure were found for risks of EPA’s epidemiology research program A Review’’ (USEPA 2003a). any birth defect, cardiac, respiratory, can be found at http:// In addition to this compilation of and urinary tract defects. For specific cfint.rtpnc.epa.gov/dwportal/cfm/ data, EPA has also prepared individual birth defects, a statistically significant dwMDBP.cfm. health criteria documents that provide association was found for a defect of the detailed summaries of the relevant new B. Reproductive and Developmental septum in the heart. information, as well as an overall Toxicology Windham et al., 2003, assessed the characterization of the human health relationship between exposure to THMs Several new reproductive and risks from exposure to certain DBPs in drinking water and characteristics of developmental toxicology studies have (USEPA 2003b-USEPA 2003h, USEPA the menstrual cycle among 403 women become available since the December 2003l). From these new evaluations, who provided daily urine samples for 1998 Stage 1 DBPR. This discussion EPA has concluded that several new an average of 5.6 cycles. Women whose presents some conclusions derived from studies on individual byproducts tap water had TTHM levels more than these studies and reports, including contribute to the weight of evidence for 0.060 mg/l had statistically significantly hazard identification, as well as an association between DBP exposure shorter menstrual cycles than women implications for the Stage 2 DBPR. and adverse effects on the developing whose tap water had lower TTHMs. On EPA conducted a literature search of fetus and reproduction. These effects average, the menstrual cycles of women animal toxicology studies on chronic include fetal loss, cardiovascular effects, with the higher levels of TTHMs were and subchronic DBP exposures and male reproductive effects and are one day shorter than cycles of women associated with reproductive and associated with bromodichloromethane with the lower levels (adjusted developmental health effects, evaluated (BDCM), dichloroacetic acid (DCAA), difference: ¥1.1 days, 95% confidence the current reproductive and trichloroacetic acid (TCAA), interval: ¥1.8 days to ¥0.4 days). This developmental toxicological database bromochloroacetic acid (BCAA), and shortening occurred during the first half for several individual DBPs, and dibromoacetic acid (DBAA). The data of the cycle, before ovulation (adjusted assessed two independent reviews (Tyl from these new studies do not change difference: ¥0.9 days; 95% confidence 2000 and WHO 2000). As a result of the MCLGs that were established as a interval: ¥1.6 days to ¥0.2 days). There these analyses, EPA has concluded that part of the Stage 1 DBPR. were no changes in bleed length or in although the database is not strong 2. Tyl 2000 the regularity of the cycles. Based on enough to quantify risk, it is sufficient their study, Windham et al., 2003, to support a hazard concern. This Tyl (2000) conducted a suggested that THM exposure may affect hazard concern supports the need to comprehensive review of the ovarian function, but since this is the address potential reproductive and reproductive and developmental first study to examine human menstrual developmental health effects in the toxicology literature on DBPs cycle variation in relation to THM Stage 2 DBPR. The following section representing over thirty-five studies. exposure, more research is needed to describes how this conclusion was Adverse effects reported by these confirm the relationship. The public reached. studies include developmental effects, health implication of a small reduction whole litter resorption, reduced fetal 1. EPA Analysis and Research in menstrual cycle length is not clear, body weights, and male reproductive but if THMs are related to disturbances Since the Stage 1 DBPR, EPA has effects (e.g., inhibited spermiation, in ovarian function, that might provide continued to support reproductive and increased abnormal sperm). Many of insight into the observed associations developmental toxicological research on these studies are categorized as high- between THMs and a variety of adverse various disinfection byproducts through dose, short-term screening studies that reproductive outcomes. extramural and intramural research can be used to assess potential hazard EPA’s epidemiology research program programs. Information on EPA’s (Table III–1), while the long term, two- continues to examine the relationship toxicology programs can be found at generation reproduction studies could between exposure to DBPs and adverse http://www.epa.gov/nheerl/. These be an appropriate basis for quantitative developmental and reproductive effects. studies, along with data on several DBPs risk assessment.

3 1 2 Two-generation Disinfectant/DBP Screening Developmental reproductive

Chlorine ...... ✔ Chlorine Dioxide ...... ✔✔ Chloramine ...... ✔ Chloroform ...... ✔✔✔ Bromoform ...... ✔✔✔ Bromodichloromethane ...... ✔✔in progress Dibromochloromethane ...... ✔✔ Monochloroacetic acid ...... ✔✔ Dichloroacetic acid ...... ✔✔ Trichloroacetic acid ...... ✔✔ Monobromoacetic acid ...... ✔✔ Dibromoacetic acid ...... ✔✔in progress Tribromoacetic acid ...... ✔ ...... Bromochloroacetic acid ...... ✔ ...... in planning stage Bromodichloroacetic acid ...... ✔ ...... Dibromochloroacetic acid ...... ✔ ......

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00013 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49560 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

3 1 2 Two-generation Disinfectant/DBP Screening Developmental reproductive

Chloroacetonitrile ...... ✔ ...... Dichloroacetonitrile ...... ✔✔ Trichloroacetonitrile ...... ✔✔ Bromoacetonitrile ...... ✔✔ Dibromoacetonitrile ...... ✔ ...... Tribromoacetonitrile ...... Bromochloroacetonitrile ...... ✔✔ Propanal ...... ✔✔ 1,1 Dichloropropanone ...... ✔ ...... Hexachloropropanone ...... ✔ ...... Dichloromethane ...... ✔ ...... MX ...... ✔✔ Bromate ...... ✔ ...... Chlorite ...... ✔✔✔ ✔ denotes the availability of at least one study in the following categories. 1 Screening studies are for hazard identification. These types of studies include the following: whole embryo culture, NTP 35-day screening studies, Chernoff-Kavlock and its modified version, and short-term male reproductive toxicity screen. 2 Developmental studies are used for dose-response determinations. 3 Two-generation reproductive studies are multi-generation reproductive toxicity studies used for dose-response determinations.

Tyl concluded that, ‘‘The screening developmental toxicants for hazard developmental effects (Table III–2) and studies, performed for a number of identification.’’ Tyl further confirms these are discussed further in the next DBPs, are ‘adequate’ and ‘sufficient’ that the database identifies certain DBPs section. only to detect potent reproductive/ with potential reproductive or

TABLE III–2.—POTENTIAL HAZARDS OF DBPS FOR REPRODUCTIVE AND DEVELOPMENTAL EFFECTS (ADAPTED FROM TYL, 2000)

Type of hazard Disinfection byproducts

Developmental defects ...... TCAA, DCAA, MCAA and chlorite. Whole litter resorption ...... Chloroform, bromoform, BDCM, DBCM, DCAA, TCAA, DCAN, and TCAN. Fetotoxicity (reduced fetal body weights, increased variations) ...... Chloroform, BDCM, DBCM, DCAA, TCAA, DCAN, TCAN, DBAN, BCAN, MCAN. Male reproductive effects (spermatotoxic) ...... DCAA, DBAA, BDCM.

a. Developmental defects. Tyl noted TCAA, DCAN, TCAN, DBAN, BCAN) From her review of the that adverse developmental effects that (Thompson et al. 1974; Schwetz et al. comprehensive animal toxicology were reported from whole embryo 1974; Murray et al. 1979; Ruddick et al. database on reproductive and culture tests on the developing heart, 1983; Narotsky et al. 1992, Balster and developmental health effects from DBP neural tube, eye, pharyngeal arch, and Borzelleca, 1982; Smith et al. 1990). exposure, Dr. Tyl concludes that ‘‘some somites tended to be associated with Again, Tyl noted a similarity in effects DBPs have an intrinsic capacity to do haloacetic acids tested at high doses observed in epidemiology studies. harm, specifically to the developing (Hunter et al. 1996; Saillenfait et al. d. Male reproductive effects. Animal conceptus and the male (and possibly 1995, Smith et al. 1989). Cardiovascular toxicology studies report increased risks the female) reproductive system’’. She effects were also observed in vivo for of adverse effects on the male concludes that ‘‘there is hazard to TCAA and DCAA from developmental reproductive system from high doses of development from the haloacetic acids segment II toxicity studies at high doses haloacetic acids and other DBPs that (TCAA, DCAA, MCAA) and acetate; to (Smith et al. 1988, 1990). have not been studied in human development from chloroform, b. Whole litter resorption. Whole litter epidemiology studies. Male bromoform, BDCM, DBCM, DCAA, resorption, likened to miscarriage or reproductive effects (e.g., inhibited TCAA, DCAN, and TCAN expressed as spontaneous abortion by Tyl 2000, was spermiation, reduced epididymus, also observed at high doses in vivo for sperm number and motility, increased full litter resorption (which most likely a range of DBPs as indicated in Table abnormal sperm, testicular damage and indicates maternal endocrine/uterine III–2 (Murray et al. 1979, Balster and inhibited in vitro fertilization) were effects); and fetotoxicity for chloroform, Borzellca, 1982, Narotsky et al. 1992; reported for DCAA, DBAA, TCAA and BDCM, DBCM, DCAA, TCAA, DCAN, 1997 a, b; Bielmeier et al. 2001; Smith BDCM (Toth et al. 1992, Linder et al. TCAN, DBAN, BCAN, CAN, et al. 1990; Smith et al. 1988). Tyl noted 1997a, b; Linder et al. 1994a, b; Cosby acetaldehyde, and possibly that similar effects were observed in and Dukelow 1992). Dr. Tyl noted that formaldehyde. Reproductive hazard several epidemiology studies. the adverse effects observed in the male exists for DCAA, DBAA, and possibly c. Fetal toxicity. Fetal toxic effects reproductive toxicity screening studies formaldehyde in males and for TCE and such as reduced fetal body weights and (Toth et al. 1992; Linder et al. 1994a, b; possibly formaldehyde in females.’’ increased variation were observed at 1997a, b) are confounded by a short high doses in vivo for a range of DBPs dosing regimen and administration of (e.g., chloroform, BDCM, DBCM, DCAA, test doses to only adult males.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00014 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49561

3. World Health Organization Review of (bones of the hand or the foot) and the the percent body weight loss after the the Reproductive and Developmental hindlimb metatarsals and phalanges. first day of dosing was comparable for Toxicology Literature (2000) These concentrations correspond to SD rats and the F344 rats that resorbed The International Programme on mean consumed doses of 45.0 mg/kg- their litters. The rats were allowed to Chemical Safety (IPCS) published an day and 82.0 mg/kg-day, respectively. deliver and pups were examined on Christian et al. (2002b) summarized evaluation of Disinfectants and DBPs in postnatal days 1 and 6. Surviving litters the results of a two-generation its Environmental Health Criteria appeared normal and no effect on post- reproductive toxicity study on monograph series (WHO 2000). In this natal survival, litter size, or pup weight bromodichloromethane conducted in review of the toxicology data on was observed. The series of experiments Sprague-Dawley rats. reproductive and developmental effects conducted by Bielmeier et al. (2001) Bromodichloromethane was identified a LOAEL of 75 mg/kg-day (the from DBP exposure, the World Health continuously provided to test animals in lowest dose tested) based on FLR in Organization (WHO) concludes that the drinking water at concentrations of F344 rats. A NOAEL was not identified. although the data on these effects are 0, 50, 150, or 450 ppm. Average daily Mechanistic studies indicate that not as robust as the cancer database, doses estimated for the 50, 150, and 450 BDCM-induced pregnancy loss is likely these effects are of potential health ppm concentrations were reportedly 4.1 to be luteinizing hormone (LH)- concern. The WHO concludes that to 12.6, 11.6 to 40.2, and 29.5 to 109 mg/ mediated (Bielmeier et al., 2001). It is reproductive effects in females have kg-day, respectively. The parental possible that BDCM alters LH levels by been principally embryolethality and NOAEL and LOAEL were 50 and 150 disrupting the hypothalamic-pituitary- fetal resorptions associated with the ppm, respectively, based on statistically gonadal axis or by altering the haloacetonitriles (trichloroacetonitrile, significant reduced body weight and responsiveness of the corpora lutea to dichloroacetonitrile, body weight gain; F1 and F2 generation LH. Since these possible mechanisms bromochloroacetonitrile, and pup body weights were reduced in the are potentially relevant to pregnancy dibromoacetonitrile) and the 150 and 450 ppm groups during the maintenance in humans, EPA believes dihaloacetates, while DCAA and DBAA lactation period after the pups began to the finding of BDCM-induced pregnancy have both been associated with adverse drink the water provided to the dams. loss in F344 rats is relevant to risk effects on male reproduction. Body weight and body weight gain were assessment, and may provide insight 4. New Studies also reduced in the 150 and 450 ppm F1 into the epidemiological finding of generation males and females. A increased risk of spontaneous abortion Christian et al. (2001) conducted a marginal effect on estrous cyclicity was associated with consumption of BDCM developmental toxicity study with observed in F1 females in the 450 ppm (Waller et al. 1998, 2001). pregnant New Zealand White rabbits exposure group. Small (≤6%), but Christian et al. (2002a) recently exposed to BDCM in drinking water at statistically significant, delays in F1 completed a two-generation drinking concentrations of 0, 15, 150, 450, and generation sexual maturation occurred water study of DBA in rats. Male and 900 ppm in drinking water on gestation at 150 (males) and 450 ppm (males and female Sprague-Dawley rats (30/sex/ days 6–29. The no observed adverse females) as determined by timing of exposure group) were administered effect level (NOAEL) and lowest vaginal patency or preputial separation. DBA in drinking water at concentrations observed adverse effect level (LOAEL) The study’s authors considered these of 0, 50, 250, or 650 ppm continuously identified for maternal toxicity in this effects to be a secondary response from initiation of exposure of the study were 13.4 mg/kg-day (150 ppm) associated with reduced body weight, parental (P) generation male and female and 35.6 mg/kg-day (450 ppm), which appears to be dehydration rats through weaning of the F2 offspring. respectively, based on decreased body brought about by taste aversion to the Based on testicular histomorphology weight gain. The developmental NOAEL compound. The results of this study indicative of abnormal spermatogenesis was 55.3 mg/kg-day (900 ppm) based on identify NOAEL and LOAEL values for in P and F1 males, the parental and absence of statistically significant, dose- reproductive effects of 50 ppm (4.1 to reproductive/developmental toxicity related effects at any tested 12.6 mg/kg-day) and 150 ppm (11.6 to LOAEL and NOAEL are 250 and 50 concentration. Christian et al. (2001) 40.2 mg/kg-day), respectively, based on ppm, respectively. also conducted a developmental study delayed sexual maturation. Previous studies by EPA have of BDCM in a second species, Sprague- Bielmeier et al. (2001) conducted a reported adverse effects of DBA, Dawley rats. Rats were exposed to series of experiments to investigate the administered via oral gavage, on BDCM in the drinking water at mode of action in spermatogenesis that impacted male concentrations of 0, 50, 150, 450, and bromodichloromethane-induced full fertility (Linder et al. 1994a, 1995, 900 ppm on gestation days 6 to 21. The litter resorption (FLR). The study 1997a) at doses-comparable to those concentration-based maternal NOAEL included a strain comparison of F344 achieved in the Christian et al. (2002a) and LOAEL for this study were 150 ppm and Sprague-Dawley (SD) rats. In the study. Based on these studies and 450 ppm, respectively, based on strain comparison experiment, female collectively, it is clear that DBA is statistically significant, persistent SD rats (13 to 14/dose group) were spermatotoxic. Moreover, reductions in maternal body weight and dosed with 0, 75, or 100 mg/kg-day by Veeramachaneni et al. (2000) reported body weight gains. Based on the mean aqueous gavage in 10% Emulphor on in an abstract that sperm from male consumed dosage of GD 6 to 10. F344 rats (12 to 14/dose rabbits exposed to DBA in utero from bromodichloromethane, these group) were dosed with 0 or 75 mg/kg- gestation days 15 and throughout life concentrations correspond to doses of day administered in the same vehicle. reduced the fertility of artificially 18.4 mg/kg-day and 45.0 mg/kg-day, The incidence of FLR in the inseminated females as evidenced by respectively. The concentration-based bromodichloromethane-treated F344 reduced conceptions. When published, developmental NOAEL and LOAEL rats was 62%, while the incidence of this study may support the evidence were 450 ppm and 900 ppm, FLR in SD rats treated with 75 or 100 that DBA is a male reproductive system respectively, based on a significantly mg/kg-day of bromodichloromethane toxicant . decreased number of ossification sites was 0%. Both strains of rats showed In addition, research on DBA by per fetus for the forelimb phalanges similar signs of maternal toxicity, and Klinefelter et al. (2001) has

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00015 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49562 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

demonstrated statistically significant toxicological studies. Tyl 2000 states 1. Population Attributable Risk Analysis delays in both vaginal opening and that ‘‘effects observed in animal studies Some epidemiological studies have preputial separation using the body included embryonic heart and neural linked the consumption of chlorinated weight on the day of acquisition tube defects from haloacetic acids in surface waters to an increased risk of (postnatal day 45) as the co-variant. This vitro and in vivo, and full litter two major causes of human mortality in was not found by Christian et al (2002a) resorption, reduced numbers of the United States, colorectal and using the body weight at weaning as the implants per litter, and reduced fetal bladder cancers (Cantor 1998). Bladder statistical covariant. However, the body weight per litter were also cancer was chosen as the primary authors analyzed the data for preputial observed from exposure to specific endpoint of concern in the Stage 1 separation and vaginal opening with trihalomethanes. Comparable effects DBPR (USEPA 1998f) economic analysis body weight on the day of weaning as were also observed in children in some because it had the most consistent a co-variant rather than body weight on (but not all) epidemiological studies, database for a possible association to the day of acquisition, i.e., the day that with exposure to trihalomethanes chlorinated surface water exposure. the prepuce separates or the day the (THMs) usually used as a surrogate for More studies have considered bladder vagina opens. It is likely that there was specific DBP classes or individual DBPs, cancer than any other cancer. EPA used an increase in body weight from as follows: increased incidences of the published mean risk estimates from postnatal day 21 (weaning) until cardiac defects (Bove et al. 1995) and of five studies to quantify the potential preputial separation (day 45) that was neural tube defects in children (Bove et range of risk for bladder cancer from independent of the delay in sexual al. 1995; Dodds et al. 1999; Klotz and DBP exposure. These risks were maturation. Pyrch 1998) were reported. Intrauterine expressed as a range of population Although the Christian et al. (2002a) growth retardation (IUGR, attributable risks (PAR) of 2–17% study was conducted in accordance approximately equivalent to reduced (USEPA 1998f). This means that if the with EPA’s 1998 testing guidelines, EPA fetal body weights per litter) was has incorporated newer, more reported to be associated with associations reported in the studies turn sophisticated measures into recent waterborne chloroform (Kramer et al. out to reflect a causal link, between 2 intramural and extramural studies that 1992; Bove et al. 1995; Gallagher et al. and 17% of new bladder cancer cases have not yet been incorporated into the 1998). Miscarriage or spontaneous could be attributable to DBPs. This PAR testing guidelines. Such measures abortion, or stillbirth (approximately range also represents that portion of the include measuring changes in specific equivalent to whole litter resorption, bladder cancer cases that would not proteins in the sperm membrane reduced numbers of total and/or live have occurred if the exposure to proteome and fertility assessments via implants per litter, and increased chlorinated drinking water were absent. in utero insemination. EPA believes that resorptions per litter) were observed by A complete discussion of the Stage 1 additional research is needed, utilizing Waller et al., 1998; Dodds et al., 1999; DBPR bladder cancer PAR evaluation, these newer toxicological measures, to and Bove et al., 1995.’’ including uncertainties and clarify the extent to which DBA poses Similarity of effects between animals assumptions, can be found in the Stage human reproductive or developmental and humans lends credence to and 2 DBPR Economic Analysis (USEPA risk. The database on male reproductive strengthens the weight of evidence for 2003i). effects from exposure to DBA is an association between adverse While EPA recognized the limitations incomplete and is not suitable for reproductive and developmental health of the epidemiological database for quantitative risk assessment at this time. effects and exposure to chlorinated making risk estimates, the Agency It does, however, identify reproductive surface water. EPA believes that the believed that it was useful for effects as an area of concern. weight of evidence of both the developing an estimate of bladder reproductive and developmental cancer risk. The PARs were derived C. Conclusions Drawn From the toxicological and epidemiological from measured risks (Odds Ratios and Reproductive and Developmental databases suggests that exposure to Relative Risk) based on the number of Health Effects Data DBPs may induce potential adverse years exposed to chlorinated surface EPA believes that the weight of health effects on reproduction and fetal water. The uncertainties associated with evidence of the best available science, in development at some DBP exposures. these PAR estimates are largely due to conjunction with the widespread However, additional toxicological work the common prevalence of both the exposure, supports regulatory changes is necessary to identify the mode of disease (bladder cancer) and exposure that target peak DBP exposures action for the effects observed. (chlorinated drinking water). EPA specifically through the Stage 2 DBPR. recognizes that risks from chlorinated Several epidemiology studies found D. Cancer Epidemiology drinking water may be lower or higher statistically significant associations Epidemiological studies on cancer than those estimated from the between exposure to chlorinated provide valuable information that epidemiological literature, and that the drinking water and fetal growth, contributes to the overall evidence on PAR range could include zero or be spontaneous abortion, stillbirth, and the potential human health hazards higher than 17%. neural tube defects. Although from exposure to chlorinated drinking Using the PARs of 2% and 17%, EPA uncertainties remain and the current water. In the area of epidemiology, a estimated that the number of possible database does not support a quantitative number of studies have been conducted bladder cancer cases per year reproductive and developmental risk to investigate the relationship between potentially associated with exposures to assessment for most of the DBPs, the exposure to chlorinated surface water DBPs in chlorinated drinking water weight of evidence provides an and cancer. While EPA cannot conclude could range from 1,100 to 9,300 cases. indication of a hazard concern that there is a causal link between exposure This was based on the estimate of warrants additional regulatory action to chlorinated surface water and cancer, 54,500 new bladder cancer cases per beyond the Stage 1 DBPR. some studies have found an association year nationally, as projected by the Biological plausibility for the effects between bladder, rectal and colon National Cancer Institute for 1997. A observed in epidemiological studies has cancer and exposure to chlorinated thorough discussion of cancer studies been demonstrated through various surface water. published prior to 1998 and possible

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00016 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49563

associations with DBP exposure can be influence of geographic and temporal analysis for this rule, either based found in the Stage 1 DBPR (USEPA differences. specifically on the Villanueva et al. 1998c). • It uses three additional high-quality publication or on the application of a studies not included in the PAR range similar approach. EPA also solicits 2. New Epidemiological Cancer Studies analysis conducted by EPA (i.e., studies comments and suggestions for use of the New studies published since the Stage by Koivusalo et al. 1998, Doyle et al. combined dose-response regression 1 DBPR continue to support an 1997, and Vena et al. 1993). slope coefficient associated with the association between bladder, colon and • It weights the individual studies increased risk of bladder cancer for each rectal cancers and exposure to according to their precision, so more additional year’s exposure to DBPs in chlorinated surface water (Yang et al. precise estimates (due principally to drinking water for estimating the drop 1998; Koivusalo et al. 1998; King et al. greater numbers of cases) carry greater in risk associated with a reduction in 2000b). Based on the weight of evidence statistical weight and therefore have DBPs as part of the benefit analysis of provided by the cancer epidemiology greater influence on the meta-estimate. this rule. EPA provides further • database, EPA has chosen to use the In addition to the primary analysis, discussion and solicitation of comment same PAR analysis to estimate the the authors conducted an evaluation of on how the slope factor might further be primary benefits from bladder cancer the robustness of their conclusions. considered in estimating the benefits of cases potentially avoided as a They examined the sensitivity of this rule in the economic section of this consequence of reducing the DBP levels estimates to decisions made with preamble. from the Stage 2 DBPR (see section VII). respect to exposure definitions, cut b. New colon cancer studies. For the Stage 2 DBPR analysis, EPA points defining exposure groups, Colorectal cancer is the third most updated the 1997 estimate of new inclusion/exclusion of individual common type of new cancer cases and bladder cancer cases per year nationally studies, and potential publication bias. deaths in both men and women in the The meta-analysis provided at least from 54,500 to 56,500 (projected by the U.S. It is estimated that 148,300 new two meta-estimates that may be useful American Cancer Society, 2002) and colorectal cancer cases will be for estimating national population accounted for the reductions in DBP diagnosed in 2002, with 56,600 attributable risk: exposure that were projected for the resulting in deaths (American Cancer • A combined odds ratio for ever- Stage 1 DBPR. Society, 2002). Human epidemiology exposure, with confidence intervals and studies on chlorinated surface water a. New bladder cancer studies. • A combined dose-response have reported associations with Bladder cancer and chlorinated DBP regression slope coefficient, relating exposure has historically been the most colorectal cancer. Since the Stage 1 increasing odds ratios to additional DBPR, two new human epidemiology strongly supported association of all the years of chlorinated drinking water possible cancers, based on human studies (Yang et al. 1998 and King et al. consumption. 2000b) have been conducted to evidence. Two new studies (Yang et al. EPA conducted an estimate of the investigate the relationship between 1998 and Koivusalo et al. 1998) also impact of using the meta-analysis to colon cancer and exposure to suggest an association of DBP exposure provide a perspective on the national chlorinated surface water. Yang et al. with bladder cancer. Yang et al. 1998 population attributable risk. This 1998 did not identify an association found a positive association between estimate is based on the author’s between consumption of chlorinated consumption of chlorinated drinking correction of a minor transcription error drinking water and colon cancer. The water and bladder cancer. Koivusalo et in their published manuscript (the King et al. (2000b) study found evidence al. (1998) found evidence of increased appropriate estimate for the King study of a DBP association with colon cancer risk as a function of increasing DBP yields corrected over-all odds ratio for among males, but no association was exposure duration. Long exposure ever-consumers of 1.2 with 95% ≥ observed among females. durations ( 45 years for Koivusalo et al. confidence interval of 1.091 to 1.320, Similarity of effects reported in 1998) were associated with about a two- personal communication from M. animal toxicity and human fold increase in risk. The new bladder Kogevinas to M. Messner, 5/19/2003). epidemiology studies strengthen the cancer studies continue to support an Assuming 70% of the U.S. population is weight of evidence for an association association and potential for a causal in the ever-consumed category (based between DBP exposure and colon relationship between exposure to on the chlorinated surface water cancer. Effects observed in animal chlorination byproducts and risk for exposed population), a point estimate of studies which included tumors in bladder cancer. the population attributable risk using BDCM exposed rats and mice at several A new publication by C.M. Villanueva the odds ratio from the meta-analysis is sites (NTP 1987); colon tumors in et al. (Villanueva et al. 2003) reports on 12% (95% interval 6% to 18%). bromoform exposed rats (NTP 1989); their meta-analysis of case-control and Although EPA’s population attributable and development of aberrant crypt foci, cohort studies. This meta-analysis may risk range (2% to 17%) was not a preneoplastic lesion of colon cancer in be useful for improving the estimate of intended to convey a quantified level of animals exposed to DBP mixtures national population attributable risk confidence, it is not vastly different (DeAngelo et al. 2002), are comparable (fraction of bladder cancer cases in the from the meta-analysis’ 95% confidence to observations in some cancer U.S. that may be attributed to range of 6% to 18%. EPA regards the epidemiological studies showing an chlorinated drinking water). Compared meta-range as additional support for association with colorectal cancer and to EPA’s current approach (i.e., EPA’s population attributable risk range. consumption of chlorinated water (King providing a range of population The meta-analysis provides continued et al. 2000b). attributable risks (PAR)), use of the support for an association between Even with the additional study meta-estimate would provide a more exposure to chlorinated surface water showing an association, the stable result because: and bladder cancer. epidemiological database on colon • It provides a single (meta) estimate EPA requests comment on the use of cancer as a whole is not as strong as that of the odds ratio from which to calculate a meta-estimated odds ratios to estimate for bladder cancer. However, this new the PAR, thereby summarizing the national population attributable risk for study increases the weight of evidence results across studies, thus reducing the the purpose of supporting the benefit of an association between DBP exposure

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00017 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49564 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

and colon cancer. The Stage 1 DBPR childhood acute lymphoblastic cancer risks of DBPs, there has been (USEPA 1998c) includes additional leukemia and THMs. There were no little change in the overall DBP discussion of colon cancer risks associations with leukemia for any of carcinogenicity database since the Stage associated with DBP exposure. the exposure indices for total THM, or 1 DBPR. c. New rectal cancer studies. The specific THMs. Therefore, the study The most significant new publication evidence for an association between does not provide evidence of an since the Stage 1 DBPR was a study of DBPs and rectal cancer is stronger than association between any of the exposure DCAA tumorigenicity in mice by for colon cancer. Yang et al. (1998) and variables and childhood leukemia. DeAngelo et al. (1999). The Agency has Hildesheim et al. (1998) both found 3. Review of the Cancer Epidemiology used the data from this study to revise associations between chlorinated Literature (WHO 2000) the slope factor for DCAA and a drinking water exposure and rectal ¥6 The International Programme on drinking water 10 lifetime cancer risk cancer, and the associations had a concentration. The slope factor is a similar magnitude in both sexes. Chemical Safety (IPCS) report on disinfectants and disinfection measure of the potency of a carcinogen Hildesheim et al. also found an ¥6 byproducts (WHO 2000) concludes that while the 10 lifetime cancer risk association in both sexes with lifetime concentration provides an estimate of average THM concentration. The results of analytical epidemiological cancer studies are insufficient to the concentration of a contaminant in consistency of the dose-response trends, drinking water that is associated with an the consistency between sexes, and the support a causal relationship for bladder, colon, rectal, or any other estimated excess lifetime cancer risk of apparent control of important potential one in a million (Table III–3). confounders in this study all support cancer and chlorinated drinking water the observed associations. or THMs. The report notes that there is Another significant advancement d. Other cancers. Two new human better evidence for an association beyond the Stage 1 DBPR was the epidemiology studies support the between exposure to chlorinated surface evaluation of the chloroform possibility of an association between water and bladder cancer than for other tumorigenicity data on the basis of its DBPs and kidney cancer. Yang et al. types of cancer. The WHO also nonlinear mode of action following the (1998) found a positive association for concludes that based on the large draft 1999 proposed Guidelines for both males and females between number of people exposed to Carcinogen Risk Assessment (USEPA consumption of chlorinated drinking chlorinated drinking water, there is a 1999a). The new chloroform assessment need to address this potential health water and kidney cancer. Koivusalo et became available on IRIS (2001) in concern. al. (1998) found a small, statistically October, 2001 (see section V for a more significant, exposure-related excess risk E. Cancer and Other Toxicology detailed discussion). for kidney cancer for males. The Few new cancer toxicology studies The Criteria Documents for association for females was not have been completed since the Stage 1 bromoform, bromodichloromethane, significant in the Koivusalo et al. 1998 DBPR was finalized in December 1998. dibromochloromethane, and study. The current database for this The information provided in the dichloroacetic acid that support the endpoint of cancer, however, is following sections adds to the Stage 2 proposal include cancer slope ¥6 insufficient to conclude an association. toxicology database and provides factors and 10 lifetime cancer risk Cantor et al. (1999) studied brain additional support for the Stage 2 DBPR concentrations that have been modified cancer, focusing on gliomas. None of the to control DBP peaks (e.g, high TTHM from their Stage 1 values in order to exposure variables were related to brain and HAA5 levels) throughout reflect the methodology proposed in the cancer among females, but males distribution systems, but does not 1996/1999 draft cancer guidelines showed a statistically significant, change the quantitative assessment of (USEPA 1999a) (Table III–3). These monotonically increasing risk associated the MCLGs. include the values based on the with duration of exposure to chlorinated Maximum Likelihood Estimate of the surface water. This study suggests a 1. EPA Criteria Documents dose producing effects in 10 percent of possible association between To date, EPA has established lifetime the animals (ED10) and from the lower chlorination byproducts and gliomas; cancer risk levels for four DBPs 95 percent confidence bound on that however, the evidence from this study (bromoform, bromodichloromethane, value (LED10). Except for is not strong enough to support a bromate, and dichloroacetic acid) dibromochloromethane, which is conclusion of a causal association. classified as ‘‘probable’’ carcinogens, as classified as a possible human Infante-Rivard et al. (2001) conducted promulgated in the Stage 1 DBPR and carcinogen, the DBPs in Table III–3 (and a population-based case-control study in reported in the Integrated Risk bromate as noted previously) are Quebec Province, Canada, to examine Information System (IRIS). Although classified as probable human possible associations between researchers have continued to assess the carcinogens.

TABLE III—3.—QUANTIFICATION OF CANCER RISK

Risk factors from LED10 Risk factors from ED10

¥6 ¥6 Slope factor 10 Risk Slope factor 10 Risk Disinfection byproduct ¥ concentration ¥ concentration (mg/kg/day) 1 (mg/kg/day) 1 (mg/L) (mg/L)

Bromodichloromethane ...... 0.034 0.001 0.022 0.002 Bromoform ...... 0.0045 0.008 0.0034 0.01 Dibromochloromethane ...... 0.04 0.0009 0.017 0.002 Dichloroacetic Acid ...... 0.048 0.0007 0.014 0.003

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00018 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49565

EPA believes that it is important to 4. WHO Review of the Cancer epidemiology and toxicology, and the pursue additional research on cancer Toxicology Literature (2000) new studies discussed in today’s from DBPs. EPA has several ongoing The IPCS report on Disinfectants and proposal. EPA solicits any additional studies in addition to a collaboration Disinfection Byproducts (WHO 2000) cancer epidemiology and toxicology with the National Toxicology Program emphasizes that the bulk of the data that need to be considered for the of the National Institute of toxicology data focuses primarily on final Stage 2 DBPR. Environmental Health Sciences. More carcinogenesis. The Task Group found EPA also solicits any health information on EPA’s toxicology BDCM to be of particular interest information available to further assess research program can be found at http:/ because it produces tumors in both rats risk to sensitive subpopulations, /www.epa.gov/nheerl. and mice at several sites. Although the especially children and the elderly. HAAs appear to be without significant 2. Other Byproducts with Carcinogenic IV. DBP Occurrence Within genotoxic activity, the brominated Distribution Systems Potential HAAs appear to induce oxidative New information on the occurrence of a. 3-chloro-4-(dichloromethyl)-5- damage to DNA, leading to tumor formation. DBPs in distribution systems raises hydroxy-2(5H)-furanone) (MX)— issues about the protection provided by multisite cancer. MX is a byproduct of F. Conclusions Drawn From the Cancer the Stage 1 DBPR. This section presents chlorination that is typically found at Epidemiology and Toxicology the new information used to identify very low concentrations (approximately EPA believes that the cancer key issues and to support the <0.000067 mg/L) in drinking water. The epidemiology and toxicology databases development of the Stage 2 DBPR. For information available on MX was provide important information that a more detailed discussion see the Stage recently compiled in the Quantitative contributes to the weight of evidence 2 Occurrence Assessment for Cancer Assessment for MX and evaluation of the potential health risks Disinfectants and Disinfection chlorohydroxyfuranones (USEPA from exposure to chlorinated drinking Byproducts (USEPA 2003o). 2000i). Overall, the weight of evidence water. At this time the cancer Under the Stage 1 DBPR, compliance indicates that MX is a direct-acting epidemiology studies are insufficient to with the DBP MCLs is determined by genotoxicant in mammals, with the establish a causal relationship between averaging, annually and system-wide, ability to induce tumors in multiple exposure to chlorinated drinking water all DBP measurements. The following sites. The primary sites for tumor and cancer, but EPA does believe there discussion shows that compliance based formation are the thyroid and liver. is a potential association. The current on system averages of DBP b. N-nitrosodimethylamine (NDMA)— database is sufficient for quantitative concentrations allows a significant multisite cancer. Health effects data analysis on the endpoint of bladder number of sampling locations within indicate that NDMA is a probable cancer, as presented previously in the distribution systems to have DBP levels PAR analysis. above the MCLs. These peak DBP human carcinogen, as described on IRIS The association between DBP occurrences are masked by averaging (1991). Risk assessments have estimated exposure and colon cancer remains with lower distribution system that the 10¥6 lifetime cancer risk level more tenuous than the link to bladder occurrence levels. The populations is 0.000007 mg/L based on induction of cancer, although similarity of effects served by portions of the distribution tumors at multiple sites. Recent studies reported in animal toxicity and human system with higher DBP concentrations have produced new information on the epidemiology studies strengthens the are not receiving the same level of occurrence and mechanism of formation weight of evidence for an association health protection. of NDMA but there is not enough between DBP exposure and colon The new information also shows that information at this time to draw cancer. Studies finding potential the highest DBP levels often do not conclusions. More research is underway relationships between exposure to occur at distribution system sites to determine the mechanism by which chlorinated drinking water and rectal, identified as representing maximum NDMA is formed in drinking water, and kidney, and brain cancer also add to the residence time. The information further the extent of its occurrence in weight of evidence for a public health shows that the highest TTHM and chloraminated systems. concern. EPA believes that the overall HAA5 levels often do not occur at the cancer epidemiology and toxicology 3. Other Toxicological Effects same site within the distribution data support the decision to pursue system. These two findings suggest that The Agency has modified the additional DBP control measures as it is appropriate to reevaluate the Stage reference dose (RfD) values for 2 of the reflected in the Stage 2 DBPR. 1 DBPR compliance monitoring sites in chlorinated acetic acids since the Stage G. Request for Comment order to target those sites with high DBP 1 DBPR. Under the Stage 1 DBPR there levels. EPA believes that distribution EPA requests comment on the was no established RfD for system compliance monitoring sites conclusions drawn from the new health need to be reevaluated to ensure monochloroacetic acid (MCAA). Data information summarized in this section. identification of sites that reflect both from a drinking water exposure study of EPA requests comment on the weight of high TTHM and HAA5 occurrence. MCAA in rats by DeAngelo et al. (1997) evidence evaluation of the potential were used to establish an RfD of 0.004 reproductive and developmental A. Data Sources mg/kg/day based on observed increases hazards from DBPs and its potential 1. Information Collection Rule Data in spleen weight. Data from DeAngelo implications for the regulatory (1997) were also used to calculate a new provisions for the final Stage 2 DBPR. The Information Collection Rule RfD of 0.03 mg/kg/day for EPA solicits any additional data on the (USEPA 1996a) established monitoring trichloroacetic acid (TCAA) based on reproductive or developmental effects and data reporting requirements for observed effects on body weight and from DBPs that need to be considered large public water systems. Under the liver effects. Detailed discussions of the for the final Stage 2 DBPR. Information Collection Rule, systems new reference doses are located in EPA requests comment on EPA’s serving at least 100,000 people were section V of this preamble. conclusions regarding cancer required to conduct DBP and DBP-

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00019 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49566 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

related monitoring. The 18 months of carbon (GAC) and membrane technology Disinfectants and Disinfection required monitoring, which began in to control for DBPs. The estimated cost Byproducts (USEPA 2003o). July 1997 and ended in December 1998, for these studies totaled approximately 2. Other Data Sources Used To Support applied to 296 public water systems $57 million (USEPA 1996a). the Proposal (500 treatment plants). In addition to the analysis of DBPs in The Information Collection Rule data distribution systems, EPA used Table IV–1 summarizes the data show the national occurrence of: (1) sources other than the Information occurrence data from the Information Influent water quality parameters; (2) Collection Rule used to support the Collection Rule to confirm selection of primary and secondary disinfectant use Stage 2 DBPR. The data from the TTHM and HAA5 as appropriate by the large plants; (3) occurrence of Information Collection Rule is from contaminants for monitoring DBPs. EPA DBPs and DBP precursors in treatment large systems. To validate the plants, finished waters, and also used occurrence data from the conclusions drawn from analysis of the distributions systems; (4) microbial Information Collection Rule to confirm Information Collection Rule for small occurrence (in subpart H systems only); differences in monitoring requirements and medium systems, EPA compared and (5) treatment plant monthly for systems using surface water versus these other data sources with the operation, and initial as well as final those using ground water, as stipulated Information Collection Rule data. EPA treatment plant design. The data were under the Stage 1 DBR. Analysis of the found that there are significant gathered after the Stage 1 DBPR was Information Collection Rule data similarities between large systems and finalized (USEPA 1998c) but well before indicates that TTHM and HAA5 medium and small systems with regard systems were required to meet Stage 1 comprise on average, across all systems, to source water quality (affecting DBP DBPR requirements. about 50% of the total mixture of formation) and use of treatment The Information Collection Rule chlorinated DBPs and that TTHM and technologies. Because of these required a significant investment for the HAA5 concentrations are much lower similarities, EPA expects that small and water treatment industry, as well as for and less variable in ground water medium systems would find DBP the EPA to analyze the data. Overall, the systems than in surface water systems. distribution system levels similar to occurrence and treatment data collected These results support the basis for those found in large systems following under the Information Collection Rule, continuing the use of TTHM and HAA5 compliance with the Stage 1 DBPR excluding microbial data, was estimated as indicators for controlling chlorinated requirements. For detailed discussion of to cost systems $54 million (USEPA DBPs. The data also reconfirmed that this analysis, see Stage 2 Occurrence 1996a). In addition, systems using ground water systems require less Assessment for Disinfectants and source waters with high DBP precursor monitoring than surface water systems Disinfection Byproducts (USEPA 2003o) levels were required to conduct bench based on lower and less variable DBP and Economic Analysis for the Stage 2 and pilot studies to evaluate the occurrence. For detailed analysis, see Disinfection Byproducts Rule (USEPA effectiveness of granular activated Stage 2 Occurrence Assessment for 2003i).

TABLE IV–1.—SUMMARY OF NON-INFORMATION COLLECTION RULE OCCURRENCE SURVEY DATA

Number of plants Data source Data collected Geographic representation (By population served)

Information Collection Rule Raw source water-(Large Systems) TOC Random national distribu- 47 serving 100,000 or Supplemental Survey. Raw source water-(Small & Medium Survey Systems) tion by SW source type 1. more. TOC, UV 254, bromide, turbidity, pH, & tempera- 40 serving 10,000–99,999. ture. 40 serving fewer than 10,000. WaterStats ...... Population served and flows Random national distribu- 219 serving 100,000 or Raw source water—Water tion. more. Quality Parameters (WQPs), 623 serving 10,000–99,999. Source water type. 30 serving fewer than Finished water-WQPs, TTHM, HAAs 10,000. Treatment-unit processes, disinfectant used. National Rural Water Asso- Population served and flows Random national distribu- 117 serving fewer than ciation Survey (NRWAS). Raw source water-temperatures, turbidity, pH, and tion. 10,000. source water type, bromide, TOC, UV 254, alka- linity, calcium, and total hardness. Finished water-residence time estimate, total and indi- vidual THMs, individual HAAs and HAA5, HAA6, HAA9,TOC, UV 254, Bromide, Temperature, pH, free and total chlorine residual levels. Treatment-unit processes, disinfectant used. State Data-Surface Water .. Distribution system TTHM occurrence data. AK, CA, IL, MN, MS, NC, 562 serving fewer than TX, WA 2. 10,000. State Data-Ground Water .. Distribution system TTHM occurrence data. AK, CA, FL, IL, NC, TX, 2336 serving fewer than WA 2. 10,000. Ground Water Supply Sur- TOC and TTHM (one sample for each parameter at Random national distribu- 979 total. vey. the entry point to distribution system.) tion. 1 Source type designations include flowing stream and lake/reservoir (Except for 7 large plants pre-selected). 2 Over 50 percent of each State’s systems are represented. EPA believes that the data reasonably represent a full range of source water qual- ity in small systems at the national level.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00020 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49567

B. DBPs in Distribution Systems Stage 1 DBPR TTHM and HAA5 MCLs 1. DBPs Above the MCL Occur at Some EPA wanted to understand DBP as an annual average. The TTHM and Locations in a Substantial Number of occurrence in distribution systems HAA5 data consist of quarterly Plants likely to exist after implementation of measurements in four locations in Figure IV–1 compares the TTHM the Stage 1 DBPR. Such an distribution systems associated with running annual average (RAA) levels understanding would enable EPA to each Information Collection Rule with the single highest TTHM recognize options on how to improve treatment plant. Two samples were concentration in the distribution protection under the Stage 2 DBPR. The collected at sites representing average analysis of occurrence data to support residence time (AVG1 and AVG2), one system. Twenty one percent (60 of 290) the Stage 2 DBPR is complicated sample at a site intended to represent of the Information Collection Rule because available national occurrence the maximum residence time (MAX), plants had single TTHM concentrations data do not reflect the changes in and one sample was reported as a higher than the 0.080 mg/L MCL. Figure occurrence resulting from the distribution system equivalent (DSE). IV–2 makes the same comparison for implementation of the Stage 1 DBPR. The DSE sample was generally HAA5. Fourteen percent (40 of 290) of Many utilities have only recently representative of average residence the plants meeting the Stage 1 DBPR changed their treatment to comply with times. EPA believes that the monitoring MCL had single HAA5 concentrations the Stage 1 DBPR (subpart H systems locations of the Information Collection higher than the 0.060 mg/L MCL. In serving 10,000 people or more were Rule, while not necessarily being the systems with a low RAA for TTHM and required to comply beginning January same as the Stage 1 DBPR compliance HAA5, the highest single TTHM and 2002) or are about to make changes in monitoring sites, provide a close HAA5 values are generally not much treatment to comply with this rule approximation of monitoring under the higher than the respective Stage 1 DBPR (subpart H systems serving fewer than Stage 1 DBPR. EPA recognizes, however, MCLs. However, as the RAAs increase, 10,000 people and ground water that data for plants that are in there is a greater likelihood of having systems are required to comply compliance with Stage 1 MCLs even peak levels above the MCLs. As the beginning January 2004). without installing additional treatment RAAs approach the Stage 1 DBPR MCLs, To address the above issue, EPA (perhaps because of low source water some of the distribution system single evaluated Stage 1 DBPR implications by TOC) are not necessarily reflective of highest concentrations approach levels using Information Collection Rule data plants that make treatment changes to that are double the Stage 1 DBPR MCLs. from plants that would not exceed the comply with the Stage 1 DBPR. BILLING CODE 6560–50–P

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00021 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49568 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00022 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.000 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49569

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00023 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.001 49570 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

2. Specific Locations in Distribution that exceeded a TTHM LRAA of 0.080 did so at two or three locations. Of the Systems Are Not Protected to MCL mg/L did so at two locations. Of the 15 8 plants, the highest LRAA was between Levels plants, the highest LRAA was between 0.060 and 0.070 mg/L at 5 plants, and 0.080 and 0.090 mg/L at 10 plants, and between 0.070 and 0.075 mg/L at 3 Data from the Information Collection between 0.090 and 0.100 mg/L at 5 plants. Among the 290 plants in the Rule show that the RAA compliance plants. Customers served at these Information Collection Rule database calculation may allow specific locations locations regularly received water with meeting the Stage 1 MCLs, 19 plants in a distribution system to regularly TTHM concentrations somewhat higher have a maximum TTHM LRAA of 0.080 receive water with DBP levels that than the MCL. mg/l or greater or a maximum HAA5 exceed the MCL. Figure IV–3 shows that Figure IV–4 shows similar results LRAA of 0.060 mg/l or greater (four five percent of plants (15 out of 290) had based on Information Collection Rule plants exceeded both MCLs), though in one or more locations that, on average, HAA5 data. Three percent of plants no case did DBP levels at a given exceeded 0.080 mg/L as a TTHM LRAA (eight of 290) exceeded 0.060 mg/L as an location consistently exceed the MCL by for that same year. One of the 15 plants LRAA, and three of these eight plants more than 20%.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00024 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49571

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00025 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.002 49572 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

BILLING CODE 6560–50–C

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00026 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.003 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49573

3. Stage 1 DBPR Maximum Residence highest TTHM and HAA5 level often Figure IV–6, based on data from the Time Location May Not Reflect the occurred at different points in the National Rural Water Survey (NRWS), Highest DBP Occurrence Levels distribution system. indicates that systems serving fewer The 1979 TTHM rule and Stage 1 Figure IV–5 illustrates that the highest than 10,000 people also frequently have DBPR monitoring locations must TTHM and HAA5 LRAAs could be at their highest TTHM and HAAS levels at include a site reflection maximum any of the four Information Collection locations other than those intended to residence time in the distribution Rule sample locations in the represent maximum residence time. The system with the intent of capturing the distribution system or, in some cases, at occurrence patterns indicated in Figures highest DBP levels in the distribution the finished water location. Fifty IV–5 and IV–6 may be due to several system. The Information Collection rule percent of the plants evaluated have the factors, such as HHA5 degrading over referred to this specific location as highest TTHM LRAA concentration time in the distribution system, MAX. The Information Collection rule occurring at a site other than the maximum residence time monitoring data indicate two important results: (1) maximum residence time monitoring sites not actually representing the that monitoring locations identified as site. over 60% of plants evaluated had maximum residence time, or that using the maximum residence time locations the highest HAA5 LRAA at a location a simple estimation of maximum often did not represent those locations other than the maximum residence time residence time cannot characterize a with the highest DBP levels and (2) the monitoring site. complex distribution system.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00027 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.004 49574 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

EPA also analyzed whether the For plants that did have their highest and HAA5 levels occurring at the same highest LRAA for TTHM and HAA5 LRAA TTHM and HAA5 concentrations location, the highest TTHM and HAA5 occurred at the same location. If TTHM at the same location, it was not LRAA simultaneously occurred at the and HAA5 occur at the same location necessarily the maximum residence maximum residence time (MAX) rather than different locations, fewer time monitoring location. Figure IV–7 monitoring location in 64% of the cases. monitoring sites would be needed to illustrates that for the Information C. Request for Comment represent TTHM and HAA5 occurrence. Collection Rule plants with the highest However, this is not the case. The TTHM and HAA5 levels occurring at the EPA requests comment on the Information Collection Rule and NRWA same location, the highest TTHM and analysis presented in this section. Is data sets, respectively, indicate that HAA5 LRAA simultaneously occurred EPA’s approach for representing post 49% and 44% of plants experienced at the maximum residence time Stage 1 DBPR occurrence appropriate? their highest LRAA TTHM and HAA5 monitoring location in 50% of the cases. What other approaches might be used? concentrations at different locations in Figure IV–8 illustrates that for the Are the conclusions that EPA derives the distribution system. NRWA plants with the highest TTHM from the analysis appropriate?

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00028 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.005 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49575

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00029 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.006 49576 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

V. Discussion of Proposed Stage 2 DBPR which is a key event in the cancer mode toxicological studies on chloroform Requirements of action of chloroform, will not occur were published and were discussed in (USEPA 2001b). A. MCLG for Chloroform two Notices of Data Availability EPA believes that the RfD used for (NODAs) (USEPA 1997a; USEPA 1. What Is EPA Proposing Today? chloroform is protective of sensitive 1998e). The 1998 NODA presented EPA is proposing an MCLG for groups, including children. This RfD substantial scientific data related to the chloroform of 0.07 mg/L based on a was developed by the EPA current mode of action as part of the chloroform cancer reference dose (RfD), an method for developing RfDs based on risk assessment and requested comment assumption that a person drinks 2 liters animal data. The method is designed to on a chloroform MCLG of 0.3 mg/L that of water per day (the 90th percentile of be protective by taking human reflected a nonlinear mode of action. intake rate for the U.S. population), and variability into account and assuming After considering comments on the a relative source contribution (RSC) of that the average human will be as NODAs, EPA determined that further 20 percent. The MCLG is proposed at a sensitive as the most responsive animal deliberations with the Science Advisory level at which no adverse effects on the species. EPA’s understanding of the Board (SAB) and stakeholders were mode of action for chloroform does not health of persons is anticipated with an needed before changing the MCLG for indicate a uniquely sensitive subgroup adequate margin of safety. This chloroform. Thus, EPA promulgated a conclusion is based on toxicological or an increased sensitivity in children. chloroform MCLG of zero in the final evidence that the carcinogenic effects of 2. How Was This Proposal Developed? Stage 1 DBPR (USEPA 1998c) and chloroform are an ultimate consequence of sustained tissue toxicity. The MCLG a. Background. EPA proposed a zero committed to conducting additional is set at a daily dose for a lifetime at MCLG for chloroform in the 1994 Stage deliberations with the SAB and which no adverse effects will occur 1 DBPR proposal (USEPA 1994b). factoring the SAB’s review into the because the sustained tissue toxicity, Following the proposal, numerous Agency’s Stage 2 DBPR rulemaking

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00030 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.007 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49577

process. The Agency consulted with the carcinogenic to humans at a dose level c. How the MCLG is derived. EPA SAB in October 1999 (USEPA 2000f). that does not cause cytotoxicity and cell continues to recognize the strength of The Stage 1 DBPR MCLG of zero for regeneration (USEPA 1998e, USEPA the science in support of a nonlinear chloroform was challenged, and the U.S. 1998b, USEPA 2001b). approach for estimating the Court of Appeals for the District of Chloroform’s carcinogenic potential is carcinogenicity of chloroform. This Columbia Circuit issued an order indicated by animal tumor evidence science was affirmed by the Chloroform vacating the zero MCLG (Chlorine (liver tumors in mice and renal tumors Risk Assessment Review Subcommittee Chemistry Council and Chemical in both mice and rats) from inhalation of the EPA SAB Executive Committee Manufacturers Association v. EPA, 206 and oral exposure. Data on metabolism, which met on October 27–28, 1999 f.3d 1286 (D.C. Circuit 2000)). EPA toxicity, mutagenicity and cellular (USEPA 2000f). The SAB Subcommittee committed to the Court to propose a proliferation contribute to an agreed that the nonlinear approach is non-zero MCLG for chloroform in the understanding of the mode of most appropriate for the risk assessment upcoming proposed Stage 2 carcinogenic action. For chloroform, of chloroform. Disinfectants and Disinfection sustained or repeated cytotoxicity with Nonzero MCLGs are scientifically and Byproducts Rule. EPA removed the secondary regenerative hyperplasia statutorily supported. The statute MCLG for chloroform from its Stage 1 precedes, and is a key event for, hepatic requires that the MCLG be set where no DBP NPDWR (USEPA 2000e). No other and renal neoplasia. known or anticipated adverse effects provision of the Stage 1 DBPR was EPA believes that a DNA reactive occur, allowing for an adequate margin affected. mutagenic mode of action is not likely of safety (56 FR 3533; USEPA 1991b). b. Basis of the new chloroform MCLG. to be the predominant influence of Historically, EPA established MCLGs of Based on an analysis of all the available chloroform on the carcinogenic process. zero for known or probable human scientific data on chloroform discussed EPA has concluded that the carcinogens based on the principle that in more detail below, EPA believes that predominant mode of action involves any exposure to carcinogens might chloroform dose-response is nonlinear cytotoxicity produced by the oxidative represent some finite level of risk. If and that chloroform is likely to be generation of highly reactive there is substantial scientific evidence, carcinogenic only under high exposure metabolites, followed by regenerative however, that indicates there is a ‘‘safe conditions. EPA’s assessment of the cell proliferation (USEPA 2001b). EPA threshold’’, then a nonzero MCLG can cancer risk associated with chloroform further believes that the chloroform be established with an adequate margin exposure (USEPA 2001b) uses the dose-response is nonlinear. The SAB of safety (56 FR 3533; USEPA 1991a)). principles of the 1999 EPA Proposed final report states ‘‘(t)he Subcommittee EPA would ideally like to use the Guidelines for Carcinogen Risk agrees with EPA that sustained or delivered dose (i.e., the amount of key Assessment (USEPA 1999a). chloroform metabolites that actually The Proposed Guidelines for repeated cytotoxicity with secondary regenerative hyperplasia in the liver reach the liver and cause cell toxicity) Carcinogen Risk Assessment, as for calculating an RfD to support the reviewed by the public and the EPA and/or kidney of rats and mice precedes, and is probably a causal factor MCLG. However, the required SAB, reflect new science and are toxicokinetic data are not currently consistent with, and an extension of, the for, hepatic and renal neoplasia’’ (USEPA 2000f). available. Thus, the RfD is calculated existing 1986 Guidelines for Carcinogen using the applied dose (i.e., the amount ii. Metabolism. The cytochrome P450 Risk Assessment (USEPA 1986). The of chloroform ingested). The RfD is isoenzyme CYP 2E1 is the primary 1986 guidelines provide for departures based on both the benchmark dose and enzyme catalyzing chloroform from default assumptions such as low the traditional no observed adverse metabolism at low concentrations. dose linear extrapolation. For example, effect level/lowest observed adverse Chloroform’s carcinogenic effects the 1986 EPA guidelines reflect the effect level (NOAEL/LOAEL) involve oxidative generation of reactive position of the Office of Science and approaches for hepatotoxicity in the and toxic metabolites (phosgene and Technology Policy (OSTP) that (OSTP most sensitive species, the dog. The hydrochloric acid [HCl]) and thus are 1985; Principle 26) ‘‘[N]o single MCLG is based on the RfD and related to its noncancer toxicities (e.g., mathematical procedure is recognized calculated as follows: as the most appropriate for low-dose liver or kidney toxicities). The extrapolation in carcinogenesis. When electrophilic metabolite phosgene could RfD×× body weight RSC relevant biological evidence on react with macromolecules such as MCLG = mechanisms of action exists (e.g, phosphotidyl inositols or tyrosine daily water consumption pharmacokinetics, target organ dose), kinases which in turn could potentially i. Reference dose. The RfD for the models or procedure employed lead to interference with signal chloroform was estimated based on should be consistent with the transduction pathways (i.e., chemical noncancer effects using both the evidence.’’ The 1985 guidelines go on to messages controlling cell division), thus benchmark dose and the traditional state ‘‘The Agency will review each leading to carcinogenesis. Likewise, it is NOAEL/LOAEL approaches. For assessment as to the evidence on also plausible that phosgene reacts with benchmark analysis, five relevant data carcinogenesis mechanisms and other cellular phospholipids, peptides and sets including target organ toxicity, biological or statistical evidence that proteins resulting in generalized tissue labeling index, histopathology in indicates the suitability of a particular injury. Glutathione, free cysteine, rodents, and liver toxicity in dogs extrapolation model.’’ histidine, methionine and tyrosine are (Heywood 1979) were evaluated. The i. Mode of action. EPA has fully all potential reactants for electrophilic effects seen in dogs are considered to be evaluated the science on chloroform and agents. early signs of liver toxicity, preceding concludes that chloroform is likely to be At high concentrations, chloroform cytotoxicity, cytolethality and carcinogenic to humans under high may undergo reductive metabolism regenerative hyperplasia. Thus, the exposure conditions that lead to which forms reactive dichloromethyl Heywood (1979) study, provides the cytotoxicity and regenerative free radicals. These free radicals can most sensitive end point in the most hyperplasia in susceptible tissue; contribute to lipid peroxidation and sensitive species and is the most chloroform is not likely to be cause cytotoxicity. appropriate basis for the RfD.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00031 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.026 49578 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

The 95% confidence lower bound on was calculated using an RSC of 80 contribute roughly 28 percent of the the dose associated with a 10% extra percent, based on the assumption that total dose of chloroform (USEPA 2001a). risk (LED10) is based on the prevalence most exposure to chloroform is likely to With assumptions as described, tap of animals demonstrating liver toxicity. come from ingestion of drinking water. water ingestion is a portion of exposure After an exposure adjustment to the In the final Stage 1 DBPR, EPA through fluid intake which contributes LED10 (1.2 mg/kg/day), an RfD of 0.01 reconsidered this assumption in about 34 percent of the total dose, mg/kg/day was calculated using an response to comments and in the light inhalation accounts for about 31 percent overall uncertainty factor of 100 (10 for of data which indicate that exposure to of the total dose, ingestion of foods interspecies extrapolation and 10 for chloroform by inhalation and dermal contributes another 27 percent of the protection of sensitive individuals) exposure may potentially contribute a overall dose, and dermal absorption (USEPA 2001b). substantial percentage of the overall (primarily during showering) adds Coincidentally, the benchmark dose exposure to chloroform depending on slightly less than 8 percent of the total and the traditional NOAEL/LOAEL the activity patterns of individuals dose. These exposure percentages are approaches yield the same RfD number (USEPA 1998e) e.g., during showering, based on average daily doses (mean (USEPA 2001b). The NOAEL/LOAEL bathing, swimming, boiling water, chloroform intake for adults) for each approach is also based on the Heywood clothes washing, and dishwashing. source and route of exposure under study (1979) which had a LOAEL of 15 There is also potential exposure to specific conditions. They do not take mg/kg/day for evidence of liver toxicity. chloroform by the dietary route. There into account the considerable variability After an exposure adjustment to the are uncertainties regarding other in several factors across the population. LOAEL (yielding 12.9 mg/kg/day), an possible highly exposed sub- For instance, intake of drinking water or RfD of 0.01 mg/kg/day was calculated populations, e.g., swimmers, those who particular foods and length of shower using an overall uncertainty factor of use humidifiers, hot-tubs, and outdoor varies from day-to-day, as do home air 1000 (10 for interspecies extrapolation, misters, persons living near industrial turnover rates and ventilation. Different 10 for protection of sensitive sources, people working in areas in the United States vary with individuals, and 10 for using a LOAEL laundromats, and persons working with respect to these factors and chloroform instead of a NOAEL) (USEPA 2001b). pesticides employing chloroform as a concentrations in food. Thus, although ii. Relative source contribution. solvent (USEPA 1998b). the 28 percent for the median individual Another factor in determining the A 1998 International Life Sciences is based on reasonable assumptions, MCLG is the relative source Institute (ILSI) report evaluated the uncertainty remains. contribution (RSC). The RSC is used uptake of drinking water contaminants Given the uncertainties of estimation, when the MCLG is set at a level above through the skin and by inhalation. The EPA believes available analyses point to zero. Its purpose is to ensure that the report noted that ‘‘(i)n the case of the RSC of 20 percent as the appropriate contribution to exposure from drinking chloroform, its high volatility leads to default (i.e., 20 percent of exposure to tap water does not cause the lifetime its rapid movement from liquid to air. chloroform comes from drinking tap daily exposure of persons to a Large water-use sources, such as water alone). EPA also believes that this contaminant to exceed RfD. The RSC is showers, become dominant sources with default is protective of public health thus a factor used to make sure that the respect to exposure’’ and ‘‘(t)he and is a more reasonable choice than MCLG is protective even if persons are inhalation route is demonstrated to be choosing any particular estimate exposed to the contaminant by other the primary route for higher-volatility because of the assumptions and routes (inhalation, dermal absorption) or compounds (e.g., chloroform)’’ (ILSI uncertainties involved with each other sources (e.g., food). If sufficient 1998). Weisel and Jo (1996) found that estimation. Hence, EPA is proposing the quantitative data are not available on ‘‘approximately equivalent amounts of MCLG based on the RSC default of 20 exposure by other routes and sources, chloroform from water can enter the percent which supports the adequacy of EPA has historically assumed that the body by three different exposure routes, the margin of safety associated with the RSC from drinking water is 20 percent inhalation, dermal absorption, and MCLG. of the total exposure, a value considered ingestion, for typical daily activities of iii. Water ingestion and body weight protective. If data indicate that drinking and bathing.’’ assumptions. In MCLG calculations, contributions from other routes and Chloroform has been found in EPA assumes the 90th percentile water sources are not significant, EPA has beverages, especially soft drinks, and ingestion of 2 liters (roughly equivalent historically assumed a less conservative food, particularly dairy products to a half gallon) per day (USEPA 2000a). RSC of 80 percent (54 FR 22,062, 22,069 (Wallace, 1997). Wallace states that The use of a conservative consumption (May 22, 1989)(USEPA 1989a), 56 FR at ‘‘ingestion (drinking tap water and soft estimate is consistent with the objective 3535 (Jan 30, 1990)(USEPA 1991a), 59 drinks and eating certain dairy foods), of setting an MCLG that is protective. FR 38,668, 38,678 (July 29, inhalation (breathing peak amounts of EPA also uses a default adult body 1994)(USEPA 1994b)). chloroform emitted during showers or weight of 70 kg (equal to 154 pounds) Today, EPA is proposing an baths, and lower levels in indoor air for the RfD since dose is calculated from assumption of a 20 percent RSC. This is from other indoor sources), and dermal lifetime studies of animals and in consideration of data which indicate absorption (during showers, baths, and compared to lifetime exposure for that exposure to chloroform by other swimming)’’ each ‘‘appear to be humans. routes and sources of exposure may potentially substantial contributors to iv. MCLG calculation. The MCLG is potentially contribute a substantial total exposure’’. calculated to be 0.07 mg/L using the percentage of the overall exposure to EPA estimates that for the median following assumptions: an adult tap chloroform. individual, ingestion of total tap water water consumption of 2 L per day for a In the 1998 Stage 1 DBPR NODA, EPA (assuming certain activity patterns, 70 kg adult, and a relative source considered an MCLG of 0.3 mg/L that habits, and home characteristics) can contribution of 20%:

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00032 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49579

0.01 mg/kg/d×× 70 kg 0.2 MCLG for Chloroform = = 007. mg/L (rounded) 2L/ day

EPA concludes that an MCLG of 0.07 a 0.07 mg/L MCLG for chloroform in included in the toxicity equivalency mg/L based on protection against liver conjunction with developing MCLs for quotient for coplanar PCBs. In the case toxicity will be protective against each of the individual TTHMs (i.e., 4 of the disinfection byproducts, EPA carcinogenicity given that the mode of MCLs and 4 MCLGs for the THMs); and believes that the THMs have different action for chloroform involves developing a single combined MCLG for modes of action and health endpoints. cytotoxicity as a key event preceding TTHM rather than developing a separate One of the THMs is a liver carcinogen tumor development. Therefore, the MCLG for each of the THMs. (chloroform) with a mode of action recommended MCLG for chloroform is EPA considered developing separate dependent on cytolethality; two are 0.07 mg/L. MCLGs and MCLs for each THM. Under DNA-reactive carcinogens v. Other considerations. The evidence this strategy, EPA would determine an (bromodichloromethane—large intestine supports similarity of potential response MCL as close to the individual MCLGs and kidney tumors, and bromoform— in children and adults. The basic as is technically feasible, taking cost large intestine tumors); and one is a biology of toxicity caused by cell into consideration, for each THM. EPA nonlinear non-carcinogen damage due to oxidative damage is would propose an MCLG of 0.07 mg/L (dibromochloromethane) which is a expected to be the same. There is for chloroform and maintain the Stage 1 liver toxicant. EPA therefore, chose not nothing about the incidence and DBPR MCLGs for BDCM, DBCM, and to develop a combined MCLG for etiology of liver and kidney cancer in bromoform (USEPA 1998c). EPA TTHM. Consequently, after considering children to indicate that they would be analyzed the impact such an MCL this alternative option in some detail, inherently more sensitive to this mode strategy would have and ultimately EPA is today proposing an MCLG of of action. Most importantly in this case, rejected this option. This approach 0.07 mg/L for chloroform. represents a fundamental shift from the children appear to be no different 3. Request for Comment quantitatively in ability to carry out the TTHM strategy agreed to by oxidative metabolism step for the stakeholders and EPA as part of the M– Based on the information presented induction of toxicity and cancer and DBP negotiation process and reflected in previously, EPA is proposing an MCLG may, as fetuses, be less susceptible the 1998 Stage 1 DBPR. In addition, one for chloroform of 0.07 mg/L. EPA (USEPA 1999c). important component of the existing requests comments on the MCLG and on Some commenters on the March 1998 single MCL is that TTHMs are an EPA’s cancer assessment for chloroform. NODA were concerned that EPA did not indicator for other DBPs. Developing a EPA also requests comments on the RfD, take drinking water epidemiology separate MCL for each THM would the default RSC of 20 percent, and the studies into account in its evaluation of move away from this indicator tap water consumption and body weight chloroform risk. EPA believes that while approach. Because precursor and DBP assumptions used in the MCLG the epidemiologic evidence suggests occurrence measurements are highly calculation. EPA solicits additional data that chlorinated drinking water may be variable, both temporally and on chloroform exposure via other associated with certain cancers and geographically, determining technical sources and routes. EPA requests reproductive, developmental effects feasibility for best available technology comment on the other options for pertinent to the risk of disinfectant (BAT) would be difficult. Compliance developing the chloroform MCLG that byproduct mixtures, it does not provide with individual THM standards would the Agency considered. insight into the risk from chloroform be very different from compliance based B. MCLGs for THMs and HAAs specifically. The SAB noted that ‘‘(t)he on a sum of the four THMs and it is not goal of the draft risk assessment (the clear what treatment technology shifts 1. What Is EPA Proposing Today? isolation of the effect of chloroform in would be needed. This problem would Today EPA is proposing new MCLGs drinking water) makes the extensive be particularly exacerbated in areas with of 0.02 mg/L for TCAA and 0.03 mg/L epidemiologic evidence on drinking high bromide, such as California. EPA for MCAA based on new toxicological water disinfection byproducts largely also projected that States would have a data. As a part of the Stage 1 DBPR, EPA irrelevant’’ to the specific question of difficult time overseeing (e.g., variances, finalized an MCLG of 0.3 mg/L for chloroform health risks because, in the exemptions, etc.) the more complicated TCAA. The Stage 1 DBPR did not available studies, chloroform cannot be rule that would result from this option. include an MCLG for MCAA (although isolated from other disinfection EPA considered establishing a single it was included as one of the five byproducts that may be in the drinking combined MCLG for TTHM. There is haloacetic acids in the HAA5 MCL). water (USEPA 2000f). The SAB noted precedent for using a toxicity With the exception of chloroform, that ‘‘the epidemiologic evidence is equivalency quotient (analogous to a discussed above, and these two HAAs, quite pertinent to the broader question combined MCLG) for dioxin and EPA is not revising any of the other of most direct regulatory concern, coplanar PCBs (USEPA 2000o, Draft MCLGs that were finalized in the Stage namely disinfection byproducts in the Dioxin Reassessment). From a scientific 1 DBPR. No significant new studies that aggregate’’. standpoint, a combined MCLG approach would change EPA’s MCLG estimates d. Feasibility of other options. During requires that the chemicals have a for BDCM, DBCM, bromoform, or DCAA the development of the MCLG for similar mode of action and health have been published since the Stage 1 chloroform, EPA considered a number endpoint. Chemicals within each of the DBPR. See section III for a summary of of options for both the chloroform dioxin and coplanar PCB classes have new health effects data. MCLG and the TTHM MCL. Today, EPA the same mode of action and endpoint is proposing the preferred option of a (target tissue). Within the PCB class, 2. How Was This Proposal Developed? 0.07 mg/L MCLG for chloroform. EPA noncoplanar PCBs have a different EPA reviewed the available literature primarily considered two other options mode of action than the coplanar PCBs. on BDCM, DBCM, bromoform, DCAA which are discussed in more detail later: Noncoplanar PCBs are, therefore, not and determined that there was no new

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00033 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18au03.027 49580 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

information that would cause EPA to clearly adverse liver toxicity endpoints, observed when pregnant rats were revise its MCLG estimates. New including increased serum levels of administered 330 mg/kg/day TCAA by toxicology studies on reproductive and liver enzymes (indicating leakage from gavage during gestation days 6 to 15 developmental effects and cancer are cells) or histopathological evidence of (Smith et al. 1989). Neither of these summarized in sections III.B. and III.D. necrosis, have been reported in rats, but studies identified a NOAEL. The results of today’s proposal. generally only at high doses. For of in vitro developmental toxicity EPA is proposing new MCLGs for example, in a rat chronic drinking water assays, including mouse and rat whole- TCAA and MCAA. The health effects study, increased hepatocyte necrosis embryo culture (Saillenfait et al. 1995; information and studies described in the was observed at a dose of 364 mg/kg/ Hunter et al. 1996) and frog embryo following two sections that support the day (DeAngelo et al. 1997). teratogenesis assay—Xenopus (FETAX) proposed MCLGs are summarized from In the DeAngelo et al.(1997) study, (Fort et al. 1993) yielded positive the Addendum to the Criteria Document groups of 50 male F344 rats were results. The Hydra test system (Fu et al. for Monochloroacetic Acid and administered TCAA in drinking water, 1990) produced negative results. Trichloroacetic Acid (USEPA 2003b). at 0, 50, 500, or 5000 mg/L, resulting in TCAA has been reported to induce The occurrence of MCAA and TCAA are time-weighted mean daily doses of 0, liver tumors in mice but not in rats discussed in the Stage 2 Occurrence 3.6, 32.5, or 364 mg/kg for 104 weeks. (USEPA 1994a). This observation has Assessment for Disinfectants and There were no significant differences in also been made in more recent drinking Disinfection Byproducts (USEPA water consumption or survival between water studies. Pereira (1996) observed 2003o). a. Trichloroacetic acid. In the the control and treatment groups. an increased incidence of hepatocellular final Stage 1 DBPR, EPA based its health Exposure to the high dose of TCAA adenomas and carcinomas in female effects assessment of TCAA on resulted in a significant decrease in B6C3F1 mice at doses of 262 mg/kg/day developmental toxicity and limited body weight of 11% at the end of the and higher after 82 weeks. In contrast, evidence of carcinogenicity (USEPA study. The absolute but not relative liver no increase in neoplastic liver lesions 1998c). Since then, the Agency has weight was decreased at the high dose. were found in F344 rats given doses up decided that the RfD based on a Complete necropsy and histopathology to 364 mg/kg/day for 104 weeks developmental LOAEL yields a less examination showed mild hepatic (DeAngelo et al. 1997). In addition, a conservative RfD than that based on cytoplasmic vacuolization in the two variety of recent mechanistic studies liver toxicity derived from the study by low-dose groups, but not in the high- have observed that TCAA either DeAngelo et al. (1997). Thus, the dose group. The severity of hepatic induced or promoted liver tumors in Agency has reassessed the health effects necrosis was increased mildly in the mice (Ferreira-Gonzalez et al. 1995; of TCAA based on liver toxicity and high-dose animals. Analyses of serum Pereira and Phelps, 1996; Tao et al. revised the RfD and MCLG. aspartate aminotransferase (AST) and 1996; Latendresse and Pereira, 1997; TCAA induces systemic, noncancer alanine aminotransferase (ALT) Stauber and Bull, 1997; Tao et al. 1998). effects in animals and humans that can activities at the end of exposure showed Recent mutagenicity data have be grouped into three categories: a significant decrease in AST activity in metabolic alterations, liver toxicity; and the mid-dose group and a significant provided mixed results (Giller et al. developmental toxicity. The primary increase in ALT level in the high-dose 1997; DeMarini et al. 1994; Harrington- site of TCAA toxicity is the liver group. Since increased serum ALT or Brock et al. 1998). TCAA did not induce (USEPA1994a; Dees and Travis, 1994; AST levels reflect hepatocellular oxidative DNA damage in mice Acharya et al. 1995; Acharya et al. 1997; necrosis, the increased ALT at the high following dosing for either 3 or 10 DeAngelo et al.1997). dose is considered an adverse effect, weeks (Parrish et al. 1996). Studies on The liver has consistently been while a non-dose related decrease of DNA strand breaks and chromosome identified as a target organ for TCAA AST is not. Peroxisome proliferation damage produced mixed results (Nelson toxicity in short-term (Goldsworthy and was increased significantly in the high- and Bull, 1988; Chang et al. 1991; Popp, 1987; DeAngelo et al. 1989; dose animals. There was no evidence of Mackay et al. 1995; Harrington-Brock et Sanchez and Bull, 1990) and longer- any exposure-related increase in al. 1998). term (Bull et al. 1990; Mather et al. hepatocyte proliferation. Based on the According to the 1999 Draft 1990; Bhat et al. 1991) studies. significant decrease in body weight Guidelines for Carcinogen Risk Peroxisome proliferation has been a (≥10%), minimal histopathology Assessment (USEPA 1999a), a primary endpoint evaluated, with mice changes, and increased serum ALT compound is appropriately classified as reported to be more sensitive to this level, the high dose of 364 mg/kg/day is ‘‘Suggestive Evidence of effect than rats. More recent studies considered the LOAEL and the mid dose Carcinogenicity, but Not Sufficient to have confirmed these earlier findings. of 32.5 mg/kg/day is considered the Assess Human Carcinogenic Potential’’ TCAA-induced peroxisome proliferation NOAEL. when ‘‘the evidence from human or was observed in B6C3F1 mice exposed There are no reproductive toxicity animal data is suggestive of for 10 weeks to doses as low as 25 mg/ studies of TCAA. The results of an in carcinogenicity, which raises a concern kg/day (Parrish et al. 1996), while in rats vitro fertilization assay indicated that for carcinogenic effects but is judged not exposed to TCAA for up to 104 weeks TCAA might decrease fertilization sufficient for a conclusion as to human (DeAngelo et al. 1997), peroxisome (Cosby and Dukelow, 1992). The carcinogenic potential’’. Based on proliferation was observed at 364 mg/ available data suggest that TCAA is a uncertainty surrounding the relevance kg/day, but not at 32.5 mg/kg/day. developmental toxicant. TCAA of the liver tumor data in B6C3F1 mice, Increased liver weight and significant increased resorptions, decreased TCAA can best be described as increases in hepatocyte proliferation implantations, and increased fetal ‘‘Suggestive Evidence of have been observed in short-term cardiovascular malformations when Carcinogenicity, but Not Sufficient to studies in mice at doses as low as 100 administered to pregnant rats at 291 mg/ Assess Human Carcinogenic Potential’’ mg/kg/day (Dees and Travis, 1994), but kg/day (Johnson et al. 1998) on gestation under the 1999 Draft Guidelines for no increase in hepatocyte proliferation days 1–22. In another study, decreased Carcinogen Risk Assessment. Thus a was noted in rats given TCAA at similar fetal weight and length, and increased quantitative estimate of cancer potency doses (DeAngelo et al. 1997). More cardiovascular malformations were is not supported.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00034 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49581

The RfD for TCAA of 0.03 mg/kg/day of tap water per day for a 70 kg adult, a role in the development of the mouse is based on the NOAEL of 32.5 mg/kg/ a relative source contribution (RSC) of tumors, rats also exhibit a peroxisomal day for liver histopathological changes 20%, and an additional safety factor to proliferative response after exposure to identified by DeAngelo et al. (1997). account for possible carcinogenicity. TCA, yet do not develop tumors. Other The RfD includes an uncertainty factor EPA has traditionally applied an data suggest that promotion of initiated of 1000 (composite uncertainty factor additional safety factor of 1–10 beyond cells and/or disrupted cell signaling consisting of three factors of 10 chosen the uncertainty factors included in the may be involved in the mode of action to account for extrapolation from a RfD to the MCLG to account for possible for the mouse tumors. Together these NOAEL in animals, inter-individual carcinogenicity in cases where there is factors argue against quantification of variability in humans, and limited evidence of carcinogenicity from the mouse liver tumors using linear insufficiencies in the database, drinking water, considering weight of extrapolation from the dose-response including the lack of full evidence, pharmacokinetics, potency curve, but are not sufficient to rule out histopathological data in a second and exposure (USEPA 1994b, p.38678). concern for a tumorigenic response. species, the lack of a developmental EPA is proposing this additional safety Accordingly, EPA has employed the ten- toxicity study in second species, and the factor of 10 for TCAA for the following fold additional safety factor in lack of a multi-generation reproductive reasons: TCAA causes liver tumors in determination of the Lifetime Health study). mice but does not do so in rats. In Advisory for TCAA. EPA requests The MCLG is calculated to be 0.02 addition, although peroxisome comment on the use of 10 as the mg/L using the following assumptions: proliferation (a mode of action of additional safety factor for possible an adult tap water consumption of 2 L limited relevance to humans) may play carcinogenicity.

(0.03 mg/kg/day)(70 kg)(20%) MCLG for TCAA = = 0.02 mg/L (rounded) (2 L/day)(10)

An RSC factor of 20% is used to was observed at 30 mg/kg/day and rodent bioassays (NTP 1992; DeAngelo account for exposure to TCAA in cardiac lesions progressed in severity et al. 1997). In chronic oral gavage sources other than tap water, such as with increasing dose. Liver and kidney studies, a LOAEL of 15 mg/kg/day (the ambient air and food. Although TCAA toxicity were only observed at higher lowest dose tested) for decreased is nonvolatile and inhalation while doses (NTP 1992). In a two-year study, survival was identified in rats. In mice showering is not expected to be a major decreased survival and nasal and the NOAEL was 50 mg/kg/day and the contribution to total dose, rain waters forestomach hyperplasia were observed LOAEL was 100 mg/kg/day for nasal contain 0.01–1.0 µg/L of TCAA in mice at 50 mg/kg/day (NTP 1992). A and forestomach epithelium hyperplasia (Reimann et al. 1996) and it can be more recent study confirms the heart (NTP 1992). In a more recent chronic assumed to be detected in the and nasal cavities as target sites for study, DeAngelo et al. (1997) reported a atmosphere. Limited data on MCAA. DeAngelo et al. (1997) noted LOAEL of 3.5 mg/kg/day in rats given concentrations of TCAA in air (NATICH decreased body weight at 26.1 mg/kg/ MCAA in their drinking water, based on 1993) indicate inhalation of TCAA in day and myocardial degeneration and increased absolute and relative spleen ambient air may contribute to overall inflammation of the nasal cavities in weight. Although spleen weight was exposure. Concentrations of TCAA that rats exposed to doses of 59.9 mg/kg/day decreased at the mid and high doses, have been measured in a limited for up to 104 weeks. this might reflect the masking effect of selection of foods including vegetables, No studies were located on the overt toxicity. As evidence for this, fruits, grain and bread (Reimann et al. reproductive toxicity of MCAA and the decreased body weight (>10%), liver, 1996) are comparable to that in water. potential developmental toxicity of kidney, and testes weight changes were About 3 to 33% of TCAA in cooking MCAA has not been adequately tested. reported beginning at the next higher water have been reported to be taken up Two developmental toxicity studies dose of 26.1 mg/kg/day. No increased by the food during cooking in a recent were identified. Johnson et al. (1998) spleen weight was reported in the NTP research summary (Raymer et al. 2001). reported markedly decreased maternal (1992) bioassays, but the lowest dose in In addition, there are uses of chlorine in weight gain, but no developmental rats caused severe toxicity, and the food production and processing, and effects, in rats exposed to 193 mg/kg/ lowest dose in mice was more than an TCAA may occur in food as a byproduct day MCAA through gestation days 1–22, order of magnitude higher than the of chlorination (USEPA 1994a). only fetal heart was examined. In LOAEL in the DeAngelo et al. (1997) Therefore, ingestion of TCAA in food contrast, in a published abstract, Smith study. may also contribute to the overall et al. (1990) reported an increase in exposure. A recent dermal absorption cardiovascular malformations when According to the 1999 Draft study of DCAA and TCAA from pregnant rats were exposed to 140 mg/ Guidelines for Carcinogen Risk chlorinated water suggested that the kg/day; this was also the LOAEL for Assessment (USEPA 1999a), a dermal contribution to the total doses of maternal toxicity, based on marked compound is appropriately classified as DCAA and TCAA from routine decreases in weight gain. MCAA was ‘‘Not Likely to be Carcinogenic to household uses of drinking water is less noted as a potential developmental Humans’’ when it has ‘‘been evaluated than 1% (Kim and Weisel, 1998). toxicant in in vitro screening assays in at least two well-conducted studies in b. Monochloroacetic acid. Subchronic using Hydra (Fu et al. 1990; Ji et al. two appropriate animal species without and chronic oral dosing studies suggest 1998). demonstrating carcinogenic effects.’’ that the primary targets for MCAA- MCAA has yielded mixed results in MCAA can best be described as ‘‘Not induced toxicity include the heart and genotoxicity assays (USEPA 1994a; Likely to be Carcinogenic to Humans’’ nasal epithelium. In a 13-week oral Giller et al. 1997), but has not induced under the 1999 Draft Guidelines for gavage study, decreased heart weight a carcinogenic response in chronic Carcinogen Risk Assessment.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00035 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.024 49582 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

The RfD for MCAA of 0.004 mg/kg/ LOAEL, and insufficiencies in the RfD because it identifies the lowest day is based on a LOAEL of 3.5 mg/kg/ database, including the lack of adequate LOAEL, and dosing was in drinking day for increased spleen weight in rats developmental toxicity studies in two water, which is more appropriate for (DeAngelo et al. 1997) and application species, and the lack of a multi- human health risk assessment. of an uncertainty factor of 1000 generation reproductive study). Two The MCLG is calculated to be 0.03 (composite uncertainty factor consisting developmental toxicity studies have mg/L using the following assumptions: of two factors of 10 chosen to account been reported (Johnson et al. 1998; for extrapolation from an animal study, Smith et al. 1990), but the NOAELs an adult tap water consumption of 2 L and inter-individual variability in yielded less conservative RfDs. The of tap water per day for a 70 kg adult, humans; as well as two factors of 3 for study by DeAngelo et al (1997) is the and a relative source contribution of extrapolation from a minimal effect most appropriate for derivation of the 20 %.

(0.004 mg/kg/day)(70 kg)(20%) MCLG for MCAA = = 0.03 mg/L (rounded) (2 L/day)

An RSC factor of 20% is used to otherwise receive finished water from sections of this preamble provide account for exposure to MCAA in other another water system (a wholesale further details as noted. sources in addition to tap water. system). As described in this section, a. Definitions. To address consecutive Although MCAA is nonvolatile and consecutive systems face particular systems in the Stage 2 DBPR, the inhalation while showering is not challenges in providing water that meets Agency must define them, along with a expected to be a major contribution to regulatory standards for DBPs and other number of related terms. total dose, rain waters contain 0.05–9 contaminants whose concentration can EPA is proposing to define a µg/L of MCAA (Reimann et al. 1996) increase in the distribution system. consecutive system in the Stage 2 DBPR and it can be assumed to be detected in Moreover, current regulation of DBP as a public water system that buys or the atmosphere. Presence of MCAA has levels in consecutive systems varies otherwise receives some or all of its also been reported in rain waters; thus, widely among States. In consideration finished water from one or more inhalation of MCAA in ambient air may of these factors, EPA is proposing wholesale systems for at least 60 days contribute to overall exposure. monitoring, compliance schedule, and per year. In addition to buying finished Concentrations of MCAA that have been other requirements specifically for water, some consecutive systems also measured in a limited selection of foods consecutive systems. These operate a treatment plant (meaning a including vegetables, fruits, grain and requirements are intended to facilitate plant that treats source water to produce bread (Reimann et al. 1996) are compliance by consecutive systems finished water). As described in section comparable to that in water. About 2.5 with MCLs for TTHM and HAA5 under V.I., monitoring requirements under the to 62% of MCAA in cooking water has the Stage 2 DBPR. Further, this Stage 2 DBPR proposal differ depending been reported to be taken up by food approach will help to ensure that on whether a consecutive system buys during cooking in a recent research consumers in consecutive systems all of its finished water year-round or, summary (Raymer et al. 2001). In receive equivalent public health alternatively, produces some of its addition, there are uses of chlorine in protection. This section begins with a finished water through treating source food production and processing, and summary of how EPA proposes to water. MCAA may occur in food as a regulate consecutive systems under the EPA proposes to define finished water byproduct of chlorination (USEPA Stage 2 DBPR. The intent of this section as water that has been introduced into 1994a). Therefore, ingestion of MCAA in is to provide an overview of all the distribution system of a public water food may also contribute to the overall consecutive system requirements in system and is intended for distribution exposure. Assuming dermal absorption today’s proposal. Detailed explanations without further treatment, except that rate of MCAA is similar to DCAA, of these requirements are provided in necessary to maintain water quality dermal contribution to the total doses of later sections of this preamble. The (such as booster disinfection). With this MCAA from routine household uses of overview of consecutive system definition, water entering the drinking water should be minor (see requirements is followed by an distribution system is finished water V.B.2.a.). explanation of why EPA has taken this even if a system subsequently applies approach to consecutive systems in 3. Request for Comment additional treatment like booster today’s proposal, including disinfection to maintain a disinfectant EPA requests comment on the new recommendations from the Stage 2 M– residual throughout the distribution MCLGs for TCAA (0.02 mg/L) and DBP Federal Advisory Committee. system. MCAA (0.03 mg/L) and all the factors 1. What Is EPA Proposing Today? In today’s proposal, EPA defines a incorporated in the derivation of the wholesale system as a public water MCLGs, including the RfDs and RSCs. As public water systems, consecutive system that treats source water and then EPA also solicits health effect systems must provide water that meets sells or otherwise delivers finished information on DBAA and the MCLs for TTHM and HAA5 under water to another public water system for monobromoacetic acid (MBAA), for the proposed Stage 2 DBPR, and must at least 60 days per year. Delivery may which MCLGs have not yet been carry out associated monitoring, be through a direct connection or established. reporting, recordkeeping, public through the distribution system of notification, and other requirements. another consecutive system. Under this C. Consecutive Systems The following discussion summarizes definition, a consecutive system that Today’s proposal includes provisions how the Stage 2 DBPR requirements passes water from a wholesaler to for consecutive systems, which are apply to consecutive systems, beginning another consecutive system, and that public water systems that purchase or with a series of definitions. Later does not also treat source water, is not

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00036 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.025 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49583

a wholesale system. Rather, the system under such an approach, each occurred. Several examples are as that actually produces the finished consecutive system must collect at least follows: water is responsible for wholesale one sample among the total number of —If a consecutive system has hired its system requirements under the samples required for the combined wholesale system under contract to proposed Stage 2 DBPR. distribution system and will base monitor in the consecutive system A consecutive system entry point is compliance on samples collected within and the wholesale system fails to defined as a location at which finished its distribution system. The consecutive monitor, the consecutive system is in water is delivered at least 60 days per system is responsible for ensuring that violation because it has the legal year from a wholesale system to a required monitoring is completed and responsibility for monitoring under consecutive system. Section V.I. the system is in compliance. The State/EPA regulations. presents the relationship between consecutive system may conduct the —If monitoring results in a consecutive consecutive system entry points and monitoring itself or arrange for the system indicate an MCL violation, the proposed Stage 2 DBPR monitoring monitoring to be done by the wholesale consecutive systems is in violation requirements. The combined system or another outside party. because it has the legal responsibility distribution system is the Whatever approach it chooses, the for complying with the MCL under interconnected distribution system consecutive system must document its State/EPA regulations. The consisting of the distribution systems of monitoring strategy as part of its DBP consecutive system may set up a wholesale systems and of the monitoring plan. contract with its wholesale system consecutive systems that receive Finally, EPA is proposing that that details water quality delivery finished water from those wholesale consecutive systems not conducting specifications. system(s). disinfectant residual monitoring comply —If a wholesale system has a violation b. Monitoring. For consecutive with the monitoring requirements and and provides that water to a systems that both purchase finished MRDLs for chlorine and chloramines. water and treat source water to produce c. Compliance schedules. EPA is consecutive system, the wholesale finished water for at least part of the proposing that consecutive systems of system is in violation. Whether the year, EPA is proposing monitoring any size comply with the requirements consecutive system is in violation will requirements under a treatment plant- of the Stage 2 DBPR on the same depend on the situation. The based approach, described in section schedule as required for the largest consecutive system will also be in V.I. This is the approach proposed for system in the combined distribution violation unless it conducted non-consecutive systems under the system. This includes the schedule for monitoring that showed that the Stage 2 DBPR as well. Under this carrying out the IDSE, described in violation was not present in the approach, the sampling requirements for section V.H, and for meeting the Stage consecutive system. consecutive systems will be influenced 2B MCLs for TTHM and HAA5, f. Public notice and consumer by both the number of treatment plants described in section V.D. As discussed confidence reports. The responsibilities operated by the system and the number later in this section, EPA is proposing for public notification and consumer of consecutive system entry points, as simultaneous compliance schedules confidence reports rest with the well as population served and source under the Stage 2 DBPR for all systems individual system. Under the Public water type. (both wholesalers and consecutive Notice Rule and Consumer Confidence For consecutive systems that purchase systems) in a combined distribution Report Rule, the wholesale system is all of their finished water year-round, system because this may allow for more responsible for notifying the EPA is proposing monitoring cost-effective compliance with TTHM consecutive system of analytical results requirements under a population-based and HAA5 MCLs. This is also consistent and violations related to monitoring approach, also described in section V.I. with the recommendations of the Stage conducted by the wholesale system. Under the population-based approach, 2 M–DBP Advisory Committee. See Consecutive systems are required to the population of the consecutive section V.J for details of compliance conduct appropriate public notification system will determine the sampling schedule requirements. after a violation (whether in the requirements. EPA believes this d. Treatment. While consecutive wholesale system or the consecutive approach is more appropriate than systems often do not need to treat system). In their consumer confidence plant-based monitoring because these finished water received from a report, consecutive systems must consecutive systems do not have wholesale system, they may need to include results of the testing conducted treatment plants. As noted in section implement procedures to control the by the wholesale system unless the V.I., EPA is requesting comment on formation of DBPs in the distribution consecutive system conducted extending population-based monitoring system. For consecutive systems, EPA is equivalent testing that indicated the to all systems, including non- proposing that the BAT for meeting consecutive system was in compliance, consecutive systems. EPA has prepared TTHM and HAA5 MCLs is in which case the consecutive system draft guidance for implementing the chloramination with management of reports its own compliance monitoring IDSE monitoring requirements hydraulic flow and storage to minimize results. (described in section V.H.) using the residence time in the distribution g. Recordkeeping and reporting. population-based approach (USEPA system. This BAT stems from the Consecutive systems are required to 2003j). recognition that treatment to remove keep all records required of PWSs EPA is also proposing that States have already-formed DBPs or minimize regulated under this rule. They are also the opportunity to specify alternative further formation is different from required to report to the State monitoring requirements for multiple treatment to prevent or reduce their monitoring results, violations, and other consecutive systems in a combined formation. See section V.F for additional actions, and are required to consult with distribution system. This option allows information on BATs and their role in the State after a significant excursion. States to consider complex consecutive compliance with MCLs. h. State special primacy conditions. system configurations for which e. Violations. Under this proposal, EPA is aware that due to the alternative monitoring strategies might monitoring and MCL violations are complicated wholesale system- be more appropriate. As a minimum assigned to the PWS where the violation consecutive system relationships that

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00037 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49584 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

exist nationally, there will be cases in consecutive systems are treated as —Whether any additional terms need to where the standard monitoring monitoring sites within the combined be defined and, if so, what the framework proposed today will be distribution system. Once fully definition should be. difficult to implement. Therefore, the implemented, this proposed rule will —Whether the criteria for States’ use of Agency is proposing to allow States to ensure similar protection for consumers the special primacy criteria and other develop, as a special primacy condition, in consecutive systems. State responsibilities are appropriate a program under which the State can EPA is proposing that consecutive and adequate. modify monitoring requirements for systems and wholesale systems be on —Whether it is necessary to require that consecutive systems. These the same compliance schedule because consecutive system treatment be modifications must not undermine generally the most cost-effective way to installed on the same compliance public health protection and all achieve compliance with TTHM and schedule as the wholesale system in systems, including consecutive systems, HAA5 MCLs is to treat at the source, cases where the size of the must comply with the TTHM and HAA5 typically through precursor removal or consecutive system might otherwise MCLs based on the LRAA. However, alternative disinfectants. For a allow it a longer compliance time such a program would allow the State wholesale system to make the best frame and the consecutive system to establish monitoring requirements decisions concerning the treatment treatment does not affect water quality that account for complicated steps necessary to meet TTHM and in any other system. distribution system relationships, such HAA5 LRAAs under the Stage 2 DBPR, D. MCLs for TTHM and HAA5 as where neighboring systems buy from both in its own distribution system and and sell to each other regularly in the distribution systems of 1. What Is EPA Proposing Today? throughout the year, water passes consecutive systems it serves, the Today, EPA is proposing use of through multiple consecutive systems wholesale system must know the DBP before it reaches a user, or a large group locational running annual averages levels throughout the combined (LRAAs) to determine compliance with of interconnected systems have a distribution system. Without this complicated combined distribution the MCLs for TTHM and HAA5. information, the wholesale system may Consistent with the Stage 2 M-DBP system. EPA intends to develop a design treatment changes that allow the guidance manual to address Advisory Committee recommendation, wholesale system to achieve EPA is proposing a phased approach for development of a State program and compliance, but leave the consecutive other consecutive system issues. LRAA implementation to allow systems system out of compliance. EPA also to identify compliance monitoring 2. How Was This Proposal Developed? recognizes that there may be cases locations for Stage 2B while facilitating where a consecutive system needs to The practice of public water systems transition to the new compliance add treatment even after a wholesale buying and selling water to each other strategy and maintaining simultaneous system has optimized its own treatment has been commonplace for many years. compliance schedules for the Stage 2 train. Reasons include saving money on DBPR and the LT2ESWTR. pumping, treatment, equipment, and In consideration of these issues, the In Stage 2A, all systems must comply personnel; assuring an adequate supply Stage 2 M–DBP Advisory Committee with MCLs of 0.120 mg/L for TTHM and during peak demand periods; acquiring recognized two principles related to 0.100 mg/L for HAA5 as LRAAs using emergency supplies; selling surplus consecutive systems: (1) Consumers in Stage 1 DBPR compliance monitoring supplies; delivering a better product to consecutive systems should be just as sites. In addition, during this time consumers; and meeting Federal and well protected as customers of all period, all systems must continue to State water quality standards. EPA systems, and (2) monitoring provisions comply with the Stage 1 DBPR MCLs of estimates that there are at least 8500 should be tailored to meet the first 0.080 mg/L TTHM and 0.060 mg/L consecutive systems nationally, based principle. Accordingly, the Advisory HAA5 as RAAs. on the definitions being proposed today. Committee recommended that all In Stage 2B, all systems, including Consecutive systems face particular wholesale and consecutive systems consecutive systems, must comply with challenges in providing water that meets comply with provisions of the Stage 2 MCLs of 0.080 mg/L TTHM and 0.060 regulatory standards for contaminants DBPR on the same schedule required of mg/L HAA5 as LRAAs using sampling that can increase in the distribution the wholesale or consecutive system sites identified under the Initial system. Examples of such contaminants serving the largest population in the Distribution System Evaluation (IDSE) include coliforms, which can grow if combined distribution system. In (discussed in section V.H.). favorable conditions exist, and some addition, the Advisory Committee Details of proposed monitoring DBPs, including THMs and HAAs, recommended that EPA solicit requirements and compliance schedules which can increase when a disinfectant comments on issues related to are discussed in preamble sections V.I. and DBP precursors continue to react in consecutive systems that the Advisory and V.J., respectively, and may be found the distribution system. Committee had not fully explored in § 141.136 and subpart V of today’s EPA is proposing requirements (USEPA 2000g). EPA agrees with these rule. specifically for consecutive systems recommendations and they are reflected because States have taken widely in today’s proposal. 2. How Was This Proposal Developed? varying approaches to regulating DBPs 3. Request for Comment a. Definition of an LRAA. The primary in consecutive systems. For example, objective of the LRAA is to reduce some States do not regulate DBP levels EPA requests comment on all exposure to high DBP levels. For an in consecutive systems that deliver consecutive system issues related to this LRAA, an annual average must be disinfected water but do not add a rule. Specifically, EPA requests computed at each monitoring site. The disinfectant. Other States determine comment on the following: RAA compliance basis of the 1979 compliance with DBP standards based —Whether the proposed definitions TTHM rule and the Stage 1 DBPR allows on the combined distribution system adequately address various wholesale a system-wide annual average under that includes both the wholesaler and system-consecutive system which high DBP concentrations in one consecutive systems. In this case, sites relationships and issues. or more locations are averaged with, and

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00038 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49585

dampened by, lower concentrations calculating compliance with the MCLs RAA, and the proposed Stage 2 DBPR elsewhere in the distribution system. for TTHM between a Stage 1 DBPR LRAA. Figure V–1 illustrates the difference in BILLING CODE 6560–50–P

BILLING CODE 6560–50–P

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00039 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.008 49586 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

b. Consideration of regulatory EPA and the Advisory Committee c. Basis for the LRAA. This section alternatives. This section will discuss reviewed vast quantities of data and discusses the data and information EPA EPA’s and the Stage 2 M-DBP Advisory many analyses that addressed health used to determine that the LRAA is an Committee’s decision-making process as effects, DBP occurrence, predicted appropriate compliance strategy for an array of alternative MCL strategies reductions in DBP levels, predicted today’s proposed rule. EPA has chosen were considered. EPA believes that the technology changes, and capital, annual, compliance based on an LRAA due to MCL alternative proposed today (MCLs and household costs. Details of the concerns about levels of DBPs above the of 0.080 mg/L TTHM, 0.060 mg/L HAA5 compliance, occurrence, and cost MCL in some portions of the as LRAAs) is supported by the best forecasts for the four alternative rule distribution system. The LRAA standard available research, data, and analysis. scenarios are described in the Stage 2 will eliminate system-wide averaging. The science related to cancer and DBPR Economic Analysis (EA) (USEPA The individuals served in areas of the reproductive and developmental health 2003i) and the Stage 2 DBPR Occurrence distribution system with above average effects that may be associated with Document (USEPA 2003o). DBP occurrence levels masked by DBPs, in conjunction with occurrence In the end, the Advisory Committee averaging under an RAA are not data that show that a significant number recommended the Preferred Alternative receiving the same level of health of high DBP levels occur under current in combination with the IDSE which protection. Although an LRAA standard regulatory scenarios, justify a change in they believed would reduce exposure to still allows averaging at a single location regulation. EPA believes that this high levels of DBPs. Today, EPA is over an annual period, EPA believes proposal achieves an appropriate proposing the Preferred Alternative in that changing the basis of compliance balance between the available science combination with the IDSE. from an RAA to an LRAA will result in and the uncertainties. EPA believes that The only difference between the decreased exposure to above average regulatory action is necessary and Preferred Alternative and Alternative 1 DBP levels (see section VII.A. for prudent in the interest of further public is the bromate MCL. The Advisory predictions of DBP reductions under the health protection and that the LRAA Committee’s recommendation to LRAA MCLs). This conclusion is based alternative in combination with the maintain the Stage 1 DBPR bromate on three considerations: IDSE is a balanced and reasonable MCL of 0.010 mg/L is discussed in (1) There is considerable evidence approach. Although it will not remove section V.G. of today’s proposal. that under the current RAA MCL all DBP peaks (individual samples with Alternatives 2 and 3 are significantly compliance monitoring requirements a values greater than the MCL), this more stringent than the Stage 1 DBPR small but significant proportion of proposed regulation will ensure that with respect to the TTHM and HAA5 monitoring locations experience high DBP exposures across a system’s requirements. Alternative 2 would DBP levels. As summarized in section distribution system are further reduced, require that all samples be below the IV of this preamble, 14 and 21% of are more equitable, and may reduce MCL. Because DBP occurrence is Information Collection Rule systems cancer and reproductive and variable across the distribution system currently meeting the Stage 1 DBPR developmental risk. and over time (as discussed in section RAA MCLs had TTHM and HAA5 single The Advisory Committee discussions IV), systems would have to base their sample concentrations greater than the primarily focused on the relative disinfectant and treatment strategies on Stage 1 MCLs and ranged up to 140 µg/ magnitude of exposure reduction versus controlling their highest DBP L and 130 µg/L respectively (Figures IV– the expected impact on the water occurrence levels. Alternative 3 1 and IV–2), though most of these industry and its customers. Initially, maintains the Stage 1 DBPR RAA exceedences were below 100 µg/L. this analysis compared expected compliance calculation, but reduces the (2) In some situations, the populations reductions in DBP levels and Stage 1 DBPR MCLs by 50 percent. Both served by certain portions of the predictions of treatment technology alternatives 2 and 3 would cause distribution system consistently receive changes associated with a wide variety significant changes in treatment for a water that exceeds the MCL even though of Stage 2 DBPR MCL alternatives. large number of systems. The estimated the system is in compliance. As After initial discussions, EPA and the costs for Alternatives 2 and 3 are discussed in section IV of this preamble, Advisory Committee primarily focused approximately an order of magnitude some Information Collection Rule on four types of alternative rule above the costs for the Preferred systems meeting the Stage 1 DBPR RAA scenarios. Alternative (see section VII.B.). MCLs had monitoring locations that Preferred Alternative.—MCLs of 0.080 Consistent with this greater stringency exceeded 0.080 mg/L TTHM and/or mg/L TTHM and 0.060 mg/L HAA5 as of alternatives 2 and 3, the predicted 0.060 mg/L HAA5 as an annual average LRAAs. Bromate MCL of 0.010 mg/L. DBP reductions and the resulting health (i.e., as LRAAs) by up to 25% (Figures Alternative 1.—MCLs of 0.080 mg/L benefits for them are greater than those IV–3 and IV–4). Five percent of plants TTHM and 0.060 mg/L HAA5 as predicted for the Preferred Alternative. that achieved compliance with the Stage LRAAs. Bromate MCL of 0.005 mg/L. Although all members of the Advisory 1 TTHM MCL of 0.080 mg/L based on Alternative 2.—MCLs of 0.080 mg/L Committee believed that the science an RAA had a particular sampling TTHM and 0.060 mg/L HAA5 as showing reproductive and location that exceeded 0.080 mg/L as an individual sample maximums (i.e., no developmental health effects that have LRAA (Figure IV–3). Figure IV–4 shows single sample could exceed the MCL). been associated with DBPs was similar results based on Information Bromate MCL of 0.010 mg/L. sufficient to cause concern and warrant Collection Rule HAA5 data. Three Alternative 3.—MCLs of 0.040 mg/L regulatory action, the Advisory percent of plants that met the Stage 1 TTHM and 0.030 mg/L HAA5 as Committee did not believe that the HAA5 MCL of 0.060 mg/L as an RAA RAAs. Bromate MCL of 0.010 mg/L. association was certain enough to justify had a sampling location that exceeded EPA and the Advisory Committee, the substantial change in treatment 0.060 mg/L as an LRAA. Customers with assistance from the Technical technologies that would be required to served at these locations consistently Workgroup, conducted an in-depth meet these alternatives. Thus, the received water with TTHM and/or analysis of these regulatory alternatives. Advisory Committee rejected HAA5 concentrations higher than the In the process of evaluating alternatives, Alternatives 2 and 3. system-wide MCL.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00040 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49587

(3) Compliance based on an LRAA facilitate compliance with the LRAA. In evaluating this recommendation, will remove the opportunity for systems EPA also requests comment on the Stage EPA believes that the Advisory to average out samples from high and 2A MCLs of 0.120 mg/L TTHM and Committee’s intent was clear with low quality water sources. Some 0.100 mg/L HAA5 as LRAAs and on the regard to the need for guidance on how systems are able to comply with an RAA long-term MCLs of 0.080 mg/L TTHM to evaluate and reduce significant MCL even if they have a plant with a and 0.060 mg/L HAA5 as LRAAs. excursions. However, the Agreement is poor quality water source (that thus less clear on how, and where, to define produces high concentrations of DBPs) E. Requirements for Peak TTHM and what constitutes a significant excursion, because they have another plant that has HAA5 Levels and how to define the scope of the a better quality water source (and thus 1. What Is EPA Proposing Today? evaluation. EPA draft guidance lower concentrations of DBPs). recommends several approaches for Today, EPA is proposing that, Individuals served by the plant with the determining whether significant concurrent with Stage 2B, systems must poor quality source will usually have excursions have occurred. While today’s specifically document occurrences of higher DBP exposure than individuals proposal requires an evaluation only of peak DBP levels, termed significant served by the other plant. distribution system operational excursions. In support of this provision, d. Basis for phasing LRAA practices, EPA believes that many EPA is proposing that States, as a compliance. EPA believes that a phased systems would benefit from a broader special primacy condition, develop approach for LRAA implementation will evaluation that includes treatment plant criteria for determining whether a facilitate transition to the new and other system operations. compliance requirements. Stage 2A of system has a significant excursion. EPA EPA recognizes that different this proposed rule does not require has developed draft guidance for stakeholders have different points of systems to conduct any additional systems and States on how systems may view on whether specific criteria that monitoring. They will continue to determine whether they have significant initiate the evaluation of significant monitor at Stage 1 DBPR locations. excursions. EPA is also proposing that excursions should be included in the Because the LRAA calculation is the a system that has a significant excursion rule or in guidance. EPA also recognizes same as the RAA calculation if there is must: (1) Evaluate distribution system that different stakeholders may have only one site, Stage 2A compliance only operational practices to identify different perspectives on how to applies to systems that monitor at more opportunities to reduce DBP levels identify a significant excursion. For this than one site and will only affect (such as tank management to reduce proposal, EPA has prepared draft medium and large surface water systems residence time and flushing programs to guidance for systems and States on how (serving at least 10,000 people) or reduce disinfectant demand), (2) to (1) determine whether a significant systems with multiple plants. Thus, the prepare a written report of the excursion has occurred, using several majority of ground water systems, small evaluation, and (3) no later than the different options, (2) conduct significant surface water systems, and some next sanitary survey, review the excursion evaluations, and (3) reduce consecutive systems are not affected by evaluation with their State. This review significant excursion occurrence. the proposed Stage 2A requirements. will take place under the sanitary e. TTHM and HAA5 as Indicators. In survey components calling for the State 3. Request for Comment part, both the TTHM and HAA5 classes to review monitoring, reporting, and EPA requests comment on the are regulated because they occur at high data verification and system proposed approach for addressing levels and represent chlorination management and operation. significant excursions and on the draft guidance. Is a special primacy condition byproducts that are produced from 2. How Was This Proposal Developed? source waters with a wide range of the appropriate means for allowing water quality. The combination of Because individual measurements flexibility in identifying significant TTHM and HAA5 represent a wide from a location are averaged over a four- excursions while ensuring that such variety of compounds resulting from quarter period to determine compliance, evaluations occur? Is the sanitary survey bromine substitution and chlorine there may be occurrence levels that the appropriate mechanism for substitution reactions (i.e., bromoform exceed the MCL even when a system is reviewing significant excursion data has 3 bromines, TCAA has 3 chlorines, in compliance with an LRAA MCL. EPA with the State? Should a system be BDCM has one bromine and two and the Advisory Committee were required to take corrective action when chlorines, etc). EPA believes that the concerned about these exposures to significant excursions occur? Should the TTHM and HAA5 classes serve as an peak levels of DBPs and the possible required scope of the evaluation be indicator for unidentified and risk they might pose. This concern was expanded beyond distribution system unregulated DBPs. EPA believes that clearly reflected in the Agreement in operations? controlling the occurrence levels of Principle, which states, EPA also requests comment on TTHM and HAA5 will control the levels ‘‘Recognizing that significant whether specific criteria that initiate the of all chlorination DBPs to some extent. excursions of DBP levels will sometimes evaluation of significant excursions occur, even when systems are in full should be included in the rule or in 3. Request for Comment compliance with the enforceable MCL, guidance. EPA requests comment on EPA requests comment on the public water systems that have how to identify significant excursions alternative MCL strategies that were significant excursions during peak (regardless of whether the criteria are in considered by the Advisory Committee periods are to refer to EPA guidance on the rule or in guidance). For example, and the determination to propose the how to conduct peak excursion should the significant excursion be Preferred Alternative in combination evaluations, and how to reduce such based on an individual measurement, with the IDSE as the preferred peaks. Such excursions will be reviewed e.g., any measurement being 25 or 50% regulatory strategy. EPA also requests as part of the sanitary survey process. over either the TTHM or HAA5 MCLs? comment on whether the proposed EPA guidance on DBP level excursions Alternatively, should the determination approach will reduce peak DBP levels. will be issued prior to promulgation of of a significant excursion be based on a EPA requests comment on the phased the final rule and will be developed in certain level of variability among MCL strategy and whether or not it will consultation with stakeholders.’’ multiple measurements? For example,

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00041 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49588 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

should the significant excursion be data from the Information Collection concentrations below approximately 6 based on the standard deviation of the Rule treatment studies and (2) EPA used mg/L (based on the Information LRAA exceeding specific numerical the Surface Water Analytical Tool Collection Rule and NRWA data, over values for either TTHM (e.g., 0.020 mg/ (SWAT), a model developed to compare 90 percent of plants have average l) or HAA5 (e.g., 0.015 mg/L)? Or should alternative regulatory strategies. Both influent TOC levels below 6 mg/L the excursion be based on a relative analyses support the BAT options (USEPA 2003o)). Of the plants that measure of variability (e.g., a relative proposed today. The results of each conducted an Information Collection standard deviation exceeding 25% or analyses are presented in the following Rule GAC treatment study, 50%) with the condition of a threshold two sections. approximately 70% of the surface water average concentration also being i. BAT analysis using the Information plants studies could meet the 0.080 mg/ exceeded (e.g., an LRAA needing to be Collection Rule treatment studies. EPA L TTHM and 0.060 mg/L HAA5 MCLs, at least 0.040 mg/l for TTHM or 0.030 analyzed data from the Information with a 20% safety factor (i.e., 0.064 mg/ mg/l for HAA5)? EPA requests comment Collection Rule treatment studies L and 0.048 mg/L, respectively) using on the above approaches or alternative (Information Collection Rule Treatment GAC with 10 minutes of empty bed approaches for determining whether a Study Database CD–ROM, Version 1.0, contact time and a 120 day reactivation significant excursion has occurred. EPA USEPA 2000m; Hooper and Allgeier frequency, and 78% of the plants could also requests comment on whether 2002). The treatment studies were meet the MCLs with a 20% safety factor different approaches may be appropriate designed to evaluate the technical using GAC with 20 minutes of empty for large and small systems. feasibility of using GAC and NF to bed contact time and a 240 day remove DBP precursors prior to the reactivation frequency. As discussed F. BAT for TTHM and HAA5 addition of chlorine-based disinfectants. previously, the treatment studies were 1. What Is EPA Proposing Today? Systems were required to conduct an conducted at plants with poorer water Information Collection Rule treatment quality than the national average. Today, EPA is proposing that the best study based on TOC levels in the source Therefore, EPA believes that much available technology (BAT) for the or finished water. Specifically, surface higher percentages of plants nationwide TTHM and HAA5 LRAA MCLs (0.080 water plants with annual average source could meet the MCLs with the proposed mg/L and 0.060 mg/L respectively) be water TOC concentrations greater than 4 GAC BATs. one of the three following technologies: mg/L and ground water plants with Among plants using GAC, larger (1) GAC adsorbers with at least 10 annual average finished water TOC systems would likely realize an minutes of empty bed contact time and concentrations greater than 2 mg/L were economic benefit from on-site an annual average reactivation/ required to conduct treatment studies. reactivation, which could allow them to replacement frequency no greater than Thus, the plants required to conduct use smaller, 10-minute empty bed 120 days, plus enhanced coagulation or treatment studies generally had waters contact time contactors with more enhanced softening. with organic DBP precursor levels that frequent reactivation (i.e., 120 days or (2) GAC adsorbers with at least 20 were significantly higher than the less). Most small systems would not minutes of empty bed contact time and Information Collection Rule national find it economically advantageous to an annual average reactivation/ plant medians of 2.7 mg/L for source install on-site carbon reactivation replacement frequency no greater than water at surface water plants and 0.2 facilities, and thus would opt for larger, 240 days. mg/L for finished water at ground water 20-minute empty bed contact time (3) Nanofiltration (NF) using a plants (USEPA 2003o). contactors, with less frequent carbon membrane with a molecular weight cut Plants that conducted GAC studies replacement (i.e., 240 days or less). off of 1000 Daltons or less (or typically evaluated performance at two The proposed reactivation/ demonstrated to reject at least 80% of empty bed contact times, 10 and 20 replacement interval for the 20 minute the influent TOC concentration under minutes, over a wide range of contactor (i.e., 240 days) is double the typical operating conditions). operational run times to evaluate the reactivation/replacement interval for 10 EPA is proposing a different BAT for variable nature of TOC removal by GAC. minute contactor (i.e., 120 days). This is consecutive systems than for wholesale This allowed GAC performance to be based on the assumption of a linear systems to meet the TTHM and HAA5 assessed with respect to empty bed relationship between empty bed contact LRAA MCLs. The proposed consecutive contact time as well as reactivation/ time and the reactivation interval (e.g., system BAT is chloramination with replacement frequency. Plants that a doubling of the empty bed contact management of hydraulic flow and conducted membrane treatment studies time will result in a doubling of the storage to minimize residence time in evaluated one or two nanofiltration reactivation interval). The data from the the distribution system. membranes with molecular weight Information Collection Rule treatment cutoffs less than 1000 Daltons. studies indicates that this linear 2. How Was This Proposal Developed? Regardless of the technology evaluated, relationship may not always hold and a. Basis for the BAT. The Safe all treatment studies evaluated DBP that doubling the empty bed contact Drinking Water Act directs EPA to formation in the effluent from the time generally results in more than a specify BAT for use in achieving advanced process under simulated doubling of the reactivation interval. compliance with the MCL. Systems distribution system conditions While there may be some operational unable to meet the MCL after representative of the average residence advantage in using larger empty bed application of BAT can get a variance time and using free chlorine as the contact times, the larger contactors will (see section V.L. for a discussion of primary and residual disinfectant. (For result in additional capital variances). Systems are not required to more information on the Information expenditures. Furthermore, the use BAT in order to comply with the Collection Rule treatment study economic optimization of a GAC MCL. They can use other technologies requirements and testing protocols, see process must also consider the number as long as they meet all drinking water USEPA 1996 a and b.) of smaller contactors in parallel, since it standards and are approved by the State. Based on the treatment study results, may be advantageous to operate a larger EPA examined BAT using two GAC is effective for controlling DBP number of smaller contactors in parallel, different methods: (1) EPA analyzed formation for waters with influent TOC allowing each individual contactor to be

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00042 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49589

operated for a longer period of time. compare alternative regulatory Economic Analysis of today’s proposed Based on these considerations, and the strategies. EPA modeled the following rule (USEPA 2003i). analysis of subject matter experts, it was BAT options: enhanced coagulation/ The compliance percentages concluded that the proposed softening with chlorine (the Stage 1 forecasted by the SWAT model are combination of GAC empty bed contact DBPR BAT); enhanced coagulation/ indicated in Table V–1. EPA estimates times and reactivation/replacement softening with chlorine and no that more than 97% of large systems intervals were reasonable for BAT. predisinfection; enhanced coagulation will be able to achieve the Stage 2B The Information Collection Rule and GAC10; enhanced coagulation and MCLs regardless of post-disinfection treatment study results also GAC20; and enhanced coagulation and choice if they were to apply one of the demonstrated that nanofiltration was chloramines. Enhanced coagulation/ proposed GAC BATs, i.e., enhanced the better DBP control technology for softening is required under the Stage 1 coagulation (EC) and GAC10 (Seidel ground water sources with high TOC DBPR at subpart H conventional Memo, 2001). As shown in the Stage 2 concentrations (i.e., above filtration plants. In the model, GAC10 DBPR Occurrence document (USEPA approximately 6 mg/L). The results of was defined as granular activated 2003o), the source water quality (e.g., the membrane treatment studies showed carbon with an empty bed contact time DBP precursor levels) in medium and that all ground water plants could meet small systems is expected to be of 10 minutes and a reactivation or the 0.080 mg/L TTHM and 0.060 mg/L comparable to or better than that for the replacement interval of 90 days or HAA5 MCLs, with a 20% safety factor large systems. Based on the large system longer. GAC20 was defined as granular (i.e., 0.064 mg/L and 0.048 mg/L, estimate, EPA believes it is conservative activated carbon with an empty bed respectively) at the average distribution to assume that at least 90% of medium system residence time using contact time of 20 minutes and a and small systems will be able to nanofiltration. Nanofiltration would be reactivation or replacement interval of achieve the Stage 2B MCLs if they were less expensive than GAC for high TOC 90 days or longer. EPA assumed that to apply one of the proposed GAC ground waters, which generally require systems would be operating to achieve BATs. EPA assumes that small systems minimal pretreatment prior to the both the Stage 2B MCLs of 0.080 mg/L may adopt GAC20 in a replacement membrane process. Also, nanofiltration TTHM and 0.060 mg/L HAA5 as an mode (with replacement every 240 days) is an accepted technology for treatment LRAA and the SWTR removal and over GAC10 because it may not be of high TOC ground waters in Florida inactivation requirements of 3-log for economically feasible for some small and parts of the Southwest, areas of the Giardia and 4-log for viruses. EPA also systems to install and operate an on-site country with elevated TOC levels in evaluated the BAT options under the GAC reactivation facility. Moreover, ground waters. assumption that plants operate to some small systems may find ii. BAT analysis using the SWAT. The achieve DBP levels 20% below the MCL nanofiltration cheaper than the GAC20 second method that EPA used to (safety factor). These assumptions along in a replacement mode if their specific examine alternatives for BAT was the with other inputs for the SWAT runs are geographic locations cause a relatively SWAT model that was developed to consistent with those used in the high cost for routine GAC shipment.

TABLE V–1.—SWAT MODEL PREDICTIONS OF PERCENT OF LARGE PLANTS IN COMPLIANCE WITH TTHM AND HAA5 STAGE 2B MCLS AFTER APPLICATION OF SPECIFIED TREATMENT TECHNOLOGIES

Compliance with 0.080 mg/L (TTHM)/0.060 mg/L Compliance with 0.064 mg/L (TTHM)/0.048 mg/L (HAA5) LRAAs (HAA5) LRAAs (MCLs with 20% safety factor) * Technology Residual disinfectant Residual disinfectant All systems All systems Chlorine Chloramine Chlorine Chloramine

Enhanced Coagulation (EC)...... 73.5 76.9 74.8 57.2 65.4 60.4 EC (no predisinfection)...... 73.4 88.0 78.4 44.1 62.7 50.5 EC & GAC10 ...... 100 97.1 99.1 100 95.7 98.6 EC & GAC20...... 100 100 100 100 100 100 EC & All Chloramines ...... NA 83.9 NA NA 73.6 NA * Enhanced coagulation/softening is required under the Stage 1 DBPR for conventional plants.

b. Basis for the Consecutive System or controlled with the BATs proposed nitrification. Nitrification, the process BAT. EPA believes that the best for wholesale systems. EPA believes the by which microbes convert free compliance strategy for consecutive proposed consecutive system BAT is an ammonia to nitrate and nitrite, is a systems is to collaborate with effective means for consecutive systems concern for systems using chloramines. wholesalers on the water quality they to meet the MCLs. Nitrification, however, can be controlled need. For consecutive systems that are Chloramination has been used for with appropriate chlorine to ammonia having difficulty meeting the MCLs, residual disinfection for many years to ratios, increasing flow in low demand EPA is proposing a BAT of minimize the formation of chlorination areas, and increasing storage tank chloramination with management of DBPs, including TTHM and HAA5 turnover. EPA proposes that systems hydraulic flow and storage to minimize (Stage 2 Technology and Cost implementing the consecutive system residence time in the distribution Document, USEPA 2003k). The BAT BAT must do the following: (1) system. EPA is proposing a different provision to manage hydraulic flow and Maintain a chloramine residual BAT than for wholesale systems because minimize residence time in the throughout the distribution system, (2) a consecutive system’s source water has distribution system is to facilitate the develop and submit a plan that already been disinfected and contains maintenance of the chloramine residual indicates actions that will be taken to DBPs that cannot be effectively removed and minimize the likelihood for minimize the residence time of water

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00043 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49590 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

within the distribution system, (3) have Because EPA is not revising the Stage without violating the MCL for chlorite, the plan approved by the Primacy 1 DBPR bromate MCL, EPA is not a byproduct of chlorine dioxide. UV Agency, and (4) implement the plan as proposing a revised BAT for bromate. light is highly effective against approved by the Primacy Agency. The Stage 1 DBPR BAT for bromate is Cryptosporidium and Giardia and most Minimum components of the defined as control of ozone treatment viruses, but has not been used management plan would include processes to reduce production of extensively to treat drinking water in periodic scheduled flushing of all dead bromate. EPA also determined that it the United States. end pipes and storage vessels through was not necessary to regulate bromate in As of early 2000, there were 332 which water is delivered to customers, non-ozone systems that use plants of various sizes using ozone and hydraulic flow control procedures hypochlorite. (Overbeck 2000) and 58 plants that were that routinely circulate water in all Finally, EPA is proposing to modify planning to install ozonation (Rice storage vessels within the distribution the criterion for a system that uses 2000—personal communication: email system. ozone (and therefore must monitor for 7/14/2000). A significant percent of bromate) to qualify for reduced bromate current ozone plants use ozone for some EPA believes that the BATs proposed monitoring from one sample per ozone portion of their disinfection objective for wholesale systems are not plant per month to one sample per plant (Rice, 2000—personal communication: appropriate for consecutive systems per quarter. email 7/14/2000). An ozone system that because each of these BATs, when could not meet a 0.005 mg/L bromate 2. How Was This Proposal Developed? applied to water with DBPs, raises other MCL would have three primary options: concerns. GAC is not cost-effective for a. Bromate MCL. Bromate is a decrease the ozone dose; switch to a removing DBPs. In addition, dioxin, a principal byproduct from ozonation of different disinfectant; or install an carcinogen, may be formed during GAC bromide-containing source waters. As advanced filtration process such as regeneration if GAC has been used to described in more detail later, making membranes, sometimes in combination adsorb chlorinated DBPs. Nanofiltration the bromate MCL more stringent has the with the first two options. Of these three is only moderately effective at removing potential to decrease current levels of options, the third is likely effective but THMs or HAAs if membranes that have microbial protection, impair the ability very expensive, while the first two a very low molecular weight cutoff and of systems to control resistant pathogens create the risk either of reducing very high cost of operation are like Cryptosporidium, and increase microbial protection for a wide range of employed. Therefore, GAC and levels of DBPs from other disinfectants microbial pathogens, or of increasing nanofiltration are not appropriate BATs that may be used instead of ozone. formation of DBPs other than bromate. for consecutive systems. EPA estimates that the 1 in 10,000 In an attempt to achieve a lower level 3. Request for Comment excess lifetime cancer risk level for of bromate, some systems might be bromate is 0.005 mg/L. EPA proposed driven to reduce the applied ozone dose EPA requests comment on the and ultimately finalized an MCL of to the minimum necessary for regulatory proposed BATs including the BAT for 0.010 mg/L in the Stage 1 DBPR, compliance or switch to other treatment consecutive systems. primarily because available analytical processes. Many systems currently detection methods for bromate could achieve more disinfection than is G. MCL, BAT, and Monitoring for only reliably measure to 0.01 mg/L required by the SWTR and if a system Bromate (USEPA 1994b). Analytical methods for were to simply lower the ozone dose, 1. What Is EPA Proposing Today? bromate are now available to quantify protection from pathogens may be bromate concentrations as low as 0.001 compromised. In addition, since EPA is proposing today that the MCL mg/L. Due to the availability of lower inactivation of Cryptosporidium for bromate for systems using ozone detection methods for bromate, as part requires much higher ozone doses than remain at 0.010 mg/L as an RAA for of the Stage 2 M–DBP Advisory Giardia inactivation, systems cannot samples taken at the entrance to the Committee deliberations, EPA achieve Cryptosporidium inactivation distribution system as established by the considered revising the MCL to 0.005 with low ozone doses. Stage 1 DBPR and as provided for under mg/L or lower. If a system were to lower the ozone the risk-balancing provisions of section As a disinfectant, ozone is highly dose and supplement with an additional 1412(b)(5) of the SDWA. EPA’s proposal effective against a broad range of disinfectant, or switch entirely to a is consistent with the recommendation microbial pathogens including bacteria, different disinfectant, the system may of the Stage 2 M–DBP Advisory viruses, and protozoa. Moreover, ozone not achieve the same level of microbial Committee, which considered the is one of the few disinfectants available protection as is afforded by ozonation. potential that reducing the bromate in water treatment that is capable of Also, other potentially harmful MCL could both increase the inactivating Cryptosporidium, a byproducts from the different concentration of other DBPs in the protozoan which can cause severe disinfectant would be produced. drinking water and interfere with the intestinal disorders and can be deadly to During the Stage 2 M–DBP Advisory efficacy of microbial pathogen those with compromised immune Committee discussions, the TWG inactivation. In addition, as required by systems. The oxidizing properties of evaluated the impact of reducing the the SDWA and as recommended by the ozone are also valuable for treatment bromate MCL from 0.010 mg/L to 0.005 Advisory Committee, EPA will review objectives like control of tastes and mg/L as an annual average. The TWG the bromate MCL as part of the 6-year odors and removal of iron and concluded that many systems currently review process and determine whether manganese. In contrast, chlorine, the using ozone or predicted to install the MCL should remain at 0.010 mg/L most common disinfectant and oxidant ozone to inactivate microbial pathogens or be reduced to a lower level. As a part in water treatment, is substantially less would have significant difficulty of that review, EPA will consider the effective for controlling maintaining bromate levels at or below increased utilization of alternative Cryptosporidium. Chlorine dioxide, 0.005 mg/L. In the Information technologies, such as UV, and whether while capable of providing low levels of Collection Rule survey of systems the risk/risk concerns reflected in inactivation for Cryptosporidium, serving greater than 100,000 people, all today’s proposal remain valid. typically cannot be used at high doses of the ozone plants had annual average

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00044 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49591

bromate concentrations below the 0.010 further reduced when periods of higher The Advisory Committee discussed mg/L level (USEPA 2003o). However, bromide levels, similar to drought these results and, since the bromate approximately 20% of these ozone conditions, were modeled. This trend is level resulting from hypochlorite plants did not meet the 0.005 mg/L further exacerbated since the proposed solutions was small compared to the level. Using the assumption that LT2ESWTR would require more MCL, did not recommend regulating systems operate their plants using a stringent ozone operating conditions bromate at systems not using ozone safety margin of 20% below the MCL, (such as higher ozone doses and longer (non-ozone systems). The Advisory about 30% of ozone plants did not contact times) than under current Committee recognized that ozone reliably attain this level (0.004 mg/L). surface water treatment requirements for systems also using hypochlorite will During the Information Collection Rule, the subset of plants with higher have to be careful about the quality of for the first half of 1998, much of the Cryptosporidium concentrations in their their stock solution. U.S. was wetter than normal (NOAA source water and would thus result in c. Criterion for reduced bromate 1998). This hydrogeological condition higher bromate formation than assumed monitoring. Because more sensitive often leads to lower than normal by the TWG. Thus, as systems are bromate methods are now available, bromide concentrations due to dilution required to meet more stringent EPA is proposing a new criterion for by higher water flows. In the second inactivation requirements, a large reduced bromate monitoring. In the half of 1998, California continued to number of systems would be forced to Stage 1 DBPR, EPA required ozone experience El Nino rains (40% of select treatment processes other than systems to demonstrate that source Information Collection Rule ozone ozone if the bromate standard were water bromide levels, as a running plants were located in California) but lowered to 0.005 mg/L. annual average, did not exceed 0.05 mg/ many other areas of the country such as The Stage 2 M–DBP Advisory L. EPA elected to use bromide as a Texas and Florida experienced a Committee considered that reducing the surrogate for bromate in determining drought. The percentage of ozone bromate MCL to 0.005 mg/L could both eligibility for reduced monitoring systems unable to achieve 0.005 mg/L increase the concentration of other DBPs because the available analytical method bromate would likely increase during in the drinking water and interfere with for bromate was not sensitive enough to years in which bromide concentrations the efficacy of microbial pathogen quantify levels well below the bromate in California were elevated as inactivation. Therefore, the Advisory MCL of 0.010 mg/L. In section V.O., EPA is proposing consequence of drought. Committee recommended, for purposes several new analytical methods for of the Stage 2 DBPR, that the bromate The ability of systems to use ozone to bromate that are far more sensitive than MCL remain at 0.010 mg/L. EPA will meet Cryptosporidium treatment the existing method. Since these review the bromate MCL as part of the requirements proposed under the methods can measure bromate to levels ongoing 6-year review process and LT2ESWTR would be diminished if the of 0.001 mg/L or lower, EPA is determine whether the MCL should bromate MCL was decreased from 0.010 proposing to replace the criterion for to 0.005 mg/L. The proposed remain at 0.010 mg/L or be reduced to reduced bromate monitoring (source LT2ESWTR will require a subset of a lower concentration based on new water bromide running annual average systems, based on source water information. not to exceed 0.05 mg/L) with a bromate pathogen levels, to provide from 1.0 to Today, EPA is proposing to leave the running annual average not to exceed 2.5 logs of additional treatment for bromate MCL at 0.010 mg/L, consistent 0.0025 mg/L. Cryptosporidium. Ozone doses required with the Advisory Committee’s In the past, EPA has often set the to inactivate Cryptosporidium are recommendation. EPA believes that this criterion for reduced monitoring substantially greater than those required is a prudent step at this time, in order eligibility at 50% of the MCL, which for Giardia and viruses. To assess the to preserve microbial protection. EPA would be 0.005 mg/L. However, as potential impact of a lower bromate will continue to analyze any new discussed before, EPA is proposing that MCL on the ability of systems to treat bromate health effects data as they the MCL for bromate remain at 0.010 for Cryptosporidium, the TWG become available. It is possible that EPA mg/L, a level that is higher than EPA’s estimated the percentage of treatment may determine that the bromate MCL usual excess cancer risk range of 10(-4) plants that could use ozone to inactivate should be decreased to 0.005 mg/L or to 10(-6) at 2x10(-4) because of risk from 0.5 to 2.5 log of Cryptosporidium lower in a future rulemaking. tradeoff considerations. EPA believes without exceeding a bromate MCL of b. Bromate in hypochlorite solutions. that the decision for reduced monitoring either 0.005 or 0.010 mg/L (USEPA The Stage 2 M–DBP Advisory is separate from these risk tradeoff 2003i). These estimations were based on Committee also discussed the issue of considerations. Risk tradeoff analyses of Information Collection Rule hypochlorite solutions contaminated considerations influence the selection of source water quality data, coupled with with bromate. This contamination can the MCL, while reduced monitoring projected ozone dose requirements for occur during the production of requirements are designed to ensure that Cryptosporidium. This analysis suggests hypochlorite solutions from natural salt the MCL, once established, is reliably that 88% of systems could use ozone to deposits. The range of bromate and consistently achieved. Requiring a achieve 1 log of Cryptosporidium concentrations in hypochlorite stock running annual average of 0.0025 mg/L inactivation and 47% could inactivate 2 solutions varies widely (Bolyard et al. for the reduced monitoring criterion log while complying with a bromate 1992; Chlorine Institute 1999, 2000). allows greater confidence that the MCL of 0.010 mg/L. With the bromate Moreover, the bromate contained in the system is achieving the MCL and thus MCL reduced to 0.005 mg/L, though, stock solution is diluted upon addition ensuring public health protection. these estimates drop to 67% of systems to the drinking water. From data on able to inactivate 1 log of Information Collection Rule ozone 3. Request for Comment Cryptosporidium with ozone and only systems that used hypochlorite versus EPA requests comment on the 14% able to inactivate 2 log. The those that used gaseous chlorine, the decision to maintain the Stage 1 DBPR number of plants predicted to be able to TWG estimated that hypochlorite bromate BAT and MCL of 0.010 mg/L. treat for Cryptosporidium with ozone solutions contributed an average of EPA also requests comment on the and meet a 0.005 mg/L standard was 0.001 mg/L bromate. decision not to require bromate

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00045 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49592 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

monitoring at non-ozone systems that sites, systems and States must have Guidance Manual will include a use hypochlorite. monitoring data collected from guidance for system-specific studies and EPA requests comment on whether throughout their distribution systems. how to determine whether site-specific the criterion for reduced bromate Therefore, under today’s proposed rule, data could be sufficient to meet the monitoring should be set at a level other systems are required to collect IDSE requirements (USEPA 2003j). monitoring data on the concentrations than 0.0025 mg/L, and a rationale for iii. 40/30 certification. Under this of these DBPs. There are three possible setting it at that level. approach, systems certify to their approaches by which a system can meet primacy agency that all required Stage H. Initial Distribution System the IDSE requirement. 1 DBPR compliance samples were Evaluation (IDSE) i. Standard monitoring program. The properly collected and analyzed during The IDSE is an important part of standard monitoring program requires the two years prior to the start of the today’s proposed regulation that is one year of monitoring on a specified IDSE, and all individual compliance intended to identify sample locations schedule throughout the distribution samples were ≤ 0.040 mg/L for TTHM for Stage 2B compliance monitoring that system. The frequency and number of and ≤0.030 mg/L for HAA5. Properly represent distribution system sites with samples required under the standard collected and analyzed compliance high DBP concentrations. monitoring program is determined by source water type, number of treatment samples are those taken at required 1. What is EPA Proposing Today? plants, and system size (see section V.J. locations at times specified in the EPA is proposing a requirement for for a more detailed discussion of the system’s Stage 1 DBPR monitoring plan systems to perform an Initial specific monitoring requirements). Prior and analyzed by certified laboratories. Distribution System Evaluation (IDSE). to commencing the standard monitoring Systems not required to collect Stage 1 Systems will collect data on DBP levels program, systems must prepare a DBPR compliance samples can not throughout their distribution system, monitoring plan. EPA’s IDSE Guidance utilize the 40/30 certification approach evaluate these data to determine which Manual will provide guidance on because they do not have data to sampling locations are most selecting monitoring sites and determine sampling locations that representative of high DBP levels and conducting the standard monitoring represent high concentrations of TTHMs compile this information into a report program (USEPA 2003j). As and HAA5. Systems that qualify for for submission to the primacy agency. recommended by the Advisory reduced monitoring for the Stage 1 a. Applicability. All community water Committee, EPA is proposing that the DBPR during the two years prior to the systems, and large nontransient standard monitoring program results are start of the IDSE, may use results of both noncommunity water systems (those not to be used for determining routine and reduced Stage 1 DBPR serving at least 10,000 people) that add compliance with MCLs and that systems monitoring to prepare the 40/30 a primary or residual disinfectant other will not be required to report IDSE certification. Large ground water than ultraviolet light, or that deliver results in the Consumer Confidence systems may not have two years of water that has been treated with a Report. HAA5 data to evaluate due to the timing primary or residual disinfectant other ii. System specific study. Under this of the Stage 1 DBPR and the IDSE than ultraviolet light (i.e., consecutive approach, systems may choose to requirements. EPA is proposing that, if systems) are required to conduct an perform a system-specific study based two years worth of HAA5 data are not IDSE under the proposed rule. The IDSE on earlier monitoring studies or other available, large ground water systems requirement for systems serving fewer data analysis in lieu of the standard evaluate the most recent two years of than 500 people may be waived if the monitoring program. These studies must TTHM data including data collected in State determines that the monitoring provide equivalent or better information accordance with the 1979 TTHM rule site approved for Stage 1 DBPR than the standard monitoring program and all available HAA5 compliance data compliance is sufficient to represent for selecting sites that represent high collected up to nine months following both high HAA5 and high TTHM TTHM and HAA5 levels. Examples of promulgation of this rule when making concentrations. The State must submit alternative studies are: (1) Recent TTHM the 40/30 certification. Similarly, small criteria for this waiver determination to and HAA5 monitoring data that wholesale and consecutive systems EPA as part of their primacy encompass a wide range of sample sites required to submit their IDSE report no application. States may decide to waive representative of the distribution later than two years after publication of the IDSE requirement for all systems system, including those judged to the final rule will evaluate all available serving fewer than 500 or some subset represent high TTHM and HAA5 Stage 1 DBPR compliance data collected of all systems serving fewer than 500 if concentrations and (2) hydraulic up to nine months following the State determines that it is modeling studies that simulate water promulgation. appropriate. EPA is developing an IDSE movement in the distribution system. c. Implementation. All systems Guidance Manual that will include Historical TTHM and HAA5 results subject to the IDSE requirement under guidance to States on situations for submitted by systems must have been the proposed rule (except those which a waiver would be appropriate generated by certified laboratories and receiving a very small system waiver (USEPA 2003j). must include the system’s most recent from the State) must submit a report to b. Data collection. IDSEs are intended data. Treatment plant and distribution the primacy agency. The requirements to help identify and select Stage 2B system characteristics at the time of for the report depend upon the IDSE compliance monitoring sites that historical data collection must reflect data collection alternative that the represent high concentrations of TTHMs the current plant operations and system selects and are listed in Table V– and HAA5. To be able to identify these distribution system. EPA’s IDSE 2.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00046 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49593

TABLE V–2.—IDSE REPORT REQUIREMENTS

IDSE data collection alternative IDSE report requirements

Standard Monitoring Pro- • All standard monitoring program TTHM and HAA5 analytical results, the original monitoring plan, and an expla- gram. nation of any deviations from that plan. • A schematic of the distribution system. • Recommendations and justification for where and during what month(s) Stage 2B monitoring should be con- ducted. System Specific Study ...... • All studies, reports, analytical results and modeling. • A schematic of the distribution system. • Recommendations and justification for where and during what month(s) Stage 2B monitoring should be con- ducted 40/30 Certification ...... • A certification that all required compliance samples were properly collected and analyzed during the two years prior to the start of the IDSE and all individual compliance samples were ≤ 0.040 mg/L for TTHM and ≤0.030 mg/L for HAA5. • Results of compliance samples taken after the IDSE was scheduled to begin and before the IDSE report was submitted. • Recommendations for where and during what month(s) Stage 2B monitoring should be conducted.

All IDSE reports must include addition to giving the system a a. Applicability. The M–DBP recommendations for the location and monitoring and reporting violation. Advisory Committee recommended that schedule for the Stage 2B monitoring. These requirements may include an IDSE be performed on all community The number of sampling locations and repeating the IDSE while conducting systems to help to identify the locations the criteria for their selection are compliance monitoring at Stage 1 in the distribution system that represent described in § 141.605 of today’s monitoring sites or conducting Stage 2 high DBP concentrations. EPA believes proposed rule, and in section V.I. compliance monitoring at sites selected that large nontransient noncommunity Generally, a system must recommend by the State. water systems (those serving at least locations with the highest LRAAs unless Consecutive systems are subject to the 10,000 people) also have distribution it provides a rationale (such as ensuring IDSE requirements of today’s proposed systems that require further evaluation geographical coverage of the rule. IDSE requirements for consecutive to determine the most representative distribution system instead of clustering systems are largely the same as for other locations of high DBP levels. Therefore, all sites in a particular section of the systems, but with two differences. First, large nontransient noncommunity distribution system) for selecting other the schedule for completion of the IDSE systems and all community systems are locations. Systems must consider both by a consecutive system is dependent required to perform an IDSE under their compliance data and IDSE data in upon the population of the wholesale today’s proposal. making this determination. In addition system. If a consecutive system serving States may waive the IDSE to specifying a protocol for identifying fewer than 10,000 buys water from a requirement for those very small recommended monitoring sites in the system that serves 10,000 or more systems (systems that serve fewer than rule language, EPA will provide people, then this consecutive system 500 people) that monitor for Stage 1 guidance for recommending compliance must comply within the same schedule DBPR compliance at the maximum monitoring sites (including rationales as that for systems ≥ 10,000. Conversely, residence time site if the State for systems to recommend sites that do if a wholesale system serves < 10,000 determines their maximum residence not have the highest LRAA but sells water to a consecutive system time Stage 1 compliance monitoring site concentrations) and preparing the IDSE serving ≥ 10,000, then both the is likely to capture both the high TTHM report. EPA will also provide a process wholesale system and the consecutive and high HAA5 levels within the to address IDSE implementation issues system must complete the IDSE within distribution system. The Advisory during the period prior to State primacy. the same schedule as that for systems ≥ Committee recommended this waiver be At the time that systems serving fewer 10,000. The second difference for included because many very small than 10,000 people conduct their consecutive systems is that the systems have small distribution systems monitoring or analyze their site-specific procedure for recommending Stage 2B and the high TTHM and high HAA5 site data, many States may have primacy. compliance monitoring locations is is at the same location. The Advisory The compliance schedules for the modified for consecutive systems Committee also recognized that not all IDSE and other requirements of the purchasing or receiving all of their very small systems have a single proposed rule are described in detail in finished water from a wholesale system. monitoring site that would represent section V.J. Systems serving at least These modified procedures are both high TTHM and high HAA5 levels 10,000 people (and those smaller described in § 141.605 of today’s (e.g., some rural systems with large wholesale and consecutive systems proposed rule, and in section V.I. distribution systems) and thus did not associated with larger systems) will be recommend a blanket IDSE waiver for collecting data for their IDSE prior to 2. How Was This Pr oposal Developed? all very small systems. State primacy. EPA intends to have an The IDSE was recommended by the b. Data collection. The data collection IDSE Guidance Manual available to Stage 2 M–DBP Advisory Committee. requirements of the IDSE are designed assist systems in performing the IDSE The Advisory Committee believed that to find both high TTHM and high HAA5 (USEPA 2003j). Primacy agencies will maintaining Stage 1 DBPR sampling sites (see section V.I. for IDSE specify requirements for systems that do sites for the Stage 2 DBPR would not monitoring site locations). The IDSE is not submit an IDSE report, or that have accomplish the objective of providing intended as a one-time requirement. not, in the determination of the primacy consistent and equitable protection High TTHM and HAA5 concentrations agency, conducted an adequate IDSE, in across the distribution system. often occur at different locations in the

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00047 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49594 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

distribution system. The Stage 1 DBPR and medium surface water systems was future. This provision requires that monitoring sites identified as the not necessary for small surface water medium and large systems conduct and maximum location are selected systems and ground water systems. The complete site-specific risk according to residence time. Because majority of small systems have determinations (i.e., the IDSE and HAAs can degrade in the distribution distribution systems with simpler LT2ESWTR Cryptosporidium system in the absence of sufficient designs than large systems. DBP monitoring) as soon as possible after disinfectant residual (Baribeau et al. occurrence in ground water systems is rule promulgation. Since small systems 2000), residence time alone is not an generally lower and less variable than in cannot begin their microbial monitoring ideal criterion for identifying high surface water systems due to lower and until after the results from the large HAA5 sites. The Information Collection less variable precursor levels and much system microbial monitoring have been Rule data show that of the four less temperature variation (see section analyzed, small systems have a longer monitoring locations sampled per IV). compliance time frame. system, the one identified as the Committee members recognized that Systems that submit a 40/30 maximum residence time location was some systems have detailed knowledge certification are required to submit that often not the location where the highest of their distribution systems by way of certification as part of the IDSE report DBP levels were found. In fact, over 60 hydraulic modeling and/or ongoing and to include a recommended Stage 2B percent of the highest HAA5 LRAAs and widespread monitoring plans (well monitoring plan. The monitoring plan is 50 percent of the highest TTHM LRAAs beyond that required for compliance required for these systems because the were found at sampling locations in the monitoring) that would provide Stage 2B MCL compliance monitoring system other than the maximum equivalent or superior monitoring site sites proposed today have residence time location (see section IV). selection compared to IDSE monitoring. fundamentally different objectives than Thus the method and assumptions used Therefore, the Advisory Committee the Stage 1 DBPR monitoring sites. to select the Information Collection Rule recommended that such systems be Additionally, many systems are monitoring sites, and the Stage 1 DBPR allowed to determine new monitoring required to have more Stage 2 compliance monitoring sites, are not sites using system-specific data such as compliance monitoring sites than Stage sufficiently reliable to select Stage 2 historical monitoring data. 1 sites because high HAA5 site may be DBPR compliance monitoring sites that Systems that certify to their State that different than high TTHM sites. all compliance samples taken in the two will capture high DBP levels. 3. Request for Comment This data analysis reveals that a years prior to the start of the IDSE were reevaluation of monitoring sites is ≤ 0.040 mg/L TTHM and ≤ 0.030 mg/L EPA requests comments on the IDSE necessary at many systems to capture HAA5 are not required to collect requirement and whether it is a good sites with high DBP levels. The additional DBP monitoring data because tool to identify sites representative of Advisory Committee recommended the Advisory Committee determined high TTHM and high HAA5 levels. sample locations (based on distribution that these systems most likely would a. Applicability. EPA requests disinfectant type) at widely distributed not have high peak DBP levels. EPA comment on requiring large (serving sites (see section V.I. for details on IDSE determined that this provision needed 10,000 or more people) nontransient monitoring requirements). Monitoring at to be more specific for three groups of noncommunity water systems to additional sites across the distribution systems: (1) Those performing Stage 1 perform an IDSE. Should NTNCWSs system increases the chance of finding DBPR reduced monitoring, (2) large serving fewer than 10,000 people be sites with high DBP levels and targets ground water systems, and (3) small required to conduct an IDSE? EPA also both DBPs that degrade, and DBPs that systems required to conduct an early requests comment upon whether States form, as residence time increases in the IDSE. Today’s proposal clarifies that should be able to waive IDSE distribution system. EPA believes that these systems may use a 40/30 requirements for very small systems the required number of monitoring certification. EPA recognizes that these (serving fewer than 500 people). Are locations plus Stage 1 monitoring systems may have less compliance data there objective criteria that the State results provides an adequate on which to base their 40/30 should use in waiving the requirement? recharacterization of DBP levels certifications. However, EPA believes Should the State be allowed to grant throughout the distribution system, at a that the data that will be available are very small system waivers based on reasonable cost. With a sufficient to make a determination on some other criterion other than serving recharacterization of distribution the most appropriate Stage 2B a population <500? For example, should systems that focuses on both high monitoring locations. the State be allowed to choose a higher TTHM and HAA5 occurrence, EPA c. Implementation. Systems are population cutoff? Should the State be believes that high occurrence sites will required to submit an IDSE report so allowed to use a non-population be better represented in this standard that primacy agencies may review the criterion such as simplicity of monitoring program. Systems will be system’s IDSE data collection efforts and distribution system to grant a very small required to take steps to address high the Stage 2B monitoring locations system waiver? If so, what should this DBP levels at points that might recommended by the system. Systems criterion be and how should otherwise have gone undetected. EPA serving at least 10,000 must submit their qualification be demonstrated? believes that the decrease in DBP IDSE report two years after rule b. Data collection. EPA requests exposure anticipated to result from the promulgation (which may be prior to comment on the requirements for each transition from an RAA to an LRAA will primacy for some States). The M–DBP of the alternatives for data collection be augmented by the IDSE. Advisory Committee recommended an under the proposed IDSE including: the The frequency and number of samples implementation schedule that would standard monitoring program, the required for the standard monitoring allow systems sufficient time to make system-specific study, and the 40/30 program decrease as system size site-specific risk determinations and certification. EPA requests comment on (population served) decreases and decisions regarding the simultaneous whether systems with less than two depend on source water type. The implementation of the Stage 2 DBPR years of routine monitoring data should Advisory Committee believed that the and LT2ESWTR but not stretch out the be considered to have sufficient data to number of samples required for large compliance time frame too far into the utilize the 40/30 certification.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00048 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49595

Specifically EPA requests comment on requirements only for IDSE and Stage sites. The following discussion provides whether systems on reduced 2B monitoring by consecutive systems details on the IDSE standard monitoring monitoring, large ground water systems, that purchase all of their finished water program. Section V.H identifies other and small systems required to conduct year-round. However, EPA is requesting approaches by which systems can meet an IDSE within the first two years after comment on applying a population- IDSE requirements of the rule. promulgation should be prohibited from based approach to all systems for the For IDSE monitoring, subpart H submitting a 40/30 certification. IDSE and Stage 2B compliance. The c. Implementation. EPA requests discussion of monitoring requirements systems serving at least 10,000 people comment on the requirement that large in this section provides details on these must collect samples approximately and medium systems must collect data two approaches. every 60 days at eight distribution and prepare their IDSE report prior to A number of factors affect DBP system sites per plant (these are in State primacy. EPA requests comment formation, including the type and addition to Stage 1 DBPR compliance from the States regarding whether they amount of disinfectant used, water monitoring sites). The distribution intend to be involved in the temperature, pH, amount and type of system residual disinfectant type consultations with systems collecting precursor material in the water, and the determines the location of the eight data for IDSE or in the review of IDSE length of time that water remains in the sites, as shown in Table V–3. reports that are submitted prior to State treatment and distribution systems. For Subpart H systems serving fewer than primacy. EPA is developing a plan to this reason, and because DBP levels can 10,000 people and all ground water be highly variable throughout the implement the IDSE during the period systems must collect IDSE samples at distribution system (as discussed in prior to State primacy. EPA requests two distribution system sites per plant comment on any issues that should be section IV), today’s proposal requires systems to collect IDSE and Stage 2B (at sites that are in addition to the Stage addressed during this period to facilitate 1 DBPR compliance monitoring sites) as the IDSE. samples at specific locations in the distribution system and in accordance shown in Table V–3. Subpart H systems I. Monitoring Requirements and with a sampling schedule. For purposes serving 500–9,999 people and ground Compliance Determination for Stage 2A of determining the number of required water systems serving at least 10,000 and Stage 2B TTHM and HAA5 MCLs samples, EPA intends to maintain the people must sample quarterly 1. What Is EPA Proposing Today? provision in the Stage 1 DBPR (approximately every 90 days); subpart (§ 141.132(a)(2)) that multiple wells H systems serving fewer than 500 Today’s proposal includes new drawing raw water from a single aquifer people and ground water systems requirements for how systems must may, with State approval, be considered serving fewer than 10,000 people must monitor TTHM and HAA5 levels in one plant, and prior approvals will sample semi-annually (approximately their distribution systems and how remain in force unless withdrawn. every 180 days). systems must assess their monitoring a. Stage 2A. For Stage 2A of the results to determine compliance with proposed rule, compliance will be based EPA is also proposing IDSE TTHM and HAA5 MCLs. The new on the compliance sampling sites and monitoring requirements for monitoring requirements are associated frequency established under the existing consecutive systems. For consecutive with the IDSE (described in section V.H) Stage 1 DBPR. Systems must continue to systems that both purchase finished and Stage 2B of the proposed rule. The monitor for TTHM and HAA5 using a water and treat source water to produce new compliance determination plant-based approach, as required under finished water, IDSE requirements are requirements relate to use of the 40 CFR 141.132. Using these monitoring the same as for non-consecutive systems locational running annual average results, systems must continue to with the same population and source (LRAA) for meeting proposed Stage 2A demonstrate compliance with Stage 1 water type (see Table V–3). For these and Stage 2B MCLs for TTHM and MCLs of 0.080 mg/L for TTHM and consecutive systems, each consecutive HAA5 (described in section V.D). This 0.060 mg/L for HAA5, based on a system entry point (defined in section section presents these proposed running annual average (see 40 CFR V.C) is counted as one treatment plant monitoring and compliance 141.133). In addition, systems must for purposes of determining sampling determination requirements for Stage comply with the Stage 2A MCLs of requirements. However, the State may 2A, the IDSE, and Stage 2B. 0.120 mg/L for TTHM and 0.100 mg/L allow a system to consider multiple An important aspect of the proposed for HAA5, based on the LRAA at each consecutive system entry points to be TTHM and HAA5 monitoring Stage 1 DBPR monitoring location. Stage considered a single point. requirements is the use of two different 1 DBPR provisions for systems to reduce approaches for determining the number the frequency of TTHM and HAA5 As noted previously, for consecutive of samples a system is required to monitoring will still apply. systems that purchase all of their collect. One approach is plant-based. Stage 2A will primarily affect surface finished water year-round, EPA is Under the plant-based approach, a water systems serving at least 10,000 proposing a population-based system’s TTHM and HAA5 sampling people or systems with multiple plants, monitoring approach (see Table V–4) requirements are determined by the because these systems are required to instead of a plant-based approach. number of treatment plants in the monitor at more than one location in the Under the population-based approach, system and, in the case of consecutive distribution system. Most other systems monitoring requirements are not systems, the number of consecutive take compliance samples at only one influenced by the number of system entry points. The second location under Stage 1 and in these consecutive system entry points, but are approach is population-based. Under cases, the calculated LRAA will be based solely on the population served the population-based approach, a equal to the calculated RAA. and the type of source water used. EPA system’s sampling requirements are b. IDSE. IDSE monitoring believes the population-based approach influenced by the number of people requirements are designed to identify is equitable and will provide served, but not by the number of locations within the distribution system representative DBP concentrations treatment plants. EPA is proposing with high TTHM and HAA5 levels, throughout distribution systems. population-based sampling which will serve as Stage 2B monitoring

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00049 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49596 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

TABLE V–3.—PROPOSED IDSE MONITORING REQUIREMENTS

Distribution system sample locations per plant per moni- Number toring period 1 System type and population Distribution system disinfectant of moni- served type toring Near Average High High periods Total entry residence TTHM HAA5 point time locations locations

Subpart H ≥10,000 ...... Chloramines ...... 26 82222 Chlorine ...... 26 81232 Subpart H 500–9,999 or Ground Any ...... 3 4 20011 Water ≥10,000. Subpart Any H <500 or Ground Any ...... 2 4 2 0 0 1 1 Water <10,000.

Consecutive Systems ...... Any ...... —Consecutive systems that purchase 100% of their finished water year-round—see Table V.4. —Consecutive systems that also treat source water to produce finished water—plant-based monitoring at same location and frequency as a non-consecutive system with the same population and source water. 1 Samples must be taken at locations other than the existing Stage 1 DBPR monitoring locations. Dual sample sets (i.e., a TTHM and an HAA5 sample) must be taken at each site. Sampling locations should be distributed throughout the distribution system. 2 Approximately every 60 days. 3 Approximately every 90 days. 4 Approximately every 180 days.

TABLE V–4. POPULATION-BASED MONITORING FREQUENCIES AND LOCATIONS UNDER IDSE FOR CONSECUTIVE SYSTEMS THAT PURCHASE 100% OF FINISHED WATER YEAR-ROUND

Distribution system sample locations 1 Source water type Population size category Monitoring periods and Near Average High High frequency Total entry residence TTHM HAA5 points 2 time locations locations

Subpart H ...... 0–499 ...... Two 2 every 180 days) ... 2 ...... 1 1 500–4,999 ...... Four (every 90 days) ...... 2 ...... 1 1 5,000–9,999 ...... 4 ...... 1 2 1 10,000-24,999 ...... Six (every 60 days) ...... 8 1 2 3 2 25,000-49,999 ...... 12 2 3 4 3 50,000-99,999 ...... 16 3 4 5 4 100,000-499,999 ...... 24 4 6 8 6 500,000-1,499,000 ...... 32 6 8 10 8 1,500,000-4,999,999 ...... 40 8 10 12 10 ≥5,000,000 ...... 48 10 12 14 12 Ground Water ...... 0–499 ...... Two (every 180 days) ..... 2 ...... 1 1 500–9,999 ...... 2 ...... 1 1 10,000-99,999 ...... Four (every 90 days) ...... 6 1 1 2 2 100,000-499,999 ...... 8 1 1 3 3 ≥500,000 ...... 12 2 2 4 4 1 Samples must be taken at locations other than the existing Stage 1 DBPR monitoring locations. Dual sample sets (i.e., a TTHM and an HAA5 sample) must be taken at each site. Sampling locations should be distributed throughout the distribution system. 2 If the number of entry points to the distribution system is less than the specified number of sampling locations, additional samples must be taken equally at high TTHM and HAA5 locations. If there is an odd extra location number, a sample at a high TTHM location must be taken. If the number of entry points to the distribution system is more than the specified number of sampling locations, samples must be taken at entry points to the distribution system having the highest water flows.

As a part of the monitoring schedule, an IDSE, Stage 2B monitoring locations Many systems on reduced monitoring all systems conducting IDSE monitoring are based on the system’s Stage 1 DBPR under the Stage 1 DBPR will conduct must collect samples during the peak compliance monitoring results and an Stage 2B compliance monitoring at historical month for TTHM levels or evaluation of the distribution system different or additional locations than water temperature. EPA will provide characteristics to identify additional those used for Stage 1 compliance guidance to assist systems in choosing monitoring locations, if required. monitoring. Such systems must conduct IDSE monitoring locations, including Consistent with the Advisory routine monitoring for at least one year criteria for selecting high TTHM and Committee recommendations, the before being eligible for reduced HAA5 monitoring locations. monitoring frequency for Stage 2B is monitoring under Stage 2B. Those c. Stage 2B. For those systems structured so that systems that monitor systems that monitor at the same required to conduct an IDSE, Stage 2B quarterly under the Stage 1 DBPR will locations under both the Stage 1 DBPR monitoring sites are based on the continue to monitor quarterly. In and Stage 2B DBPR and have qualified system’s IDSE results and Stage 1 DBPR addition, the monitoring schedule must for reduced monitoring under Stage 1 compliance monitoring results. For include the month with the highest may remain on reduced monitoring at those systems not required to conduct historical DBP concentrations. the beginning of Stage 2B.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00050 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49597

EPA is proposing to require all discussed in section V.C) must develop requirements for Stage 2B of today’s rule systems to develop and maintain a DBP and maintain its own monitoring plan. for non-consecutive systems and for monitoring plan that must include the To comply with the requirement for a consecutive systems that also treat following information: monitoring monitoring plan, systems may develop a source water. Tables V–6 and V–7 locations, monitoring dates, compliance new plan or update the monitoring plan summarize proposed routine and calculation procedures, and copies of required under the Stage 1 DBPR (see reduced Stage 2B monitoring any permits, contracts, or other § 141.132(f)). In either case, the system requirements for consecutive systems agreements with third parties to sample, must follow the monitoring plan, which that purchase all of their finished water will be based on the IDSE report analyze, report, or perform any other year-round. The proposed reduced submitted to the State, modified by any monitoring requirement. Each system in monitoring requirements are consistent changes required by the State. a combined distribution system (as Table V–5 summarizes proposed with the approach taken in the Stage 1 routine and reduced monitoring DBPR.

TABLE V–5.—PROPOSED STAGE 2B ROUTINE AND REDUCED MONITORING REQUIREMENTS FOR NON-CONSECUTIVE SYSTEMS AND FOR CONSECUTIVE SYSTEMS THAT ALSO TREAT SOURCE WATER TO PRODUCE FINISHED WATER 1

System size and source Routine monitoring (per Requirements to qualify for reduced Reduced monitoring Trigger for returning to water type plant) 2 monitoring (per plant) routine monitoring

Subpart H systems serv- Four dual sample sets One year of completed routine moni- Two dual sample sets TOC >4.0 mg/L as an ing ≥10,000 people. per quarter. toring and all TTHM and HAA5 per quarter. RAA, or TTHM LRAA LRAAs are no more than 0.040 >0.040 mg/L or HAA5 mg/L and 0.030 mg/L, respec- LRAA >0.030 mg/L. tively, and TOC running annual average ≤4.0 mg/L. Subpart H systems serv- Two dual sample sets One year of completed routine moni- Two dual sample sets TOC >4.0 mg/L as an ing 500 to 9,999 peo- per quarter 3. toring and all TTHM and HAA5 per year 4. RAA, or Single Sam- ple. LRAAs are no more than 0.040 ple of TTHM >0.060 mg/L and 0.030 mg/L, respec- mg/L or HAA5 >0.045 tively, and TOC running annual mg/L.5 average ≤4.0 mg/L. Subpart H systems serv- One dual sample set Monitoring may not be reduced ...... NA ...... NA. ing <500 people. per year 56. Ground water systems Two dual sample sets One year of completed routine moni- Two dual sample sets Single Sample of TTHM serving ≥10,000 peo- per quarter 3. toring and all TTHM and HAA5 per year 4. >0.060 mg/L or HAA5 ple 7. LRAAs are no more than 0.040 >0.045 mg/L.5 mg/L and 0.030 mg/L, respectively. Ground water systems Two dual sample sets One year of completed routine moni- Two dual samples every Single sample of TTHM serving 500 to 9,999 per year 35. toring and all TTHM and HAA5 third year 4. >0.040 mg/L or HAA5 people 7. LRAAs are no more than 0.040 >0.030 mg/L.5 mg/L and 0.030 mg/L, respectively. Ground water systems One dual sample set One year of completed routine moni- Two dual samples every Single sample of TTHM serving <500 people 7. per year 56. toring and all TTHM and HAA5 third year 4. >0.040 mg/L or HAA5 LRAAs are no more than 0.040 >0.030 mg/L 5 mg/L and 0.030 mg/L, respectively.

Consecutive systems System must meet the routine and reduced monitoring requirements of a non-consecutive system with the same pop- that also treat source ulation and source water. Monitoring may be reduced to the level required of that non-consecutive system. water. 1 Samples must be taken during representative operating conditions. Quarterly samples must be taken approximately every 90 days. 2 Systems will use the results of their IDSEs and Stage 1 DBPR compliance monitoring to recommend Stage 2B monitoring locations rep- resentative of high TTHM and HAA5 concentrations to the State in their IDSE reports. Systems must monitor at the recommended locations un- less the State requires other locations. 3 If site and quarter of highest individual TTHM and HAA5 measurement are the same, monitoring is only required at one location if State ap- proves. 4 If site and quarter of highest individual TTHM and HAA5 measurement are the same, monitoring is only required at one location. 5 If any single sample of TTHM >0.080 mg/L or HAA5 >0.060 mg/L, system must go to increased monitoring of quarterly dual samples at each routine monitoring location and can return to routine monitoring when TTHM ≤0.060 mg/L and HAA5 ≤0.045 mg/L as LRAAs. 6 If the site or month of highest TTHM is not the same as the site or month of highest HAA5, the system must monitor for TTHM at the location of the highest TTHM LRAA during the month of highest TTHM single measurement and for HAA5 at the location of the highest HAA5 LRAA dur- ing the month of highest HAA5 single measurement. 7 Ground water systems are those not under the direct influence of surface water. For the purpose of determining the required number of sam- ples, multiple wells drawing water from a single aquifer may, with State approval, be considered one treatment plant.

i. Subpart H systems serving 10,000 or report, unless the State has required of the highest HAA5 levels, and two more people. other locations. Most systems must take dual sample sets at points representative Routine monitoring: Systems must samples at each plant in the system as of the highest TTHM levels. take four dual sample sets (i.e., a TTHM follows: One dual sample set at the Systems must schedule monitoring so and an HAA5 sample must be taken at existing Stage 1 DBPR average residence that one quarter’s monitoring is each sampling site) per treatment plant time monitoring location with the conducted during the peak historical per quarter. Systems must monitor at highest TTHM or HAA5 LRAA, one month for high TTHM concentration locations recommended in the IDSE dual sample set at a point representative and the other quarterly monitoring is

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00051 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49598 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

conducted approximately every 90 days must be taken during the month(s) of monitor quarterly at the previously on a predetermined schedule included peak historical TTHM and HAA5 levels determined monitoring locations. in the system’s monitoring plan. at the same locations specified under Reduced monitoring: These systems Reduced monitoring: Only systems routine monitoring. Systems that have may not reduce monitoring to less with source water TOC ≤4.0 mg/L as an their highest TTHM and HAA5 levels in frequently than annually. Systems on RAA that have completed at least one the same month must take dual sample increased (quarterly) monitoring may year of routine monitoring may qualify sets at both the high TTHM and high return to routine monitoring if the for reduced monitoring (see Table V–5). HAA5 sites. If the high months for LRAAs of quarterly samples are no more Systems that have a TTHM LRAA TTHM and HAA5 are not the same, the than 0.060 mg/L for TTHM and 0.045 ≤0.040 mg/L and an HAA5 LRAA system must take dual sample sets in mg/L for HAA5. ≤0.030 mg/L at all sites, in addition to both the high TTHM and high HAA5 Compliance determination: A PWS is a source water TOC RAA ≤ 4.0 mg/L, months. Systems on a reduced in compliance when the annual sample may reduce the monitoring frequency monitoring schedule may remain on (or LRAA of quarterly samples, if for TTHM and HAA5 to two dual that reduced schedule as long as the increased or additional monitoring is sample sets (one each at sites annual sample taken at each location is conducted) is less than or equal to the representative of the highest HAA5 and no more than 0.060 mg/L for TTHM and MCL. If the annual sample exceeds the TTHM LRAAs) per treatment plant per 0.045 mg/L for HAA5 or if source water MCL, the system must conduct quarter. Systems on a reduced TOC exceeds 4.0 mg/L as an RAA. increased (quarterly) monitoring but is monitoring schedule may remain on Systems that do not meet these levels not immediately in violation of the that reduced schedule as long as the must revert to routine monitoring in the MCL. The system is out of compliance LRAA of all samples taken in the year quarter immediately following the if the LRAA of the quarterly samples for is no more than 0.040 mg/L for TTHM quarter in which the system exceeded the past four quarters exceeds the MCL. and 0.030 mg/L for HAA5 or if source 0.060 mg/L for TTHM or 0.045 mg/L for iv. Ground water systems serving water TOC exceeds 4.0 mg/L as an RAA. HAA5. Additionally, the State may 10,000 or more people. Routine Systems must revert to routine return a system to routine monitoring at monitoring: Systems are required to monitoring in the quarter immediately the State’s discretion. monitor quarterly for each treatment following any quarter in which the Compliance determination: A PWS is plant in the system by taking two dual LRAA for any monitoring location in compliance with Stage 2B when the sample sets, one each at sites exceeds 0.040 mg/L for TTHM or 0.030 LRAAs of each sample location, representative of high HAA5 levels and mg/L for HAA5. Additionally, the State computed quarterly, are less than or high TTHM levels (as recommended in may return a system to routine equal to the MCLs. Otherwise, the the IDSE report). However, if the State monitoring at the State’s discretion. system is out of compliance. If the determines that the sites representative Compliance determination: A PWS is annual sample taken under reduced of the high TTHM and HAA5 levels are in compliance with Stage 2B when the monitoring exceeds the MCL, the system the same, the State may determine that TTHM and HAA5 LRAAs for each must resume quarterly monitoring but is the system only has to monitor at one sample location, computed quarterly, not immediately in violation of the site per treatment plant. One quarterly are less than or equal to the Stage 2B MCL. The system is out of compliance sample must be taken during the peak MCLs of 0.080 mg/L and 0.060 mg/L, if the LRAA of the quarterly sample for historical month for TTHM, with respectively. Otherwise, the system is the past four quarter exceeds the MCL. subsequent quarterly samples taken out of compliance. iii. Subpart H systems serving fewer approximately every 90 days. ii. Subpart H systems serving 500 to than 500 people. Routine monitoring: Reduced monitoring: To qualify for 9,999 people. Routine monitoring: Systems are required to sample annually reduced monitoring, systems must meet Systems must monitor quarterly for each for each treatment plant at the location certain requirements (see Table V–5). treatment plant by taking two dual with high TTHM and HAA5 values Systems eligible for reduced monitoring sample sets, one each at sites during the month of peak historical may reduce the monitoring frequency representative of high HAA5 levels and TTHM levels. The system must take one from quarterly to annually. Samples high TTHM levels (as recommended in dual sample set at the site representative must be taken during the month(s) of the IDSE report). However, if the State of the high HAA5 and TTHM levels (at peak historical TTHM and HAA5 levels determines that the sites representative the Stage 1 DBPR monitoring site or as at the same locations specified under of the high TTHM and HAA5 levels are recommended in the IDSE report), routine monitoring. Systems that have at the same location, the State may unless the State determines that the their highest TTHM and HAA5 levels in determine that the system is only highest TTHM site and the highest the same quarter must take dual sample required to monitor at one site per HAA5 site are not at the same location sets at both the high TTHM and high treatment plant. or are not during the same quarter. If the HAA5 sites. If the quarter for high Systems must conduct quarterly State determines that the highest TTHM TTHM and high HAA5 are not the same, monitoring during the peak historical and highest HAA5 do not occur in the the system must take dual sample sets month for TTHM with quarterly same location, the system is required to in both the high TTHM and high HAA5 samples taken approximately every 90 take two samples, an HAA5 sample at quarters. Systems on a reduced days on a predetermined schedule the site representative of the high HAA5 monitoring schedule may remain on specified in the system’s monitoring levels and a TTHM sample at the site that reduced schedule as long as the plan. All samples must be taken as dual representative the high TTHM levels. If annual sample taken at each location is sample sets (i.e., a TTHM and an HAA5 the State determines that the highest no more than 0.060 mg/L for TTHM and sample must be taken at each site). TTHM and highest HAA5 do not occur 0.045 mg/L for HAA5. Systems that do Reduced monitoring: To qualify for in the same quarter, the systems is not meet these levels must revert to reduced monitoring, systems must meet required to take one sample in the high routine monitoring in the quarter certain prerequisites (see Table V–5). TTHM quarter and one sample in the immediately following the quarter in Systems eligible for reduced monitoring high HAA5 quarter. If the annual which the system exceeded 0.060 mg/L may reduce the monitoring frequency sample exceeds the MCL for either for TTHM or 0.045 mg/L for HAA5. from quarterly to annually. Samples TTHM or HAA5, the system must Additionally, the State may return a

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00052 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49599

system to routine monitoring at the are not at the same location. If the State increased (quarterly) monitoring but is State’s discretion. makes this determination, the system is not immediately in violation of the Compliance determination: A PWS is required to take samples at two MCL. The system is out of compliance in compliance with Stage 2B when the locations, an HAA5 sample at the site if the LRAA of the quarterly samples for locational running annual average of representative of the high HAA5 levels the past four quarters exceeds the MCL. each sample location, computed and a TTHM sample at the site vi. Consecutive systems. Routine quarterly, is less than or equal to the representative of the high TTHM levels. monitoring: Monitoring requirements MCL. Otherwise, the system is out of If the State makes a determination that are determined by whether the compliance. If the annual sample high TTHM and high HAA5 occur in consecutive system purchases all of its exceeds the MCL, the system must different quarters, the system must finished water year-round or also treats conduct increased (quarterly) monitor accordingly. If the annual source water, along with the population monitoring but is not immediately in sample exceeds the MCL for either served and source water type of the violation of the MCL. The system is out TTHM or HAA5, the system must wholesale system (unless the of compliance if the LRAA of the monitor quarterly at the previously consecutive system also has a surface quarterly sample for the past four determined monitoring locations. water or ground water under the direct quarter exceeds the MCL. Reduced monitoring: To qualify for v. Ground water systems serving reduced monitoring, systems must meet influence of surface water (GWUDI) fewer than 10,000 people. Routine certain prerequisites (see Table V–5). source and the wholesale system is only monitoring: Systems serving 500 to Systems eligible for reduced monitoring ground water, in which case the 9,999 people are required to take two may reduce the monitoring frequency consecutive system is classified as a dual sample sets annually, one each at for TTHM and HAA5 to every third subpart H system). Section V.C. of sites representative of high HAA5 levels year. Systems are required to take two today’s document provides a more and high TTHM levels (as water samples, at sites representative of detailed discussion of consecutive recommended in the IDSE report). high HAA5 and TTHM levels (as system issues. However, if the State determines that discussed under routine monitoring) As noted earlier, for consecutive the sites representative of the high during the month of peak TTHM levels. systems that purchase all their finished TTHM and HAA5 levels are the same, Systems on a reduced monitoring water year-round, EPA is proposing the State may allow the system to schedule may remain on that reduced population-based monitoring. The monitor at only one site per treatment schedule as long as the sample taken proposed number of monitoring plant. If the State makes a determination every third year is no more than 0.040 locations is based on the source water that high TTHM and high HAA5 occur mg/L for TTHM and 0.030 mg/L for type of the wholesale system and in different quarters, the system must HAA5. Systems that do not meet these consecutive system population. monitor accordingly. If the annual levels must resume routine annual Proposed Stage 2B compliance sample exceeds the MCL for either monitoring until their annual average is monitoring requirements for TTHM or HAA5, the system must no more than 0.040 mg/L for TTHM and consecutive systems that purchase all monitor quarterly at the previously 0.030 mg/L for HAA5. their finished water are contained in determined monitoring locations. Compliance determination: A PWS is Table V–6. Consecutive systems that Systems serving fewer than 500 in compliance when the annual sample also treat source water to produce people are required to take one dual (or LRAA of quarterly samples, if finished water must monitor at the same sample set at the site representative of increased or additional monitoring is locations and same frequency as a non- both high HAA5 and TTHM levels, conducted) is less than or equal to the consecutive system with the wholesale unless the State determines that the MCL. If the annual sample exceeds the system’s source water type and the high TTHM site and the high HAA5 site MCL, the system must conduct consecutive system’s population.

TABLE V–6.—PROPOSED POPULATION-BASED ROUTINE MONITORING ROUTINE FREQUENCIES AND LOCATIONS UNDER STAGE 2B FOR CONSECUTIVE SYSTEMS THAT PURCHASE ALL THEIR FINISHED WATER YEAR-ROUND

Distribution system sample location 2

Monitoring Existing Source water type Population size category 1 Highest Highest frequency Total TTHM HAA5 stage 1 locations locations compliance locations 3

Subpart H ...... 0–499 ...... per year ...... 2 4 1 1 ...... 500–4,999 ...... per quarter ...... 2 4 1 1 ...... 5,000–9,999 ...... per quarter ...... 2 1 1 ...... 10,000–24,999 ...... per quarter ...... 4 2 1 1 25,000–49,999 ...... per quarter ...... 6 3 2 1 50,000–99,999 ...... per quarter ...... 8 4 2 2 100,000–499,999 ...... per quarter ...... 12 6 3 3 500,000–1,499,000 ...... per quarter ...... 16 8 4 4 1,500,000–4,999,999 ..... per quarter ...... 20 10 5 5 ≥5,000,000 ...... per quarter ...... 24 12 6 6 0–499 ...... per year ...... 2 4 1 1 ...... 500–9,999 ...... per year ...... 2 1 1 ...... Ground Water ...... 10,000–99,999 ...... per quarter ...... 4 2 1 1 100,000–499,999 ...... per quarter ...... 6 3 2 1 ≥500,000 ...... per quarter ...... 8 4 2 2 1 All systems must take at least one dual sample set during month of highest DBP concentrations. Systems on quarterly monitoring must take dual sample sets approximately every 90 days.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00053 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49600 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

2 Locations based on system recommendations for Stage 2B monitoring locations in IDSE report to the State, unless State requires different or additional locations. Locations should be distributed through distribution system to the extent possible. 3 Alternate between highest HAA5 LRAA and highest TTHM LRAA locations among the existing Stage 1 compliance locations. If the number of existing Stage 1 compliance locations is fewer than the specified number for Stage 2B, alternate between highest HAA5 LRAA locations and highest TTHM LRAA locations from the IDSE. 4 System is required to take individual TTHM and HAA5 samples at the locations with the highest TTHM and HAA5 concentrations, respec- tively. Only one location with a dual sample set per monitoring period is needed if highest TTHM and HAA5 concentrations occur at the same location.

Reduced monitoring: Consecutive Consecutive systems that purchase all of monitoring at the same locations and systems can qualify for reduced their finished water year-round may same frequency as a non-consecutive monitoring if the LRAA at each location reduce their monitoring as specified in system with the wholesale system’s is ≤0.040 mg/L for TTHM and ≤0.030 Table V–7. Consecutive systems that source water type and the consecutive mg/L for HAA5 based on at least one also treat source water to produce system’s population. year of monitoring at Stage 2B locations. finished must conduct reduced

TABLE V–7.—REDUCED MONITORING FREQUENCY FOR CONSECUTIVE SYSTEMS THAT BUY ALL THEIR WATER

Population served Reduced monitoring frequency and location

Subpart H systems

<500 ...... Monitoring may not be reduced. 500 to 4,999 ...... 1 TTHM and 1 HAA5 sample per year at different locations or during different quarters if the highest TTHM and HAA5 occurred at different locations or different quarters or 1 dual sample per year if the highest TTHM and HAA5 occurred at the same location and quarter. 5,000 to 9,999 ...... 2 dual sample sets per year; one at the location with the highest TTHM single measurement during the quarter that the highest single TTHM measurement occurred, one at the location with the highest HAA5 single meas- urement during the quarter that the highest single HAA5 measurement occurred. 10,000 to 24,999 ...... 2 dual sample sets per quarter at the locations with the highest TTHM and highest HAA5 LRAAs. 25,000 to 49,999 ...... 2 dual sample sets per quarter at the locations with the highest TTHM and highest HAA5 LRAAs. 50,000 to 99,000 ...... 4 dual sample sets per quarter at the locations with the two highest TTHM and two highest HAA5 LRAAs. 100,000 to 499,999 ...... 4 dual sample sets per quarter at the locations with the two highest TTHM and two highest HAA5 LRAAs. 500,000 to 1,499,999 ...... 6 dual sample sets per quarter at the locations with the three highest TTHM and three highest HAA5 LRAAs. 1,500,000 to 4,999,999 ...... 6 dual sample sets per quarter at the locations with the three highest TTHM and three highest HAA5 LRAAs. >=5,000,000 ...... 8 dual sample sets per quarter at the locations with the four highest TTHM and four highest HAA5 LRAAs.

Ground water systems

<500 ...... 1 TTHM and 1 HAA5 sample every third year at different locations or during different quarters if the highest TTHM and HAA5 occurred at different locations or different quarters or 1 dual sample every third year if the highest TTHM and HAA5 occurred at the same location and quarter. 500 to 9,999 ...... 1 TTHM and 1 HAA5 sample every year at different locations or during different quarters if the highest TTHM and HAA5 occurred at different locations or different quarters or 1 dual sample every year if the highest TTHM and HAA5 occurred at the same location and quarter. 10,000 to 99,000 ...... 2 dual sample sets per year; one at the location with the highest TTHM single measurement during the quarter that the highest single TTHM measurement occurred and one at the location with the highest HAA5 single measurement during the quarter that the highest single HAA5 measurement occurred. 100,000 to 1,499,999 ...... 2 dual sample sets per quarter; at the locations with the highest TTHM and highest HAA5 LRAAs. ≥1,500,000 ...... 4 dual sample sets per quarter; at the locations with the two highest TTHM and two highest HAA5 LRAAs.

Systems may remain on reduced either 0.060 mg/L for TTHM or 0.045 2. How Was This Proposal Developed? monitoring as long as the TTHM LRAA mg/L for HAA5, or if the source water The proposed monitoring ≤ 0.040 mg/L and the HAA5 LRAA annual average TOC level, before any requirements for the IDSE and Stage 2B ≤ 0.030 mg/L at each monitoring location treatment, exceeds 4.0 mg/L at any primarily follow a plant-based approach for systems with quarterly reduced treatment plant treating surface water or that was adopted from the 1979 TTHM monitoring. If the LRAA at any location ground water under the direct influence Rule and the Stage 1 DBPR. This exceeds either 0.040 mg/L for TTHM or of surface water, the system must approach includes differences in 0.030 mg/L for HAA5 or if the source resume routine monitoring. monitoring frequency between surface water annual average TOC level, before Compliance determination: A PWS is water and ground water sources, and any treatment, exceeds 4.0 mg/L at any in compliance when the annual sample between large and small systems. of the system’s treatment plants treating or LRAA of quarterly samples is less However, the proposed monitoring surface water or ground water under the than or equal to the MCLs. If an annual requirements differ from Stage 1 DBPR direct influence of surface water, the sample exceeds the MCL, the system requirements in certain areas, including system must resume routine monitoring. (a) sampling intervals for quarterly must conduct increased (quarterly) For systems with annual or less frequent monitoring, (b) reduced monitoring monitoring but is not immediately in reduced monitoring, systems may frequency, (c) different sampling remain on reduced monitoring as long violation of the MCL. The system is out locations by disinfectant type (for the as each TTHM sample ≤0.060 mg/L and of compliance if the LRAA of the IDSE), and (d) population-based each HAA5 sample ≤0.045 mg/L. If the quarterly samples for the past four monitoring requirements for certain annual sample at any location exceeds quarters exceeds the MCL. consecutive systems. This subsection

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00054 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49601

presents the basis for these disinfectant residual used in the for ground water systems. The basis requirements. distribution system. Systems that use for this is that, in general, systems a. Sampling intervals for quarterly chloramines are required to select more using surface water or mixed source monitoring. Today’s proposal requires near-entry point monitoring sites for the water supplies are likely to systems conducting routine quarterly IDSE than chlorinated systems of experience higher and more variable monitoring to sample approximately similar size and source water type. This DBP occurrence over time than every 90 days. This provision modifies is due to differences in DBP formation systems using ground water the approach used in the 1979 TTHM under chloraminated and chlorinated exclusively. rule and the Stage 1 DBPR, which conditions. Chloramine residuals are —Smaller systems should be allowed to requires certain systems to conduct more stable than chlorine residuals and monitor less frequently per location monitoring based on calendar quarters. do not react as readily with organic than larger systems because their When systems are required to monitor compounds in the water. Based on distribution systems are generally less based on calendar quarters, systems can evaluation of Information Collection complex and monitoring costs on a choose to cluster samples during times Rule data, DBP concentrations in per capita basis are much higher. of the year when DBP levels are lower chloraminated systems vary less —For systems using surface water, the (DBPs tend to form more slowly in throughout the distribution system than ratio of IDSE sample locations to the colder water temperatures). For in chlorinated systems. HAA5, in number of routine sample locations example, a system could sample in particular, can peak at or near the entry required for Stage 2B should be December (at the end of the fourth point to the distribution system in a approximately 2:1 (consistent with quarter) and again in January (at the chloraminated system (USEPA 2003o). Advisory Committee beginning of the first quarter) when the d. Population-based monitoring recommendations for plant-based water is the coldest and sample in April requirements for certain consecutive monitoring). IDSE sampling is (at the beginning of the second quarter) systems. While the Advisory Committee intended to identify distribution and September (at the end of the third recommended basic principles for how system locations with high DBP levels quarter) at either end of the hot summer consecutive systems should be and should, therefore, be more months. regulated, it did not recommend how thorough than routine monitoring. To address the concern with systems —Because ground water systems have EPA should explicitly address some of not sampling during months with the much less variable DBP levels within the unique situations that pertain to highest DBP levels, EPA is proposing to the distribution system than surface consecutive systems. In this regard, require systems to monitor during the water systems (see section IV), a consecutive systems that purchase all of month of highest historical DBP smaller number of additional IDSE their finished water year-round are concentrations and require that systems monitoring locations is warranted. different than other systems in that they monitor approximately every 90 days. —Distribution system sampling do not have a treatment plant. Rather, EPA believes that this new monitoring locations should be approximately these systems often receive water from strategy will improve public health consistent with the proposed plant- protection because systems will be multiple wholesale systems or through based approach as recommended by required to monitor when high DBP multiple consecutive system entry the Advisory Committee. This will levels are expected, and the LRAA will points on a seasonal or intermittent capture the locations with the high better reflect actual exposure during the basis. Because a plant-based monitoring TTHM and HAA5 LRAAs identified year. approach (which counts treated water in the IDSE, but also include Stage 1 b. Reduced monitoring frequency. distribution system entry points from compliance locations with high Today’s proposal contains provisions different entities as plants) would be TTHM and HAA5 for historical allowing reduced routine monitoring very difficult to implement for tracking. consecutive systems that purchase all of when certain criteria are fulfilled Consistent with the first two their finished water year-round, EPA is (shown in Table V–5 and V–7). EPA guidelines, the proposed population- proposing a population-based approach believes that more stringent standards based monitoring requirements for such systems. are necessary to ensure public health maintain the same monitoring frequency Under a population-based approach, protection when systems reduce the per sample location as proposed under the frequency of monitoring is based on frequency of their DBP monitoring. the plant-based approach. The following the population served and remains the Under the reduced monitoring subsection provides further discussion same regardless of how many entities provisions, systems must collect of the population-based approach as it are providing water to the consecutive samples during the months of highest might apply to all systems. DBP occurrence. For systems sampling system at different times of the year. annually under the reduced monitoring The population categories and 3. Request For Comment provisions, EPA believes that public associated monitoring frequencies in EPA is requesting comments on the health protection would likely be Tables V–4 and V–6 for IDSE and Stage proposed monitoring requirements. This ensured throughout the year if the high 2B routine monitoring reflect EPA’s subsection begins with a list of specific values are known to be below 0.060 mg/ consideration that distribution system questions related to the proposed L for TTHM and 0.045 mg/L for HAA5. complexity generally increases as the requirements for IDSE and Stage 2B Systems monitoring every three years population served grows. Increasing monitoring. This is followed by a must maintain single samples under distribution system complexity warrants discussion of issues associated with 0.040 mg/L for TTHM and 0.030 mg/L more monitoring to represent DBP plant-based monitoring requirements for HAA5 to ensure adequate public occurrence. and a request for comment on potential health protection over the course of the EPA developed the proposed approaches to address these issues, three years. population-based monitoring including the extension of population- c. Different IDSE sampling locations requirements in accordance with certain based monitoring requirements to all by disinfectant type. Today’s proposal guidelines. These are stated as follows: systems under the Stage 2 DBPR. contains different requirements for IDSE —The sampling frequency for surface a. Proposed IDSE and Stage 2B monitoring locations based on the water systems should be greater than monitoring requirements. EPA is

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00055 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49602 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

requesting comment on a number of part, with provisions such as the the Stage 1 DBPR and derives from the specific aspects of the proposed following: assumption that as systems increase in monitoring requirements. —Should EPA set a limit on the size, they will tend to have more plants —Should EPA require all systems that maximum number of IDSE and (with different sources and treatment) are on reduced monitoring to revert to routine monitoring samples that could and increased complexity. This routine monitoring during the IDSE be required? Should this limit be warrants increased monitoring to monitoring period to allow for more different for systems using ground represent DBP occurrence in the data to be evaluated in the IDSE water or surface water or mixed distribution system. report to better select Stage 2B systems? For different system size EPA has identified a number of issues monitoring locations? Or should EPA categories? What rationale should be related to the use of a plant-based require a system to be on routine used to specify maximum sample monitoring approach under the Stage 2 monitoring during the IDSE numbers? DBPR. The following discussion monitoring period in order to be —Should a provision be included that presents these issues and solicits eligible for an IDSE waiver? What would allow States to reduce the comment on approaches to address limitations, if any, should EPA put on sampling frequency, beyond those them, including the use of population- system eligibility for an IDSE waiver? currently proposed (i.e., common based monitoring requirements. aquifer determinations and low DBP i. Issues with plant-based monitoring —Should EPA require different IDSE levels)? If so, should specific criteria requirements. One issue with a plant- monitoring locations for subpart H for systems to qualify for State based monitoring approach is that it can systems based on the residual approval of reduced monitoring be result in disproportionate monitoring disinfectant (chlorine or chloramines) specified in the rule? requirements for systems serving the in light of the possible difficulties for —What, if any, criteria should be set by same number of people. This occurs implementation and data which systems with very large because the required number of management? Should EPA specify distribution systems but few plants sampling sites increases with the monitoring locations in the rule would be required to conduct number of plants that feed disinfected language for samples intended to additional IDSE or routine water into a distribution system. represent exposure for people in high- monitoring, beyond that currently Consequently, some systems, depending rise buildings? Should monitoring proposed? upon their size, the number of treatment location selection be addressed in —For subpart H mixed systems, should plants, and the nature of their guidance? Where should these States be given discretion to reduce distribution system, will be required to locations be so that they are truly routine compliance monitoring collect relatively large or small numbers representative of the levels of DBPs in samples intended to represent ground of TTHM and HAA5 samples relative to water actually being consumed in water sources, since such sources their population served. these kinds of structures? typically have lower precursor levels Table V–8 reflects EPA estimates of —Is a population-based monitoring and produce lower DBP the number of plants per system by approach (instead of a plant-based concentrations? system size category for systems using monitoring approach) for consecutive —Should EPA allow or require systems ground water and subpart H systems. systems that purchase all of their to reallocate plant-based IDSE Subpart H systems include systems that finished water year-round appropriate monitoring locations from small use ground water as a source because and, if so, is the population-based plants to large plants? From plants under the proposal, ground water plants approach proposed today adequate? with better water quality (based on in subpart H systems are treated as EPA solicits comment on the expected lower DBP formation) to surface water plants for purposes of significance of monitoring and poorer water quality? What criteria determining monitoring requirements. implementation issues such as common should be used? While the proposed plant-based aquifer determinations, consecutive b. Plant-based vs. population-based requirements distinguish sampling system entry point determinations, monitoring requirements. The proposed requirements by three systems sizes seasonal plants, and monitoring monitoring requirements incorporate a (<500 people, 500–9999 people, and inequities, and whether the proposed plant-based approach for all systems 10,000 or more people), Table V–8 monitoring requirements should be other than consecutive systems that includes additional size categories to modified. EPA also solicits comment on purchase all of their finished water year- reflect the potential inequities in modifying the proposed monitoring round. The plant-based approach was sampling requirements among different- requirements to address these issues, in adopted from the 1979 TTHM Rule and sized systems.

TABLE V–8.—NUMBER OF TREATMENT PLANTS PER SYSTEM (BASED ON DATA FROM 1995 CWSS (1))

No. of sys- No. of treatment plants per system Source water type Population served tems in 10th 90th 95th database percentile Median Mean percentile percentile Maximum

Subpart H ...... 0–499 ...... 124 1 1 1.4 2 3 5 500–4,999 ...... 146 1 1 1.3 2 3 6 5,000–9,999 ...... 64 1 1 1.7 3 4 6 10,000–24,999 ...... 59 1 1 2.0 3 4 18 25,000–49,999 ...... 46 1 1 2.2 4 6 9 50,000–99,999 ...... 76 1 2 3.4 6 12 34 100,000–499,999 ... 51 1 2 3.0 5 10 21 ≥500,000 ...... 23 2 4 5.8 10 13 56 Ground Water ...... 0–499 ...... 181 1 1 1.4 3 4 11 500–9,999 ...... 332 1 1 1.8 3 4 13

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00056 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49603

TABLE V–8.—NUMBER OF TREATMENT PLANTS PER SYSTEM (BASED ON DATA FROM 1995 CWSS (1))—Continued

No. of sys- No. of treatment plants per system Source water type Population served tems in 10th 90th 95th database percentile Median Mean percentile percentile Maximum

10,000–99,999 ...... 128 1 4 4.2 9 11 18 ≥100,000 ...... 21 1 3 9.9 31 32 33 (1) Results from analysis of 1995 CWSS data (Question Q18). The analysis uses a statistical bootstrapping approach to generate the number of plants per system. Details of this analysis are described in the 2002 revisions to the Model Systems Report [to be published]. The maximums reflect the maximum number of plants per system among the respondents to the 1995 CWSS. Since the 1995 CWSS database only reflects a fraction of all the systems in the respective size categories, some systems are likely to have a higher number of plants per system than the maximums listed in this table.

Noteworthy in Table V–8 are the wide Another issue with plant-based samples that could be required. EPA ranges of number of plants per system monitoring requirements is when plants believes that this limit should be in the various size categories for both or consecutive system entry points are different for systems using ground ground water and surface water systems operated seasonally or intermittently. A water or surface water or mixed and, consequently, the wide range of monitoring location that represents a systems and for different system size potential monitoring implications. Since plant or entry point during a monitoring categories. However, the Agency has the number of treatment plants directly period when it is in operation will not not developed a rationale to specify influences the number of samples be representative when that plant or maximum sample numbers for required, systems serving the same entry point it is not in operation. specific system categories. number of people may have more than A third issue is requirements for —Include a provision that would allow a 10-fold difference in required consecutive systems. For consecutive States to reduce the required number sampling, depending on the numbers of systems that also treat source water to of samples for reasons other than plants in their systems. For example, produce finished water, each those currently proposed (i.e., Table V–8 indicates that for ground consecutive system entry point is common aquifer determinations and water systems serving at least 10,000 considered a treatment plant for the low DBP levels). EPA would have to people, at least 10% of the systems had purpose of determining monitoring develop specific criteria in the rule for only one treatment plant, while 10% requirements, except when the State systems to qualify for State approval (90th percentile) had 10 or more allows multiple entry points to be of reduced monitoring. For example, treatment plants. treated as a single plant (see section V.C. in subpart H mixed systems, States for further discussion). Each entry point could be given discretion to reduce While Table V–8 does not take into is treated as a separate plant to routine compliance monitoring for account factors that may reduce recognize different source waters and ground water sources, since such monitoring requirements, such as treatment (resulting in different DBP sources typically have lower DBP common aquifer determinations, EPA levels) from the wholesale system(s) and concentrations. believes these data indicate that DBP the treatment plants(s) operated by the —Develop criteria by which systems sampling requirements based on the consecutive system. However, under with very large distribution systems number of water treatment plants per this plant-based approach, State but with few plants would be required system may be excessive for many determinations of monitoring to conduct additional IDSE or routine systems. This is particularly the case for requirements for consecutive systems monitoring in order to better those systems with many ground water will be complicated, especially in large characterize DBP exposure throughout plants, since their DBP levels are often combined distribution systems with the distribution system. low and relatively stable. many connections between systems. These provisions would allow for Conversely, for other systems, such as ii. Approaches to addressing issues some issues to be addressed, but would large surface water systems with one with plant-based monitoring. EPA is make implementation complex and plant, plant-based monitoring requesting comment on two approaches could add a significant burden to States. requirements may not require enough to address the issues with plant-based An alternative approach to addressing samples to fairly represent DBP monitoring requirements described in the issues with plant-based monitoring occurrence in the distribution system. this subsection. One approach is to keep requirements is to apply population- For example, under the plant-based the proposed plant-based monitoring based monitoring requirements to all approach, a system with only one plant approach and add new provisions to systems. Under a population-based serving 100,000—499,000 people would address specific concerns. Another monitoring approach, the total system have the same sampling requirements as approach is to base monitoring population served and the source water a system with one plant serving 11,000 requirements on population served in type would determine the number of people. The larger of these two systems lieu of the number of water treatment IDSE and routine monitoring samples is likely to have much more pipe length plants per system. The following taken. Monitoring requirements would and other complex factors influencing paragraphs describe each approach. not be based on the number of plants DBP formation (such as number of EPA could maintain a plant-based per system or consecutive system entry storage tanks or booster chlorination monitoring approach and try to address points. States would not be required to points in the distribution system). Also, the related issues described in this make common aquifer determinations or systems with multiple plants must take subsection through modifying the address whether plants are combined the same number of samples per plant, proposed monitoring requirements with into a single pipe prior to entering the even if one plant provides a much provisions like the following: distribution system. higher percentage of the water than —Set a limit on the maximum number Proposed population-based another. of IDSE and routine monitoring monitoring requirements for

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00057 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49604 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

consecutive systems that purchase all categories than under the plant-based and mean number of plants per system their finished water year-round are approach, population-based monitoring given in Table V–8 for each of the size shown in Tables V–4, V–6, and V–7. could result in an equitable categories. For surface water systems, Also, the proposed rule language in proportioning of DBP sampling the median provides a better indicator subparts U and V contains requirements requirements. Tables V–9 and V–10 of the typical number of required for population-based monitoring similar compare the proposed numbers of sampling locations under the plant- to what might be required for all sampling locations per system under a based approach because it is much less systems. EPA believes that through population-based approach with a sensitive to systems with a very large using a broader array of system size plant-based approach, using the median number of plants.

TABLE V–9.—COMPARISON OF MONITORING LOCATIONS PER SYSTEM UNDER IDSE FOR PLANT-BASED AND POPULATION- BASED APPROACHES

Plant-based Population-based Number of monitoring locations per Number system Source water type Population size category of sam- Number of Number of moni- pling monitoring lo- Based on me- toring locations periods cations per Based on mean 3 1 dian number of per system plant plants per number of plants per system 2 system 2

Subpart H ...... 0–499 ...... 2 2 2 3 2 500–4,999 ...... 4 2 2 3 2 5,000–9,999 ...... 4 2 2 3 4 10,000–24,999 ...... 6 8 8 16 8 25,000–49,999 ...... 6 8 8 18 12 50,000–99,999 ...... 6 8 16 27 16 100,000–499,999 ...... 6 8 16 24 24 500,000–1,499,000 ...... 32 1,500,000–4,999,999 ..... 6 8 32 46 40 ≥5,000,000 ...... 48 Ground Water ...... 0–499 ...... 2 2 2 2 2 500–9,999 ...... 2 2 2 4 2 10,000–99,999 ...... 4 2 8 9 6 100,000–499,999 ...... 4 2 6 20 8 ≥500,000 ...... 12 1 From Table V–5. 2 Calculated from the number of locations per plant multiplied by number of plants per system (Table V–8). 3 From Table V–4.

TABLE V–10.—COMPARISON OF ROUTINE MONITORING LOCATIONS PER SYSTEM UNDER STAGE 2B FOR PLANT-BASED AND POPULATION-BASED APPROACHES

Plant-based Population-based Number of monitoring locations per Frequency system Source water type Population size category of Number of Number of moni- monitoring lo- Based on me- toring locations monitoring cations per Based on mean dian number of per system 3 1 number of plants plant plants per per system 2 system 2

Subpart H ...... 0–499 ...... 1 1 1 1 2 500–4,999 ...... 4 2 2 3 2 5,000–9,999 ...... 4 2 2 3 2 10,000–24,999 ...... 4 4 4 8 4 25,000–49,999 ...... 4 4 4 9 6 50,000–99,999 ...... 4 4 8 14 8 100,000–499,999 ...... 4 4 8 12 12 500,000–1,499,000 ...... 16 1,500,000–4,999,999 ..... 4 4 16 23 20 ≥5,000,000 ...... 24 Ground Water ...... 0–499 ...... 1 1 1 1 2 500–9,999 ...... 1 2 2 4 2 10,000–99,999 ...... 4 2 8 9 4 100,000–499,999 ...... 4 2 6 20 6 ≥500,000 ...... 8 1 From Table V–5. 2 Calculated from the number of locations per plant multiplied by number of plants per system (Table V–8). 3 From Table V–6.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00058 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49605

Under the population-based prepared an economic analysis Stage 2 DBPR and the LT2ESWTR, approach, the number of required addressing monitoring impacts using which will ensure continued microbial sampling locations for systems of the population-based approach protection as systems implement different size and source water type (Economic Analysis for the Stage 2 changes to decrease DBP levels and approximates the number of sampling DBPR, EPA 2003i) and guidance on how minimize risk-risk tradeoffs. locations that would be required for the plant-based monitoring requirements IDSE schedule. Subpart H and ground majority of systems under the plant- would be affected if a population-based water systems covered by today’s based approach. However, systems in approach were used instead (Draft IDSE proposed rule that serve a population of the tail ends of the distribution of Guidance Manual, EPA 2003j). 10,000 or more must submit the results number of plants per system would be EPA requests comments on alternative of their IDSE to the primacy agency two required to take more or fewer samples DBP monitoring requirements that are years after rule promulgation. In than under the plant-based approach. population-based versus plant-based; addition, wholesale or consecutive EPA used the median number of plants specifically on the merits of a systems serving fewer than 10,000 that in a given size category as the primary population-based monitoring approach are part of a combined distribution basis for establishing the number of for all systems for the purpose of system with at least one system serving monitoring locations for the population- addressing the issues raised in this ≥10,000 must meet this same schedule. based approach. section. Specifically: These systems must begin IDSE EPA adjusted the number of sampling —Should alternative system size monitoring early enough to collect and locations for systems in population sizes categories be specified under the analyze 12 months of data and prepare 25,000 to 49,999, 100,000–499,999, and suggested population-based an IDSE report, which includes greater than 1,500,000 to provide a more approach? recommendations for Stage 2B even upward trend in proportion to —What potential issues might be unique monitoring locations (see section V.H). population increase. Consistent with the for a population-based monitoring Subpart H and ground water systems plant-based approach, ground water approach and how might they be covered by today’s proposed rule that systems serving 10,000 people or greater addressed? serve a population of fewer than 10,000 would be required to sample at —Should alternative numbers of (except those noted before) must submit approximately 1⁄3 to 1⁄2 the frequency monitoring locations or frequencies be the results of their IDSE to the primacy required for surface water systems required in the IDSE or for Stage 2B agency four years after rule under the population-based approach. monitoring? EPA suggests that the monitoring promulgation. These systems must —Are reduced monitoring requirements frequencies for the IDSE and Stage 2B begin IDSE monitoring early enough to adequate to ensure continued compliance proposed for consecutive collect and analyze the data and prepare protection relative to the MCL? systems that purchase all of their the IDSE report. —What are the transition costs and finished water year-round (as presented Stage 2A schedule. All systems must issues associated with moving from a in Tables V–4 and V–6) are appropriate comply with the Stage 2A MCLs for plant-based to a population based for all systems if a population-based TTHM and HAA5 three years after rule approach and how might they be approach were used in lieu of a plant- promulgation. addressed? based approach in the final rule. EPA Stage 2B schedule. Systems required believes that the population-based J. Compliance Schedules to submit an IDSE report due two years approach would ensure more equal and after the rule is promulgated must 1. What is EPA Proposing? rational monitoring requirements among comply with Stage 2B six years after systems serving similar populations Today’s proposed rule establishes rule promulgation. Subpart H systems than the plant-based approach does, compliance deadlines for public water required to submit IDSE reports four while providing generally improved systems to implement the requirements years after rule promulgation and representation of DBP occurrence in this rulemaking. EPA is proposing a required to do Cryptosporidium throughout the distribution system. phased strategy for MCLs and monitoring under the LT2ESWTR must Such an approach would simplify simultaneous compliance with the comply with Stage 2B 8.5 years after implementation and reduce LT2ESWTR consistent with the rule promulgation. Small systems not transactional costs to States by recommendation of the M-DBP required to Cryptosporidium monitoring facilitating determination of the number Advisory Committee and to comply must be in compliance with Stage 2B of sampling locations. with SDWA requirements for risk 7.5 years after rule promulgation. Figure To further evaluate the potential balancing. Central to the determination V–2 contains several examples of how implications of monitoring under the of these deadlines is the principle of to determine IDSE and Stage 2B population-based approach, EPA has simultaneous compliance between the compliance dates.

FIGURE V–2. SCHEDULE EXAMPLES

—Wholesale system (pop. 64,000) with three consecutive systems (pops. 21,000; 15,000; 5,000): —IDSE report due for all systems two years after promulgation since wholesale system serves at least 10,000 —Stage 2B compliance beginning six years after promulgation for all systems —Wholesale system (pop. 4,000) with three consecutive systems (pops. 21,000; 5,000; 5,000): —IDSE report due for all systems two years after promulgation since one consecutive system in combined distribution system serves at least 10,000 —Stage 2B compliance beginning six years after promulgation for all systems —Wholesale system (pop. 4,000) with three consecutive systems (pops. 8,000; 5,000; 5,000): —IDSE report due for all systems four years after promulgation since no system in combined distribution system exceeds 10,000 (even though total population exceeds 10,000) —Stage 2B compliance beginning 7.5 years after promulgation if no Cryptosporidium monitoring under the LT2ESWTR is required or be- ginning 8.5 years after promulgation if Cryptosporidium monitoring under the LT2ESWTR is required

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00059 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49606 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

2. How Did EPA Develop This Proposal? serving ≥10,000 to complete IDSE The first, the ‘‘Alternative IDSE’’ EPA is proposing provisions for monitoring and prepare and submit the option, would delay the schedule for simultaneous rule compliance with the IDSE report two years after the rule is each IDSE requirement for two years. LT2ESWTR to maintain a balance finalized. This requirement for Since the compliance date for Stage 2B between DBP and microbial risks. wholesale systems and consecutive would not be delayed, systems would Simultaneous compliance was systems serving fewer than 10,000 that need to implement changes necessary mandated by the 1996 SDWA are part of a combined distribution for compliance on a much shorter Amendments which require that EPA system with at least one system serving schedule. ‘‘minimize the overall risk of adverse at least 10,000 to conduct an ‘‘early The second, the ‘‘Concurrent IDSE’’ allows the wholesale system to be health effects by balancing the risk from Compliance Monitoring’’ option, would aware of compliance challenges facing the contaminant and the risk from other eliminate the IDSE but require the consecutive system and to contaminants, the concentrations of compliance monitoring at an increased implement treatment plant capital and which may be affected by the use of a number of sites during the first year of operational improvements as necessary treatment technique or process that compliance monitoring as a way to to ensure compliance. The Advisory would be employed to attain the identify sites with high DBP levels. This Committee and EPA both recognized maximum contaminant level’’ (Sec. option would reduce government that DBPs, once formed, are difficult to 1412(b)(5)(B)(i)). oversight and management and, as with remove and are generally best addressed If systems were required to comply other rules, leave compliance by treatment plant improvements. with the Stage 2 DBPR prior to the determinations and preparations to While this schedule allows for LT2ESWTR, systems could lower their individual systems (with guidance disinfectant dose or switch to a less systems to have the three years to comply with Stage 2B following State available from States). In addition to effective disinfectant in an attempt to compliance monitoring at Stage 1 DBPR decrease DBP levels. This practice could review of the IDSE report, it begins prior to States being required to obtain compliance monitoring sites during the leave segments of the population first year under Stage 2B, systems would exposed to greater microbial risks. primacy to implement the IDSE. States also monitor at additional compliance Therefore, simultaneous compliance have two years from promulgation to monitoring sites equal in number to the was a consensus recommendation of the adopt and implement new regulations IDSE requirement and selected using the Stage 2 M-DBP Advisory Committee to and may request a two year extension. same criteria that systems use to select ensure that systems would not While EPA is preparing to support IDSE monitoring sites. Following one compromise microbial protection while implementation of those IDSE year of concurrent compliance attempting to meet more stringent DBP requirements that must be completed monitoring, the system would select requirements. prior to States achieving primacy, routine Stage 2B compliance monitoring The Advisory Committee supported several States have expressed concern locations using a protocol similar to the the Initial Distribution System about EPA providing guidance and Evaluation, as discussed in section V.H, reviewing reports from systems that the one used to recommend Stage 2B and EPA is proposing an IDSE schedule State has permitted, inspected, and compliance monitoring locations in the consistent with the Advisory worked with for a long time. These IDSE report. Committee’s recommendations, in States believe that their familiarity with Neither alternative would extend the which systems are required to submit the systems enables them to make the compliance dates for either Stage 2A or their IDSE reports to the State either two best decisions to implement the rule Stage 2B. As with the proposed IDSE, or to four years following rule and protect public health. systems would be eligible for the 40/30 promulgation. The Advisory Committee As specific rule requirements were certification approach if all TTHM and recommended this to allow enough time developed and implementation HAA5 compliance monitoring results in for the State to review (and revise, if schedules and resource burdens the two years prior to the effective date necessary) systems’ recommendations determined, States also expressed were below 0.040 mg/L and 0.030 mg/ for Stage 2B monitoring locations and to concerns about the challenges that early L, respectively. States would be able to allow systems three years after implementation posed. In response to grant very small system waivers to completion of the State review to these concerns, EPA has developed systems serving <500 with a State comply with Stage 2B MCLs as LRAAs several alternatives to the IDSE schedule finding that Stage 1 DBPR compliance at Stage 2B monitoring locations. and provisions that may meet the goals monitoring locations sites are adequate This schedule requires systems of the IDSE, but allow for greater State to represent both high TTHM and high serving ≥10,000 people and smaller involvement, lower implementation HAA5 concentrations. Table V–11 wholesale and consecutive systems that burden, and no delay of the public contains a comparison of the proposed are part of a combined distribution health protection assured by compliance IDSE schedule and the schedules for the system that includes at least one system with Stage 2B. alternatives.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00060 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49607

TABLE V–11.—COMPARISON OF IDSE AND IDSE ALTERNATIVE SCHEDULES [Dates in italics are not in today’s proposed rule, but reflect EPA’s recommendation and guidance]

1 ‘‘Alternative IDSE’’ Requirement Today’s proposal option ‘‘Concurrent compliance monitoring’’ option

IDSE start date for systems ≥10,000 ...... 0.5 years after 2.5 years after Requirement is for system to conduct concur- IDSE start date for systems <10,000 ...... publication. publication rent compliance monitoring (generally IDSE report due for systems ≥10,000 ...... 2.5 years after 4.5 years after equal to number of samples required under IDSE report due for systems <10,000 ...... publication. publication Stage 1 plus number under IDSE) during State review of IDSE report complete for sys- 2 years after 4 years after first year of compliance monitoring. Based tems ≥10,000. publication. publication on results in first year, system would iden- State review of IDSE report complete for sys- 4 years after 6 years after tify routine compliance monitoring locations tems <10,000. publication. publication using a procedure similar to that in IDSE 3 years after 5 years after report and begin routine monitoring. publication. publication 4.5 years after publi- 6.5 years after publi- cation. cation Stage 2B compliance for systems ≥10,000 ..... 6 years after publication 2 Stage 2B compliance for systems <10,000 ..... 7.5 years after publication if system is not required to conduct Cryptosporidium monitoring; 8.5 years after publication if system required to conduct Cryptosporidium monitoring 2 1 Systems serving ≥10,000 also include wholesale systems and consecutive systems serving <10,000 that are part of a combined distribution system in which at least one system serves ≥10,000. 2 State may grant up to two additional years for capital improvements necessary to comply.

3. Request for Comments notices under Subpart Q. Because of the reasonably available to the system. At EPA requests comments on today’s uncertainties in the health data the time this type of variance is granted, proposed compliance schedules. discussed in section III of today’s the State must prescribe a compliance document, EPA is not proposing to Specifically: schedule and may require the system to include information about reproductive implement additional control measures. —Should EPA promulgate an alternative and developmental health effects in Furthermore, before EPA or the State approach to the IDSE recommended public notices at this time. may grant a general variance, it must in section V.H. that achieves the same 2. Request for Comments find that the variance will not result in goal of identifying Stage 2B an unreasonable risk to health to the compliance monitoring locations and EPA requests comment on the public served by the public water does not delay compliance with Stage proposed public notification system. In this proposed rule, EPA is 2B MCLs, but allows for the States to requirements, including whether specifying BATs for general variances receive primacy and be more involved information about the possible under section 1415(a) (see section V.F). in IDSE implementation? Do either reproductive or fetal development Section 1415(e) authorizes the the ‘‘Alternative IDSE’’ option or the effects that may be associated with high primacy agency to issue variances to ‘‘Concurrent Compliance Monitoring’’ levels of DBPs should be provided. small public water systems (those option achieve this goal? Does one L. Variances and Exemptions serving fewer than 10,000 people) where achieve the goal better than the other? the primacy agent determines (1) that States may grant variances in Why? Are there either changes to the system cannot afford to comply with these alternatives or other alternatives accordance with sections 1415(a) and 1415(e) of the SDWA and EPA’s an MCL or treatment technique and (2) not presented that achieve this goal? that the terms of the variances will —Should EPA allow small consecutive regulations. States may grant exemptions in accordance with section ensure adequate protection of human systems to meet Stage 2B compliance health (63 FR 1943–57; USEPA 1998d). deadlines corresponding to their size 1416 of the SDWA and EPA’s regulations. These variances may only be granted (and later than the deadlines for their where EPA has determined that there is wholesale system) provided they 1. Variances no affordable compliance technology complete their IDSE on the same and has identified a small system schedule as the wholesale system and The SDWA provides for two types of variances—general variances and small variance technology under section provided their water quality does not 1412(b)(15) for the contaminant, system affect the water quality of any other system variances. Under section 1415(a)(1)(A) of the SDWA, a State that size and source water quality in system? has primary enforcement responsibility question. As discussed below, small K. Public Notice Requirements (primacy), or EPA as the primacy system variances under section 1415(e) agency, may grant general variances are not available because EPA has 1. What is EPA Proposing? from MCLs to those public water determined that affordable compliance SDWA section 1414(c) requires PWSs systems of any size that cannot comply technologies are available. to provide notice to their customers for with the MCLs because of The 1996 Amendments to the SDWA certain violations or in other characteristics of the water sources. A identify three categories of small public circumstances. EPA’s public notification variance may be issued to a system on water systems that need to be addressed: rule was published on May 4, 2000 (65 condition that the system install the best (1) Those serving a population of 3301– FR 25982), and is codified at 40 CFR technology, treatment techniques, or 10,000; (2) those serving a population of 141.201–141.210 (Subpart Q). Today’s other means that EPA finds available 500–3300; and (3) those serving a proposal does not alter the existing and based upon an evaluation population of 25–499. The SDWA TTHM and HAA5 health effects satisfactory to the State that indicates requires EPA to make determinations of language that is required in most public that alternative sources of water are not available compliance technologies and,

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00061 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49608 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

if needed, variance technologies for EPA has completed an analysis of the the Safe Drinking Water Act’’ (USEPA each size category. A compliance affordability of DBP control 1998g) and the ‘‘Variance Technology technology is a technology that is technologies for each of the three size Findings for Contaminants Regulated affordable and that achieves compliance categories. Based on this analysis, Before 1996’’ (USEPA 1998h). with the MCL and/or treatment multiple affordable compliance 2. What Are the Affordable Treatment technique. Compliance technologies can technologies were found for each of the Technologies for Small Systems? include point-of-entry or point-of-use three system sizes (USEPA 2003i) and therefore variance technologies were not treatment units. Variance technologies The treatment trains considered and are only specified for those system size/ identified for any of the three size categories. The analysis was consistent predicted to be used in EPA’s source water quality combinations for with the methodology used in the compliance forecast for systems serving which there are no listed compliance document ‘‘National-Level Affordability under 10,000 people, are listed in Table technologies. Criteria Under the 1996 Amendments to V–12.

TABLE V–12.—TECHNOLOGIES CONSIDERED AND PREDICTED TO BE USED IN COMPLIANCE TECHNOLOGY FORECAST FOR SMALL SYSTEMS 1

SW water plants GW water plants

• Switching to chloramines as a residual disinfectant ...... • Switching to chloramines as a residual disinfectant • Chlorine dioxide (Not for systems serving fewer than 100 people) ..... • UV • UV ...... • Ozone (not for systems serving fewer than 100 people) 2 • Ozone (not for systems serving fewer than 100 people) ...... • GAC20 2 • Micro-filtration/Ultra-Filtration 2 ...... • Nanofiltration 2 • GAC20 2 ...... • GAC20 + Advanced disinfectants ...... • Membranes (Micro-Filtration/Ultra-Filtration + Nanofiltration) ...... 1 Based on exhibits 6.8a and 6.8b in Economic Analysis for the proposed Stage 2 DBPR (USEPA 2003i) 2 Italicized technologies are those predicted to be used in the compliance forecast.

The household costs for these purchasing bottled water or a home based on 1995 estimates (USEPA 1998h) technologies were compared against the treatment device. The complete and adjusted upward for anticipated national-level affordability criteria to rationale for EPA’s selection of 2.5% as costs attributed to new drinking water determine the affordable treatment the affordability threshold is described regulations since 1995, i.e., the IESWTR, technologies. The Agency’s national- in ‘‘Variance Technology Findings for Stage 1 DBPR, Filter Backwash level affordability criteria were Contaminants Regulated Before 1996’’ Recycling Rule, Arsenic Rule, published in the August 6, 1998 Federal (USEPA 1998h). LT1ESWTR, Public Notification Rule, Register (USEPA 1998g). In this The Variance Technology Findings and Consumer Confidence Rule.1 document, EPA discussed the procedure document also describes the derivation Median household income estimates are for affordable treatment technology of the baselines for median household based on estimates made in 1995 determinations for the contaminants income, annual water bills, and annual (USEPA 1998h) and adjusted upward regulated before 1996. household consumption. Data from the for inflation to represent 2000 incomes The following section provides a Community Water System Survey (USEPA 2003i). description of how EPA derived the (CWSS) were used to derive the annual national-level affordability criteria water bills and annual water usage 1 EPA is currently receiving input from a National pertinent to this rule. First, EPA values for each of the three small system Drinking Water Advisory Council (NDWAC). This calculated an ‘‘affordability threshold’’ size categories. The data on zip codes process is expected to conclude in the fall of 2003 (i.e., the total annual household water were used with the 1990 Census data on with a report that will be sent by the NDWAC. EPA has also received a report from the Science bill that would be considered median household income to develop Advisory Board’s Environmental Economics affordable). The total annual water bill the median household income values Advisory Committee on its review of the national- includes costs associated with water for each of the three small-system size level affordability criteria (USEPA 2002c). One of treatment, water distribution, and categories. The median household- the charges given to both groups was to evaluate the operation of the water system. In income values used for the affordable process used by EPA to adjust the baseline water bills to account for costs attributable to regulations developing the threshold of 2.5% technology determinations are not based promulgated after 1996. Because the Stage 2 DBPR median household income, EPA on the national median income. The affordability analysis is being conducted before EPA considered the percentage of median value for each size category is a national can complete a comprehensive reassessment of household income spent by an average median income for communities served affordability, today’s estimate for the increase to the household on comparable goods and by small water systems within that average water bill to account for regulations after 1996 reflects existing Agency affordability criteria services and on cost comparisons with range. Table V–13 presents the baseline and methodology. This estimate may change in the other risk reduction activities for values for each of the three small-system future. drinking water such as households size categories. Annual water bills are

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00062 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49609

TABLE V–13.—BASELINE VALUES FOR SMALL SYSTEMS CATEGORIES AND AVAILABLE EXPENDITURE MARGIN FOR AFFORDABLE TECHNOLOGY DETERMINATIONS

Annual HH consumption Median 2.5% me- Current annual Available ex- System size category (pop. served) (1000 gallons/ HH in- dian HH water bills penditure mar- yr) come ($) income(s) ($/yr) gin ($/hh/year)

25–500 ...... 72 35,148 878 290 588 501–3,300 ...... 74 30,893 772 230 542 3,301–10,000 ...... 77 31,559 789 219 570

For each size category, the threshold available expenditure margin for the from the 25–100 and 101–500 value was determined by multiplying three size categories is presented in categories. A similar approach was used the median household income by 2.5 Table V–13. to derive design and average flows from percent. The annual household water The size categories specified in the 501–1000 and 1001–3300 categories bills were subtracted from this value to SDWA for affordable technology for the 501–3300 category. The Variance obtain the available expenditure margin. determinations are different from the Technology Findings document (USEPA Projected treatment costs were size categories typically used by EPA in 1998h) describes this procedure in more compared against the available the Economic Analysis. A weighted detail. Table V–14a lists the design and expenditure margin to determine if average procedure was used to derive average flows for the three size there were affordable compliance design and average flows for the 25–500 categories. technologies for each size category. The category using design and average flows

TABLE V–14A.—DESIGN AND AVERAGE DAILY FLOWS USED FOR AFFORDABLE TECHNOLOGY DETERMINATIONS

Design flow Average flow System size category (population served) (mgd) (mgd)

25–500 ...... 0.058 0.015 501–3,300 ...... 0.50 0.17 3,301–10,000 ...... 1.8 0.70

Capital and operating and forecast for the Stage 2 DBPR and that expenditure margins as a result of maintenance costs were derived for each costs slightly more than the available adding advanced technology for the treatment technology used in the expenditure margin is GAC20 (240 day Stage 2 DBPR. This prediction is compliance forecast for small systems carbon replacement) with advanced probably overestimated because small using the flows listed previously and disinfectants for systems serving 500 systems have other compliance the cost equations in the Technology people or fewer. As shown in the alternatives available to them besides and Cost Document (USEPA 2003k). Economic Analysis (USEPA 2003i), 13 adding treatment. For example, some of Capital costs were amortized using the systems (less than 1 percent) among these systems currently may be operated 7 percent interest rate preferred by systems serving fewer than 500 people on a part-time basis; therefore, they may Office of Management and Budget are predicted to use GAC20 with be able to modify the current (OMB) for benefit-cost analyses of advanced disinfection to comply with operational schedule or use excessive government programs and regulations the proposed Stage 2 DBPR. However, capacity to avoid installing a costly rather than a 3 percent interest rate. alternate affordable technologies are technology to comply with the Stage 2 The annual system treatment cost in available. Thus, EPA believes that DBPR. The system also may identify dollars per year was converted into a compliance by these systems will be another water source that has lower rate increase using the average daily affordable. In some cases, the TTHM and HAA5 precursor levels. flow. The annual water consumption compliance data for these systems under Systems that can identify such an values listed in Table V–13 were the Stage 2 DBPR is the same as under alternate water source may not have to multiplied by 1.15 to account for water the Stage 1 DBPR (because many treat that new source water as intensely lost due to leaks. Since the water lost to systems serving fewer than 500 people leaks is not billed, the water bills for the will have the same single sampling site as their current source, resulting in actual water used were adjusted to cover under both rules); these systems will lower treatment costs. Systems may this lost water by increasing the have already installed the necessary elect to connect to a neighboring water household consumption. The rate compliance technology to comply with system. While connecting to another increase in dollars per thousand gallons the Stage 1 DBPR. It is also possible that system may not be feasible for some used was multiplied by the adjusted less costly technologies such as those remote systems, EPA estimates that annual consumption to determine the for which percentage use caps were set more than 22 percent of all small water annual cost increase for the household in the decision tree may actually be systems are located within metropolitan for each treatment technology. used to achieve compliance (e.g., regions (USEPA 2000c) where distances With very few exceptions, the chloramines, UV). between neighboring systems will not household costs for all predicted As shown in Table V–14b, the cost present a prohibitive barrier. More compliance technologies in Table V–12 model (USEPA 2003i) predicts that discussion of household cost increases are below the available expenditure households served by very small is presented in a later section (Section margin. The only technology that was systems will experience household cost VII) and the Economic Analysis (USEPA predicted to be used in the compliance increases greater than the available 2003i).

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00063 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49610 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

EPA is currently reviewing its and modify, as required, their compliance with Stage 2A MCLs based national-level affordability criteria, and qualification standards to take into on LRAAs (0.120 mg/L TTHM and 0.100 has solicited recommendations from account new technologies (e.g., mg/HAA5), as well as continue to report both the NDWAC and the SAB as part ultraviolet (UV) disinfection) and new compliance with 0.080 mg/L TTHM and of this review. If the national-level compliance requirements (including 0.060 mg/L HAA5 as RAAs. Systems affordability criteria are revised prior to simultaneous compliance and must report compliance with the Stage promulgation of the final Stage 2 DBPR, consecutive system requirements). 2B TTHM and HAA5 MCLs (0.080 mg/ EPA may reevaluate the affordability of L TTHM and 0.060 mg/L HAA5 as N. System Reporting and Recordkeeping the identified small system compliance LRAAs) according to the compliance Requirements technologies based on the revised schedules outlined in section V.J. of criteria and may revise its determination 1. Confirmation of Applicable Existing today’s proposal. Reporting for DBP of whether to list any variance Requirements monitoring, as described previously, technologies as a result. EPA requests Today’s proposed Stage 2 DBPR, will remain generally consistent with comment on the application of its consistent with the current system current public water system reporting affordability criteria in this rulemaking reporting regulations under 40 CFR requirements (§ 141.31 and § 141.134); and on its determination that there are 141.131, requires public water systems systems will be required to calculate affordable small system compliance to report monitoring data to States and report each LRAA (instead of the technologies for all three statutory small within ten days after the end of the system’s RAA) and each individual system size categories. compliance period. In addition, systems monitoring result (as required under the are required to submit the data required Stage 1 DBPR). Systems will also be M. Requirements for Systems To Use required to consult with the State about Qualified Operators in § 141.134. These data are required to be submitted quarterly for any each peak excursion event no later than EPA believes that systems that must monitoring conducted quarterly or more the next sanitary survey for the system, make treatment changes to comply with frequently, and within ten days of the as discussed in section V.E. requirements to reduce microbiological end of the monitoring period for less 3. Request for Comment risks and risks from disinfectants and frequent monitoring. disinfection byproducts should be EPA requests comment on all system operated by personnel who are qualified 2. Summary of Additional Reporting reporting and recordkeeping to recognize and respond to problems. Requirements requirements. Subpart H systems were required to be EPA proposes that two years after rule O. Analytical Method Requirements operated by qualified operators under promulgation, systems serving 10,000 or the SWTR (40 CFR 141.70). The Stage 1 more people (plus consecutive systems 1. What Is EPA Proposing Today? DBPR added requirements for all that are part of a combined distribution The Stage 2 DBPR proposed today disinfected systems to be operated by system with a system serving at least does not add any new disinfectants or qualified personnel who meet the 10,000) be required to report the results disinfection byproducts to the list of requirements specified by the State, of their IDSE to their State, unless the contaminants currently covered by which may differ based on system size State has waived this requirement for MRDLs or MCLs. However, additional and type. The rule also required that systems serving fewer than 500. Systems methods have become available since States maintain a register of qualified are also required to report to the State the analytical methods in the Stage 1 operators (40 CFR 141.130(c)). While the recommended long-term (Stage 2B) DBPR were promulgated (USEPA proposed Stage 2 DBPR requirements do compliance monitoring sites as part of 1998c). EPA is proposing to add to 40 not supercede or modify the the IDSE report. While the IDSE options CFR 141.131 one method for chlorine requirement that disinfected systems be discussed in section V.J. would delay dioxide and chlorite, one method for operated by qualified personnel, the the timing of this requirement, EPA HAA5 which can also be used to Stage 2 DBPR re-emphasizes the believes that the burden would not analyze for the regulated contaminant important role that qualified operators change. dalapon, three methods for bromate, play in delivering safe drinking water to Beginning three years after rule chlorite, and bromide, one method for the public. States should also review promulgation, systems must report bromate only, and one method for total

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00064 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.009 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49611

organic carbon (TOC) and specific have been included in publications that citation for one method for bromate, ultraviolet absorbance (SUVA). One of were issued after December 1998. EPA chlorite, and bromide. the methods that is currently approved is proposing to approve the use of the EPA is also proposing to standardize for bromate, chlorite, and bromide can recently published versions of three the HAA5 sample holding times and the be used to determine chloride, fluoride, methods for determining free, bromate sample preservation procedure nitrate, nitrite, orthophosphate, and combined, and total chlorine residuals, and holding time. EPA is clarifying sulfate, so EPA is proposing to add it as two methods for total chlorine only, one which methods are approved for an approved method for those method for free chlorine only, one magnesium hardness determinations in contaminants in 40 CFR 141.23 and 40 method for chlorite and chlorine 40 CFR 141.131 and 40 CFR 141.135. CFR 143.4. EPA is also proposing to add dioxide, one method for chlorine Analytical methods that are proposed the HAA5 method that includes dalapon dioxide only, one method for HAA5, for approval or for which changes are to 40 CFR 141.24 for dalapon three methods for TOC and dissolved proposed in today’s rule are compliance monitoring. organic carbon (DOC), and one method summarized in Table V–15 and are Several of the methods that were for ultraviolet absorption at 254nm described in more detail later in this promulgated with the Stage 1 DBPR (UV 254). EPA is proposing to update the section.

TABLE V–15.—ANALYTICAL METHODS ADDRESSED IN TODAY’S PROPOSED RULE

Analyte EPA method Standard method 1 Other

§ 141.23 Fluoride ...... 300.1 Nitrate ...... 300.1 Nitrite ...... 300.1 Orthophosphate ...... 300.1 § 141.24 Dalapon ...... 552.3 § 141.131—Disinfectants Chlorine (free, combined, total) ...... 4500–Cl D 4500–Cl F 4500–Cl G (total) 4500–Cl E 4500–Cl I (free) 4500–Cl H Chlorine Dioxide ...... 327.0 4500–ClO 2 D 4500–ClO 2 E § 141.131—Disinfection Byproducts HAA5 ...... 552.1 2 6251 B 2 552.3 Bromate ...... 300.1 3 ASTM D 6581–00 317.0 Revision 2 321,8 4 326.0 Chlorite (monthly or daily) ...... 300.1 3 ASTM D 6581–00 317.0 Revision 2 326.0 (daily) ...... 327.0 4500–ClO 2 E § 141.131—Other parameters Bromide ...... 300.1 3 ASTM D 6581–00 317.0 Revision 2 326.0 TOC/DOC ...... 415.3 5310 B 5310 C 5310 D UV 254 ...... 415.3 5910 B SUVA ...... 415.3 § 143.4 Chloride ...... 300.1 Sulfate ...... 300.1 1 EPA is proposing to cite both the 20th edition and the 2003 On-Line Version of Standard Methods for the Examination of Water and Waste Water in addition to the currently cited 19th editions for all methods listed in this column with the exception of 4500–ClO2 D for chlorine dioxide which is not available in the 2003 On-Line Version. 2 EPA is proposing to change the sample holding time to 14 days. 3 EPA is proposing to update the citation. 4 EPA is proposing that samples be preserved with 50 mg ethylenediamine/L and analyzed within 28 days.

2. How Was This Proposal Developed? data of comparable or better quality than EPA reviewed the new publications of EPA evaluated the performance of the the methods that are currently methods from consensus organizations new methods for their applicability to approved. Methods currently approved such as Standard Methods and compliance monitoring. The primary for DBPs were also examined to American Society for Testing and purpose of this evaluation was to determine applicability to other Materials (ASTM). As a result, EPA determine if the new methods provide regulated contaminants. identified one new method from ASTM

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00065 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49612 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

which is suitable for compliance was calculated and multiplied by the EPA Method 552.3 for HAA5, other monitoring. EPA also determined that student’s t-value at 99% confidence and haloacetic acids, and the regulated the newer editions of Standard Methods n–1 degrees of freedom (3.143 for 7 contaminant dalapon allows two did not change the individual methods replicates) to determine the detection extraction options. The first option approved under the Stage 1 DBPR. limit. The accuracy reported in the involves an acidic extraction with method for laboratory fortified blanks at methyl tertiary butyl ether (MTBE) 3. Which New Methods Are Proposed which is the same solvent used in the for Approval? concentrations of 0.2–1.0 mg/L is 103– 118 % for chlorite and 102–124 % for currently approved HAA5 methods. The a. EPA Method 327.0 for chlorine chlorine dioxide with relative standard analytes (HAA5, other HAAs, and dioxide and chlorite. EPA is proposing deviations between 2.9 and 16 %. dalapon) are then converted to their to add a new method for the Replicate analyses of drinking water methyl esters by the addition of acidic measurement of chlorine dioxide samples from surface and ground water methanol to the extract followed by residuals and daily chlorite sources fortified at concentrations of heating. The amount of acidic methanol concentrations. EPA Method 327.0 approximately 1 and 2 mg/L chlorite that is added to the extract is increased (USEPA 2003q) is an enzymatic/ and chlorine dioxide showed average in the new method resulting in spectrophotometric method in which a recoveries of 91–110 % with relative increased methylation efficiency for total chlorine dioxide plus chlorite standard deviations of 1–9 %. some of the analytes. The increased concentration is determined in an methylation efficiency is significant for EPA is proposing to approve EPA unsparged sample and the chlorite the additional HAAs and thus provides Method 327.0 as an additional method concentration is determined in a greater sensitivity, precision, and for monitoring chlorine dioxide and for sparged sample. The chlorine dioxide accuracy for them when compared to making the daily determination of concentration is then calculated by EPA Method 552.2. The acidic extract is chlorite at the entry point to the subtracting the chlorite concentration neutralized with a saturated solution of distribution system. It will provide from the total. sodium bicarbonate and the target water systems with additional flexibility The pH of the samples (sparged and analytes are identified and measured by unsparged) and blank are adjusted to 6.0 in monitoring the application of gas chromatography using electron with a citric acid/glycine buffer. The chlorine dioxide. EPA believes that capture detection (GC/ECD). chromophore Lissamine Green B (LGB) many water plant operators will prefer The second option in the new EPA and the enzyme horseradish peroxidase the new method over the currently Method 552.3 involves an acidic are added. The enzyme reacts with the approved methods due to its ease of use. extraction with tertiary amyl methyl chlorite in the sample to form chlorine b. EPA Method 552.3 for HAA5 and ether (TAME). The HAAs are then dioxide which then reacts with the dalapon. EPA is proposing to add a new converted to their methyl esters by the chromophore LGB to reduce the method (EPA Method 552.3) for HAA5 addition of acidic methanol to the absorbance at 633nm of the sample. The that provides comparable sensitivity, extract followed by heating. The use of absorbance of the samples and blank are accuracy, and precision to the TAME instead of MTBE as the determined spectrophotometrically. The previously approved methods. EPA extraction solvent allows the use of a difference in absorbance between the Method 552.3 (USEPA 2003p) has the higher temperature during the samples and the blank is proportional to added benefit of allowing laboratories to methylation process. This increases the the chlorite and total chlorine dioxide/ more easily measure four additional methylation efficiency and thus chlorite concentrations in the samples. haloacetic acids (bromochloroacetic provides significant increases in EPA Method 327.0 offers advantages acid, bromodichloroacetic acid, sensitivity, precision, and accuracy for over the currently approved methods in chlorodibromoacetic acid, and the additional HAAs. The acidic extract that it is not subject to positive tribromoacetic acid) at the same time is neutralized with a saturated solution interferences from other chlorine the HAA5 compounds are being of sodium bicarbonate and the target species and it is easier to use. measured, without compromising the analytes are identified and measured by The single laboratory detection limits quality of data for the HAA5 gas chromatography using electron presented in the method are 0.08–0.11 compounds. Of the currently approved capture detection (GC/ECD). mg/L for chlorite and 0.04–0.16 mg/L methods for HAA5, only EPA Method The performance of EPA Method for chlorine dioxide. The detection 552.2 (USEPA 1995) provides method 552.3 is comparable to the currently limits are based on the analyses of sets performance data for all of these approved methods for determining the of seven replicates of reagent water that additional compounds, but the reaction HAA5 analytes. A comparison of the were fortified with low concentrations conditions must be carefully controlled. performance of EPA Method 552.3 to of chlorite with and without the EPA believes that analyses for these the currently approved HAA5 methods presence of chlorine dioxide and low additional HAAs can be accomplished is shown in Table V–16. The data are concentrations of chlorine dioxide with more easily without compromising the taken from the individual methods, so and without the presence of chlorite. quality of data for the HAA5 the precision, accuracy, and detection The standard deviation of the mean compounds by using EPA Method data were not generated using the same concentration for each set of samples 552.3. samples or by the same laboratory.

TABLE V–16.—PERFORMANCE OF HALOACETIC ACID METHODS

QC Parameter MCAA DCAA TCAA MBAA DBAA

Precision (Max %RSD in fortified drinking water samples) 1 EPA 552.1 ...... 15 14 28 11 7 EPA 552.2...... 13 6 15 6 5 EPA 552.3 (MTBE option) ...... 6 4 1 4 5 EPA 552.3 (TAME option) ...... 10 4 2 4 5 SM 6251 B ...... 8 7 6 8 7

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00066 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49613

TABLE V–16.—PERFORMANCE OF HALOACETIC ACID METHODS—Continued

QC Parameter MCAA DCAA TCAA MBAA DBAA

Accuracy (Range of % Recoveries in fortified drinking water samples) 2 EPA 552.1 ...... 76–100 75–126 56–106 86–97 94–103 EPA 552.2 ...... 84–97 96–105 62–82 86–100 72–112 EPA 552.3 (MTBE option) ...... 98–126 96–103 89–100 99–113 101–111 EPA 552.3 (TAME option) ...... 97–131 97–107 89–103 99 101–105 SM 6251 B ...... 99–103 96–103 100–103 97–101 102 Detection Limit (µg/L) 3 EPA 552.1...... 0.21 0.45 0.07 0.24 0.09 EPA 552.2...... 0.27 0.24 0.08 0.20 0.07 EPA 552.3 (MTBE option) ...... 0.17 0.02 0.02 0.03 0.01 EPA 552.3 (TAME option) ...... 0.20 0.08 0.02 0.13 0.02 SM 6251 B ...... 0.08 0.05 0.05 0.09 0.06 1 The highest relative standard deviation (%RSD) for replicate analyses of fortified drinking water samples as shown in each method. 2 The range of recoveries reported for replicate analyses of fortified drinking water samples as shown in each method. 3 The detection limit as determined by analyzing seven or more replicates of reagent water that is fortified with low concentrations of the haloacetic acids. The standard deviation of the mean concentration for each analyte is calculated and multiplied by the student’s t-value at 99% confidence and n-1 degrees of freedom (3.143 for 7 replicates).

Two of the currently approved HAA5 a synthetic organic chemical. The new comparable or better method methods (EPA Methods 552.1 (USEPA HAA5 method can also be used to performance than the approved 1992) and 552.2 (USEPA 1995)) are also determine dalapon in drinking water. methods. approved for analyses of water samples As shown in Table V–17, both solvent for the regulated contaminant dalapon, options in EPA Method 552.3 provide

TABLE V–17.—PERFORMANCE OF DALAPON METHODS

EPA 552.3 Dalapon performance characteristic EPA 552.1 EPA 552.2 MTBE TAME

Precision1 (% RSD) ...... 14 11 2 4 Accuracy2 (% Recovery) ...... 88–102 86–100 98–112 87–103 Detection Limit3 (µg/L) ...... 0.32 0.12 0.02 0.14 1 The highest relative standard deviation (%RSD) for replicate analyses of fortified drinking water samples as shown in each method. 2 The range of recoveries reported for replicate analyses of fortified drinking water samples as shown in each method. 3 The detection limit as determined by analyzing seven or more replicates of reagent water that is fortified with low concentrations of dalapon. The standard deviation of the mean dalapon concentration is calculated and multiplied by the student’s t-value at 99% confidence and n-1 de- grees of freedom (3.143 for 7 replicates).

EPA is proposing to approve EPA instrument time. EPA is not proposing ASTM Method D 6581–00 is an Method 552.3 for dalapon to withdraw the other dalapon methods, acceptable consensus standard and it is (§ 141.24(e)(1)) in addition to HAA5 because that would reduce flexibility for published in the 2001, 2002, and 2003 even though dalapon is not a the laboratories and place an editions of The ASTM Annual Book of contaminant that is addressed in this unnecessary burden on laboratories that Standards. EPA is proposing to approve proposed rule. EPA believes that do not need to use EPA Method 552.3. ASTM Method D 6581–00 in order to extending approval to all the regulated c. ASTM D 6581–00 for bromate, provide additional flexibility to contaminants covered by the method chlorite, and bromide. ASTM Method D laboratories. Any edition containing the provides more flexibility to laboratories. 6581–00 (ASTM 2002) for the cited version may be used. It allows the laboratories the option of determination of bromate, chlorite, and d. EPA Method 317.0 revision 2 for reducing the number of methods that bromide was adopted by ASTM in 2000. bromate, chlorite, and bromide. EPA they need to keep in operation for their This method uses the same procedures Method 317.0 Revision 2 (USEPA clients, because the new method can be as EPA Method 300.1 (USEPA 2000l) 2001d) is an extension of the currently used for dalapon and HAA5 compliance (the method promulgated in the Stage 1 approved EPA Method 300.1 for monitoring samples and for determining DBPR) and thus is considered bromate, chlorite, and bromide. It uses the additional HAAs for non-regulatory equivalent to the approved method the EPA Method 300.1 technology, but purposes. EPA recognizes that (Hautman et al. 2001). The ASTM it adds a postcolumn reactor that laboratories will probably not be method includes interlaboratory study provides a more sensitive and specific determining dalapon concentrations for data that were not available when EPA analysis for bromate than is obtained compliance purposes in the same Method 300.1 was published. The study using EPA Method 300.1. As with EPA samples as used for HAA5 compliance data demonstrate good precision and Method 300.1, the anions are separated monitoring. However, EPA believes low bias for all analytes. by ion chromatography and detected allowing the same method to be used Under section 12(d) of the National using a conductivity detector. (Bromate, even if the samples are not the same is Technology Transfer and Advancement chlorite, and bromide concentrations more cost effective for laboratories, Act, the Agency is directed to consider determined by the conductivity detector because switching between methods whether to use voluntary consensus are equivalent to those measured using results in increased analyst and standards in its regulatory activities. EPA Method 300.1.) After the sample

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00067 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49614 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

passes through the conductivity EPA believes EPA Method 317.0 laboratories using this method will detector, it enters a postcolumn reactor Revision 2.0 should be approved as an increase as utilities become aware of the chamber in which o-dianisidine additional method for bromate, chlorite, method’s sensitivity and begin to dihydrochloride (ODA) is added to the and bromide compliance monitoring. request it be used for their samples. sample. This compound forms a EPA anticipates that water systems will f. EPA Method 321.8 for bromate. EPA chromophore with the bromate that is prefer to have their bromate samples is proposing to add EPA Method 321.8 present in the sample and the analyzed by this new method, because (USEPA 2000d) specifically for bromate chromophore concentration is it provides higher quality data than the compliance monitoring. It involves an determined using a ultraviolet/visible currently approved method when ion chromatograph coupled to an (UV/Vis) absorbance detector. There are bromate concentrations are below the inductively coupled plasma mass several advantages of this method: MCL of 0.010 mg/L (10 µg/L). Only a spectrometer (IC/ICP–MS). The ion (1) Very few ions react with ODA to few laboratories are currently chromatograph separates bromate from form compounds that are detected by performing analyses using the other ions present in the sample and the UV/Vis detector. This makes the postcolumn reactor technology included then bromate is detected and method less susceptible to interferences in the method, but the number is quantitated by the ICP–MS. Mass 79 is for bromate. increasing as more laboratories become used for quantitation while mass 81 (2) The UV/Vis detector is very aware of the advantages. provides isotope ratio information that sensitive to the chromophore, so lower e. EPA Method 326.0 for bromate, can be used to screen for potential concentrations of bromate can be chlorite, and bromide. EPA Method polyatomic interferences. The advantage detected and quantitated. (Bromate 326.0 (USEPA 2002a) is based on the of this method is that it is very specific concentrations can be reliably procedure reported by Salhi and von and sensitive to bromate. The single quantitated as low as 1 µg/L using this Gunten (1999) and uses an approach laboratory detection limit presented in detector versus 5 µg/L for EPA Method that is similar to EPA Method 317.0 the method is 0.3 µg/L. The average 300.1.) Revision 2.0. The method involves the accuracy reported in the method for (3) Since the front part of the analysis separation of the anions (bromate, laboratory fortified blanks is 99.8% is the same as EPA Method 300.1, chlorite, and bromide) following the recovery with a three sigma control bromate, chlorite, and bromide can be scheme outlined in EPA Methods 300.1 limit of 10.2%. Average accuracy and determined in the same analysis. and 317.0 Revision 2.0. (Bromate, precision in fortified drinking water The first version of this method, EPA chlorite, and bromide data from the samples are reported as 97.8% recovery Method 317.0 has been evaluated in a conductivity detector are equivalent to and 2.9% relative standard deviation, multiple laboratory study (Wagner et al. data generated using EPA Method respectively. 2001; Hautman et al. 2001). The results 300.1.) The eluent stream exiting the During the Information Collection from the study indicate high precision conductivity detector is mixed with a Rule, thirty-three samples were and very low bias in data generated postcolumn reagent consisting of an analyzed by this method in addition to using this method. The interlaboratory acidic solution of potassium iodide with the selective anion concentration (SAC) precision for bromate, chlorite, and a catalytic concentration of method used by EPA for the low-level bromide using the conductivity detector molybdenum (VI). Bromate reacts with bromate analyses. EPA Method 321.8 and bromate using the UV/Vis detector the iodide to form triiodide which is provided comparable data to that are 12%, 4.2%, 6.9%, and 9.6% relative measured by its UV absorption at 352 generated by the SAC method (Fair standard deviation (RSD), respectively. nm. 2002). The interlaboratory bias for bromate, EPA Method 326.0 has similar EPA Method 321.8 has undergone chlorite, and bromide using the accuracy, precision, and sensitivity for second laboratory validation (Day et al. conductivity detector and bromate using bromate compared to EPA Method 317.0 2001) and the results indicate the the UV/Vis detector are 0.35%, Revision 2.0. Thirty drinking water method can be successfully performed ¥0.98%, ¥0.87%, and 4.8%, samples fortified with 1–7 µg bromate/ in non-EPA laboratories. The calculated respectively. The average detection L were analyzed using both methods. detection limit determined by the levels for bromate, chlorite, and Accuracy, expressed as % recovery, second laboratory is 0.4 µg/L. The bromide using the conductivity detector ranged from 78.0 to 129% for both average accuracy achieved for laboratory and bromate using the UV/Vis detector methods and precision, expressed as % fortified blanks at 5 µg/L is 93% are 2.2, 1.6, 2.8, and 0.24 µg/L, RSD ranged from 3.7 to 13.5% (Wagner recovery with a relative standard respectively. et al. 2002). The detection limit of EPA deviation of 8.9%. Average accuracy Subsequent to the interlaboratory Method 326.0 is 0.17 µg/L as and precision in fortified drinking water study of EPA Method 317.0, a problem determined by analyzing seven or more samples are reported as 101% recovery with ODA was discovered. The purity of replicates of reagent water that is and 9% relative standard deviation, the reagent can vary from lot to lot and fortified with low concentrations of respectively. this affects the performance of the bromate. The standard deviation of the The IC/ICP–MS instrumentation used method. EPA has evaluated the method mean bromate concentration is in EPA Method 321.8 is a new performance using ODA obtained from calculated and multiplied by the technology in the drinking water several commercial sources and from student’s t-value at 99% confidence and laboratory community. Even though the different lots from the same supplier. n-1 degrees of freedom (3.143 for 7 technology is not yet widely used, EPA Based on that new information, EPA replicates). believes that approving this new revised Method 317.0 to document how EPA is proposing EPA Method 326.0 method will provide laboratories with to detect and correct problems that can as an additional method for bromate, the flexibility to adopt the new result from a contaminated ODA supply. chlorite, and bromide compliance technology if they have additional The revised method is designated EPA monitoring. It provides higher quality applications for it. The instrumentation Method 317.0 Revision 2.0 and this is bromate data than the currently is especially promising in the area of the version that is being proposed today. approved EPA Method 300.1 when trace metal speciation. Laboratories that The performance of the revised method bromate concentrations are below 10 µg/ are performing that type of analysis is identical to the original version. L. EPA anticipates the number of would find it very useful to also be able

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00068 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49615

to perform bromate compliance quality control (QC) requirements were mg/L and the relative standard monitoring analyses by EPA Method included for these measurements, deviations across the analyses ranged 321.8. EPA believes that advances in because the methods did not contain the from 35% RSD (for the lowest analytical technology should be necessary criteria. The rule included concentration sample) to ≤13% RSD for encouraged when they provide instructions for calculating SUVA based the remainder of the samples. additional options for obtaining on UV254 and DOC analyses. The new EPA Method 415.3 includes a accurate and precise data for EPA Method 415.3 includes the procedure to ensure that inorganic compliance monitoring. Approval of additional QC necessary to achieve carbon does not interfere with the this method would not require reliable determinations for TOC, DOC, organic carbon analyses. Since this is laboratories to adopt the new and UV254. It describes a procedure for critical to obtaining accurate organic technology; it strictly offers the choice removing inorganic carbon from the carbon determinations, EPA is for laboratories that would like to use sample prior to the organic carbon proposing to add a requirement at the latest technology. analysis. The method uses the same §§ 141.131(d)(3) and (4)(i) to remove EPA is proposing to add sample technologies as already approved. The inorganic carbon prior to performing collection and holding time requirement advantage of this new method is that it TOC or DOC analyses. Laboratories will to EPA Method 321.8. The current documents the precision and accuracy have the option of using the procedure method does not address the potential that can be expected when proper QC described in EPA Method 415.3 or for changes in bromate concentrations procedures are implemented and it verifying that the process used by their after the sample is collected as a result places all the necessary information for TOC instrument adequately removes the of reactions with hypobromous acid/ SUVA in one place. inorganic carbon prior to the organic hypobromite ion. Hypobromous acid/ EPA Method 415.3 provides carbon measurement. Determination of hypobromite ion are intermediates sensitivity, precision and accuracy data organic carbon by subtracting the formed as byproducts of the reaction of for TOC and DOC measured using five inorganic carbon from the total carbon either ozone or hypochlorous acid/ different technologies: is not acceptable for compliance hypochlorite ion with bromide ion. If (1) Catalyzed 680°C combustion purposes, because the percentage of not removed from the sample matrix, oxidation of organic carbon to carbon inorganic carbon is usually large in further reactions may form bromate ion. relation to the organic carbon of the dioxide (CO2) followed by The reactions can be prevented by nondispersive infrared detection sample and the subtraction process adding 50 mg of ethylenediamine (NDIR). introduces a large potential for error. (EDA)/L of sample. This is the (2) High temperature (700 to 1100°C) The manufacturer of one of the preservation technique specified in the combustion oxidation followed by instruments that was used during the development of EPA Method 415.3 other methods both approved and NDIR. recommends that hydrochloric acid be proposed for bromate compliance (3) Elevated temperature (95–100°C) used to acidify TOC and DOC samples analyses. The fortified drinking water catalyzed persulfate digestion of organic prior to analysis. EPA confirmed that samples analyzed in the second carbon to CO followed by NDIR. 2 use of this acid is critical for proper laboratory validation study of EPA (4) UV catalyzed persulfate digestion operation of the instrument. However, Method 321.8 (Day et al. 2001) and the followed by NDIR. use of hydrochloric acid is in conflict Information Collection Rule samples (5) UV catalyzed persulfate digestion with the current regulation at that were analyzed using the SAC followed by membrane permeation into method and EPA Method 321.8 were §§ 141.131(d)(3) and (4)(i) which specify a conductivity detector. preserved with EDA, thus phosphoric or sulfuric acid. The type of demonstrating that EDA can be used in These technologies are included in the acid used to preserve samples and to samples analyzed by IC/ICP–MS. EPA currently approved Standard Methods treat the samples to remove inorganic believes that adding this sample 5310 B and 5310 C (APHA, 1996). The carbon prior to the organic carbon preservation requirement to EPA new method indicates these analysis should be based on the Method 321.8 will help ensure sample technologies can provide detection analytical method or the instrument integrity. It will also simplify the limits between 0.02 mg/L and 0.12 mg/ manufacturer’s specification. Therefore, sampling protocols that water systems L. Accuracy and precision data from EPA is proposing to remove the must follow, because all sampling for analyses of fortified reagent water and specification of acid type from bromate, regardless of the method natural waters indicate the technologies §§ 141.131(d)(3) and (4)(i). employed to analyze the sample, will can produce acceptable data for EPA Method 415.3 specifies that TOC require the same sample preservation determining compliance with the samples be acid preserved at the time of technique. treatment technique for control of collection in order to prevent microbial EPA Method 321.8 does not include disinfection byproduct precursors degradation of the organic carbon. This information concerning how long a specified in § 141.135. Seven natural is consistent with the sampling sample may be stored prior to analysis. waters were fortified with organic instructions in the currently approved EPA is proposing to specify a maximum carbon from potassium hydrogen methods (Standard Methods 5310 B, of 28 days for the sample holding time. phthalate and analyzed by each of the 5310 C, and 5310 D). EPA proposes to This would make the method consistent five technologies. The average amend § 141.131(d)(3) by removing the with the other bromate methods recoveries ranged from 97% to 103% for phrase ‘‘not to exceed 24 hours’’ in the proposed today and the method that is TOC and 98% to 106% for DOC. description of when samples must be currently approved. The method presents data from the preserved, so that the rule is consistent g. EPA 415.3 for TOC and SUVA analyses of seven different waters and with the method specifications. (DOC and UV254). Today’s rule proposes demonstrates that comparable analytical Analyses for both DOC and UV254 are to add EPA Method 415.3 (USEPA results are obtained regardless of the required for a SUVA determination. The 2003r) as an approved method for TOC technology used as long as all inorganic DOC measurement is identical to the and SUVA. The Stage 1 DBPR included carbon is removed from the sample TOC measurement after the sample is three Standard Methods for TOC and prior to the analysis. The samples filtered through a 0.45 µm pore size one method for UV254. Additional ranged in concentration from 0.4 to 3.6 filter. The filtration step must be

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00069 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49616 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

performed using a prewashed filter in measurement, the UV254 analysis is 2000l) can also be used to determine order to eliminate positive interferences performed on a sample that has been chloride, fluoride, nitrate, nitrite, from material that can leach from filtered through a prewashed 0.45 µm orthophosphate, and sulfate in drinking improperly cleaned filters. EPA Method pore size filter. In addition to verifying water. A comparison of the performance 415.3 contains a description of how to that the filter blank is low enough, the of EPA Method 300.1 to the currently properly rinse the filters and how to method also includes a approved EPA Method 300.0 (USEPA verify that the filter blank is acceptable. spectrophotometer check procedure to 1993) is shown in Table V–18 and The method demonstrates that it is ensure that the spectrophotometer is demonstrates that EPA Method 300.1 feasible to have a filter blank with a operating properly. provides comparable or better precision, DOC concentration <0.2 mg/L. The 4. What Additional Regulated accuracy, and sensitivity for these method also provides performance data Contaminants Can Be Monitored by contaminants based on the single for DOC. Extending Approval of EPA Method laboratory data presented in each

The UV254 analysis that is part of the 300.1? method. SUVA determination is also described In addition to bromate, chlorite, and in EPA Method 415.3. As with the DOC bromide, EPA Method 300.1 (USEPA

TABLE V–18.—COMPARISON OF EPA METHODS 300.0 AND 300.1

Phosphate- QC parameter Chloride Fluoride Nitrate Nitrite P Sulfate

Precision (Max % RSD in fortified water samples) 1

EPA 300.0...... 5.7 18 4.8 3.6 3.5 7.1 EPA 300.1...... 0.22 0.85 0.41 0.77 4.7 0.39

Accuracy (Range of % Recoveries in fortified water samples) 2

EPA 300.0 ...... 86–114 73–95 93–104 92–121 95–99 95–112 EPA 300.1 ...... 93–98 80–89 88–96 72–87 61–92 89

Detection Limit (mg/L) 3

EPA 300.0 ...... 0.02 0.01 0.002 0.004 0.003 0.02 EPA 300.1...... 0.004 0.009 0.008 0.001 0.019 0.019 1 The highest relative standard deviation (%RSD) reported in the method for replicate analyses of fortified water samples in a single laboratory. 2 The range of recoveries reported for replicate analyses of fortified water samples in a single laboratory as shown in the method. 3 The detection limit as determined by analyzing seven or more replicates of reagent water that is fortified with low concentrations of the anions. The standard deviation of the mean concentration for each analyte is calculated and multiplied by the student’s t-value at 99% con- fidence and n-1 degrees of freedom (3.143 for 7 replicates).

EPA is proposing to extend approval 5. Which Methods in the 20th Edition The On–Line Version of Standard of EPA Method 300.1 for fluoride, and 2003 On-Line Version of Standard Methods is an effort to provide the nitrate, nitrite, and orthophosphate Methods Are Proposed for Approval? consensus methods to the public prior (§ 141.23(k)(1)) and for chloride and to the release of the next full sulfate (§ 143.4(b)) even though these The Stage 1 DBPR approved eight publication. Standard Methods is contaminants are not addressed in methods (4500–Cl D, 4500–Cl F, 4500– making sections of the next version today’s proposed rule. As discussed Cl G, 4500–Cl E, 4500–Cl I, 4500–Cl H, available for purchase in both electronic 4500–ClO D, and 4500–ClO E) for before for dalapon, EPA believes that 2 2 or printed format. EPA has reviewed the determining disinfection residuals from extending approval to all the regulated applicable sections and determined that the 19th edition of Standard Methods contaminants covered in a method ten of the methods are identical to the (APHA, 1995). Standard Methods 6251 provides greater flexibility to currently approved versions from the B and 4500–CIO E in the 19th edition laboratories and allows them to reduce 2 19th editions. Section 4500–Cl contains of Standard Methods (APHA, 1995) analytical costs. EPA recognizes that the methods for determining chlorine were approved for HAA5 and daily laboratories will probably not be chlorite analyses, respectively. Three residuals and it includes the 4500–Cl D, determining concentrations of these TOC methods (5310 B, 5310 C, and 5310 4500–Cl F, 4500–Cl G, 4500–Cl E, 4500– non-DBP anions for compliance D) from the Supplement to the 19th Cl I, and 4500–Cl H. Section 4500–ClO2 purposes in the same samples as used edition of Standard Methods (APHA, contains the methods for determining for chlorite or bromate compliance chlorine dioxide residuals and chlorite 1996) and one UV254 method (5910 B) monitoring. However, EPA believes from the 19th edition of Standard and it includes method 4500–ClO2 E. allowing the same method to be used Methods (APHA, 1995) were also Section 5310 contains the methods for even if the samples are not the same is approved in the Stage 1 DBPR. determining TOC and it includes more cost effective for laboratories. EPA methods 5310 B, 5310 C, and 5310 D. These thirteen methods are is not proposing to withdraw any Because the ten listed methods in these unchanged in the 20th edition of methods for the non-DBP anions, sections are unchanged from the Standard Methods (APHA, 1998), so because that would place an EPA proposes to cite the 20th edition for versions that were published in the 19th unnecessary burden on laboratories that these analyses in addition to the 19th editions, EPA is proposing to cite the do not need to use EPA Method 300.1. editions. On–Line Version for these analyses in

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00070 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49617

addition to the currently approved 19th chlorite, and bromide citation for this was based on studies conducted on editions and the proposed 20th edition. method to the August 2000 methods fortified reagent water samples rather Section 6251 includes method 6251 B manual in today’s rule so that the users than drinking water samples. Because for HAA5. The method has been are directed to the correct source of the HAAs have been shown to biodegrade updated for the On–Line Version to method. in some distribution systems (Williams include precision and accuracy data et al. 1995), EPA believes that some 7. How Is the HAA5 Sample Holding from the Information Collection Rule samples may not be stable for 28 days. Time Being Standardized? and the sample holding time has been Today’s rule proposes reducing the extended from 9 days to 14 days. The The analytical methods approved for holding time to 14 days when EPA additional quality control data does not HAA5 compliance monitoring (EPA Method 552.1 is used in order to better technically change the method from the 552.1, EPA 552.2, and Standard Method ensure sample stability. During the previously approved version in the 19th 6251 B) all specify the use of Information Collection Rule, EPA only edition; it simply demonstrates the ammonium chloride to eliminate the allowed the 14–day sample holding performance that can be expected when free chlorine residual in samples and time for all HAA samples (regardless of the method is used. The change in they require samples be iced/ the method used to analyze the sample holding time is consistent with refrigerated after collection. Even samples), so laboratories and water EPA’s proposal to standardize the HAA5 though the sampling parameters agree in systems have demonstrated their sample holding time at 14 days (See the three methods, the methods specify capability to implement this method discussion in section V.O.7). Thus EPA different sample holding times (time change. is proposing to cite the On–Line Version between sample collection and EPA believes that by standardizing for this analysis in addition to the extraction). EPA Methods 552.1 (USEPA the sample holding times allowed in the currently approved 19th edition and the 1992) and 552.2 (USEPA 1995) allow at various HAA5 methods, the burden for proposed 20th edition. least 14 days while Standard Method laboratories and water systems will be Section 5910 includes method 5910 B 6251 B (APHA 1995 and 1998) specifies reduced. Sampling considerations will for determining UV254. The method has that samples must be extracted within be simplified, because all HAA5 been updated for the On–Line Version nine days of sample collection. The samples will be collected and stored the to include precision data from the holding time for the Standard Method is same way. Information Collection Rule. Because based on data which indicated an the additional quality control data does increase in DCAA concentration to 8. How Is EPA Clarifying Which not technically change the method from slightly greater than 120% of the initial Methods Are Approved for Magnesium the previously approved version in the concentration after the sample was Determinations? 19th edition, EPA is proposing to cite stored for 14 days (Krasner et al. 1989). The Stage 1 DBPR allows systems the On–Line Version for this analysis in All other HAA5 compounds were well practicing enhanced softening that addition to the currently approved 19th within the 80–120% criteria set by the cannot achieve the specified level of edition and the proposed 20th edition. researchers. The decision was made to TOC removal, to meet instead one of The On–Line Version of Standard use a conservative approach to be sure several alternative performance criteria, Methods will not include method 4500– that the concentrations of all HAAs including the removal of 10 mg/L ClO2 D, so it is not being proposed with were stable, and nine days was the magnesium hardness (as CaCO3) from the other twelve methods cited in the closest data point to the 14 day–data the source water. Analytical methods for On–Line Version. point in question. Subsequent to measuring magnesium hardness were EPA is proposing to add a citation to Krasner’s study, EPA conducted not included in the rule, but they were the 20th edition and the On–Line additional sample holding time studies later promulgated in a Methods Update Version of Standard Methods for as part of the EPA methods Rule (USEPA 1999b). The December thirteen and twelve methods, development process. EPA has 1999 Methods Rule cited the respectively. EPA believes these should published data to support the 14–day magnesium methods at § 141.23(k)(1), be cited in addition to the 19th editions sample holding time for the HAA5 but it did not identify that these in order to allow flexibility for the water compounds (Pawlecki–Vonderheide et methods were to be used to demonstrate systems performing the analyses. al. 1997; USEPA 2003p). Since there is compliance with the alternative Withdrawal of the older editions would no technical reason for the holding performance criteria specified in require all systems to purchase one of times to be different between the HAA5 § 141.135(a)(3)(ii). EPA is proposing to the newer editions, which could impose methods addressed in this rule, EPA clarify this today by referencing the an unnecessary burden on systems that proposes to allow a 14–day sample approved magnesium methods at use the reference for only a few holding time for samples being analyzed § 141.131(d)(6) and § 141.135(a)(3)(ii). methods. by Standard Method 6251 B. This would provide consistency across methods and 9. Which Methods Can Be Used To 6. What Is the Updated Citation for EPA it would simplify sampling Demonstrate Eligibility for Reduced Method 300.1? considerations for water systems. EPA is Bromate Monitoring? EPA Method 300.1 (USEPA 2000l) for only proposing to standardize the Today’s rule proposes to change the bromate, chlorite and bromide is now holding time allowed for the samples. monitoring requirements for included in an EPA methods manual Due to differences in the sample demonstrating eligibility to reduce that was published August 2000. The preparation (i.e., extraction) procedures bromate monitoring from monthly to manual titled ‘‘Methods for the in the various methods, the extract quarterly. The Stage 1 DBPR allows the Determination of Organic and Inorganic holding times cannot be standardized. monitoring to be reduced if the system Compounds in Drinking Water’’ is a Laboratories must follow the individual demonstrates that the average source compilation of methods developed by method requirements when determining water bromide concentration is less than EPA for drinking water analyses. EPA storage conditions and holding times for 0.05 mg/L based upon monthly bromide Method 300.1 was previously only the extracts. measurements for one year. Today’s rule available as an individual method. EPA EPA Method 552.1 specifies a 28–day proposes to change that requirement to proposes to update the bromate, holding time for HAA samples. This a demonstration that the finished water

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00071 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49618 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

bromate concentration is <0.0025 mg/L optimization of ozone application to 2. What Changes Are Proposed for the as a running annual average. If this control for bromate formation.) PE Acceptance Criteria? change is implemented, there will no 10. Request for Comments The Stage 1 DBPR specified that in longer be a need for bromide order to be certified the laboratory must compliance monitoring methods. EPA is EPA requests comments on whether pass an annual performance evaluation proposing additional bromide methods the methods proposed today should be (PE) sample approved by EPA or the today in order to provide flexibility to approved for compliance monitoring. State using each method for which the the laboratories and water systems in EPA solicits comments as to whether laboratory wishes to maintain the interim period before the Stage 2 standardizing the sample holding times certification. The acceptance criteria for DBPR compliance monitoring for the HAA5 methods is appropriate. the DBP PE samples were set as requirements becomes effective. Specifically, should the sample holding statistical limits based on the In order to qualify for reduced time for Standard Method 6251 B be performance of the laboratories in each bromate monitoring, EPA is proposing extended from 9 days to 14 days and study. This was done because EPA did that the samples must be analyzed for should the sample holding time for EPA not have enough data to specify fixed bromate using either EPA Method 317.0 Method 552.1 be shortened from 28 acceptance limits. Revision 2.0 (UV/Vis detector), EPA days to 14 days? Subsequent to the 1998 promulgation, Method 326.0 (UV/Vis detector), or EPA EPA requests comments as to whether EPA evaluated the results for the EPA Method 321.8. These three methods can laboratories should be required to Water Supply (WS) PE studies and the provide quantitative data for bromate switch to one of the more sensitive Information Collection Rule PE studies concentrations as low as 0.001 mg/L, bromate methods for compliance to determine if fixed acceptance limits thus ensuring that a bromate running monitoring sample analyses. Should could now be applied. (Fixed limits annual average of <0.0025 mg/L can be EPA Method 300.1 be withdrawn as a were used during the Information reliably demonstrated. Laboratories that compliance monitoring method for Collection Rule). analyze samples by these three methods ± bromate and be replaced by EPA Four different fixed limits ( 20%, must report quantitative data for ± ± ± Methods 317.0 Revision 2.0, 326.0, and 30%, 40%, and 50% of the true bromate concentrations as low as 0.001 321.8 which provide reliable data for value) were applied to each analyte in mg/L. bromate concentrations as low as 1µg/L? the WS PE study TTHM, HAA5, Since EPA Methods 317.0 Revision bromate, and chlorite samples. 2.0, 326.0, and 321.8 offer significantly P. Laboratory Certification and Successful analysis of the sample was greater sensitivity for bromate analyses, Approval defined as passing all four THMs EPA considered whether these should 1. What Is EPA Proposing Today? individually in the TTHM PE sample; be the only methods approved for passing four of the five HAAs in the bromate compliance monitoring. EPA recognizes that the effectiveness HAA5 PE sample; and passing bromate However, the new methods using of today’s proposed regulation depends and chlorite individually. The number postcolumn reactions with UV/Vis on the ability of laboratories to reliably and percentage of laboratories that detection (EPA Methods 317.0 Revision analyze the regulated disinfection successfully passed each study sample 2.0 and 326.0) or IC/ICP–MS (EPA byproducts at the proposed MCLs. EPA were determined for the four fixed Method 321.8) require greater analyst has established a drinking water limits. These results were then skill than is necessary for the standard laboratory certification program that evaluated to determine how narrow the ion chromatographic (IC) methodology States must adopt as part of primacy. criteria could be set in order to achieve (EPA Method 300.1 and ASTM Method Laboratories must be certified in order accurate data and also provide enough D 6581–00). They also require to analyze samples for compliance with certified laboratories to meet the instrumentation that may not be the MCLs. EPA has also specified capacity needs. Only the last six WS PE currently owned by many laboratories laboratory requirements for analyses, Studies administered by EPA (WS36– that perform bromate analyses. As a such as alkalinity, bromide, disinfectant WS41 conducted between 1996–1998) result of these factors and because the residuals, magnesium, TOC, and SUVA, were used in the final recommendation, standard IC methods are adequate for that must be conducted by parties because they provided a better estimate determining compliance with the approved by EPA or the State. EPA’s of current laboratory capabilities. Table bromate MCL that was promulgated as ‘‘Manual for the Certification of V–19 summarizes the results of this WS part of the Stage 1 DBPR, EPA decided Laboratories Analyzing Drinking Water’’ PE Study evaluation. not to propose withdrawal of the (USEPA 1997b) specifies the criteria The number of laboratories that currently approved method (EPA that EPA uses to implement the analyzed WS TTHM PE samples was Method 300.1). In addition, EPA drinking water laboratory certification significantly larger than for the other decided to propose ASTM Method D program. Today’s proposed rule DBPs, because a laboratory certification 6581–00, because it is equivalent to EPA maintains the requirements of program for TTHM has been in effect Method 300.1. EPA strongly encourages laboratory certification for laboratories since the promulgation of the THM rule laboratories to expand their services by performing analyses to demonstrate in 1979 (USEPA 1979). Most of the adding the capability to perform compliance with MCLs and all other analytical methods for TTHM have been analyses using one of the more sensitive analyses to be conducted by approved in use for many years, and the methods for bromate. EPA believes that parties. It revises the acceptance criteria laboratories are experienced in their there will be a shift to the more for performance evaluation (PE) studies use. The Stage 1 DBPR established the sensitive methods as water systems and proposes reporting limits for the first requirements to monitor for the realize that the analytical capabilities DBPs as part of the certification other DBPs and certification was not are available for a slightly increased program. Today’s rule also proposes that required until December 2001. analytical cost. (The ability to determine TTHM and HAA5 analyses that are Therefore, the WS PE results for HAA5, bromate concentrations as low as 1 µg/ performed for the IDSE or system- chlorite, and bromate were from L will provide water systems more specific study be conducted by laboratories that were not part of a information concerning the laboratories certified for those analyses. certification process and the laboratories

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00072 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49619

were using methods that were relatively EPA Method 300.1 in the Stage 1 DBPR, PE samples using Method 300.1 and the new. In addition, the method used for because Method 300.1 is more sensitive. bromate methods that are being bromate during the WS studies was EPA Laboratories would be expected to have proposed in today’s rule. Method 300.0 which was replaced by greater success in passing the bromate

TABLE V–19.—FIXED LIMIT EVALUATION OF WS PE STUDIES 36—41 [Average # and % of labs successfully completing studies]

±20% of TV ±30% of TV ±40% of TV ±50% of TV DBP Sample #Labs %Labs #Labs %Labs #Labs %Labs #Labs %Labs

TTHM ...... 609 73 731 88 773 93 788 94 HAA5 1 ...... 50 37 83 61 103 75 115 84 chlorite ...... 55 63 68 78 72 82 74 85 bromate ...... 45 50 52 57 57 64 60 68 1 Study 38 was excluded from this analysis, because a valid DCAA true value was not available for the HAA sample.

Based on the results from the analyses DBPR. During the Information EPA believes chlorite PE samples described previously, EPA believes it is Collection Rule, laboratories that were should be evaluated using a ±30% reasonable to set the TTHM acceptance approved using the ±40% criteria were criteria. Over 70% of the laboratories criteria at ±20% around the study true able to provide accurate and precise could meet this requirement for chlorite values. The number of laboratories data as evidenced by the quality control in the WS studies. capable of performing TTHM analyses is data collected when the Information The percentage of passing labs for large and the results described Collection Rule samples were analyzed bromate is almost 60% when a ±30% previously show that in the time frame (Fair et al. 2002). Of the 1,250 criteria is applied to the WS study data. of 1996–1998, over 70% of the Information Collection Rule samples Since the data do not accurately reflect laboratories could successfully meet the that were fortified with known amounts the bromate methods that are now being ±20% criteria. The PE studies of HAAs, the median recovery was used by laboratories, EPA believes a conducted during the Information 103% and the recoveries ranged greater percentage of laboratories would Collection Rule used the same between 89% and 120% in 80% of the pass the bromate PE study using today’s acceptance criteria (USEPA 1996b). fortified samples. There were 1,211 technology. Unfortunately, EPA does The data indicate that ±40% are Information Collection Rule samples not have the data to verify this probably the tightest criteria that could that were analyzed in duplicate and the assumption, because EPA no longer be used to evaluate HAA5 PE samples. median relative percent difference for conducts PE studies. Even if the Setting this criteria balances the need those HAA5 analyses was 4%. Ninety assumption is flawed, a 57% acceptance for approval of enough labs to meet percent of the analyses had RPDs less rate would still provide enough certified monitoring capacity and the need to than 21%. EPA believes laboratories laboratories to handle the number of provide data of acceptable accuracy. that are certified using the ±40% criteria bromate samples required for The ±40% criteria is consistent with the in PE studies should be capable of compliance monitoring under the Stage Information Collection Rule PE study performing at a level comparable to the 1 DBPR. acceptance criteria and it is tighter than Information Collection Rule The proposed acceptance criteria are the criteria established in the Stage 1 laboratories. listed in Table V–20.

TABLE V–20.—PROPOSED PERFORMANCE EVALUATION (PE) ACCEPTANCE CRITERIA

Acceptance DBP limits Comments (percent)

TTHM Chloroform ±20 Laboratory must meet all 4 individual THM acceptance limits in Bromodichloromethane ±20 order to successfully pass a PE sample for THMs. Dibromochloromethane ±20 Bromoform ±20 HAA5 Monochloroacetic Acid ±40 Laboratory must meet the acceptance limits for 4 out of 5 of Dichloroacetic Acid ±40 the HAA5 compounds in order to successfully pass a PE Trichloroacetic Acid ±40 sample for HAA5. Monobromoacetic Acid ±40 Dibromoacetic Acid ±40 Chlorite ±30 Bromate ±30

EPA is also proposing that the PE allow the laboratory certification acceptance criteria would be subject to acceptance limits described previously program to implement the fixed limits the new criteria when it is time for them become effective within 60 days of as soon as possible. Laboratories that to analyze their annual DBP PE promulgation of the final rule. This will were certified under the Stage 1 PE samples(s).

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00073 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49620 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

3. What minimum reporting limits are are based on PE data, if available, or are expected to contain concentrations being proposed? set at five or ten times the method greater than the specified MRLs. detection level. The Consumer Confidence Reports EPA evaluated the data from the In today’s rule, EPA is proposing an Information Collection Rule to Rule (USEPA 1998i) requires that all alternate approach for establishing the determine if the laboratories were able detected regulated contaminants be lowest concentration for which to reliably measure down to the reported in the annual reports, but laboratories are expected to report required MRL concentrations. Precision detection is not defined for the DBP quantitative data for DBPs. The and accuracy data from the calibration contaminants. This rule addresses the approach is based on a unique data set check standards prepared at the MRL deficiency by proposing reporting limits from the Information Collection Rule. concentrations (listed in Table V–21) for the regulated DBPs. Laboratories were required to meet Laboratories that analyze compliance specific quality control criteria when were examined. The data indicated most samples must be able to reliably they analyzed samples for the laboratories were able to provide measure the DBPs at concentrations Information Collection Rule. The rule quantitative data for samples with these below the MCL. Laboratories must also established a regulatory minimum concentrations. be able to measure the individual TTHM reporting level (MRL) for each analyte Because laboratories demonstrated the and HAA5 compounds at levels that are and laboratories were required to capability to meet the Information much lower than the MCLs for these demonstrate they could accurately Collection Rule MRLs, EPA believes it is compound classes, because the MCLs measure at these concentrations each reasonable to expect similar are based on the sum of the individual time a set of samples was analyzed. The performance during the analyses of DBP compound concentrations. regulatory MRLs were based on compliance monitoring samples. In Historically, EPA has used practical recommendations from experts who today’s rule, EPA is proposing to quantitation levels to estimate the were experienced in DBP analyses and incorporate MRL requirements into the lowest concentration at which were set at concentrations for which laboratory certification program for laboratories can be expected to provide most laboratories were expected to be DBPs and to use regulatory MRLs as the data within specified limits of precision able to meet the precision and accuracy minimum concentrations that must be and accuracy during routine operating criteria under normal operating reported as part of the Consumer conditions (USEPA 1985). The estimates conditions. Most samples were also Confidence Reports (§ 141.151(d)).

TABLE V–21.—PROPOSED MINIMUM REPORTING LEVEL (MRL) REQUIREMENTS

MRL (µg/L) DBP Information Comments collection Proposed stage rule 2 DBPR

TTHM Chloroform ...... 1.0 1.0 Bromodichloromethane ...... 1.0 1.0 Dibromochloromethane ...... 1.0 1.0 Bromoform ...... 1.0 1.0 HAA5 Monochloroacetic Acid ...... 2.0 2.0 Dichloroacetic Acid ...... 1.0 1.0 Trichloroacetic Acid ...... 1.0 1.0 Monobromoacetic Acid ...... 1.0 1.0 Dibromoacetic Acid ...... 1.0 1.0 Chlorite ...... 20.0 200.0 Bromate ...... 5.0 5.0 or 1.0 Laboratories that use EPA Methods 317.0 Revision 2.0, 326.0, or 321.8 must meet a 1.0 µg/L MRL for bromate.

As part of the request for certification, specified MRL concentration daily or precision and accuracy criteria at lower EPA is proposing to require laboratories each time samples are analyzed.) The concentrations, so EPA believes that to demonstrate they can reliably measured concentration for this check each laboratory should have the measure concentrations at least as low standard must be within ±50% of the flexibility to continue using its own as the ones listed in Table V–21 in order expected value. Laboratories may reporting limits as long as the laboratory to be certified for those parameters. This choose to report quantitative data at MRLs are not higher than the regulatory would mean that the calibration curve concentrations lower than the proposed ones proposed in this rule. This must encompass the proposed regulatory MRLs as long as the required flexibility would minimize the cost of regulatory MRL concentration and that accuracy criteria (±50% of the expected implementing the regulatory MRL the laboratory must verify the accuracy value) is met by daily analyzing requirements, because the laboratory of the calibration curve at the lowest standards at the lowest reporting limit would not have to make changes in its concentration for which quantitative chosen by the laboratory. established quality control procedures data are reported by analyzing a Laboratories were not given the unless its procedures are less stringent calibration check standard at that opportunity to report concentrations than those being proposed today. concentration prior to analyzing each lower than the specified MRLs during Requiring a laboratory to adopt batch of samples. (Laboratories would the Information Collection Rule. Some regulatory MRLs that are higher than the analyze a check standard at the laboratories indicated they have met the laboratory reporting limits currently in

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00074 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49621

use offers no advantage and could <200. µg/L. If a lower reporting limit is certification requirement, EPA does not increase analytical costs. The capability used, then the laboratory will be believe this significantly increases the to provide quantitative data at the required to meet the precision and time required to review a laboratory laboratory’s MRL or the regulatory MRL accuracy requirements at that lower prior to certification. Each DBP method would need to be demonstrated on a concentration by daily successfully requires the laboratory to generate a daily basis by analyzing a check analyzing a check standard at the similar set of data at a higher standard at that concentration and by laboratory reporting limit concentration concentration and to meet specific achieving a recovery in the range of 50 prior to analyzing compliance samples. accuracy and precision criteria as part of to 150%. EPA believes very few water systems the initial demonstration of laboratory The proposed regulatory MRL for will request more sensitive chlorite capability to perform the method; MCAA is 2.0 µg/L based on the analyses, because their samples won’t review of the MRL data set will be Information Collection Rule have low enough concentrations to comparable to what is already being performance data. However, MCAA was require it. done. This new requirement will ensure not present at concentrations higher EPA is proposing two regulatory that laboratories can reliably analyze than this in more than half of the MRLs for bromate analyses in today’s samples that contain low concentrations samples analyzed for HAAs during the rule. This is because the traditional ion of DBPs on an on-going basis. Information Collection Rule (USEPA chromatographic (IC) methods using EPA is also proposing to require the 2003o). Some laboratories reported that conductivity detection (EPA Method regulatory MRLs be used for compliance they could have provided quantitative 300.1 and ASTM Method 6581–00) are reporting by the Public Water Systems. data for MCAA down to concentrations only capable of quantitating down to 5.0 Finally, the regulatory MRLs would be as low as 1.0 µg/L. µg/L while the new IC methods using used when Public Water Systems inform EPA is proposing a regulatory MRL either post column reactions with UV/ customers of their water quality relative for chlorite that is much higher than can Vis detection (EPA Methods 317.0 to DBP concentrations in the annual easily be achieved using the approved Revision 2.0 and 326.0) or IC followed Consumer Confidence Reports. or proposed methods. The MRL by ICP–MS detection (EPA Method 4. What Are the Requirements for specified during the Information 321.8) can reliably quantitate bromate Analyzing IDSE Samples? Collection Rule was 20. µg/L and concentrations as low as 1.0 µg/L. EPA laboratories were able to successfully believes it is appropriate to set the EPA is proposing that the TTHM and obtain quantitative data at that level. regulatory MRL based on the capability HAA5 samples collected for the Initial However, in the context of this rule, of the method. (EPA has published Distribution System Evaluations (IDSE) EPA believes that requiring laboratories detection limits for inorganic and the system specific studies to verify their calibration curves down contaminants based on method conducted in lieu of IDSEs be analyzed to 20. µg/L each time samples are capability (§ 141.23(a)(4)(i)), so the by certified laboratories. EPA recognizes analyzed is unnecessary. This is because approach proposed today is consistent that this will require additional chlorite analyses are only performed on with previous regulations.) If the laboratory capacity during the time samples from water plants that use regulatory MRL is based on the most period in which these studies are chlorine dioxide and most of the sensitive method, then the routine IC conducted. The largest challenge will be applied chlorine dioxide is converted to methods could no longer be used even in developing the capacity to analyze chlorite, so the concentrations that are though they are adequate for the samples for the water systems that expected in most compliance demonstrating compliance with the must complete the studies, analyze the monitoring samples will be much higher bromate MCL. If the regulatory MRL is data, and recommend Stage 2 DBP than 20. µg/L. (The Information set using the least sensitive method, sampling sites within two years of the Collection Rule data showed a median then the feasibility for reduced bromate promulgation date of the rule. However, chlorite concentration of 380 µg/L in the monitoring based on a running annual EPA believes commercial laboratories, finished water and 333 µg/L as the average of <0.0025 µg/L (<2.5 µg/L) in particular, will be able to expand distribution system average in systems would not be adequately demonstrated their capacity to meet the demand based using chlorine dioxide (USEPA 2003o).) based on data reported with a reporting in the information presented below. EPA is proposing a regulatory MRL of limit of 5.0 µg/L. Assuming no waivers or system- 200. µg/L for chlorite, because most of EPA is proposing MRLs as part of the specific studies, the number of IDSE the samples are expected to contain certification process. Laboratories samples is estimated to be between concentrations higher than 200. µg/L. would be required to demonstrate they 14,000 and 21,000 per month in the first The MCL for chlorite is 1.0 mg/L (1,000 can reliably quantitate at the specified round of IDSE monitoring, depending µg/L). EPA recognizes that setting the MRL concentration when their current on whether the monitoring requirements regulatory MRL for chlorite based on the DBP certification is subject to renewal are based on population or number of concentrations expected to be found in or if they are applying for certification treatment plants, respectively. the samples rather than the sensitivity for DBP methods for the first time. Laboratories should easily be able to of the analytical method is inconsistent (Demonstration would be accomplished accommodate this increase in TTHM with the approach taken for other by providing precision and accuracy samples, because experience performing compounds in this rule. Nevertheless, data from the analyses of check TTHM analyses is spread across a large EPA believes setting the MRL based on standards at or below the regulatory number of laboratories. Hundreds of occurrence information is appropriate MRL concentration over a several day laboratories have been certified for because it will not compromise the period. The laboratory’s standard TTHM analyses, since certification was compliance data. Water systems would operating procedure for HAA5 analyses first required in 1979. There were close have the option of requiring that would include a requirement to daily to 600 laboratories certified to perform laboratories establish a lower reporting meet the MRL accuracy criteria for a TTHM analyses in 1991. In the 1996– limit when their samples are analyzed check standard at or below the 1998 period, there were over 800 and EPA would encourage this in cases regulatory MRL concentration.) laboratories participating in the PE in which the samples consistently Although ensuring laboratories can meet studies for TTHMs and 600 of those contain chlorite concentrations that are the regulatory MRLs is a new laboratories were capable of meeting the

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00075 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49622 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

TTHM PE acceptance criteria proposed 5. Request for Comments To receive primacy for the Stage 2 in today’s rule. Many water system EPA requests comments concerning DBPR, when final, States will be laboratories are certified to perform the appropriateness of the proposed PE required to adopt the following new or TTHM analyses and will be able to acceptance criteria. revised requirements under their own incorporate the IDSE TTHM samples EPA solicits comments as to whether regulations: from their systems into the laboratory an MRL lower than 2 µg/L is feasible for —Section 141.33(a) and (f), Record schedule. It is reasonable to expect that MCAA and if so, what should that MRL maintenance; commercial laboratories will be able to concentration be? —Section 141.64, MCLs for disinfection handle the remainder of the TTHM EPA requests comments concerning byproducts; samples. (EPA does not have a current whether the MRL for chlorite should be —Subpart L, Disinfectant Residuals, estimate of the number of laboratories based on the sensitivity of the method Disinfection Byproducts, and certified to perform TTHM analyses. (i.e., 20. µg/L) or on the expected Disinfection Byproduct Precursors; However, if the number of IDSE samples concentration range of the samples (i.e., —Subpart O, Consumer Confidence from large systems was evenly spread 200. µg/L). Reports; over the 600 laboratories that were EPA solicits comments concerning —Subpart Q, Public Notification of certified in 1991, this would be less which MRL approach should be Drinking Water Violations; than 40 additional samples per month considered for bromate. Specifically, —Subpart U, Initial Distribution System for each laboratory. Analysis of 40 should EPA set the MRL based on the Evaluation; and TTHM samples would involve less than capability of the method which would —Subpart V, Stage 2B Disinfection Byproducts Requirements. two days of analyst and instrument time mean that two different MRLs are which does not seem unreasonable for defined or should one MRL be In addition to adopting basic primacy commercial laboratories to established based on either the least or requirements specified in 40 CFR part accommodate.) most sensitive method? 142, States are required to address EPA requests comments concerning applicable special primacy conditions. Analyses of the HAA5 samples will the appropriateness of the MRL Special primacy conditions pertain to present a greater challenge, because certification requirements and whether specific regulations where certification is relatively new for this additional certification requirements implementation of the rule involves measurement. EPA anticipates that most should be considered. activities beyond general primacy of the HAA5 samples will be analyzed EPA solicits comments on the provisions. The purpose of these special by commercial and State laboratories, availability of laboratory capacity to primacy requirements in today’s because the methods are more complex perform TTHM and HAA5 analyses for proposal is to ensure State flexibility in than the TTHM analyses and monitoring IDSE studies. implementing a regulation that: (1) was not widely required until January Applies to specific system VI. State Implementation 2002. Laboratories were not required to configurations within the particular be certified to perform HAA5 analyses This section describes the regulations State and (2) can be integrated with a until January 2002. However, the PE and other procedures and policies States State’s existing Public Water Supply Study results from 1996–1998 indicate would have to adopt to implement the Supervision Program. States must that over 130 laboratories were Stage 2 DBPR, if finalized as proposed include these rule-distinct provisions in performing HAA5 analyses during that today. States must continue to meet all an application for approval or revision time frame and approximately 100 of other conditions of primacy in 40 CFR of their program. These primacy those laboratories were capable of part 142. requirements for implementation meeting the HAA5 PE acceptance The SDWA establishes requirements flexibility are discussed in the following criteria proposed in today’s rule. that a State or eligible Indian Tribe must section. Ninety-four laboratories were approved meet to assume and maintain primary enforcement responsibility (primacy) for A. State Primacy Requirements for to perform HAA analyses during the Implementation Flexibility Information Collection Rule; twenty- its public water systems. These SDWA seven of them were commercial requirements include: (1) adopting To ensure that a State program laboratories and nine were State drinking water regulations that are no includes all the elements necessary for laboratories. EPA anticipates that large less stringent than Federal drinking an effective and enforceable program commercial laboratories already water regulations, (2) adopting and within that State under today’s rule, a certified to perform HAA5 analyses will implementing adequate procedures for State primacy application must include recognize this market potential and add enforcement, (3) keeping records and a description of how the State will staff and instrumentation to making reports available on activities review IDSE reports and approve new or accommodate the increased demand. that EPA requires by regulation, (4) revised monitoring sites for long-term issuing variances and exemptions (if DBP compliance monitoring. If a State Most systems serving <10,000 people allowed by the State), under conditions will use the authority to grant blanket will not begin their IDSE studies until no less stringent than allowed under the waivers for IDSE requirements to very after the large systems have completed SDWA, and (5) adopting and being small systems, it must comply with the their studies. Even though the potential capable of implementing an adequate special primacy provision for granting number of samples is greater, the small plan for the provision of safe drinking such waivers. A State that intends to use systems have two additional years in water under emergency situations. the authority for addressing consecutive which to complete their studies, so General rule implementation activities system monitoring requirements must there is more opportunity to schedule include notifying systems of rule include a description of how it intends the sampling in such a manner that requirements, updating internal and to implement that authority. A State laboratory capacity is maintained. The external databases, providing training primacy application must also include a laboratory capacity should be readily and technical assistance, and reviewing description of how the State will require available by the time analyses of these (and, if necessary, approving) systems to identify significant samples are required. monitoring and other reports and plans. excursions.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00076 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49623

B. State Recordkeeping Requirements NPDWRs must obtain primacy for this changes and changes in DBP The current regulations in § 142.14 rule. occurrence, and input and analysis from expert technical review panels to assist require States with primacy to keep E. IDSE Implementation with model validation and technology various records, including analytical As discussed in section V.J., many results to determine compliance with selection. A brief description of the systems will be performing certain IDSE process is outlined in section VII.E. but MCLs, MRDLs, and treatment technique activities prior to their State receiving a more detailed explanation of the requirements; system inventories; State primacy. During that period, EPA will analytical process is in the EA for the approvals; enforcement actions; and the act as the primacy agency, but will proposed Stage 2 DBPR (USEPA 2003i). issuance of variances and exemptions. consult and coordinate with individual The Stage 2 DBPR economic impact The proposed Stage 2 DBPR requires States to the extent practicable and to analysis uses a model, (referred to as the that the State keep records related to the extent that States are willing and Surface Water Analytical Tool or any decisions made pursuant to the able to do so. In addition, prior to SWAT) and information collected under requirements in subparts U and V, plus primacy, States may be asked to assist the Information Collection Rule to make copies of IDSE reports submitted by EPA in identifying and confirming predictions about finished water and systems until those reports are reversed systems that are required to comply delivered water DBP levels, as well as or revised in their entirety. Today’s with certain IDSE activities. Once the predicting technology changes proposal also includes a revision to the State has received primacy, it will necessary for systems to comply with State recordkeeping requirements that become responsible for IDSE rule alternatives. Specifically, SWAT requires States to maintain records of implementation activities. estimates post-Stage 1 DBPR (pre-Stage DBP monitoring plans submitted by 2) and post-Stage 2 DBPR DBP levels public water systems until superceded F. State Burden and likely technology choices by the by a new system monitoring plan. Section VII of today’s document industry to achieve compliance. For C. State Reporting Requirements contains an analysis of the burden that smaller systems and for all ground water this rule will place on States in systems, expert panels considered EPA currently requires in § 142.15 receiving primacy and implementing occurrence data and current treatment that States report information such as this rule. technology specific to these systems and violations, variance and exemption used this information to predict G. Request for Comment status, and enforcement actions to EPA. technology treatment changes that may The proposed Stage 2 DBPR will not EPA requests comment on the State result from this proposed rule. add any additional reporting implementation requirements including Both benefits and costs are presented requirements. the special primacy requirements. as annualized values. The process allows comparison of cost and benefit D. Interim Primacy VII. Economic Analysis streams that are variable over a given On April 28, 1998, EPA amended its This section summarizes the Health time period. The time frame used for State primacy regulations at 40 CFR Risk Reduction and Cost Analysis both benefit and cost comparisons is 25 142.12 to incorporate the new process (HRRCA) in support of the Stage 2 DBPR years; approximately five years account identified in the 1996 SDWA as required by section 1412(b)(3)(C) of for rule implementation and 20 years for Amendments for granting primary the 1996 SDWA. In addition, under the average useful life of the equipment. enforcement authority to States while Executive Order 12866, Regulatory The Agency uses social discount rates of their applications to modify their Planning and Review, EPA must both three percent and seven percent to primacy programs are under review (63 estimate the costs and benefits of the calculate present values from the stream FR 23362) (USEPA 1998j). The new Stage 2 DBPR in an Economic Analysis of benefits and costs and also to process grants interim primary (EA). EPA has prepared an EA to annualize the present value estimates. enforcement authority for a new or comply with the requirements of this The EA for the proposed rule (USEPA revised regulation during the period in order and the SDWA Health Risk 2003i) also shows the undiscounted which EPA is making a determination Reduction and Cost Analysis (HRRCA) stream of both benefits and costs over with regard to primacy for that new or (USEPA 2003i). SDWA (Section 1412 the 25 year analysis period. revised regulation. This interim (b)(4)(C)) also requires the Agency to enforcement authority begins on the determine that the benefits of the A. Regulatory Alternatives Considered date of the complete primacy promulgated rule would justify the costs by the Agency application submission or the effective of compliance. The proposed EA is Today’s proposed Stage 2 DBPR date of the new or revised State available in the docket and is also represents the second of a set of rules regulation, whichever is later, and ends published on the Agency’s web site: that address public health risks from when EPA makes a final determination. http://www.epa.gov/edocket. DBPs. The Stage 1 DBPR was However, this interim primacy authority It is important to note that the promulgated to decrease average is only available to a State that has regulatory options considered by the exposure to DBPs and associated health primacy for every existing NPDWR in Agency are the direct result of an risks by focusing compliance on MCLs effect when the new regulation is Advisory Committee process that based on average concentrations of promulgated. involved various drinking water TTHM and HAA5 within the As a result, States that have primacy stakeholders. More information on this distribution system. Today’s proposed for every existing NPDWR already in process is discussed in sections II and Stage 2 DBPR further reduces exposure effect may obtain interim primacy for V of today’s preamble. to chlorinated DBPs by basing this rule, beginning on the date that the In order to analyze both benefits and compliance on the LRAA of TTHM and State submits the application for this costs of the proposed rule and other HAA5 concentrations at each sampling rule to EPA, or the effective date of its regulatory alternatives considered by point within the distribution system. revised regulations, whichever is later. the Agency, EPA relied on several data Section V illustrated the LRAA concept In addition, a State which wishes to sources to understand DBP occurrence, and differences in the two compliance obtain interim primacy for future an analytical model to predict treatment calculation methodologies. In addition,

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00077 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49624 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

section V provided a comparison of the The Advisory Committee considered alternative rule scenarios illustrated regulatory options considered. This the DBP occurrence estimates and next. subsection will summarize the characteristics of these distributions to Preferred Alternative comparison of options and subsection be important in understanding the VII.B. will outline the exposure analyses nature of public health risks. Although —Long-term MCLs of 0.080 mg/L for that led EPA to propose the preferred the Information Collection Rule data TTHM and 0.060 mg/L for HAA5 as option and will present the predicted were collected prior to promulgation of LRAAs. national occurrence distributions that the Stage 1 DBPR, the data support the —Bromate MCL remaining at 0.010 mg/ were used to quantify predicted concept that a system could be in L. exposure reductions from today’s compliance with the Stage 1 DBPR proposed rule. A detailed discussion of MCLs of 0.080 mg/L and 0.060 mg/L for Alternative 1 EPA’s exposure analyses can be found TTHM and HAA5, respectively, and yet —Long-term MCLs of 0.080 mg/L for in the Economic Analysis for the Stage have points in the distribution system TTHM and 0.060 mg/L for HAA5 as 2 DBPR (USEPA 2003i). with either periodically or consistently LRAAs. There are two components in the higher DBP levels (see section IV). —Bromate MCL of 0.005 mg/L. Agency’s M–DBP regulatory Based on these findings, and in order development process that are to address disproportionate risk within Alternative 2 particularly relevant to evaluation of distribution systems, the Advisory —Long-term MCLs of 0.080 mg/L for options discussed in today’s proposal: Committee discussed an array of options TTHM and 0.060 mg/L for HAA5 as (1) the data synthesis and evaluation that would base compliance on absolute maximums for individual resulting from the Information exposure at specific sampling locations measurements. rather than on average exposures for the Collection Rule; and (2) the analysis and —Bromate MCL remaining at 0.010 mg/ entire distribution system. These recommendations of the M–DBP L. Advisory Committee. Data from the included options for determining Information Collection Rule were used compliance as an LRAA (requiring Alternative 3 with the SWAT model to estimate the systems to meet the MCL at individual sampling locations as a running annual —Long-term MCLs of 0.040 mg/L for national distributions of DBP TTHM and 0.030 mg/L for HAA5 as occurrence. The Advisory Committee average) or as absolute maximums (requiring that no samples taken exceed an RAA. considered several questions during the —Bromate MCL remaining at 0.010 mg/ negotiation process, including: the MCL concentration), in addition to a combination of these approaches. For L. —What are the remaining health risks example, the Advisory Committee Figure VII–1 shows how compliance after implementation of the Stage 1 reviewed the exposure reductions for a would be determined under each of the DBPR? number of approaches based on TTHM/HAA5 alternatives described and —What are approaches to addressing different LRAA and absolute maximum the Stage 1 DBPR for a hypothetical these risks? incremental MCL levels, and large surface water system. This —What are the risk tradeoffs that need combinations of an LRAA approach hypothetical system has one treatment to be considered in evaluating these with a companion absolute maximum plant and measures TTHM in the approaches? for a variety of different concentration distribution system in four locations per —How do the estimated costs of the levels. The Advisory Committee also quarter (the calculation methodology approach compare to reductions in evaluated the associated technology shown would be the same for HAA5). peak occurrences and overall changes and costs for these alternatives. Ultimately, the Advisory Committee exposure for that approach? How does In the process of narrowing down recommended the Preferred Alternative this measure (ratio of costs to alternatives based on this vast amount in combination with an IDSE exposure reduction) compare among of information, the Advisory Committee requirement. the approaches? primarily focused on four types of BILLING CODE 6560–50–P

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00078 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49625

The Preferred Alternative, coupled system without compromising microbial concentrations for some portion of the with the IDSE’s refocused sampling (see protection. At the same time, it will year. In addition, this alternative will section V), was recommended by the only require a few higher risk systems further reduce average DBP levels as Advisory Committee because this to face the cost of employing additional systems make changes to reduce these approach addresses the objective of advanced technologies. While this peak concentrations. Subsection VII.B. reducing potential adverse reproductive alternative controls the occurrence of will show how today’s proposed and developmental health risks. It consistently high DBP levels, it is still requirements are predicted to decrease achieves this objective by controlling possible that individual samples could exposure risks. The benefits and costs of peak TTHM and HAA5 concentrations exceed the MCL, and consumers could each alternative are presented in at sites throughout the distribution thus be exposed to higher DBP subsections VII.C. through VII.E.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00079 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.010 49626 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

B. Rationale for the Proposed Rule average more than 0.080 mg/L and need to install treatment to comply with Option 0.060 mg/L; the Stage 1 DBPR may be different than —Chronic exposures at all locations in in those that do not, perhaps due to low DBP concentrations can be highly the distribution system by reducing source water TOC concentrations. variable throughout a distribution overall system average DBP However, EPA does not have data system and over time at the same concentrations; and representing DBP levels post-Stage 1. location in a distribution system —Chronic and peak exposures in EPA requests comment on its approach (USEPA 2003o). The determination of consecutive systems (systems that of using data from plants in compliance compliance with an RAA under the purchase treated water from another with Stage 1 DBPR requirements Stage 1 DBPR requires a system to system). without implementing additional average all of their spatially-distributed treatment as a proxy for post-Stage 1 samples collected in one quarter of the Under the Stage 1 DBPR, high DBP DBP levels. year and to combine this average concentrations at specific locations in concentration with the three prior the distribution system could be masked 2. Reducing Average Exposure quarterly averages determined by the by spatial and temporal averaging. As To quantify the benefits of today’s system. Thus, the RAA-based standard discussed in subsection VII.C, short proposed rule, EPA compared predicted allows utilities to average spatial and term exposures resulting from these post-Stage 2 DBPR occurrence and temporal variability in TTHM and high concentrations may be of concern compared this to the predicted baseline HAA5 samples to determine in regard to potential adverse concentrations after the Stage 1 DBPR to compliance, as shown in figure VII–1. reproductive and developmental health determine reductions in exposure This allows lower results found, effects. Chronic exposures at locations resulting from the Stage 2 DBPR. The perhaps, nearer a water treatment plant having repeated high DBP SWAT model was the main tool used in to offset higher results that might be concentrations may be of concern for this analysis. SWAT results were used found at the ends of the distribution cancer endpoints as well. The directly for medium and large surface system. In addition, systems with remainder of this subsection will water systems. For small surface water multiple plants of differing water illustrate how today’s proposed rule is systems and all ground water systems. quality (either multiple surface water expected to reduce ‘‘peak’’ and average Adjustments were made to the SWAT plants or surface and ground water exposures to address these health results to account for different plants) may have particular plant concerns. percentages of plants changing distribution system sampling locations 1. Reducing Peak Exposure technology to meet Stage 2 DBPR with high DBPs that are offset by lower EPA used Information Collection Rule requirements. The Economic Analysis measurements observed in the portion data to estimate the reduction in for today’s proposed rule (USEPA 2003i) of the distribution network served by exposure to DBP peaks resulting from provides an in-depth discussion of this other plants. the Stage 2 DBPR. Because the analysis. Under the Stage 2 DBPR proposed Information Collection Rule data Table VII–2 shows the reduction in today, TTHM and HAA5 MCLs will represent pre-Stage 1 DBPR conditions, average plant-level TTHM and HAA5 remain the same, but compliance will be subsets of those plants already in concentrations estimated to result from based on a locational running annual compliance with the Stage 1 DBPR and the Stage 2 DBPR. EPA expects average average (LRAA) for each of the sampling Stage 2 DBPR were used to estimate pre- DBP levels to decline by 4.7 percent for sites in the distribution system. In Stage 2 and post-Stage 2 occurrence all surface water systems. DBP averages addition, the IDSE requirement will respectively. By comparing these are expected to decline by 2.2 percent increase the probability that the subsets of data, EPA estimated that for all large ground water systems and compliance sampling sites will capture approximately 69% of plant locations 1.7 percent for all small ground water the highest DBP levels in the having TTHM peaks greater than 0.080 systems. These estimates include both distribution system. Thus, the reduction mg/L remaining after the Stage 1 DBPR systems already in compliance with the in DBP exposure from the Stage 1 DBPR could be reduced through Stage 2 DBPR and systems making to the proposed Stage 2 DBPR results implementation of the Stage 2 DBPR. treatment changes to comply with the from the revised requirements for EPA conducted this additional peak rule. The Agency uses these national compliance calculations combined with reduction analysis only for TTHMs and average reductions to quantify the new compliance monitoring sites. not HAA5s because current primary benefit of this rule which is the EPA expects the Stage 2 DBPR, as epidemiological data only considers the estimated range of reduction in bladder proposed, will result in health benefits association between TTHM exposure cancer cases nationally. Systems making by reducing the estimated health risks and adverse health impacts (see treatment changes to comply with the associated with the following exposures: subsection VII.C). Additional rule will experience significantly greater —Individual TTHM/HAA5 occurrences information on reduction of peak estimated average reductions than the significantly exceeding 0.080 mg/L exposures can be found in section 5.4.1 national average for all systems. Chapter and 0.060 mg/L; of the Economic Analysis (USEPA 5 of the EA (USEPA 2003i) includes a —Chronic exposures at individual 2003i). EPA recognizes that temporal more detailed discussion of this distribution system locations that and spatial variability in systems that analysis.

TABLE VII–2.—REDUCTION IN AVERAGE DBP LEVELS FROM PRE-STAGE 2 TO POST-STAGE 2 (ALL PLANTS)

Average plant-level TTHM concentrations Average plant-level HAA5 concentrations System size (µg/L) (µg/L) Source water (population served) Post-stage Percent Post-stage Percent Pre-stage 2 2 reduction Pre-stage 2 2 reduction

SW ...... ≤ 10,000 35.5 33.8 4.7 25.0 23.8 4.7

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00080 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49627

TABLE VII–2.—REDUCTION IN AVERAGE DBP LEVELS FROM PRE-STAGE 2 TO POST-STAGE 2 (ALL PLANTS)—Continued

Average plant-level TTHM concentrations Average plant-level HAA5 concentrations System size (µg/L) (µg/L) Source water (population served) Post-stage Percent Post-stage Percent Pre-stage 2 2 reduction Pre-stage 2 2 reduction

> 10,000 35.5 33.8 4.7 25.0 23.8 4.7

GW ...... ≤ 10,000 16.0 15.6 2.2 8.5 8.3 2.2 10,000 16.2 16.0 1.7 8.6 8.5 1.7 Note: Due to rounding, percent reductions calculated from data in the tables may differ from the actual values presented here Source: Economic Analysis (USEPA 2003i) Exhibit 5.22b

C. Benefits of the Proposed Stage 2 preamble. For one endpoint, fetal loss, reductions in TTHM and HAAs will DBPR the Agency provides an illustrative correspond to the percent reductions in As described previously, the Stage 2 calculation to explore the implications bladder cancer risk attributed to DBPR is expected to reduce both peak of some published results for potential populations receiving chlorinated and long-term exposure to DBPs, benefits associated with reducing fetal drinking water as indicated by various thereby reducing the potential risk of losses that may be attributable to certain epidemiology studies (USEPA 1998a). both adverse reproductive and DBP exposures. Zero is included in this range because developmental health effects and In addition to achieving greater of the inconsistent evidence regarding bladder cancer. As discussed in section protection from possible adverse the association between exposure from III of this preamble, both reproductive and developmental health DBPs and cancer. effects, the rule may provide additional epidemiological and toxicological Other regulatory alternatives reduction in bladder cancer cases as the evidence suggest a possible increased considered by the FACA committee and risk for pregnant women and their overall level of DBPs in distribution the Agency could provide greater fetuses who are exposed to DBPs in systems nation-wide decreases. The benefits but with greater technology cost drinking water. The Agency believes Agency estimated and monetized the implications. Table VII–3 presents and the Advisory Committee concluded potential benefits from reduction in that the weight of evidence is enough to bladder cancers resulting from this rule. benefits estimates of the proposed Stage take regulatory action to help address Reductions in bladder cancer (including 2 DBPR using two population the potential reproductive and both fatal and non-fatal cases) provide a attributable risks derived from developmental endpoints in the Stage 2 range of annualized present value published studies (2% and 17%) and DBPR. However, data are not available benefits from $0 to $986 million using assuming there is a causal link between at this time to conduct a traditional a three percent discount rate ($0 to $854 DBP exposure and bladder cancer. In quantitative risk assessment. Instead, million using a seven percent discount subsection VII.G., Table VII–14 shows the benefits from reducing most rate) depending on the risk level potential benefits of all regulatory reproductive and developmental risks assumed. These estimates are based on alternatives considered by the Agency. are discussed qualitatively in this the assumption that the percent BILLING CODE 6560–50–P

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00081 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49628 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

It is important to note that the nonquantifiable health effects and other be found in the Stage 2 DBPR EA monetized benefits only reflect benefits potentially associated with DBP (USEPA 2003i). estimated benefits from reductions in regulation, monetized benefits estimates For additional perspective EPA used bladder cancer. As shown in subsection could be significantly higher than what updated cancer risk factors for four VII.C.1.and in Table VII–3, there may be is shown in the table. A complete DBPs for which we have toxicological significant nonquantifiable benefits discussion of how EPA calculated the data. Table III–3 (see section III of this associated with regulating DBPs in risks and the corresponding health preamble) shows the estimated pre- drinking water. Were EPA able to benefits potentially associated with Stage 2 concentrations of these four quantify some of the currently exposure to DBPs in drinking water can compounds and the estimated number

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00082 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.028 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49629

of people exposed to them. The Agency we believe it is important to provide avoided per year as a result of the Stage used these four DBPs to calculate an some quantitative indication of the 2 rule. alternative baseline number of annual potential risk suggested by some of the Caution is required in interpreting the pre-Stage 2 cancer cases. The published results on reproductive/ numbers because many experts calculations use the linearized developmental endpoints, despite the recommend that population attributable multistage model and predict 37 cases absence of certainty regarding a causal risk analysis should not be conducted for the ED10 risk factors and 87 cases for link between disinfection byproducts unless causality has been established. the LED10 risk factors. The ED10 risk and these risks. To do this, we have Causality has not been established factors (also known as the maximum adapted illustrative PAR calculations between exposure to disinfection likelihood estimate) are based on the from several studies on the relationship byproducts and fetal loss. The estimates estimated dose that the model predicts between chlorinated water exposure and presented here are not part of EPA’s will result in a carcinogenic response in fetal loss and applied these to national quantitative benefits analysis, and the 10 percent of the subjects, while LED10 statistics on annual incidence of fetal ranges are not meant to suggest upper risk factors correspond to the lower 95% loss. and lower bounds. Rather, they are confidence bound on the dose that the Specifically, we calculate the intended to illustrate quantitatively the model predicts will result in a unadjusted population attributable risk potential risk implications of some of carcinogenic response in 10% of the associated with each of the three the published results. subjects (LED10 is EPA’s more distinct population-based EPA has not monetized the value of conservative and more commonly used epidemiological studies of fetal loss potential reductions in fetal loss, but expression of toxicologically based published: Waller et al. 2001, King et al. recognizes that there is a significant cancer risk). Assuming that DBP risk 2000a, and Savitz et al. 1995. All three value associated with improvements in reductions for Stage 2 for the entire are high quality studies that have reproductive and developmental health. population average 4.2% (corresponding sufficient sample sizes and high In the absence of valuation studies to the reduction in average TTHM response rates, adjust for known specific to the health endpoints of levels), Stage 2 cancer cases avoided confounders 2, and have exposure concern, the Agency typically draws based on the toxicological data range assessment information from water upon existing studies of similar health from 1.7 to 4.0 cases per year. Section treatment data, residential histories, and endpoints to estimate benefits. The 5.2.2.2 of the Economic Analysis THM measurements. Because the ‘‘transfer’’ of the results of these studies (USEPA 2003i) presents a more detailed populations in these three studies to value similar health endpoints must basis for the derivation of these appear to have TTHM exposures be done carefully and methodically, estimates. It is important to note that significantly greater than those of the controlling for differences in the health these estimates do not include risks general U.S. population, we have endpoints and in the relevant from dermal or inhalation exposure nor chosen to scale the results using populations. Some researchers have do they account for many other DBPs (or Information Collection Rule data to attempted to transfer values using the mixture of DBPs seen in actual allow us to derive population sophisticated analytical techniques such PWSs) for which occurrence or attributable risks that may be more as preference calibration methods (e.g., toxicological risk data do not exist. relevant to the general U.S. population Smith et al. 2002). Regardless of the (USEPA 2003i). approach used, ‘‘benefit transfer’’ 1. Non-Quantifiable Health and Non- These three studies (using unadjusted requires systematic comparison of the Health Related Benefits data to allow for comparability, and differences in the health effects in the Although there are significant scaled to the TTHM levels reported in studies and those resulting from the monetized benefits that may result from the Information Collection Rule data regulation. Application of benefit this rule from the reduction in bladder base) yield median PARs of 0.4%, 1.7%, transfer leads to a detailed qualitative cancer, other important potential and 1.7% (with 95% confidence examination of the implications of using benefits of this rule are not quantified intervals for each of the studies of 0 to those studies and potentially to including potential reductions in 4%) 3. Using the prevalence of fetal loss empirical adjustments to the results of adverse reproductive and reported by CDC, the median PARs for the existing studies. developmental effects and other these three studies suggest that the The Agency is investigating further cancers. incidence of fetal loss attributable to work specific to the case of fetal loss The primary purpose of the Stage 2 exposure to chlorinated drinking water valuation. One possible area of further DBPR is to address potential adverse could range from 3,900 to 16,700 research is the value that prospective reproductive and developmental health annually. As part of the analysis to parents attach to reducing risks during effects that might be associated with evaluate potential reduction in fetal loss pregnancy. In this regard, the DBP exposure. EPA concludes that, ‘‘the for the Stage 2 DBPR, EPA assumed that substantial lifestyle changes that epidemiologic data, although not reductions in risk are proportional to prospective parents often undertake conclusive, are suggestive of potential the 28 percent reductions in the number during pregnancy suggests that reducing developmental, reproductive, or of locations having one or more these kinds of risks is of value. A second carcinogenic health effects in humans quarterly TTHM measurements that possible area of further investigation exposed to DBPs’’ (Simmons et al 2002). exceed the study population cut-offs would be work on benefit transfer EPA does not believe the available (>75 to >81 ug/l, depending on study). methodologies that address how evidence provides an adequate basis for This analysis implies that a range of existing studies can inform the quantifying potential reproductive/ 1,100 to 4,700 fetal losses could be estimation of the benefits of reduced developmental risks. Nevertheless, fetal loss. given the widespread nature of exposure 2 Use of unadjusted PAR estimates has the effect EPA has not monetized the potential to DBPs and the priority our society of removing the adjustments for known reductions in fetal loss. Without more places on reproductive/developmental confounders, however, EPA believes the unadjusted information and discussion on these estimates are adequate for purposes of the health, and the large number of fetal illustrative calculations presented here. subjects the Agency cannot fully losses experienced each year in the U.S. 3 The negative lower 95% confidence intervals for consider and describe the implications (nearly 1 million (Ventura et al. 2000)), all three studies was truncated at zero. of relying upon existing studies.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00083 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49630 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

However, research on valuation and December (4th quarter) and January (1st confidence intervals used to value benefit transfer continues to progress quarter) when the waters in the reductions in bladder cancer. and the Agency anticipates new distribution system are colder and DBP To calculate the total value of benefits research and future efforts to value formation generally lower. The derived from reductions in bladder reproductive and developmental proposed Stage 2 DBPR addresses this cancer cases as a result of the Stage 2 endpoints. issue by requiring that the samples must DBPR, a stream of estimated monetary EPA was also unable to quantify or be taken about 90 days apart. The benefits is calculated by combining the monetize the benefit from potential benefits of these provisions include the annual cases avoided with valuation reductions in other cancers, such as greater certainty that health protection inputs using Monte Carlo simulation. colon and rectal, that may result from is actually achieved because it is more Use of a Monte Carlo simulation allows this rule. Both toxicology and likely that a system’s high DBP levels the characterization of uncertainty epidemiology studies indicate that other will be identified. In addition, the rule around final modeling outputs based on cancers may be associated with DBP will reduce variability in the DBP levels the uncertainty underlying the various exposure but currently there is not throughout the distribution system, valuation inputs. The Stage 2 DBPR enough data to quantify or monetize ensuring greater equity in public health benefits model uses distributions of these cancer risks. protection. value of statistical life (VSL), Other potential non-health related willingness-to-pay (WTP), and income benefits not quantified or monetized in 2. Quantifiable Health Benefits elasticity values to attribute monetary today’s proposed rule include reduced Although DBPs in drinking water values (with uncertainty bounds) to the uncertainty about becoming ill from have been associated with non- number of bladder cancer cases avoided. consumption of DBPs in drinking water, cancerous health effects discussed Several of the inputs needed in the the ability for some treatment previously, the quantified benefits that benefit analysis, such as the VSL and technologies to eliminate or reduce result from today’s rule are associated WTP estimates, are based on older multiple contaminants, and monitoring only with estimated reductions in DBP- studies that were updated to current changes that will ensure that systems related bladder cancer. A complete dollar values. In addition, both the VSL can effectively measure their DBP levels discussion of risk assessment and WTP values are dependent on resulting in greater equity in protection methodology and assumptions can be income levels. Therefore, these values from DBPs. First, the reduced found in Chapter 5 of the Stage 2 DBPR also have to be adjusted for increases in uncertainty concept depends on several Economic Analysis (USEPA 2003i). real income growth from when the factors including consumer’s degree of Section III of this preamble also studies were conducted. The valuation risk aversion, their perceptions about discusses the health effects that have inputs and an explanation of the update drinking water quality (degree to which been associated with DBP exposure. factors used to bring these values to they will be affected by the regulatory The annualized present value benefits current price levels and discussed in the action), and the expected probability for reductions in bladder cancer that are following two sections. and severity of human health effects the result of today’s rule for both Valuation inputs. In order to monetize associated with DBPs in drinking water. community water system (CWS) and the benefit from the bladder cancer This effect could be positive or negative non-transient non-community water fatalities, EPA applied a VSL estimate to depending on whether knowledge of the systems (NTNCWSs) range from $0 to the cancer cases that result in mortality. rule decreases or increases their concern $986 million using a three percent EPA assumed a 26 percent mortality rate about DBPs in drinking water and discount rate ($0 to $854 million using for bladder cancer (USEPA 1999d). The potentially associated health effects. a seven percent discount rate). Overall, Agency uses a distribution of VSL Another nonquantified potential the Stage 2 DBPR may reduce on values which are based on 26 wage-risk benefit is the impact of technology average 0 to 182 bladder cancer cases studies. The mean VSL value from these selection to address DBPs on a system’s per year. studies is $4.8 million in 1990 dollars. ability to address other contaminants. The lower estimate of zero is included The mean value reflects the best For example, membrane technology because of inconsistent evidence estimate in the range of plausible values (depending on pore size), can be used to regarding the association between reflected by the 26 studies. A more lower DBP formation but it can also exposure to DBPs and cancer. The upper detailed discussion of these studies and remove other contaminants that EPA is estimate of monetized benefits and cases the VSL estimate can be found in EPA’s in the process of regulating or avoided is based on a population Guidelines for Preparing Economic considering regulating. Therefore, by attributable risk (PAR) of 17 percent. Analyses (USEPA 2000b). installing membrane technology, a Table VII–3 also presents monetized The VSL represents the value of system may not have to make new benefits based on a PAR value of 2%. reducing the risk of a premature death. capital improvement to comply with The PAR estimates are derived from an This valuation, however, does not take future regulations. analysis of five epidemiological studies into account the medical costs Last, today’s proposed rule makes which indicate that perhaps 2 to 17 associated with the period of illness changes to Stage 1 monitoring percent of bladder cancers may be (morbidity increment) leading up to a requirements. The IDSE monitoring attributable to DBP exposure. These death. In its review of the Arsenic Rule, provision of the proposed Stage 2 DBPR PAR estimates are described in more the Science Advisory Board (SAB) will help systems identify locations to detail in section III of today’s document. suggested that the appropriate measure conduct their routine monitoring to These are the same PAR values that EPA to use in valuing the avoidance of the capture high DBP occurrence levels. used in the Stage 1 DBPR benefits morbidity increment is the medical cost Also, the proposed Stage 2 DBPR will analysis, as discussed in the Regulatory attributable to a cancer case (USEPA prevent a system from conducting Impact Analysis for the Stage 1 DBPR 2001e). Based on available medical data, sampling designed to avoid monitoring (USEPA 1998f). Table VII–3 shows the EPA estimates the medical costs for a when DBP formation is generally higher. estimated benefits associated with fatal bladder cancer case to be $93,927 For example, the Stage 1 DBPR required bladder cancer reduction as a result of at a 1996 price level (USEPA 1999d). systems to take quarterly samples but the proposed rule. Table VII–4 This medical cost value (updated to samples could conceivably be taken in summarizes the mean, median and 2000 price levels) is applied as a point

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00084 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49631

estimate to each fatal case of bladder attributed to reducing the chances of tradeoff estimate and has a mean of cancer in the benefits model. facing a sudden death and are derived $587,500, standard deviation of A review of the available literature from the Magat et al. study (1996). $264,826, and a maximum value of $1.5 did not reveal any studies that Therefore, the Agency applies the 58.3 million at 1998 price values. specifically measured the WTP to avoid percent to the VSL distribution to derive Update factors. All valuation risks of contracting nonfatal cases of a range of value for non-fatal cancers bladder cancer. Instead, two alternates parameters must be updated to the same with a mean WTP value of $2.8 million price level so comparisons can be made were used, the WTP to avoid the risk of ($4.8 million * 58.3 percent) at a 1990 contracting a case of curable lymph in real terms. Values for VSL, WTP, and price level. The WTP for avoiding a case cancer (lymphoma) and the WTP to the morbidity increment used in the of chronic bronchitis is based on the avoid a case of chronic bronchitis. The model are updated based on adjustment SAB suggested this approach in their same methodology used for the Stage 1 factors derived from Bureau of Labor review of the Arsenic Rule (USEPA DBPR (see Stage 2 DBPR EA (USEPA Statistics (BLS) consumer price index 2001e). The median risk-risk trade-off 2003i) for a complete discussion). The (CPI) data so that each represents a year for a curable case of lymphoma was estimate is based on a lognormal 2000 price level. Table VII–4 equivalent to 58.3 percent of the risk distribution that uses the risk-dollar summarizes these updates.

Although the price level (year 2000) is The development of cancer due to individual risks become more reflective held constant throughout the benefits exposure to environmental carcinogens of the new lower exposures than the model, projections of benefits in future involves a complex set of processes that past higher exposures. years are subject to income elasticity are not well-understood for most Recently, the Arsenic Rule Benefits adjustments. Income elasticity specific substances. In general, however, Review Panel of the EPA Science adjustments represent changes in the development of cancer involves Advisory Board (SAB) addressed this valuation in relation to changes in real some time period, usually referred to as issue in detail and provided some income. For fatal cancers, the Agency the latency period, between the initial guidance for computing benefits to used a triangular distribution with a exposure and the manifestation of account for this transition period central estimate of 0.40 (low end: 0.08; disease. Defining a latency period is between higher and lower steady-state highly uncertain because the mode of high end:1.00) to represent the risks (USEPA 2003s). The Arsenic Rule action for most chemical contaminants uncertainty of the income elasticity Benefits Review Panel coined the term are poorly understood. Latency periods value. For non-fatal cancers, the Agency ‘‘cessation-lag’’ to emphasize the focus in humans often involve many years, uses a triangular distribution with a on the timing of the attenuation of risk even decades. central estimate of 0.45 (low end: 0.25; after reduction in exposures to avoid high end: 0.60). These distributions are EPA recognizes that despite confusion with the more traditional used as assumptions in the Monte Carlo uncertainties in the latency period term of ‘‘latency’’ that reflects the simulation to further characterize associated with different types of increased risk 4 from the time of initial uncertainty in benefits estimates. carcinogens, it is unlikely that all cancer exposure. reduction benefits would be realized In order to apply the income elasticity immediately upon exposure reduction. 4 SAB included the following in its report on values in the model, they are combined If it is assumed that lower risk is arsenic to emphasize this difference: ‘‘An important with projections of real income growth attained immediately upon reduction in point is that the time to benefits from reducing over the time frame for analysis. arsenic in drinking water may not equal the exposure, this would tend to estimated time since first exposure to an adverse Population and real gross domestic overestimate the benefits. On the other effect. A good example is cigarette smoking: the product (GDP) projections are combined hand, assuming that no risk reduction latency between initiation of exposure and an to calculate per-capita real GDP values. occurs for some period of time following increase in lung cancer risk is approximately 20 A more detailed discussion of these years. However, after cessation of exposure, risk for exposure reduction may lead to an lung cancer begins to decline rather quickly. A adjustments is in Chapter 5 of the EA underestimation of the benefits. There benefits analysis of smoking cessation programs (USEPA 2003i). will likely be some transition period as Continued

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00085 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.011 49632 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

Although the focus of the cessation indicator to characterize the influence of disinfected water responsible for the lag discussion in the SAB review was on cessation lag in calculating benefits. The observed effect have not been identified. reducing levels of arsenic in drinking carcinogen for which the most extensive The use of the tobacco smoke cessation water, much of their consideration of database was available for lag model reflecting a mixture of this issue has more general applications characterizing cessation lag was for initiators and promoters would be beyond just the arsenic issue at hand. In cigarette smoking. EPA examined expected to attenuate a possible bias in particular, SAB noted the following: several extensive epidemiological either direction if the DBPs responsible • The same model should be used to studies on the comparison of the risks for bladder cancer are acting estimate the time pattern of exposure of adverse health effects, including lung predominately as either initiators or and response as is used to estimate the cancer, for smokers and former smokers. promoters. potency of the carcinogen. EPA selected the Hrubek and Another factor to consider is that the • If possible, information about the McLaughlin (1997) study as the most cessation lag model used is based upon mechanism by which cancer occurs appropriate study for development of a exposure to tobacco smoke where lung should be used in estimating the statistical model of disease response to cancer is the end-point but is being cessation lag (noting that late-stage smoking cessation. This was a applied to exposure to disinfection by- mechanisms in cancer formation imply comprehensive study involving a 26- products where the end-point is bladder a shorter cessation lag than early stage year follow-up of almost 300,000 U.S. cancer. Of concern here is that there is mechanisms). male military veterans. More detail a more direct correlation between • If specific data are not available for about this study and how it is applied inhalation and the site of cancer for characterizing the cessation lag, an to estimate the cessation lag can be smoking than there is for ingestion and upper bound for benefits can be found in Chapter 5 of the EA (USEPA inhalation of drinking water and the provided based on the assumption of 2003i) and the cessation lag document sites of cancer for DBP exposure. immediately attaining steady-state (USEPA 2003s). Unfortunately, EPA does not have data results. The smoking cessation lag data imply on which to develop a cessation lag • In the absence of specific cessation that the majority of the potential steady model using data specific to how lag data, other models should be state cases avoided occur within the changes in DBP exposures affect the considered to examine the influence of first several years, but with diminishing risks of developing bladder cancer. the lag. incremental increases in later years. For Another divergence, and perhaps the Following the release of the SAB example, the cessation lag model most important, between the smoking report on arsenic, EPA initiated an effort indicates that approximately 40 percent model and the DBP application is that to explore approaches to including the of the steady-state cases avoided are the smoking model is based on complete cessation lag in modeling risk reduction achieved by the end of the second year, cessation of exposure, whereas in the and calculating benefits for the arsenic with 70 percent achieved by the end of case of DBP exposure is only being regulation. EPA recognized, however, the fifth year, and approximately 80 reduced. In some water systems the that the concept of cessation lag is not percent by the tenth year. By the reduction is only 10 percent, whereas in only applicable to arsenic but to other twentieth year, 90 percent of the steady others it may be as high as 60 percent, drinking water contaminants having a state cases are avoided. with an average of approximately 30%. cancer end-point as well. EPA recognizes that there are several This moderate reduction in exposure In response to the SAB cessation lag factors that contribute to the uncertainty may prevent full DNA repair, which recommendations, EPA has: in the application of the specific some scientists interpret as the basis for • Conducted a study using data on cessation lag model used in the the short cessation lag associated with lung cancer risk reductions following estimation of the benefits of the smoking. cessation of smoking that resulted in the proposed Stage 2 regulation. A key Currently, smoking is the only January 2003 report Arsenic in Drinking factor to consider in assessing this contaminant for which enough data Water: Cessation Lag Model (USEPA impact is the likely mode of action of exist to estimate a cessation lag. In the 2003s). DBPs in eliciting bladder cancer versus absence of a reliable cessation lag model • Conducted an expert scientific peer the mode of action of tobacco smoke in based specifically on DBPs and bladder review of that draft report. producing lung cancer, and in particular cancer, EPA used the cessation lag • Initiated development of general whether they behave as initiators or model based on smoking to provide a criteria for incorporating cessation lag promoters of the carcinogenic process. means of estimating the rate at which modeling in benefits analyses for other As discussed in the SAB report and the bladder cancer risk in the exposed drinking water regulations. EPA Cessation Lag report (USEPA population falls from the pre-Stage 2 In the effort to develop a cessation lag 2001e, USEPA 2003s), carcinogens that levels to the post-Stage 2 levels. model specific to DBPs, EPA reviewed act solely or primarily as initiators However, this model is derived from the available epidemiological literature would tend to show a longer cessation data involving notable differences from for information relating to the timing of lag (lower rate of risk reduction DBPs in drinking water, including exposure and response, but could not following reductions in exposure) than different cancer sites (lung versus identify any studies that were adequate, carcinogens that act solely or primarily bladder), different exposure pathways alone or in combination, to support a as promoters. The available information (inhalation versus a combination of specific cessation lag model for DBPs in on tobacco smoke and lung cancer ingestion, inhalation and dermal), drinking water. Thus, in keeping with suggests that it involves a mixture of different risk levels, and, perhaps most the SAB recommendation to consider both initiators and promoters, and importantly, complete cessation for other models in the absence of specific therefore the cessation lag derived from smoking versus small exposure cessation lag information, EPA explored smoking data is expected to reflect the decreases for DBPs. For these reasons, the use of information on other combined influence of these divergent the extent to which the smoking / lung carcinogens that could be used as a mechanisms. There are no data available cancer model is directly transferable to on the mechanism of action for DBPs DBP / bladder cancer is uncertain. It is based on the observed latency would greatly and bladder cancer; indeed the specific not possible to know, however, whether underestimate the actual benefits.’’ carcinogenic agent(s) present in and to what degree the tobacco smoke

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00086 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49633

cessation lag model either over-states or as a result of cessation lag effects, there methodology used to estimate the under-states the rate at which is a delay in exposure reduction quantified benefits of the Stage 1 DBPR. population risk reduction for bladder resulting from the Stage 2 DBPR cancer occurs following DBP exposure implementation. In general, EPA D. Costs of the Proposed Stage 2 DBPR reductions. assumes that a fairly uniform increment In estimating the costs of today’s EPA is currently examining the of systems will complete installation of proposed rule, the Agency considered recently published meta-analysis by new treatment technologies each year, impacts on water systems (CWSs and Villanueva et al. (2003) to determine if with the last systems installing NTNCWSs) and on States (including the information provided on increases treatment by 2013. EPA recognizes that territories and EPA implementation in in risk as a function of duration of more systems may start in early or later non-primacy States). EPA assumed that exposure can provide any insight on years, but believes that a uniform systems would be in compliance with how reductions in risk over time might schedule is a reasonable assumption. the Stage 1 DBPR, which has a occur following reductions in exposure. Appendix D of the EA presents detailed compliance date of January 2004 for Villanueva et al. (2003) demonstrated information regarding the rule activity ground water systems and small surface that the risk associated with chlorinated schedule assumptions (USEPA 2003i). water systems and January 2002 for drinking water and bladder cancer are The delay in exposure reduction large surface water systems. Therefore, related to exposure duration. resulting from the rule implementation the cost estimate only considers the Specifically, they estimated a unit schedule is incorporated into the additional requirements that are a direct increase in the odds ratio of 1.006 per benefits model by adjusting the result of the Stage 2 DBPR. More year (95% CI of 1.004 to 1.009). The cessation lag weighting factor. For detailed information on cost estimates model suggests a cumulative odds ratio example, if ten percent of systems are described later and a complete of 1.13 after 20 years of exposure (95% install treatment equipment (and start discussion can be found in Chapter 6 of CI of 1.08 to 1.20), and 1.27 (95% CI of realizing reductions in cancer cases) in the Stage 2 DBPR EA (USEPA 2003i) 1.17 to 1.43) after 40 years. This result year one, only that portion of the cases is consistent with most of the individual are modeled to begin the cessation lag 1. National cost estimates studies which do not show statistically equilibrium process in that year. Thus, significant risk increases until at least the resulting ‘‘weighted weighting EPA estimates that the mean 30–40 years of exposure. However, these factor’’ is higher relative to the base annualized cost of the proposed rule studies provide indirect evidence only factor. Appendix E in the EA (USEPA ranges from approximately $59.1 about the latency of potential effects. 2003i) presents detailed breakdowns of million using a three percent discount For perspective, it is important to note all weighting factor adjustments and rate to $64.6 million using a seven that the latency between initiation of resulting cancer cases avoided, by year, percent discount rate. Drinking water exposure and an increase in lung cancer for each rule alternative based on the utilities will incur approximately 98 risk is approximately 20 years. As noted application of the cessation lag percent of the rule’s costs. States will above, latency is not the same as the methodology. incur the remaining rule cost. Tables cessation lag. EPA is requesting VII–5 a and b summarize the total comment on (a) the potential 3. Benefit Sensitivity Analyses annualized cost estimates for the application of the Villanueva et al. The Agency performed one other proposed Stage 2 DBPR. In addition to (2003) model to estimate reductions in benefit sensitivity analysis which is mean estimates of costs, the Agency bladder cancer risk that might included in the EA to allow for calculated 90 percent confidence accompany decreased exposure to DBPs comparison with the benefit estimates bounds by considering the uncertainty as a result of the Stage 2 Rule; (b) the calculated for the Stage 1 DBPR. This around the mean unit technology costs. advantages and disadvantages of using analysis assumes that there is not a Table VII–6 shows the undiscounted the current approach—i.e., application cessation lag or latency adjustment capital cost and all one-time costs of the smoking cessation lag model; and associated with bladder cancer broken out by rule component. A table (c) suggestions for alternative data sets reductions that result from the rule. In comparing total annualized costs among or approaches to characterize cessation this case, the analysis assumes that the the regulatory alternatives considered lag. steady state reduction in bladder cancer by the Agency is located in subsection In addition to the delay in reaching a occurs immediately with rule VII.G. new steady-state level of risk reduction implementation. This is the same BILLING CODE 6560–50–P

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00087 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49634 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00088 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18au03.012 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49635

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00089 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18au03.014 49636 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00090 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18au03.013 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49637

2. Water system costs baseline number of ground water plants being used to estimate monitoring costs. The proposed Stage 2 DBPR applies to is larger than that of surface water Section V describes the population- all community or nontransient plants, so there is a larger number of based approach in more detail and a noncommunity water systems that add ground water plants adding treatment. discussion of how this approach may a chemical disinfectant other than UV or Subsection VII.F. provides a more influence costs is provided in Appendix distribute water that has been treated detailed explanation of treatment H of the EA (USEPA 2003i). A small with a disinfectant other than UV. EPA changes that may occur as a result of the percentage of systems (approximately has estimated the cost impacts for both proposed rule. 3.0 percent of surface water CWSs and types of public water systems. As shown All systems will incur costs for rule 0 percent of ground water systems) are in Tables VII–5 a and b, the total implementation. Some will need to expected to experience significant annualized present value costs for CWSs conduct a one-time Initial Distribution excursions. is approximately $55.8 million and for System Evaluation (IDSE) and others (a different subgroup depending on the A complete discussion of the rule NTNCWSs, $2.2 million, using a three provisions is located in section V of this percent discount rate ($60.8 million and system size) may incur additional costs preamble; the Stage 2 DBPR Economic $2.2 million using a seven percent for routine DBP monitoring. Some Analysis includes a complete analysis of discount rate). systems may also have to conduct a Although the number of systems peak excursion evaluation if single rule impacts (USEPA 2003i). Table VII– adding treatment is small, treatment samples indicate high DBP levels. 8 summarizes the number of systems costs make up a significant portion of Sixty-nine percent of surface water subject to non-treatment related rule the total costs of the rule (more than 75 and 7 percent of ground water CWSs are activities. Column D indicates the percent of total rule costs). Table VII–7 predicted to conduct the IDSE number of systems expected to use the shows the baseline number of plants monitoring. EPA estimates that a very standard monitoring program to and the estimated percent of those small portion of systems (approximately implement the IDSE. Column F plants adding treatment. The estimated 16 percent overall) will conduct indicates the number of systems percent of plants adding advanced additional routine monitoring beyond expected to increase monitoring sites treatment or converting to chloramines the Stage 1 DBPR requirements. beyond that required by Stage 1. The is 2.8 percent of all systems. A higher However, fewer samples overall would last two columns show the number and percentage of surface water plants are be required if a population-based percent of plants estimated to predicted to add treatment compared to approach is implemented instead of the experience significant excursions each ground water plants. However, the plant-based approach that is currently year.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00091 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49638 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00092 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18au03.015 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49639

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00093 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18au03.016 49640 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

In addition to using distributions to the EA, USEPA 2003i, for details of this systems. EPA estimated that the mean develop unit cost estimates, the Agency analysis). Table VII–9 lists the high-end annualized costs at the 3% discount rate conducted sensitivity analyses to further estimates of the number of systems could be as high as $77.5 million (IDSE explore uncertainty regarding system adding treatment in IDSE sensitivity Sensitivity Analysis No. 1) or $108.8 compliance estimates. The first two analyses No. 1 and No. 2. For both IDSE million (IDSE Sensitivity Analysis No. sensitivity analyses were prepared to sensitivity analyses, only small 2) versus the Preferred Alternative evaluate the possibility that the IDSE additional impacts were assumed analysis estimate of $57.4 million. At monitoring requirement will result in possible for systems serving 10,000 the 7% discount rate these estimates more systems needing to install people or fewer because such systems would respectively correspond to $86.1 treatment beyond what is predicted in generally have much less complicated million, $120.7 million, and $63.3 the current cost model (see chapter 7 of distribution systems than larger million.

EPA believes that the percentage of DBPR, as many are expected to do, cost impact to States, EPA considered systems estimated to add treatment calculation of compliance will produce initial implementation costs, costs for under IDSE sensitivity analyses No. 1 the same results for the running annual assisting systems in evaluating IDSE and No. 2 are overestimates and that the average (RAA) and locational running information, and for annual rule estimate for the Preferred Alternative is annual average (LRAA) measure, implementation activities. EPA likely to already capture the influence of implying that they are not likely to add considered the incremental change in the IDSE because of the conservative treatment for the Stage 2 DBPR if they activities that result from the Stage 2 assumptions used in the analysis. For comply with the Stage 1 DBPR. DBPR. For example, States may have to example, the compliance forecast EPA conducted a third sensitivity update their databases to track the new analysis assumes that systems will try to analysis to evaluate the possibility that Stage 2 DBPR monitoring strategy but meet the LRAA MCLs with a 20% small systems will continue to monitor could modify the system they developed margin of safety. Systems complying by at one point in their distribution system. for the Stage 1 DBPR. EPA accounted for switching to chloramines may choose to In this sensitivity analysis, EPA the cost of a Stage 1 DBPR database in meet the new MCLs with a much assumed that no surface water plants the Stage 1 Regulatory Impact Analysis smaller margin of safety since serving fewer than 10,000 people and no (USEPA 1998f). State costs are not chloramines dampen the variability of ground water plants would add expected to change dramatically DBP concentrations within the treatment to meet Stage 2 DBPR between alternatives. distribution system. Furthermore, EPA requirements (i.e., only costs are 4. Non-quantifiable believes that the number of ground associated for large surface water water and small surface water systems systems). Under this analysis, the EPA has identified and quantified adding chloramines or changing average cost figures are reduced costs that it believes are likely to be technology in the baseline analysis may dramatically from $57.4 million or $63.3 significant. In some instances, EPA did be overestimated because their million to $22.9 million or $25.7 million not include a potential cost element monitoring requirements are expected to using a 3 percent or 7 percent discount because it believes the effects are be very similar from Stage 1 to Stage 2. rate, respectively, for the Preferred relatively minor and difficult to The Stage 1 DBPR required only one Regulatory Alternative. Chapter 7 of the estimate. For example, the Stage 2 DBPR compliance monitoring location (at the Economic Analysis (USEPA 2003i) may be the determining factor in the point of maximum residence time) for contains a detailed explanation of the decision by some small water systems to producing surface water systems serving aforementioned sensitivity analysis. merge with neighboring systems. Such between 500 and 10,000 people and for changes have both costs (legal fees and all ground water systems. The Stage 2 3. State Costs connecting infrastructure) and benefits DBPR requires that these systems add an The Agency estimates that the States (economies of scale). Likewise, costs for additional site if they determine that and primacy agencies will incur an procuring a new source of water would their high TTHM and high HAA5 annualized present value cost of $1.1 have costs for new infrastructure but concentrations do not occur at the same million to $1.5 million (using a three could result in lower treatment costs. location. If systems maintain a single percent and seven percent discount rate, Also, EPA was unable to quantify monitoring location for the Stage 2 respectively). In order to estimate the several distribution system-related

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00094 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18au03.017 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49641

changes that can reduce TTHM and characterize water quality and treatment 2. Predicted Technology Distributions HAA5 levels. Activities such as looping practices for the nation’s public water Post-Stage 2 DBPR distribution systems and optimizing systems. EPA also used information in The treatment compliance forecast for storage can minimize retention times the Water Industry Baseline Handbook the Stage 2 DBPR has two components— and help to control DBP formation. (USEPA 2000j) to develop the industry 1) the percent of plants that must add Costs for these activities range from profile. The Baseline Handbook uses treatment to comply with Stage 2 DBPR almost zero (modifying retention time) data derived from the 1995 Community requirements, and 2) the treatment to more substantial costs for modifying Water Systems Survey and the Safe technologies these plants are predicted distribution systems. In the absence of Drinking Water Information System to to select. This information, coupled detailed information needed to make characterize the U.S. drinking water with the baseline data discussed before, cost evaluations for situations such as systems. Another EPA study, provides an estimate of the total number these, EPA has included a discussion of Geometries and Characteristics of Water of plants using specific technologies to possible effects where appropriate. Systems Report (USEPA 2000k), also meet the requirements of the proposed E. Expected System Treatment Changes provided information for the industry Stage 2 DBPR. National costs are then In order to quantify the effects of the profile. generated using technology unit cost Stage 2 DBPR, it is necessary to predict EPA developed and used a model information. how plants will modify their treatment (SWAT) to characterize treatment The four step process EPA used to processes to meet the proposed following the Stage 1 DBPR and Stage 2 develop a Stage 2 DBPR compliance requirements. To estimate the DBPR options considered. SWAT served forecast is summarized in table VII–10. The difference between the Stage 1 incremental impacts of the Stage 2 as the primary tool to predict changes in DBPR Technology Selections and Stage DBPR, relative to the Stage 1 DBPR, EPA treatment and DBP occurrence. The compared predicted ‘‘ending 2 DBPR Technology Selections (Step 4— model used a series of algorithms and Incremental Technology Selections) was technologies’’ (types of treatment in use decision rules to predict the type of after implementation of the Stage 2 used to develop national cost estimates treatment a large surface water plant for today’s proposed rule. Tables VII–11 DBPR) to the distribution of baseline will use given a specific regulatory technologies predicted to be in place a and b (surface water) and VII–12 a and alternative and source water quality. b (ground water) show the incremental after the implementation of the Stage 1 Other tools were used to estimate DBPR. This subsection outlines the technology selections shown as the practices at large ground water systems process for deriving baseline and ending percent change between Stage 1 and or any medium or small systems. A Stage 2 technology distributions that are Stage 2 DBP rules. Delphi process (a detailed technical the basis for the national cost estimates of today’s proposed rule. treatment characterization and DBP TABLE VII–10.—STAGE 2 DBPR occurrence review by drinking water COMPLIANCE FORECAST SUMMARY 1. Pre-Stage 2 DBPR Baseline Conditions experts) was used to predict treatment Development of the Pre-Stage 2 changes for large ground water systems Step Description of Step baseline (i.e., conditions following the (those serving 10,000 or more people). The results of the SWAT analyses and 1 ...... Model a pre-Stage 1 baseline sce- Stage 1 DBPR) consists of the following nario using Information Collection processes: the Delphi process were extrapolated to Rule data to allow consistent com- • Compiling an industry profile— the medium surface water and ground parison between different rule al- identifying and collecting information water systems based on analysis of ternatives. on the segment(s) of the water supply source water treatment characteristics 2 ...... Model technology selection to meet industry subject to the Stage 2 DBPR; and treatment decision trees. For the Stage 1 DBPR requirements • Characterizing influent water small surface and ground water systems (Stage 1 DBPR Technology Selec- tion). quality—summarizing the relevant analyses, a group of experts provided 3 ...... Model technology selection to meet characteristics of the raw water treated predictions for a pre-Stage 2 baseline Stage 2 DBPR requirements by the industry; and and resulting treatment and water (Stage 2 DBPR Technology Selec- • Characterizing treatment for the quality conditions under the Stage 2 tion). Stage 1 DBPR—predicting what the DBPR regulatory alternatives. A detailed 4 ...... Subtract the results in Step 2 from industry will do to comply with the description of these analyses can be Step 3 and adjust to obtain the in- provisions of the Stage 1 DBPR. cremental impact of an alternative found in the Economic Analysis for the (Stage 2 DBPR incremental tech- Section IV of this document details Stage 2 DBPR (USEPA 2003i). nology selection). the data sources EPA used to

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00095 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49642 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00096 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18au03.018 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49643

F. Estimated Household Costs of the estimated to be $8.38 for those systems cost is $0.51. The last two columns of Proposed Rule that need to install technology to Table VII–13 show the potential impact This analysis considers the potential comply with this rule. Table VII–13 as the percent of households that will increase in a household’s water bill if a shows the range of household costs for incur either less than a $1 or less than system passed the entire cost increase all surface and ground water systems a $10 increase in their monthly water resulting from this rule on to their subject to the rule and also only for bills (shown in the table as annual customers. It is a tool to gauge potential those systems installing technology to values). For systems adding treatment, impacts and should not be construed as comply with this rule. For all systems, 84% of households will face less than precise estimates of potential changes to including those that may not have to a $1 increase in their monthly bill, individual water bills. take any additional action to comply while 99% are expected to face less than Overall, the potential increase in with this rule but are still subject to its a $10 increase. mean annual water bill per household is provisions, the mean annual household

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00097 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18au03.019 49644 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

Both household cost estimates reflect compliance forecast. For example, the Also, more small systems than costs for rule implementation (e.g., system may identify another water projected in the primary analysis may reading and understanding the rule), source that may form lower levels of already be in compliance with Stage 2 IDSE, additional routine monitoring, TTHM and HAA5. Systems that can DBPR. A sensitivity analysis discussed and treatment changes. Although identify such an alternate water source in the subsection VII.D.2 describes this implementation and the IDSE represent may not have to treat that water as much issue in more detail. Also, certain relatively small, one-time costs, they as their current source, resulting in technologies installed to treat DBPs may have been annualized and included in lower treatment costs that may offset the treat many other contaminants thus the analysis to provide a complete costs of obtaining water from the eliminating the need to install picture of household costs. alternate source. Systems may also be additional equipment to comply with Overall, EPA estimates that 99 percent able to connect to a neighboring water future drinking water regulations. of the 98 million households that are system. While connecting to another G. Incremental Costs and Benefits of the provided disinfected drinking water system may not be feasible for some Proposed Stage 2 DBPR would face less than $1 increase in their remote systems, EPA estimates that monthly water bill. Approximately 86 more than 22 percent of all small water Incremental costs and benefits are percent of the households impacted by systems are located within metropolitan those that are incurred or realized in the rule are served by systems serving regions (USEPA 2000c) where distances reducing DBP exposures from one at least 10,000 people; these systems between potential connecting water alternative to the next more stringent experience the lowest increases in costs systems may not present a prohibitive alternative. Estimates of incremental due to significant economies of scale. barrier. Consolidation was not an costs and benefits are useful in Households served by small systems element used in developing the considering the economic efficiency of that install advanced technologies will compliance forecasts for small systems. different regulatory options considered face the greatest increases in annual Costs for consolidation may be either by the Agency. However, as pointed out costs. The cumulative distributions of greater or less than the costs for by the Environmental Economics household costs for all systems are changing technologies, and Advisory Committee of the Science presented in the Economic Analysis consolidation may have other benefits Advisory Board, efficiency is not the (USEPA 2003i). (e.g., lower costs for compliance with only appropriate criterion for social When interpreting the results of the future regulations). In addition, decision making (USEPA 2000n). household cost analysis, it is important potentially lower cost alternatives such Generally, the goal of an incremental to remember that systems, especially as controlling water residence time in analysis is to identify the regulatory small systems, may have other options the distribution systems were not option where net social benefits are that were not included in the included in the compliance forecast. maximized. If net incremental benefits

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00098 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18au03.020 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49645

are positive, society is incurring greater incremental cost of achieving that estimated costs and benefits for each costs as a result of the health damages reduction. However, the usefulness of alternative. Evaluation of the compared to the costs society could pay this analysis is constrained when major incremental changes between different to reduce those health damages (i.e. benefits and/or costs are unquantified or rows in the tables shows that society would be better off to invest not monetized. incremental costs generally fall within more in controlling the health damage). For the proposed Stage 2 DBPR, the range of incremental benefits for If net incremental benefits are negative, presentation of incremental quantitative each more stringent alternative. Equally than the cost of the additional control is benefit and cost comparisons may be important, the addition of any benefits higher than the value of the additional unrepresentative of the true net benefits attributable to the non-quantified health damages avoided. Therefore, the of the rule because a significant portion categories would add to the benefits ‘‘efficient’’ regulatory level is where the of the rule’s potential benefits are non- without any increase in costs. next additional incremental reduction quantifiable (see section C.1). Tables in health damages equals the VII–14 and VII–15 show the total

TABLE VII–14.—TOTAL ANNUALIZED PRESENT VALUE COSTS BY RULE ALTERNATIVE ($millions, 2000$)

Total annualized cost ($millions) 3 Percent discount rate 7 Percent discount rate Rule alternative 90 Percent confidence bound 90 Percent confidence bound

Mean estimate Lower (5th % Upper (95th % Mean estimate Lower (5th % Upper (95th % tile) tile) tile) tile)

Preferred ...... $59.1 $54.3 $63.9 $64.6 $59.2 $70.0 Alt. 1...... 182.2 165.1 199.6 195.1 175.9 214.3 Alt. 2...... 409.6 383.6 435.7 442.7 413.4 472.2 Alt. 3...... 594.3 556.3 631.9 644.2 601.1 686.9 Note: Costs represent values in millions of 2000 dollars. Estimates are discounted to 2003—90 percent Confidence Intervals reflect uncertainty in technology unit cost estimates Source: Economic Analysis (USEPA 2003i) exhibit 6.24

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00099 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 49646 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

The range of quantified benefits 2 and 3. However, the associated costs presented in Table VII–14 show values increases significantly with Alternatives also increase significantly—cost figures approaching or exceeding $500 million

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00100 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.029 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49647

per year. Although the estimated brominated DBP species, should K. Uncertainties in Baseline, Risk, benefits for Alternatives 2 and 3 are decrease as a result of this rule. Benefit, and Cost Estimates potentially significant, EPA rejected EPA’s analysis shows that a large Today’s proposal models the current these alternatives because the Agency portion of systems that do not currently baseline risk from DBP exposure as well believes that the uncertainty about the meet Stage 2 requirements will do so by as the reduction in risk and the cost for health effects data does not warrant the switching from chlorination to various rule options. There is additional expense associated with chloramination; approximately 5% of uncertainty regarding many aspects of these regulatory alternatives. surface water plants and 1.3% of ground this analysis including the risk Given the uncertainty in the health water plants in systems serving greater calculation, the benefit estimate, and the effects, and the resulting rejection of than 10,000 are estimated to convert to cost estimates. EPA has tried to capture Alternatives 2 and 3, a comparison of chloramination in order to comply with much of the uncertainty and also the Alternative 1 with the Preferred the Stage 2 DBPR from the Stage 1 DBPR variability associated with many of the Alternative shows that Alternative 1 (USEPA 2003i). A potential would have approximately the same inputs used in the economic analysis by chloramination byproduct is N- using distributions or ranges as model benefits as the Preferred Alternative but nitrosodimethylamine (NDMA), a with greater costs. This results from the inputs instead of point estimates probable human carcinogen. The whenever possible. The Stage 2 DBPR inability of the Agency to estimate the concern over the formation of NDMA in additional benefits of reducing the EA contains a more extensive the treatment process is based on the discussion of the modeling techniques bromate MCL. Alternative 1 was also compound’s ability to persist for a long determined to be unacceptable due to used to address uncertainty and period of time in the distribution variability (USEPA 2003i). the potential for increased risk of system. The mechanism of formation of microbial exposure. See section VII.A of In addition, the Agency conducted NDMA, however, is still under sensitivity analyses to address today’s action for a description of examination. A number of ongoing regulatory alternatives. uncertainty. The sensitivity analyses studies will also evaluate occurrence, focus on various benefit and cost factors H. Benefits From the Reduction of Co- factors that affect NDMA formation, that may have a significant influence on Occurring Contaminants mechanisms, treatment effectiveness the outcome of the rule. All of these Installing certain technologies to and improved analytical methods for sensitivity analyses are explained in control DBPs also has the added benefit measuring NDMA. more detail in the EA for the Stage 2 of controlling other drinking water Another contaminant of concern to DBPR (USEPA 2003i). contaminants. For example, some the Agency is chlorite. Levels may The major source of benefit membrane technologies (depending on increase slightly because of technology uncertainty is the scientific uncertainty pore size) installed to reduce DBP shifts to chlorine dioxide resulting from regarding the impact of DBP exposure precursors can also reduce or eliminate this rule but very few systems (<0.1 on reproductive and developmental many other drinking water percent) are predicted to install this outcomes. However, the Agency contaminants, including arsenic and technology. However, individual believes that the monetized value of microbial pathogens. EPA has finalized systems will not shift to chlorine these outcomes could be significant. As a rule to further control arsenic level in dioxide unless they can meet the discussed in subsection VII.C.1, EPA drinking water and has proposed the chlorite MCL (established under the performed an illustrative calculation Ground Water Rule to address microbial Stage 1 DBPR) which is considered that explored the potential implications contamination. The Stage 2 DBPR is also protective of public health. for the proposed rule using some of the being concurrently proposed with the EPA also considered the impact this published results on fetal loss, but did Long Term 2 Enhanced Surface Water rule may have on microbial not attempt to quantify benefits Treatment Rule. Because of the contamination that may result from associated with reducing other difficulties in establishing which altering disinfection practices. To reproductive and developmental systems would have multiple problems address this concern, the Agency endpoints potentially associated with such as microbial contamination, developed this rule jointly with the DBP exposure. arsenic, and DBPs (or any combination Long Term 2 Enhanced Surface Water Another possible underestimation of of the three), no estimate was made of Treatment Rule (LT2ESWTR). EPA today’s monetized benefits results from the potential cost savings from expects that the LT2ESWTR provisions the inability of the Agency to quantify addressing more than one contaminant will prevent significant increases in or monetize the potential benefit from simultaneously. microbial risk resulting from the Stage avoiding other cancers associated with 2 DBPR. EPA also expects the Ground DBP exposure such as colon and rectal I. Are There Increased Risks From Other Water Rule, scheduled for promulgation cancers. Furthermore, while the Agency Contaminants? in 2003, to prevent any increases in estimated the range of bladder cancer Today’s proposed rule may slightly microbial risk in ground water systems risks avoided to be 0 to 182 cases per shift the distribution of TTHM and deemed vulnerable to source water year, the true risk of bladder cancer HAAs to brominated species. Some contamination. avoided from decreased DBP exposure systems, depending on bromide and may be higher than this range. J. Effects on General Population and organic precursor levels in the source While EPA believes it has accounted Subpopulation Groups water and technology selection, may for the significant costs of today’s experience a shift to higher ratios or Section III of today’s proposed rule proposed rule, there are uncertainties concentrations of brominated DBPs discusses the health effects associated about some of the cost inputs. As while the overall TTHM or HAA5 with DBPs on the general population as discussed in subsection VII.D.4, cost concentration decreases. However, EPA well as the effects on pregnant women estimates do not include some anticipates that this phenomenon may and fetuses. In addition, health effects alternatives to installing treatment (e.g., only occur in a small percentage of associated with children and pregnant improving management of distribution systems affected. For most systems, women are discussed in greater detail in system residence time) that may be a overall levels of DBPs, as well as subsection VIII.G of this preamble. less costly means of complying with the

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00101 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49648 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

Stage 2 DBPR. The Agency also rule to the annual number of bladder President’s priorities, or the principles explored two additional uncertainties cancer cases potentially avoided. For set forth in the Executive Order. which might have the greatest impact on bladder cancer reduction, the cost per Pursuant to the terms of Executive our current estimates by conducting case avoided for the proposed rule Order 12866, it has been determined sensitivity analyses. These include the would be $0.3 million if the PAR is that this rule is a ‘‘significant regulatory impact of IDSE monitoring and the 17%, and $3.1 million if the PAR is 2%, action.’’ As such, this action was possibility that the primary analysis and also varies depending on the submitted to OMB for review. Changes overestimates the compliance forecast discount rate used. made in response to OMB suggestions or for small surface water systems and all recommendations will be documented M. Request for Comment ground water systems. A detailed in the public record. The Agency requests comment on all discussion of these analyses can be B. Paperwork Reduction Act found in chapter 7 of the Economic aspects of the rule’s economic impact Analysis (USEPA 2003i). analysis. Specifically, EPA seeks input The information collection Last, EPA has recently proposed or into the following issues: (1) To what requirements in this proposed rule have finalized new regulations for arsenic, extent can systems install treatment to been submitted for approval to the radon, and microbials in ground water address multiple contaminants?; (2) Are Office of Management and Budget systems (Ground Water Rule); there methods for monetizing potential (OMB) under the Paperwork Reduction Cryptosporidium in small surface water reproductive and developmental Act, 44 U.S.C. 3501 et seq. The systems and filter backwash in all endpoints associated with DBP Information Collection Request (ICR) system sizes (LT1ESWTR and Filter exposure?; (3) To what extent will use document prepared by EPA has been Backwash Rule); as well as concurrently of chloramination increase levels of assigned ICR No. 2068.01 (USEPA proposing additional microbial control NDMA and potentially associated health 2003m). in surface water systems (Long Term 2 risks, and how should this be The information collected as a result Enhanced Surface Water Treatment considered in this rule making; and (4) of this rule will allow the States and Rule). These rules may have How should the Agency value nonfatal EPA to determine appropriate overlapping impacts on some drinking cancers? Specifically, EPA uses a range requirements for specific systems, and water systems but it is not possible to of severities to calculate the WTP to evaluate compliance with the rule. estimate these because of lack of estimate to avoid a case of chronic For the first 3 years after Stage 2 DBPR information on co-occurrence. However, bronchitis. Should the Agency only promulgation, the major information it is possible for a system to choose consider the most severe case of chronic requirements involve monitoring treatment technologies that would bronchitis as a better proxy for a non- activities, which include conducting the address multiple contaminants. fatal cancer? Also, should the Agency IDSE and submission of the IDSE report, Therefore, the total cost impact of these use the risk-risk trade-off estimate of and tracking compliance. The drinking water rules is uncertain; WTP to avoid a case of chronic information collection requirements are however, it may be less than the bronchitis instead of the risk-dollar mandatory (Part 141), and the estimated total cost of all individual trade-off estimate (see the EA (USEPA information collected is not rules combined. 2003i) for a complete discussion of confidential. these issues)? The estimate of annual average L. Benefit/Cost Determination for the burden hours for the Stage 2 DBPR for Proposed Stage 2 DBPR VIII. Statutory and Executive Order systems and States is 248,568 hours. The Agency has determined that the Reviews This estimate covers the first three years quantified and unquantified benefits of A. Executive Order 12866: Regulatory of the Stage 2 DBPR and includes the proposed Stage 2 DBPR justify the Planning and Review implementation of Stage 2A and most of costs. As discussed previously, the main Under Executive Order 12866, (58 FR the IDSE (small system reports are not concern for the Agency and the 51735, October 4, 1993) the Agency due until the fourth year). The annual Advisory Committee involved in the must determine whether the regulatory average aggregate cost estimate is $18.0 Stage 2 rulemaking process was to action is ‘‘significant’’ and therefore million for operation and maintenance address potential reproductive and subject to OMB review and the as a purchase of service for lab work, developmental impacts associated with requirements of the Executive Order. and $6.8 million is associated with exposure to high DBP levels. The The Order defines ‘‘significant labor. The annual burden hour per proposed rule achieves this objective regulatory action’’ as one that is likely response is 2.59 hours. The frequency of using the least cost alternative by to result in a rule that may: response (average responses per modifying how the annual average DBP (1) Have an annual effect on the respondent) is 11.8 annually. The level is calculated. This will reduce economy of $100 million or more or estimated number of likely respondents both average DBP levels associated with adversely affect in a material way the is 8,131 per year (the product of burden bladder cancer (and possibly other economy, a sector of the economy, hours per response, frequency, and cancers) and peak DBP levels which are productivity, competition, jobs, the respondents does not total the annual potentially associated with reproductive environment, public health or safety, or average burden hours due to rounding). and developmental effects. In addition, State, local, or Tribal governments or Because disinfecting systems have this rule may reduce uncertainty about communities; already purchased basic monitoring drinking water quality and may allow (2) Create a serious inconsistency or equipment to comply with the Stage 1 some systems to avoid installing otherwise interfere with an action taken DBPR, EPA assumes no capital start-up additional technology to meet future or planned by another agency; costs are associated with the Stage 2 drinking water regulations. (3) Materially alter the budgetary DBPR ICR. Compared to other rule options impact of entitlements, grants, user fees, Burden means the total time, effort, or consider by the Agency, the proposed or loan programs or the rights and financial resources expended by persons rule option is also the most cost- obligations of recipients thereof; or to generate, maintain, retain, or disclose effective. The cost-effectiveness analysis (4) Raise novel legal or policy issues or provide information to or for a compares the annual dollar cost of the arising out of legal mandates, the Federal agency. This includes the time

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00102 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49649

needed to review instructions; develop, to OMB is best assured of having its full Water Act for small system flexibility acquire, install, and utilize technology effect if OMB receives it by September provisions. In accordance with the RFA and systems for the purposes of 17, 2003. The final rule will respond to requirements, EPA proposed using this collecting, validating, and verifying any OMB or public comments on the alternative definition in the Federal information, processing and information collection requirements Register (63 FR 7620 (February 13, maintaining information, and disclosing contained in this proposal. 1998)), requested public comment, and providing information; adjust the consulted with the Small Business C. Regulatory Flexibility Act existing ways to comply with any Administration (SBA), and expressed its previously applicable instructions and The Regulatory Flexibility Analysis intention to use the alternative requirements; train personnel to be able (RFA), as amended by the Small definition for all future drinking water to respond to a collection of Business Regulatory Enforcement regulations in the Consumer Confidence information; search data sources; Fairness Act (SBREFA) of 1996, 5 U.S.C. Reports regulation (63 FR 44511 (August complete and review the collection of 601 et seq., generally requires an agency 19, 1998)). As stated in that final rule, information; and transmit or otherwise to prepare a regulatory flexibility the alternative definition is applied to disclose the information. analysis of any rule subject to notice this regulation. An Agency may not conduct or and comment rulemaking requirements After considering the economic sponsor, and a person is not required to under the Administrative Procedure Act impacts of today’s proposed rule on respond to a collection of information or any other statute, unless the Agency small entities, I certify that this action unless it displays a currently valid OMB certifies that the rule will not have a will not have a significant economic control number. The OMB control significant economic impact on a impact on a substantial number of small numbers for EPA’s regulations in 40 substantial number of small entities. entities. We have determined that 75 CFR are listed in 40 CFR part 9. Small entities include small businesses, small systems using surface water or To comment on the Agency’s need for small organizations, and small ground water under the direct influence this information, the accuracy of the governmental jurisdictions. of surface water (GWUDI), which are provided burden estimates, and any The RFA provides default definitions 1.67% of all such systems affected by suggested methods for minimizing for each type of small entity. It also the Stage 2 DBPR, will experience an respondent burden, including the use of authorizes an agency to use alternative impact of greater than or equal to 1% of automated collection techniques, EPA definitions for each category of small their revenues, and 49 small systems has established a public docket for this entity, ‘‘which are appropriate to the using surface water or GWUDI, which rule, which includes this ICR, under activities of the agency’’ after proposing are 1.09% of all such systems affected Docket ID No. OW–2002–0043. Submit the alternative definition(s) in the by the Stage 2 DBPR, will experience an any comments related to the ICR for this Federal Register and taking comment. 5 impact of greater than or equal to 3% of proposed rule to EPA and OMB. See U.S.C. 601(3) through (5). In addition to their revenues; further, 109 small ADDRESSES section at the beginning of the above, to establish an alternative ground water systems, which are 0.28% this notice for where to submit small business definition, agencies must of all such systems affected by the Stage comments to EPA. Send comments to consult with SBA’s Chief Counsel for 2 DBPR, will experience an impact of OMB at the Office of Information and Advocacy. greater than or equal to 1% of their Regulatory Affairs, Office of For purposes of assessing the impacts revenues, and 38 small ground water Management and Budget, 725 17th of today’s proposed rule on small systems, which are 0.10% of all such Street, NW., Washington, DC 20503, entities, EPA considered small entities systems affected by the Stage 2 DBPR, Attention: Desk Office for EPA. Since to be public water systems serving will experience an impact of greater OMB is required to make a decision 10,000 or fewer persons. This is the cut- than or equal to 3% of their revenues concerning the ICR between 30 and 60 off level specified by Congress in the (see Tables VIII–1 and VIII–2). days after August 18, 2003, a comment 1996 Amendments to the Safe Drinking BILLING CODE 6560–50–P

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00103 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49650 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00104 Fmt 4701 Sfmt 4725 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.021 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49651

BILLING CODE 6560–50–C they will effectively base their exempt from IDSE monitoring, systems As a result of the input received from compliance with the Stage 1 DBPR on serving fewer than 500 people may stakeholders, the EPA workgroup, the an LRAA and therefore will not be receive a waiver from their States, and Advisory Committee, and other significantly affected by the Stage 2 other systems are eligible for a 40/30 interested parties, EPA has developed DBPR. In addition to meeting the MCLs certification if all compliance MCLs using locational running annual for TTHM and HAA5, systems will be monitoring samples have been ≤ 0.040 averages (LRAA) of 0.080 and 0.060 mg/ required to conduct IDSEs. The purpose and ≤ 0.030 mg/L for TTHM and HAA5 L for TTHM and HAA5 respectively, in of the IDSE is to identify compliance respectively during the previous two combination with Initial Distribution monitoring sites representing high years. A large number of small ground Systems Evaluations (IDSE), as the TTHM and HAA5 levels in the water systems will qualify for this preferred Stage 2 DBPR option. LRAAs distribution system. According to the certification. This provision is described are running annual averages calculated Stage 2 DBPR Economic Analysis in more detail in section V.H. of this for each sample location in the (USEPA 2003i), only 17% of all small preamble. distribution system. Since many small water systems will conduct IDSE Although not required by the RFA to systems only monitor at one location, monitoring because small NTNCWSs are convene a Small Business Advocacy

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00105 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.022 49652 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

Review (SBAR) Panel because EPA Standards and Risk Reduction Branch of the Stage 2 DBPR will not impose any determined that this proposal would not the Office of Ground Water and additional burden on them. have a significant economic impact on Drinking Water within EPA’s Office of The Panel also noted the concern of a substantial number of small entities, Water, the Administrator of the Office of several SERs that flexibility should be EPA did convene a panel to obtain Information and Regulatory Affairs provided in the compliance schedule of advice and recommendations from within the Office of Management and the rule. SERs noted the technical and representatives of the small entities Budget, and the Chief Counsel for financial limitations that some small potentially subject to this rule’s Advocacy of the Small Business systems will have to address, the requirements. Administration. The Panel convened on significant learning curve for operators Before convening the SBAR Panel, April 25, 2000, and met five times with limited experience, and the need to EPA consulted with a group of 24 SERs before the end of the 60-day Panel continue providing uninterrupted likely to be impacted by the Stage 2 M– period on June 23, 2000. The SBAR service as reasons why additional DBP Rules. The SERs included small Panel’s report, ‘‘Final Report of the compliance time may be needed for system operators, local government small systems. The panel encouraged Small Business Advocacy Review Panel officials, and small nonprofit EPA to keep these limitations in mind on Stage 2 Disinfectants and organizations. The SERs were provided in developing the proposed rule and Disinfection Byproducts Rule (Stage 2 with background information on the provide as much compliance flexibility Safe Drinking Water Act, Stage 1 DBPR, DBPR) and Long-Term 2 Enhanced to small systems as is allowable under IESWTR, and Stage 2 DBPR alternatives Surface Water Treatment Rule the SDWA. EPA believes that the and unit cost analyses resulting from (LT2ESWTR)’’, the Small Entity proposed compliance schedules using different technologies to meet the Representatives (SERs) comments on provides sufficient time for small required MCLs in preparation for the components of the Stage 2 MDBP Rules, systems to achieve compliance. teleconferences on January 28, 2000, and the background information Under the proposed LT2ESWTR, February 25, 2000, and April 7, 2000. provided to the SBAR Panel and the certain subpart H systems with low This information package included data SERs are available for review in the levels of indicators such as E. coli will on options and preliminary unit costs Office of Water Docket. not have to monitor for for treatment enhancements under Today’s proposal takes into Cryptosporidium. The efficacy of E. coli consideration. It is important to note consideration the recordkeeping and as an indicator will be evaluated using that, since EPA did not consider the reporting concerns identified by the the large system data. Thus, small IDSE requirements until after these Panel and the SERs. The Panel systems E. coli monitoring cannot be consultations with SERs and the SBAR recommended that EPA evaluate ways initiated until large and medium system panel, no comments were received on to minimize the rule recordkeeping and monitoring has been completed. The the IDSE requirements from the SERs or reporting burdens by ensuring that LT2ESWTR compliance time line for the SBAR panel. However, small system States have appropriate capacity for rule small systems thus lags 1.5 to 2.5 years representatives were included in the implementation and that EPA provide behind the large and medium systems; Advisory Committee that recommended as much monitoring flexibility as timeline. Because the Stage 2 DBPR the IDSE. possible to small systems. Continuity must be implemented on a simultaneous During these conference calls, the with the Stage 1 DBPR was maintained schedule, the compliance timeline is information was discussed and EPA to the extent possible to ease the similarly delayed 1.5 to 2.5 years behind large and medium systems. In addition, provided feedback and noted these transition to the Stage 2 DBPR, initial SER comments. Following the if capital improvements are necessary especially for small systems. EPA’s calls, the SERs were asked to provide for a particular PWS to comply, a State decision to maintain the same MCLs for input on the potential impacts of the may allow the system up to an TTHM and HAA5 will also help to rule. Seven SERs provided written additional two years to comply with the minimize the additional comments on these materials. These MCL. The Agency is developing comments were provided to the SBAR implementation burden. Generally, guidance manuals to assist small Panel when the Panel convened in April routine monitoring will be similar in entities with their compliance efforts. 25, 2000. After a teleconference between frequency to monitoring for the Stage 1 The Panel considered a wide range of the SERs and the Panel on May 25, DBPR, and systems with low DBP levels options and regulatory alternatives for 2000, the SERs were invited to provide will still be eligible for reduced providing small businesses with additional comments on the information monitoring. Many small systems will flexibility in complying with the Stage provided. Seven SERs provided conduct the same amount of monitoring 2 DBPR. The Panel recognized the additional comments on the rule for the Stage 2 DBPR as for the Stage 1 concern shared by most stakeholders components. DBPR. Surface and ground water regarding the need to reduce DBP In general, the SERs consulted on the community water systems (CWSs) variability in the distribution system. Stage 2 M–DBP rules were concerned serving 500 to 9,999 people and ground This concern comes from recent studies about the impact of these proposed rules water systems serving at least 10,000 that, while not conclusive, suggest that on small water systems. They were people may be required to add one there may be adverse reproductive particularly concerned with acquiring sampling site and take an additional effects associated with relatively short- the technical and financial capability to quarterly TTHM/HAA5 sample at that term exposure to high levels of DBPs. implement requirements, maintaining site. Also, EPA has specified Many small systems will be monitoring flexibility to tailor requirements to their consecutive system requirements; these at only a single point in the distribution needs, and the limitations of small will be new requirements in States system (designed to represent the point systems. where consecutive systems are not of maximum TTHM and HAA5 The Small Business Advocacy Review required to comply with some or all exposure), and many small systems will (SBAR) Panel members for the Stage 2 Stage 1 DBPR requirements. As noted be monitoring only once during the DBPR were: the Small Business before, since some small systems will be year, at a time which corresponds to the Advocacy Chair of the Environmental effectively complying with such season with the highest potential Protection Agency, the Chief of the requirements under the Stage 1 DBPR, occurrence.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00106 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49653

Since there is a chance for this single The Panel noted the strong concerns generally must prepare a written sample to exceed an MCL, today’s expressed by some SERs about the statement, including a cost-benefit proposal requires systems that exceed uncertainty in the current scientific analysis, for proposed and final rules an MCL on an annual or less frequent evidence regarding health effects from with ‘‘Federal mandates’’ that may sample to begin increased (quarterly) exposure to DBPs, particularly regarding result in expenditures by State, local, monitoring rather than immediately short term exposure. A Panel member and Tribal governments, in the being in violation of the MCL. The recommended that EPA give further aggregate, or by the private sector, of system must comply with the MCL as an serious consideration to making a $100 million or more in any one year. LRAA once it has collected four determination that the currently Before promulgating an EPA rule for quarterly samples. This allows small available scientific evidence does not which a written statement is needed, systems to generally monitor less warrant imposing additional regulatory section 205 of the UMRA generally frequently (to reduce their monitoring requirements beyond those in the Stage requires EPA to identify and consider a burden) during the period when the 1 DBPR at this time. This Panel member reasonable number of regulatory highest DBP levels are expected (to recommended that EPA instead alternatives and adopt the least costly, protect public health) without continue to vigorously fund ongoing most cost-effective or least burdensome penalizing them (by requiring them to research in health effects, occurrence, alternative that achieves the objectives meet an MCL that would effectively be and appropriate treatment techniques of the rule. The provisions of section based on a single highest value if the for DBPs, and reconsider whether 205 do not apply when they are systems were immediately in violation additional requirements are appropriate inconsistent with applicable law. after a single sample exceeds an MCL). during its next SDWA required six-year Moreover, section 205 allows EPA to This compliance determination is review of the standard. This panel adopt an alternative other than the least consistent with requirements for member also recommended that EPA costly, most cost-effective or least systems that monitor quarterly for separately explore whether adequate burdensome alternative if the whom compliance is based on the data exist to warrant regulation of Administrator publishes with the final compliance monitoring results of the NTNCWSs at a national level at this rule an explanation why that alternative previous four quarters. time. was not adopted. It is important to note that based on EPA has considered these Before EPA establishes any regulatory the IDSE results, some small systems recommendations and believes the Stage requirements that may significantly or will have a high TTHM site that is 2 DBPR is needed at this time to protect uniquely affect small governments, different from the high HAA5 site. public health. EPA’s main mission is the including Tribal governments, it must These systems will need to monitor at protection of human health and the have developed, under section 203 of two sites under the Stage 2 DBPR. EPA environment. When carrying out this the UMRA, a small government agency believes that an approach based on mission, EPA must often make plan. The plan must provide for compliance with 0.080 mg/L TTHM and regulatory decisions with less than notifying potentially affected small 0.060 mg/L HAA5 LRAAs is an effective complete information and with governments, enabling officials of way of addressing concerns regarding uncertainties in the available affected small governments to have locational variability. information. EPA believes it is meaningful and timely input in the In addressing seasonal variability, the appropriate and prudent to err on the development of EPA regulatory Panel was concerned about a regulatory side of public health protection when proposals with significant Federal alternative requiring compliance with there are indications that exposure to a intergovernmental mandates and 0.080 mg/L TTHM and 0.060 mg/L contaminant may present risks to public informing, educating, and advising HAA5 single highest value MCL health, rather than take no action until small governments on compliance with (Alternative 2), because it would impose risks are unequivocally proven. the regulatory requirements. significant additional cost on some Therefore, while recognizing the EPA has determined that this rule small systems. The Panel recommended uncertainties in the available does not contain a Federal mandate that that EPA instead explore an approach information, EPA believes that the may result in expenditures of $100 under which individual high values weight of evidence represented by the million or more for State, local and might trigger additional assessment and/ available epidemiology and toxicology Tribal governments, in the aggregate, or or notification requirements, rather than studies on chlorinated water and DBPs the private sector in any one year. Based an MCL violation. supports a hazard concern and a on total estimated nominal costs EPA agrees with the panel protective public health approach to incurred by year, costs for public or recommendations on a single highest regulation. In addition, EPA has an private systems are not expected to value MCL. Under today’s proposal, ongoing research program to study DBP exceed $100 million in any one year. In public water systems are required to health effects, occurrence, and addition, total estimated annualized maintain a record of TTHM and HAA5 treatment. costs of this rule are $59 to $65 million concentrations detected at each sample EPA continues to be interested in the for all systems, including labor burdens location. As part of the sanitary survey potential impacts of the proposed rule that States would face, such as training process, systems are required to conduct on small entities and welcome employees on the requirements of the an evaluation and consult with their comments on issues related to such Stage 2 DBPR, responding to PWS State regarding significant excursions in impacts. reports, and record keeping. Thus, TTHM and HAA5 occurrence that have today’s proposed rule is not subject to occurred. EPA is developing guidance D. Unfunded Mandates Reform Act the requirements of sections 202 and for public water systems and States on Title II of the Unfunded Mandates 205 of the UMRA. how to identify significant excursions Reform Act of 1995 (UMRA), Public EPA has determined that the Stage 2 and conduct significant excursion Law 104–4, establishes requirements for DBPR contains no regulatory evaluations, and how to reduce DBP Federal agencies to assess the effects of requirements that might significantly or levels through actions such as their regulatory actions on State, local, uniquely affect small governments (see distribution system operational changes and Tribal governments and the private Tables VIII–1 and VIII–2). Since the (USEPA 2003n) (Section V.E.). sector. Under UMRA section 202, EPA Stage 2 DBPR affects all size systems

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00107 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49654 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

and the impact on small entities will be Although Executive Order 13132 does substantial direct compliance costs, and 0.00 to 0.11 percent of revenues, the not apply to this rule, EPA did consult that is not required by statute, unless Stage 2 DBPR is not subject to the with State and local officials in the Federal government provides the requirements of section 203 of UMRA. developing this proposed regulation. On funds necessary to pay the direct Nevertheless, in developing this rule, February 20, 2001, EPA held a dialogue compliance costs incurred by Tribal EPA consulted with small governments on both the Stage 2 DBPR and governments, or EPA consults with (see sections VIII.B., VIII.C. and VIII.F.). LT2ESWTR with representatives of Tribal officials early in the process of In preparation for the proposed Stage 2 State and local governmental developing the proposed regulation and DBPR, EPA conducted an analysis of organizations including those that develops a Tribal summary impact small government impacts and included represent elected officials. statement. small government officials or their Representatives from the following EPA has concluded that this proposed designated representatives in the organizations attended the consultation rule may have Tribal implications rulemaking process. As noted meeting: Association of State Drinking because it may impose substantial direct previously, a variety of stakeholders, Water Administrators (ASDWA), the compliance costs on Tribal including small governments, had the National Governors’ Association (NGA), governments, and the Federal opportunity for timely and meaningful the National Conference of State government will not provide the funds participation in the regulatory Legislatures (NCSL), the International necessary to pay those costs. development process through the City/County Management Association Total Tribal costs are estimated to be SBREFA process, public stakeholder (ICMA), the National League of Cities approximately $199,372 per year (at a 3 meetings, and Tribal meetings. (NLC), the County Executives of percent discount rate) and this cost is Representatives of small governments America, and health departments. At distributed across 559 Tribal systems. took part in the SBREFA process for this the consultation meeting, questions The cost for individual systems depend rulemaking and they attended public ranged from a basic inquiry into how on system size and source water type. stakeholder meetings. Through such Cryptosporidium gets into water to more Of the 559 Tribes that may be affected participation and exchange, EPA detailed queries about anticipated in some form by the Stage 2 DBPR, 502 notified several potentially affected implementation guidance, procedures, use ground water as a source and 57 small governments of requirements and schedules. No concerns were systems use surface water or GWUDI. under consideration and provided expressed. Some of the State and local Since the majority of Tribal systems are officials of affected small governments organizations who attended the ground water systems serving fewer with an opportunity to have meaningful governmental dialogue on upcoming than 500 people, less than 10 percent of and timely input into the development microbial and disinfection byproduct all Tribal systems will likely have to of this regulatory proposal. rulemakings were also participants in conduct an IDSE. As a result, the Stage The Agency has developed fact sheets the Advisory Committee meetings and 2 DBPR is most likely to have an impact that describe requirements of the signed the Agreement in Principle. In on Tribes using surface water or GWUDI proposed Stage 2 DBPR. These fact addition, EPA consulted with a mayor serving more than 500 people. sheets are available by calling the Safe in the SBREFA consultation described Accordingly, EPA provides the Drinking Water Hotline at 800–426– in section VIII B. following Tribal summary impact 4791. In the spirit of Executive Order 13132, statement as required by section 5(b) of Executive Order 13175. EPA provides E. Executive Order 13132: Federalism and consistent with EPA policy to promote communications between EPA further detail on Tribal impact in the Executive Order 13132, entitled and State and local governments, EPA Economic Analysis for the Stage 2 ‘‘Federalism’’ (64 FR 43255, August 10, specifically solicits comment on this Disinfectants and Disinfection 1999), requires EPA to develop an proposed rule from State and local Byproduct Rule (USEPA 2003i). accountable process to ensure officials. EPA consulted with Tribal officials ‘‘meaningful and timely input by State early in the process of developing this and local officials in the development of F. Executive Order 13175: Consultation regulation to permit them to have regulatory policies that have federalism and Coordination With Indian Tribal meaningful and timely input into its implications.’’ ‘‘Policies that have Governments development. Consistent with Executive federalism implications’’ is defined in Executive Order 13175, entitled Order 13175, EPA engaged in outreach the Executive Order to include ‘‘Consultation and Coordination with and consultation efforts with Tribal regulations that have ‘‘substantial direct Indian Tribal Governments’’ (65 FR officials in the development of this effects on the States, on the relationship 67249, November 9, 2000), requires EPA proposed regulation. The most long- between the national government and to develop ‘‘an accountable process to term participation of Tribes was on the the States, or on the distribution of ensure meaningful and timely input by Advisory Committee through a power and responsibilities among the tribal officials in the development of representative of the All Indian Pueblo various levels of government.’’ regulatory policies that have tribal Council (AIPC), which is associated This proposed rule will not have implications.’’ ‘‘Policies that have tribal with approximately 20 Tribes. federalism implications. It will not implications’’ is defined in the In addition to obtaining Tribal input impose substantial direct effects on the Executive Order to include regulations during the Advisory Committee States, on the relationship between the that have ‘‘substantial direct effects on negotiations, EPA presented the Stage 2 national government and the States, or one or more Indian tribes, on the DBPR at the 16th Annual Consumer on the distribution of power and relationship between the Federal Conference of the National Indian responsibilities among the various government and the Indian tribes, or on Health Board, the Environmental levels of government, as specified in the distribution of power and Council’s Annual Conference, and the Executive Order 13132. The proposed responsibilities between the Federal EPA/Inter-Tribal Council of Arizona, rule has one-time costs for government and Indian tribes.’’ Inc. Over 900 attendees representing implementation of approximately $68.5 Under Executive Order 13175, EPA Tribes from across the country attended million. Thus, Executive Order 13132 may not issue a regulation that has the National Indian Health Board’s does not apply to this rule. Tribal implications, that imposes Consumer Conference and over 100

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00108 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49655

Tribes were represented at the annual G. Executive Order 13045: Protection of low birth weight. This rule is designed conference of the National Tribal Children From Environmental Health to lower those risks. EPA has provided Environmental Council. Representatives and Safety Risks an illustrative calculation of potential from 15 Tribes participated at the EPA/ Executive Order 13045: ‘‘Protection of fetal losses avoided in section VII.C.1. Inter-Tribal Council of Arizona meeting. Children From Environmental Health Section V.D of this preamble presents the regulatory alternatives that EPA At the first two conferences, an EPA Risks and Safety Risks’’ (62 FR 19885, evaluated for the proposed Stage 2 representative conducted workshops on April 23, 1997) applies to any rule that: DBPR, and the Economic Analysis EPA’s drinking water program and (1) Is determined to be ‘‘economically (USEPA 2003i) provides a more detailed upcoming regulations, including the significant’’ as defined under Executive discussion. The Agency considered four Stage 2 DBPR. EPA sent the presentation Order 12866, and; (2) concerns an alternatives involving different MCLs materials and a meeting summary to environmental health or safety risk that and different compliance calculations. over 500 Tribes and Tribal EPA has reason to believe may have a The proposed alternative was organizations. disproportionate effect on children. If recommended by the Advisory the regulatory action meets both criteria, Fact sheets describing the Committee and selected by EPA as the requirements of the proposed rule and the Agency must evaluate the Preferred Regulatory Alternative requesting Tribal input were distributed environmental health or safety effects of because it provides significant public at an annual EPA Tribal meeting in San the planned rule on children, and health benefits for an acceptable cost. Francisco, and at a Native American explain why the planned regulation is EPA’s analysis of benefits and costs Water Works Association meeting in preferable to other potentially effective indicates that the proposed alternative Scottsdale, Arizona. EPA also worked and reasonably feasible alternatives is superior among those evaluated with through its Regional Indian considered by the Agency. respect to maximizing net benefits, as While this proposed rule is not Coordinators and the National Tribal shown in the Economic Analysis subject to the Executive Order because Operations Committee to raise (USEPA 2003i). The result of the Stage it is not economically significant as awareness of the development of the 2 DBPR may include a reduction in defined in Executive Order 12866, EPA proposed rule. EPA mailed fact sheets reproductive and developmental risk to nonetheless has reason to believe that on the Stage 2 DBPR to all of the children and pregnant women and a the environmental health or safety risk reduction in cancer risk. federally recognized Tribes in (i.e., the risk associated with DBPs) November 2000, as well as the Tribal It should also be noted that the addressed by this action may have a LT2ESWTR, which will be implemented Caucus of the National Tribal disproportionate effect on children. As Operations Committee. at the same time as this proposed rule, a matter of EPA policy, we have provides better controls of pathogens A few Tribes responded by requesting therefore assessed the environmental and achieves the goal of increasing more information and expressing health or safety effect of DBPs on microbial drinking water protection for concern about having to implement too children. EPA has consistently and children. The public is invited to submit many regulations. Some members of the explicitly considered risks to infants or identify peer-reviewed studies and Tribal Caucus noted that the rule would and children in all assessments data, of which EPA may not be aware have a benefit. They also expressed a developed for this rulemaking. The that assessed results of early life concern about infrastructure costs and results of the assessments are contained exposure to DBPs. the lack of funding attached to the rule. in section III of this preamble, Health In response to one Tribal Risks to Fetuses, Infants, and Children: H. Executive Order 13211: Actions That representative’s comments on the A Review (USEPA 2003a), and in the Significantly Affect Energy Supply, November 2000 mailout, EPA explained Economic Analysis (USEPA 2003i). A Distribution, or Use the health protection benefit expected to copy of all documents has been placed The proposed Stage 2 DBPR is not a be gained by this proposed rule. EPA in the public docket for this action. ‘‘significant energy action’’ as defined in also directed those who asked for more EPA’s Office of Water has historically Executive Order 13211, ‘‘Actions information to the Agreement in considered risks to sensitive Concerning Regulations That Principle on the EPA Web site. subpopulations (including fetuses, Significantly Affect Energy Supply, infants, and children) in establishing Distribution, or Use’’ (66 FR 28355 (May EPA also held a teleconference for drinking water assessments, health 22, 2001)) because it is not likely to Tribal representatives on January 24, advisories or other guidance, and have a significant adverse effect on the 2002. Prior to the teleconference, standards (USEPA 1989c and USEPA supply, distribution, or use of energy. invitations were sent to all of the 1991a). Waterborne disease from This determination is based on the Federally-recognized Tribes, along with pathogens in drinking water is a major following analysis. fact sheets explaining the rule. Twelve concern for children and other The first consideration is whether the Tribal representatives and four regional subgroups (elderly, immune Stage 2 DBPR would adversely affect the Tribal Program Coordinators attended. compromised, pregnant women) supply of energy. The Stage 2 DBPR The Tribal representatives requested because of their increased does not regulate power generation, further explanation of the rule and vulnerabilities (Gerba et al. 1996). There either directly or indirectly. The public expressed concerns about funding is a concern for potential reproductive and private utilities that the Stage 2 sources. EPA also received calls from and developmental risks posed by DBPs DBPR regulates do not, as a rule, Tribes after the teleconference which to children and pregnant women generate power. Further, the cost provided EPA with further feedback. In (USEPA 1994b; USEPA 1998c, Reif et al. increases borne by customers of water the spirit of Executive Order 13175, and 2000; Tyl, 2000). Specific to this action, utilities as a result of the Stage 2 DBPR consistent with EPA policy to promote human epidemiology and animal are a low percentage of the total cost of consultation between EPA and Tribal toxicology studies on DBPs have shown water, except for a very few small governments, EPA specifically solicits potential increased risks for systems that might install advanced additional comment on this proposed spontaneous abortion, still birth, neural technologies that must spread that cost rule from Tribal officials. tube defects, cardiovascular effects and over a narrow customer base. Therefore,

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00109 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49656 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

the customers that are power generation add treatment technologies that use dividing the total energy cost per plant utilities are unlikely to face any electrical power, this potential impact is for a range of flows by an average significant effects as a result of the Stage evaluated in more detail. The analyses national cost of electricity of $0.076/ 2 DBPR. In sum, the Stage 2 DBPR does that underlay the estimation of costs for kilowatt hours per year (kWh/yr) (U.S. not regulate the supply of energy, does the Stage 2 DBPR are national in scope Department of Energy, Energy not generally regulate the utilities that and do not identify specific plants or Information Administration (USDOE supply energy, and is unlikely utilities that may install treatment in EIA) 2002). These calculations are significantly to affect the customer base response to the rule. As a result, no shown in detail in Chapter 8 of the of energy suppliers. Thus, the Stage 2 analysis of the effect on specific energy Economic Analysis for the Stage 2 DBPR DBPR would not translate into adverse suppliers is possible with the available (USEPA 2003i). The energy use per effects on the supply of energy. data. The approach used to estimate the plant for each flow range and The second consideration is whether impact of energy use, therefore, focuses the Stage 2 DBPR would adversely affect on national-level impacts. The analysis technology is then multiplied by the the distribution of energy. The Stage 2 estimates the additional energy use due number of plants predicted to install DBPR does not regulate any aspect of to the Stage 2 DBPR, and compares that each technology in a given flow range. energy distribution. The utilities that are to the national levels of power The energy requirements for each flow regulated by the Stage 2 DBPR already generation in terms of average and peak range are then added to produce a have electrical service. As derived later loads. national total. No electricity use is in this section, the proposed rule is The first step in the analysis is to subtracted to account for the projected to increase peak electricity estimate the energy used by the technologies that may be replaced by demand at water utilities by only 0.007 technologies expected to be installed as new technologies, resulting in a percent. Therefore, EPA estimates that a result of the Stage 2 DBPR. Energy use conservative estimate of the increase in the existing connections are adequate is not directly stated in Technologies energy use. Table VIII–3 shows the and that the Stage 2 DBPR has no and Costs for Control of Microbial estimated energy use for each Stage 2 discernable adverse effect on energy Contaminants and Disinfection By- DBPR compliance technology in distribution. Products (USEPA 2003k), but the annual kilowatt hours per year (kWh/yr). The The third consideration is whether cost of energy for each technology incremental national annual energy the Stage 2 DBPR would adversely affect addition or upgrade necessitated by the usage is 0.08 million megawatt-hours the use of energy. Because some Stage 2 DBPR is provided. An estimate (mWh). drinking water utilities are expected to of plant-level energy use is derived by

To determine if the additional energy Administration, electricity producers In addition to average energy use, the required for systems to comply with the generated 3,800 million mWh of impact at times of peak power demand rule would have a significant adverse electricity in 2001 (USDOE EIA 2002). is important. To examine whether effect on the use of energy, the numbers Therefore, even using the highest increased energy usage might in Table VIII–3 are compared to the assumed energy use for the Stage 2 significantly affect the capacity margins national production figures for DBPR, the rule when fully implemented of energy suppliers, their peak season electricity. According to the U.S. would result in only a 0.002 percent generating capacity reserve was Department of Energy’s Information increase in annual average energy use. compared to an estimate of peak

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00110 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 EP18AU03.023 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49657

incremental power demand by water use American Society for Testing and The Stage 1 DBPR has served as a utilities. Materials (ASTM) Method D 6581–00 template for the development of the Both energy use and water use peak for chlorite, bromide, and bromate Stage 2 DBPR. As such, the Agency built in the summer months, so the most compliance monitoring, which can be on the efforts conducted during the significant effects on supply would be found in the Annual Book of ASTM development of the Stage 1 DBPR to seen then. In the summer of 2001, U.S. Standards Volume 11.01. In the Stage 1 comply with Executive Order 12898. On generation capacity exceeded DBPR, EPA approved 13 methods from March 12, 1998, the Agency held a consumption by 15 percent, or the Standard Methods Committee for stakeholder meeting to address various approximately 120,000 mW (USDOE measuring disinfectants, DBPs, and components of pending drinking water EIA 2002). Assuming around-the-clock other parameters. Today’s rule proposes regulations and how they might impact operation of water treatment plants, the to add the most recent versions of these sensitive subpopulations, minority total energy requirement can be divided 13 methods as approved methods. These populations, and low-income by 8,760 hours per year to obtain an consist of Standard Methods 4500–Cl D, populations. This meeting was a average power demand of 8.3 mW. A 4500–Cl F, 4500–Cl G, 4500–Cl E, 4500– continuation of stakeholder meetings more detailed derivation of this value is Cl I, 4500–Cl H, 4500–ClO2 D, 4500– that started in 1995 to obtain input on shown in Chapter 8 of the Economic ClO2 E, 6251 B, 5310 B, 5310 C, 5310 the Agency’s Drinking Water Programs. Analysis for the Stage 2 DBPR (USEPA D, and 5910 B for chlorine, chlorine Topics discussed included treatment 2003i). Assuming that power demand is dioxide, HAA5, chlorite, TOC/DOC, and techniques, costs and benefits, data proportional to water flow through the UV254. These methods can be found in quality, health effects, and the plant and that peak flow can be as high the 19th and 20th editions of Standard regulatory process. Participants were as twice the average daily flow during Methods for the Examination of Water national, State, Tribal, municipal, and the summer months, about 16.6 mW and Waste Water (APHA 1995; APHA individual stakeholders. EPA conducted could be needed for treatment 1996; APHA 1998). Standard Methods the meeting by video conference call technologies installed to comply with 4500–Cl D, 4500–Cl F, 4500–Cl G, 4500– between eleven cities. The major the Stage 2 DBPR. This is only 0.014 Cl E, 4500–Cl I, 4500–Cl H, 4500–ClO2 percent of the capacity margin available objectives for the March 12, 1998, E, 6251 B, 5310 B, 5310 C, 5310 D, and meeting were the following: at peak use. 5910 B for chlorine, chlorine dioxide, Although EPA recognizes that not all • Solicit ideas from stakeholders on HAA5, chlorite, TOC/DOC, and UV254 areas have a 15 percent capacity margin are also available in the On-Line known issues concerning current and that this margin varies across Version of Standard Methods for the drinking water regulatory efforts; regions and through time, this analysis Examination of Water and Waste Water • Identify key areas of concern to reflects the effect of the rule on national (APHA 2003). stakeholders; and energy supply, distribution, and use. • While certain areas, notably California, EPA welcomes comments on this Receive suggestions from have experienced shortfalls in aspect of the proposed rulemaking and, stakeholders concerning ways to generating capacity in the recent past, a specifically, invites the public to increase representation of communities peak incremental power requirement of identify potentially applicable voluntary in OGWDW regulatory efforts. consensus standards and to explain why 16.6 mW nationwide is not likely to In addition, EPA developed a plain- such standards should be used in this significantly change the energy supply, English guide for this meeting to assist regulation. distribution, or use in any given area. stakeholders in understanding the Considering this analysis, EPA has J. Executive Order 12898: Federal multiple and sometimes complex issues concluded that Stage 2 DBPR will not Actions to Address Environmental surrounding drinking water regulations. have any significant effect on the use of Justice in Minority Populations or Low The Stage 2 DBPR and other drinking energy, based on annual average use and Income Populations water regulations promulgated or under on conditions of peak power demand. Executive Order 12898 establishes a development are expected to have a I. National Technology Transfer and Federal policy for incorporating positive effect on human health Advancement Act environmental justice into Federal regardless of the social or economic Section 12(d) of the National agency missions by directing agencies to status of a specific population. The Technology Transfer and Advancement identify and address disproportionately Stage 2 DBPR serves to provide a similar Act (NTTAA) of 1995, Pub. L. No. 104– high and adverse human health or level of drinking water protection to all 113, 12(d) (15 U.S.C. 272 note) directs environmental effects of its programs, groups. Where water systems have high EPA to use voluntary consensus policies, and activities on minority and DBP levels, they must reduce levels to standards in its regulatory activities low-income populations. The Agency meet the MCLs. Thus, the Stage 2 DBPR unless to do so would be inconsistent has considered environmental justice meets the intent of Federal policy with applicable law or otherwise related issues concerning the potential requiring incorporation of impractical. Voluntary consensus impacts of this action and consulted environmental justice into Federal standards are technical standards (e.g., with minority and low-income agency missions. materials specifications, test methods, stakeholders. The Stage 2 DBPR applies uniformly sampling procedures, and business Two aspects of the Stage 2 DBPR to community water systems and practices) that are developed or adopted comply with the order that requires the nontransient noncommunity water by voluntary consensus standard bodies. Agency to consider environmental systems that apply a chemical The NTTAA directs EPA to provide justice issues in the rulemaking and to disinfectant or deliver water that has Congress, through OMB, explanations consult with stakeholders representing a been chemically disinfected. when the Agency decides not to use variety of economic and ethnic Consequently, the health protection available and applicable voluntary backgrounds. These are: (1) The overall from DBP exposure that this rule consensus standards. nature of the rule, and (2) the convening provides is equal across all income and This proposed rulemaking involves of a stakeholder meeting specifically to minority groups served by systems technical standards. EPA proposes to address environmental justice issues. regulated by this rule.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00111 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49658 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

K. Consultations with the Science APHA 1995. Nineteenth Edition of Standard Tumor Induction is B6C3F1 Mice by Advisory Board, National Drinking Methods for the Examination of Water and Dichloroacetate and Trichloracetate. Water Advisory Council, and the Wastewater, American Public Health Toxicology. 63: 341–359. Secretary of Health and Human Services Association, 1015 Fifteenth Street, NW., Cantor, K.P., C.F. Lunch, M. Hildesheim, M. Washington, DC 20005. Dosemeci, J. Lubin, M. Alavanja, G.F. In accordance with sections 1412 (d) APHA 1996. Supplement to the Nineteenth Craun. 1998. Drinking Water Source and and (e) of SDWA, the Agency has Edition of Standard Methods for the Chlorination Byproducts. I. Risk of Bladder consulted with the Science Advisory Examination of Water and Wastewater, Cancer. Epidemiology; 9(1):21–28. Board (SAB), the National Drinking American Public Health Association, 1015 Cantor KP, Lynch CF, Hildesheim ME, Water Advisory Council (NDWAC), and Fifteenth Street, NW., Washington, DC Dosemeci M, Lubin J, Alavanja M, Craun 20005. G., 1999. Drinking Water Source and will consult with the Secretary of Health APHA 1998. Twentieth Edition of Standard Chlorination Byproducts in Iowa. III. Risk and Human Services regarding the Methods for the Examination of Water and of Brain Cancer. Am J Epidemiol. proposed Stage 2 DBPR during the Wastewater, American Public Health 150(6):552–60. public comment period. Association, 1015 Fifteenth Street, NW., Chang, L.W., F. B. Daniel and A. B. EPA met with the SAB to discuss the Washington, DC 20005. DeAngelo. 1991. Analysis of DNA Strand Stage 2 DBPR on June 13, 2001 APHA 2003. On-Line Version of Standard Breaks Induced in Rodent Liver in vivo, (Washington, DC), September 25–26, Methods for the Examination of Water and Hepatocytes in Primary Culture, and a 2001 (teleconference), and December Wastewater, American Public Health Human Cell Line by Chloroacetic Acids 10–12, 2001 (Los Angeles, CA). Written Association, 1015 Fifteenth Street, NW., and Chloroacetaldehydes. Environ. Molec. Washington, DC 20005. Mutagen, 20:277–288. comments from the December 2001 Aschengrau, A., Zierler S. and Cohen A. Chlorine Institute 1999. Bromate in Sodium meeting of the SAB addressing the 1989. Quality of Community Drinking Hypochlorite solutions. occurrence analysis and risk assessment Water and the Occurrence of Spontaneous Chlorine Institute 2000. Bromate in Sodium were generally supportive. EPA met Abortions. Arch. Environ. Health. 44:283– Hypochlorite. with the NDWAC on November 8, 2001, 90. Christian, M.S., R.G. York, A.M. Hoberman, in Washington, DC to discuss the Stage Aschengrau, A, Zierler S. and Cohen A. 1993. R.M. Diener, and L.C. Fisher. 2001. Oral 2 DBPR proposal. The Advisory Quality of Community Drinking Water and (Drinking Water) Developmental Toxicity Committee generally supported the need the Occurrence of Late Adverse Pregnancy Studies of Bromodichloromethane (BDCM) for the Stage 2 DBPR based on health Outcomes. Arch. Environ. Health. 48:105– in Rats and Rabbits. International Journal 113. of Toxicology 20(4):225–237. and occurrence data, but also stressed ASTM 2002. Method D 6581–00. Annual Christian M.S., York R.G., Hoberman A.M., the importance of providing flexibility Book of ASTM Standards. Vol. 11.01, Frazee, L.C., Fisher L.C., Brown W.R., and to the systems implementing the rule. American Society for Testing and D.M. Creasy. 2002a. Oral (drinking water) The results of these discussions are Materials. Two Generation Reproductive Toxicity included in the docket for this rule. Balster, R.L., and J.F. Borzelleca, 1982. Study of Dibromoacetic Acid (DBA) in Behavioral Toxicology of Trihalomethane Rats. International Journal of Toxicology L. Plain Language Contaminants of Drinking Water in Mice. 21(4) 237–76. Executive Order 12866 encourages Environmental Health Perspectives. 46, Christian M.S., York R.G., Hoberman A.M., Federal agencies to write rules in plain 127–136. Diener R.M., Fisher L.C. 2002b. Oral language. EPA invites comments on Baribeau, H., S.W. Krasner, R., Chin, R., and (drinking water) Two Generation how to make this proposed rule easier P.C. Singer. 2000. Impact of Biomass on the Reproductive Toxicity Study of Stability of Haloacetic Acids and Bromodichloromethane (BDCM) in Rats. to understand. For example: Has EPA Trihalomethanes in a Simulated International Journal of Toxicology 21 organized the material to suit Distribution System. Proc. Of the Water (2):115–146. commenters’ needs? Are the Quality Technology Conference, Denver, Cosby, N. C. and W. R. Dukelow. 1992. requirements in the rule clearly stated? CO. AWWA. Toxicology of Maternally Ingested Does the rule contain technical language Bhat, H.K., M.F. Kanz, G.A. Campbell and Trichloroethylene (TCE) on Embryonal and or jargon that is not clear? Would a G.A.S. Ansari. 1991. Ninety Day Toxicity Fetal Development in Mice and of TCE different format (grouping and ordering Study of Chloroacetic Acids in Rats. Metabolites on in vitro Fertilization. of sections, use of headings, paragraphs) Fundam. Appl. Toxicol. 17:240–253. Fundam. Appl. Toxicol. 19(2): 268–74. make the rule easier to understand? Bielmeier, S.R., D.S. Best, D.L. Guidici, and Day, J.A., Vonderheide, A.P., and Caruso, J.A. M.G. Narotsky. 2001. Pregnancy Loss in the Second Laboratory Validation of U.S. EPA Could EPA improve clarity by adding Rat Caused by Bromodochloromethane. Method 321.8: Determination of Bromate in tables, lists, or diagrams? What else Toxicol Sci. Feb; 59(2):309–15. Drinking Waters by Ion Chromatography could EPA do to make the rule easier to Bolyard, M..G. and M.B. Stricklen. 1992. Inductively Coupled Plasma Mass understand? Expression of a modified Dutch elm Spectrometry. University of Cincinnati, disease toxin in Escherichia coli. Mol Plant January 2001. IX. References Microb Interact 1992. 5(6):520–4. D.C. Circuit 2000. Chlorine Chemistry Acharya, S., K. Mehta, S. Rodrigues, J. Bove, F.J., M.C. Fulcomer, J.B. Koltz, J. Council and Chemical Manufacturers Pereira, S. Krishman and C.V. Rao. 1995. Esmart, E.M. Dufficy, R.T. Zagraniski and Association v. EPA, 206 F.3d 1286. Administration of Subtoxic Doses of T- J.E. Savrin. 1992. Report on Phase IV-B: DeAngelo, A.B., F.B. Daniel, L. McMillan, P. butyl Alcohol and Trichloroacetic Acid to Public Drinking Water Contamination and Wernsing and R. E. Savage. 1989. Species Male Wistar Rats to Study the Interactive Birthweight and Selected and Birth and Strain Sensitivity to the Induction of Toxicity. Toxicol. Lett. 80: 97–104. Defects, a Case-Control Study. New Jersey Peroxisome Proliferation by Chloroacetic Acharya, S., K. Mehta, S. Rodrigues, J. Dept. of Health. Acids. Toxicol. Appl. Pharmacol. 101:285– Pereira, S. Krishman and C.V. Rao. 1997. Bove, F.J. et al. 1995. Public Drinking Water 289. A Histopathological Study of Liver and Contamination and Birth Outcomes. Amer. DeAngelo, A.B., F.B. Daniel, B.M. Most and Kidney in Male Wistar Rats Treated with J. Epidemiol., 141(9), 850–862. G.R. Olson. 1997. Failure of Subtoxic Doses of T-butyl Alcohol and Bove, F.J.; Shim, Y.; and Zeitz, P. 2002. Monochloroacetic Acid and Trichloroacetic Acid. Exp. Toxicol. Pathol. Drinking Water Contaminants and Adverse Trichloroacetic Acid Administered in the 49: 369–373. Pregnancy Outcomes: A Review. Drinking Water to Produce Liver Cancer in American Cancer Society. 2002. Cancer Facts Environmental Health Perspectives Male F344/N rats. J. of Toxicol. and and Figures. http://www.cancer.org/ 110(Suppl. 1):61–74. Environ. Health. 52:425–445. downloads/STT/ Bull, R.J.; I.M. Sanchez, M.A. Nelson, J.L. DeAngelo, A.B., M.H. George and D.E. House. CancerFacts&Figures2002TM.pdf. Larson and A.J. Lansing. 1990. Liver 1999. Hepatocarcinogenicity in the Male

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00112 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49659

B6C3F1 Mouse Following a Lifetime Peroxisomal Enzyme Activity in Relation Potential of Chlorinated Drinking Water Exposure to Dichloroacetic Acid in the to Species and Organ Carcinogenicity. Disinfection By-products by using Hydra Drinking Water: Dose-Response Toxicol. Appl. Pharmacol. 88:225–233. Regeneration Assay. J. of Environ. Determination and Modes of Action. J. Harrington-Brock, K. C.L. Doerr and M.M. Sciences. 10(1):110–112. Toxicol. Environ Health. 58(8):485–507. Moore. 1998. Mutagenicity of Three Johnson, P.D., B.V. Dawson, and S.J. DeAngelo, A.B., Geter D.R., Rosenberg D.W., disinfection by-products; di- and Goldberg. 1998. Cardiac Teratogenicity of Crary C.K., George M.H. 2002. The trichloroacetic acid and chloral hydrate in Trichloroethylene Metabolites. J. American induction of aberrant crypt foci (ACF) in L5178Y/TK+/¥3.7.2C mouse lymphoma College of Cardiology. 32(2):540–545. the colons of rats by trihalomethanes cells. Mutation Research. 413:265–276. Ka¨lle´n, B.A.J .and E. Robert. 2000. Drinking administered in the drinking water. Cancer Hautman, D.P., Munch, D.J., Frebis, C.P., water Chlorination and Delivery Letters 187(1–2):25–31. Wagner, H.P., and Pepich, B.V. 2001. Outcome—a Registry Based Study in Dees, C. and C. Travis. 1994. Trichloroacetate Review of the Methods of the U.S. Sweden. Reprod. Toxicol. 14:303–309. Stimulation of Liver DNA Synthesis in Environmental Protection Agency for Kanitz S, Franco Y, Patrone V, Caltabellotta Male and Female Mice. Toxicol. Lett. Bromate Determination and Validation of M, Raffo E, Riggi C, Timitilli D, Ravera G. 70:343–355. Method 317.0 for Disinfection By-Product 1996, Association between drinking water DeMarini, D.M., E. Perry and M.L. Sheldon. Anions and Low-Level Bromate. Journal of disinfection and somatic parameters at 1994. Dichloroacetic Acid and Related Chromatography A, 920 (2001) 221–229. birth. Environ Health Perspect 104(5):516– Compounds: Induction of Prophage in E. Heywood, R., R.J. Sortwell, P.R.B. Noel, A.E. 520. coli and Mutagenicity and Mutation Street, D.E. Prentice, F.J.C. Roe, P.F. Kim, H. and C. P. Weisel. 1998. Dermal Spectra in Salmonella TA 100. Wadsworth, A.N. Worden, N.J. Van Abbe. Absorption of Dichloro- and Mutagenesis. 9:429–437. 1979. Safety Evaluation of Toothpaste Trichloroacetic Acids from Chlorinated Dodds, L., W. King, C. Wolcott and J. Pole. Containing Chloroform. III. Long-term Water. J. of Exposure Anal. and Environ. 1999. Trihalomethanes in Public Water Study in Beagle Dogs. J. Environ. Pathol. Epidem. 8(4):555–575. Supplies and Adverse Birth Outcomes. Toxicol. 2:835–851. King, W., L. Dodds and A. Allen. 2000a. Epidemiology. 10:233–237. Hildesheim, M.E., K.P. Cantor, C.F. Lynch, Relation between Stillbirth and Specific Dodds, L. and W.D. King. 2001. Relation M. Dosemeci, J. Lubin, M. Alavanja, and Chlorination By-products in Public Water Between Trihalomethane Compounds and G.F. Craun. 1998. Drinking Water Source Supplies. Environ. Health Birth Defects. Occup Environ Med., and Chlorination Byproducts: Risk of Perspect.108:883–886. 58(7):443–46. Colon and Rectal Cancers. Epidemiology. King, W.D., L.D. Marrett and C.G. Woolcott. Doyle, Timothy J; Zheng, Wei; Cerhan, James 9(1):29–35. 2000b. Case-Control Study of Colon and R; Hong, Ching-Ping. 1997. The association Hooper and Allgeier, 2002. Information Rectal Cancers and Chlorination By- of drinking water source and chlorination Collection Rule Treatment Studies. products in Treated Water. Cancer by-products with cancer incidence among AwwaRF. Epidemiology, Biomarkers & Prevention postmenopausal women in Iowa: A Hrubek Z. and J.K. McLaughlin. 1997. 9:813–818. prospective cohort study. American ‘‘Former cigarette smoking and mortality Klinefelter, G.R., Hunter, E.S., and Narotsky, Journal of Public Health, 87(7):1168–1176. among U.S. veterans: a 26-year follow-up.’’ M. 2001. Reproductive and Developmental Fair, P.S. Memo to the record. February 2002. In: Changes in cigarette related disease Fair, P.S., R.K. Sorrell, M. Stultz-Karapondo, risks and their implication for prevention Toxicity Associated with Disinfection By- et al., 2002. Quality of Information and control. D.M. Burns, L. Garfinkel, J.M. Products of Drinking Water, In: Microbial Collection Rule Monitoring Data. In Samet (eds.). NIH Monograph No. 8, Pathogens and Disinfection By-Products of Information Collection Rule Data Analysis, National Institutes of Health. Washington, Drinking Water, ILSI Press, 309–323. M.J. McGuire, J. McLain, and A. Obolensky DC: National Cancer Institute, pp.501–530. Klotz J.B. and L.A. Pyrch, 1998. A Case (eds), AwwaRF, Denver, CO. Hunter, III, E.S., E.H. Rogers, J.E. Schmid and Control Study of Neural Tube Defects and Ferreira-Gonzalez, A., A.B. DeAngelo, S. A. Richard. 1996. Comparative Effects of Drinking Water Contaminants. U.S. Nasim and C.T. Garrett. 1995. Ras Haloacetic Acids in Whole Embryo Department of Health and Human Services, Oncogene Activation during Culture. Teratology. 54:57–64. Agency for Toxic Substances and Disease Hepatocarcinogenesis in B6C3F1 Male Hwang, B., P. Magnus, and J.K. Jaakkola, Registry (ATSDR). Mice by Dichloroacetic and Trichloroacetic 2002. Risk of Specific Birth Defects in Klotz, J.B. and L.A. Pyrch, 1999. Neural Tube Acids. Carcinogenesis. 16(3):495–500. Relation to Chlorination and the Amount Defects and Drinking Water Disinfection Fort, D., E. Stover, J. Rayburn, M. Hull and of Natural Organic Matter in the Water Byproducts. Epidemiology 10:383–390. J. Bantle. 1993. Evaluation of the Supply. Am J Epidemiol 2002; 156:374– Koivusalo M, T. Hakulinen, T. Vartiainen, et Developmental Toxicity of 382. al., 1998. Drinking Water Mutagenicity and Trichloroethylene and Detoxification ILSI 1998. International Life Sciences Urinary Tract Cancers: a Population-Based Metabolites using Xenopus. Teratogenesis, Institute. Exposure to Contaminants in Case-Control Study in Finland. American Carcinogenesis, and Mutagenesis. 13:35– Drinking Water Estimating Uptake through Journal of Epidemiology 148(7):704–12. 45. the Skin and by Inhalation. Kramer M.D., C.F. Lynch, P. Isacson, J.W. Fu, L., E.M. Johnson and L.M. Newman. Infante-Rivard, C., E. Olson, L. Jacques and Hanson, 1992. The Association of 1990. Prediction of the Developmental P. Ayotte. 2001. Drinking Water Waterborne Chloroform with Intrauterine Toxicity Hazard Potential of Halogenated Contaminants and Childhood Leukemia. Growth Retardation. Epidemiology 3:407– Drinking Water Disinfection By-products Epidemiology 12(1):13–19. 413. Tested by the in vitro Hydra Assay. Reg. IRIS 1991. Integrated Risk Information Krasner, S.W., Sclimenti, M.J., and Hwang, Toxicol. and Pharmacol. 11:213–219. System (IRIS). N-nitrosodimethylamine C.J. 1989. Experiences with Implementing Gallagher, M.D., J.R. Nuckols, L. Stallones (NDMA). Washington, DC: U. S. a Laboratory Program to Sample and and D.A. Savitz. 1998. Exposure to Environmental Protection Agency. http:// Analyze for Disinfection By-Products in a Trihalomethanes and Adverse Pregnancy www.epa.gov/iris/subst/0045.htm National Study. In Disinfection By- Outcomes. Epidemiology. 9:484–489. IRIS 2001. Integrated Risk Information Products: Current Perspectives, AWWA, Gerba, C.P., J.B. Rose and C.N. Haas. 1996. System (IRIS). Chloroform. Washington, Denver, CO. Sensitive Populations: Who is at the DC: U. S. Environmental Protection Latendresse, J.R. and M.A. Pereira. 1997. Greatest Risk. Int. J. Food and Agency. http://www.epa.gov/iris/subst/ Dissimilar Characteristics of N-methyl-N- Microbiology. 30:113–123. 0025.htm nitrosourea-initiated Foci and Tumors Giller, S., F. Le Curieux, F. Erb and D. Jaakkola JJK, Magnus P, Skrondal A, Hwang Promoted by Dichloroacetic Acid or Marzin. 1997. Comparative Genotoxicity of B–F, Becher G, Dybing E. 2001. Foetal Trichloroacetic Acid in the Liver of Female Halogenated Acetic Acids Found in growth and duration of gestation relative to B6C3F1 Mice. Toxicol. Pathol. 25(5): 433– Drinking Water. Mutagenesis. 12(5):321– water chlorination. Occup Environ Med 440. 328. 58:437–442. Linder, R.E., G.R. Klinefelter, L.F. Strader, Goldsworthy, T.L. and J.A. Popp. 1987. Ji, Y., C. Qin-Yao, W. Xiao-fei, L. Yi and L. J.D. Suarez, and C.J. Dyer. 1994a. Acute Chlorinated Hydrocarbon-Induced Hong-mei. 1998. Prescreening Teratogenic Spermatogenic Effects of Bromoacetic

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00113 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49660 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

Acids. Fundamental and Applied Nelson, M.A. and R. J. Bull. 1988. Induction Canada, Ottawa, Ontario, Canada. Toxicology. 22: 422–430. of Strand Breaks in DNA by Executive summary available at http:// Linder, R.E., G.R. Klinefelter, L.F. Strader, Trichloroethylene and Metabolites in Rat www.hc-sc.gc.ca/pphb-dgspsp/publicat/ J.D. Suarez, N.L. Roberts, and C.J. Dyer. and Mouse livers in vivo. Toxicol. Appl. reif/index.html. 1994b. Spermatotoxicity of Dibromoacetic Pharmacol. 94:45–54. Reimann, S., K. Grob and H. Frank. 1996. Acid in Rats after 14 Daily Exposures. Nieuwenhuijsen, M.J., M.B. Toledano, N.E. Environmental chloroacetic acids in foods Reproductive Toxicology. 8(3): 251–259. Eaton, J. Fawell and P. Elliott. 2000. analyzed by GC–ECD. Mitteilungen Aus Linder, R.E., G.R. Klinefelter, L.F. Strader, Chlorination Disinfection By-products in Dem Gebiete der M.G. Narotsky, J.D. Suarez, N.L. Roberts Water and Their Association with Adverse Lebensmitteluntersuchung und Hygiene. and S.D. Perreault. 1995. Dibromoacetic Reproductive Outcomes: A Review. Occup. 87 (2):212–222. Acid Affects Reproductive Competence Environ. Med., 57(2):73–85. Rice, 2000—personal communication: e-mail and Sperm Quality in the Male Rat. NOAA 1998. Palmer Drought Severity Index 7/14/2000. Fundamental and Applied Toxicology. 28: Maps http://www.cpc.noaa.gov/products/ Ruddick, J.A., D.C. Villeneuve, and I. Chu, 9–17. monitoring_and_data/drought.html. 1983. A Teratological Assessment of Four Linder, R.E., G.R. Klinefelter, L.F. Strader, NTP 1987. National Toxicology Program. Trihalomethanes in the Rat. J. Environ. Sci. J.D. Suarez, and N.L. Roberts. 1997a. Toxicity and carcinogenesis studies of Health B18(3), 333–349. Spermatotoxicity of Dichloroacetic Acid. bromodichloromethane (CAS No. 75–27–4) Saillenfait, A. M., I. Langonne and J. P. Reproductive Toxicology. 11(5): 681–688. in F344/N rats and B6C3F1 mice (gavage Sabate. 1995. Developmental Toxicity of Linder, R.E., G.R. Klinefelter, L.F. Strader, studies). Technical Report Series No. 321. Trichloroethylene, Tetrachloroethylene D.N. Veeramachaneni, N.L. Roberts and Research Triangle Park, NC: U.S. and Four of Their Metabolites in Rat Whole J.D. Suarez, 1997b. Histopathologic Department of Health and Human Services. Embryo Culture. Arch. Toxicol. 70:71–82. Changes in the Testes of Rats Exposed to NTP 1989. Toxicology and carcinogenesis Salhi, E. and von Gunten, U. 1999. Dibromoacetic Acid. Reprod. Toxicol. studies of tribromomethane (bromoform) in Simultaneous Determination of Bromide, 11(1), 47–56. F344/N rats and B6C3F1 mice (gavage Bromate and Nitrite in Low µg l-1 Levels by Mackay, J.M., V. Fox, K. Griffiths, D.A. Fox, studies). Technical Report Series No. 350. Ion Chromatography without Sample C.A. Howard, C. Coutts, I. Wyatt and J.A. Research Triangle Park, NC: U.S. Pretreatment. Water Research. 33 Styles. 1995. Trichloroacetic Acid: Department of Health and Human Services. (15):3239–3244. Investigation into the Mechanism of NTP 1992. NTP technical Report on the Sanchez, I. M. and R. J. Bull. 1990. Early Chromosomal Damage in the in vitro Toxicology and Carcinogenesis Studies of Induction of Reparative Hyperplasia in Human Lymphocyte Cytogenetic Assay Monochloroacetic Acid (CAS No. 79–11–8) B6C3F1 Mice Treated with Dichloroacetate and the Mouse Bone Marrow Micronucleus in F344/N rats and B6C3F1 Mice (Gavage and Trichloroacetate. Toxicology. 64:33– Test. Carcinogenesis. 16(5): 1127–1133. Studies). NTP TR 396. NTIS Publication 46. Magat, W.A., W.K. Viscusi, and J. Huber. No. PB92–189372. Savitz, D. A., K.W. Andrews and L. M. 1996. ‘‘A Reference Lottery Metric for OSTP 1985. Chemical Carcinogens; A Review Pastore. 1995. Drinking Water and Valuing Health.’’ Management Science of the Science and Its Associated Pregnancy Outcome in Central North 42:1118–1130. Principles, February 1985. Presented in Carolina: Source, Amount, and Magnus, P., J.J.K. Jaakkola, A. Skrondal, J. Alexander, G. Becher, T. Krogh and E. Risk Analysis: A guide to Principles and Trihalomethane levels. Environ. Health Dybing. 1999. Water Chlorination and Methods for Analyzing Health and Perspectives. 103(6), 592–596. Birth Defects. Epidemiology. 10:513–517. Environmental Risks. Appendix G. Fed. Schwetz, B.A., K.J. Leong, and P.J. Gehring, Malley, J., J. Show, and J. Ropp. 1996. Reg., Pages 10371–10442. (March 14, 1974. Embryo- and Fetotoxicity of Inhaled Evaluation of the by-products produced by 1985). Chloroform in Rats. Toxicol. Appl. the treatment of groundwaters with Overbeck, P.K. 2000. WQA Ozone Task Pharmacol. 28(3), 442–451. ultraviolet radiation. American Water Force—An Update. Water Conditioning Seidel, C. 2001. BAT Memorandum on Works Association Research Foundation, and Purification. 42(3) 76–78. SWAT Runs for Stage 2 BAT Evaluation. Denver, CO. Parrish, J.M., E.W. Austin, D.K. Stevens, D.H. (June 25, 2001). Mather, G.G, J.H. Exon and L.D. Koller. 1990. Kinder and R.J. Bull. 1996. Haloacetate- Simmons, J.E.; S Richardson, T. Speth, R. Subchronic 90-day Toxicity of Induced Oxidative Damage to DNA in the Miltner, G. Rice, K. Schenck, E.S. Hunter Dichloroacetic and Trichloroacetic Acid in Liver of Male B6C3F1 Mice. Toxicology. III, and L. Teuschler. 2002. Development of Rats. Toxicology 64: 71–80. 110:103–111. a Research Strategy for Integrated Murray, F.J., B.A. Schwetz, J.G. McBride, and Pawlecki-Vonderheide, A.M., Munch, D.J., Technology-Based Toxicological and R.E. Staples, 1979. Toxicity of Inhaled and Munch, J.W. 1997. Research Chemical Evaluation of Complex Mixtures Chloroform in Pregnant Mice and Their Associated with the Development of EPA of Drinking Water Disinfection Byproducts. Offspring. Toxicol. Appl. Pharmacol. 50(3), Method 552.2. J. of Chromatographic Environmental Health Perspectives Vol. 515–522. Science. 35:293–301. 110 Supplement 6, 1013–1024. Narotsky, M.G., and R.J. Kavlock. 1992. Pereira, M. A. 1996. Carcinogenic Activity of Smith, M.K., J.L. Randall, and J.A. Stober. Effects of Bromoform and Dichloroacetic Acid and Trichloroacetic 1988. Developmental effects of Bromodichloromethane in an in vivo Acid in the Liver of Female B6C3F1 Mice. trichloroacetic acid in Long-Evans rats. Developmental Toxicity Screen. EPA Fundam. Appl. Toxicol. 31:192–199. Teratology 37(5), 495. report to Office of Water. Pereira, M.A. and J.B. Phelps. 1996. Smith, M.K., J.L. Randall, E.J. Read and J.A. Narotsky, M.G., B.T. Hamby, and D.S. Best, Promotion by Dichloroacetic Acid and Stober. 1989. Teratogenic Effects of 1997a. Developmental Effects of Trichloroacetic Acid of N-methyl-N- Trichloroacetic Acid in the Rat. Teratology. Dibromoacetic Acid (DBA) in a Segment II nitrosourea-initiated cancer in the Liver of 40: 445–451. Study in Mice. Teratology 55 (1), 67. Female B6C3F1 Mice. Cancer Lett. Smith, M.K., J.L. Randall, E.J. Read, and J.A. Narotsky, M.G., R.A. Pegram, and R.J. 102:133–141. Stober. 1990. Developmental effects of Kavlock. 1997b. Effect of dosing Vehicle on Personal communication from M. Kogevinas Chloroacetic acid in the Long-Evans Rat. the Developmental Toxicity of to M. Messner, 5/19/2003. Teratology 41 (5), 593 (Abstract No. P164). Bromodichlomethane and Carbon Raymer, J.H., Pellizzari, E.D., Hu, Y. et al. Smith, V.K., G. Van Houtven and S.K. Tetrachloride in Rats. Fundamental and (2001). Assessment of Human Dietary Pattanayak. 2002. Benefit transfer via Applied Tox. 40:30–36. Ingestion Exposures to Water Disinfection preference calibration: ‘Prudential algebra’ NATICH 1993. National Air Toxics Byproducts via Food. Star Drinking Water for policy. Land Economics, 78(1):132–152. Information Clearinghouse. Acceptable Progress Review Meeting, February 22–23, Stauber, A.J. and R.J. Bull. 1997. Differences ambient concentration guidelines or 2001, Silver Spring, MD. in Phenotype and Cell Replicative standards by pollutants: Trichloroacetic Reif, J.S., A. Bachand and M. Andersen. Behavior of Hepatic Tumors Induced by acid. Washington, DC: U.S. Environmental 2000. Reproductive and Developmental Dichloroacetate (DCA) and Protection Agency, Office of Air Quality Effects of Disinfection By-Products. Bureau Trichloroacetate (TCA). Toxicol. Appl. Planning and Standards. April 22, 1993. of Reproductive and Child Health, Health Pharmacol. 144(2): 235–46.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00114 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49661

Tao, L., K. Li, P.M. Kramer and USEPA 1993. EPA Method 300.0. In Methods Before 1996. Office of Water. EPA 815–R– M.A.Perei.1996.Loss of Heterozygosity on for the Determination of Inorganic 98–003. Chromosome 6 in Dichloroacetic Acid and Substances in Environmental Samples. USEPA 1998i. National Primary Drinking Trichloroacetic Acid-Induced Liver EPA/600/R/93/100. Water Regulations: Consumer Confidence Tumors in Female B6C3F1 Mice. Cancer USEPA 1994a. Draft Drinking Water Health Reports; Final Rule. FR 63:160:44512– Lett. 108: 257–261. Criteria Document for Chlorinated Acetic 44536. Tao, L., P.M. Kramer, R. Ge and M.A. Pereira. Acids/Alcohols/Aldehydes and Ketones. USEPA 1998j. Revisions to State Primacy 1998. Effect of Dichloroacetic Acid and Office of Science and Technology, Office of Requirements to Implement Safe Drinking Trichloroacetic Acid on DNA Methylation Water. Water Act Amendments; Final Rule. FR in Liver and Tumors of Female B6C3F1 USEPA 1994b. National Primary Drinking 63:81:23362–23368. Mice. Toxicol. Sciences. 43: 139–144. Water Regulations; Disinfectants and USEPA 1999a. Guidelines for carcinogen risk Thompson, D.L., S.D. Warner, and V.B. Disinfection Byproducts; Proposed Rule. assessment. July SAB Review draft. Office Robinson, 1974. Teratology Studies in FR 59:145:38668–38829. (July 29, 1994). of Research and Development, Washington, Orally Administered Chloroform in the Rat USEPA 1995. EPA Method 552.2. In Methods DC. USEPA NCEA–F–0644. http:// and Rabbit. Toxicol. Appl. Pharmacol. 29, for the Determination of Organic www.epa.gov/ncea/raf/crasab.htm 348–357. Compounds in Drinking Water. USEPA 1999b. National Primary and Toth, G.P., K.C. Kelty, E.L. George, E.J. Read, Supplement III. EPA–600/R–95/131. NTIS, Secondary Drinking Water Regulations: and M.K. Smith, 1992. Adverse Male PB95261616. Analytical Methods for Chemical and Reproductive Effects Following Subchronic USEPA 1996a. National Primary Drinking Microbiological Contaminants and Exposure of Rats to Sodium Water Regulation: Monitoring Revisions to Laboratory Certification Dichloroacetate. Fund. Appl. Toxicol. 19, Requirements for Public Drinking Water Requirements; Final Rule. FR 57–63. Supplies: Cryptosporidium, Giardia, 64:230:67449. (December 1, 1999). Tyl, R.W. 2000. Review of Animal Studies for Viruses, Disinfection Byproducts, Water USEPA 1999c. Chloroform Mode of Action Reproductive and Developmental Toxicity Treatment Plant Data and Other Analysis. Prepared for the Science Assessment of Drinking Water Information Requirements. Final Rule. FR Advisory Board by Office of Science and Contaminants: Disinfection By-Products 61:94:24354–24388. (May 14, 1996). Technology, Office of Water. October 1999. (DBPs). RTI Project No. 07639. Research USEPA 1996b. DBP/ICR Analytical Methods http://www.epa.gov/sab/chloro00.htm Triangle Institute. Manual. EPA 814–B–96–002. NTIS, PB96– USEPA 1999d. Cost of Illness Handbook. USDOE Energy Information Administration 157516. Office of Pollution Prevention and Toxics. 2002. Table 7.1 Electricity Overview USEPA 1997a. National Primary Drinking Chapter 1 II.8. Cost of Bladder Cancer. (Billion Kilowatthours). http:// Water Regulations; Disinfectants and September, 1999. http://www.epa.gov/ www.eia.doe.gov/emeu/mer/txt/mer7–1 Disinfection Byproducts; Notice of Data oppt/coi USEPA 1979. National Interim Primary Availability; Proposed Rule. FR USEPA 2000a. Estimated per Capita Water Drinking Water Regulations; Control of 62:212:59388–59484. (November 3, 1997). Ingestion in the United States. EPA– Trihalomethanes in Drinking Water. FR USEPA 1997b. Manual for the Certification of 82200–008. http://www.epa.gov/ 44:231:68624. (November 29, 1979). Laboratories Analyzing Drinking Water. waterscience/drinking/percapita/ USEPA 1985. National Primary Drinking EPA 815–B–97–001. http://www.epa.gov/ USEPA 2000b. Guidelines for Preparing Water Regulations; Volatile Synthetic OGWDW/certlab/labindex.html Economic Analyses. Washington, DC. EPA Organic Chemicals; Final Rule and USEPA 1998a. Quantification of Bladder 240R–00–003, September 2000. Proposed Rule. FR 50:219:46880 Cancer Risk from Exposure to Chlorinated USEPA 2000c. Information Collection Rule (September 13, 1985). Surface Water. Office of Science and Auxiliary 1 Database, Version 5, EPA 815– USEPA 1986. Guidelines for Carcinogen Risk Technology, Office of Water. November 9, C–00–002, April 2000. Assessment, FR 51:185:33992–34003. EPA/ 1998. USEPA 2000d. EPA Method 321.8. In 600/8–87/045. NTIS PB88–123997. http:// USEPA 1998b. Health Risk Assessment/ Methods for the Determination of Organic www.epa.gov/ncea/raf/rafguid.htm Characterization of the Drinking Water and Inorganic Compounds in Drinking USEPA 1989a. National Primary Drinking Disinfection Byproduct Chloroform. Office Water, Volume 1. ORD–NERL, Cincinnati, Water Regulations; Filtration, Disinfection, of Science and Technology, Office of OH. EPA 815–R–00–014. Available on Turbidity, Giardia lamblia, Viruses, Water. EPA 815–B–98–006. PB 99–111346. ORD–NERL Web site at http:// Legionella, and Heterotrophic Bacteria; USEPA 1998c. National Primary Drinking www.epa.gov/nerlcwww/ordmeth.htm. Final Rule. Part II. FR 54:124: 27486. (June Water Regulations: Disinfectants and USEPA 2000e. Removal of the Maximum 29, 1989). Disinfection Byproducts; Final Rule. FR Contaminant Level Goal for Chloroform USEPA 1989b. National Primary Drinking 63:241:69390–69476. (December 16, 1998). From the National Primary Drinking Water Water Regulations; Total Coliforms http://www.epa.gov/safewater/mdbp/ Regulations. FR 65:104:34404–34405. (May (Including Fecal Coliform and E. coli); dbpfr.pdf 30, 2000). http://www.epa.gov/safewater/ Final Rule. FR 54:124: 27544. (June 29, USEPA 1998d. National Primary Drinking regs/chlorfr.html 1989). Water Regulations: Interim Enhanced USEPA 2000f. Review of the EPA’s Draft USEPA 1989c. Review of Environmental Surface Water Treatment Rule; Final Rule. Chloroform Risk Assessment by a Contaminants and Toxicology. U.S. EPA. FR 63:241:38832–38858. (December 16, Subcommittee of the Science Advisory Office of Drinking Water Health 1998). http://www.epa.gov/safewater/ Board. Science Advisory Board, Advisories. Volume 106. 225 pp. mdbp/ieswtrfr.pdf Washington, DC. EPA–SAB–EC–00–009. USEPA 1991a. National Primary Drinking USEPA 1998e. National Primary Drinking USEPA 2000g. Stage 2 Microbial and Water Regulations; Synthetic Organic Water Regulations; Disinfectants and Disinfection Byproducts Federal Advisory Chemicals and Inorganic Chemicals; Disinfection Byproducts; Notice of Data Committee Agreement in Principle. FR Monitoring for Unregulated Contaminants; Availability; Proposed Rule. FR 65:251:83015–83024. (December 29, 2000). National Primary Drinking Water 63:61:15606–15692. (March 31, 1998). http://www.epa.gov/fedrgstr/EPA–WATER/ Regulations Implementation; National USEPA 1998f. Regulatory Impact Analysis of 2000/December/Day-29/w33306.htm Secondary Drinking Water Regulations. Final Disinfectant/Disinfection By- USEPA 2000h. National Primary Drinking Final rule, January 31, 1991. FR 56:20: Products Regulations. Washington, DC. Water Regulations: Ground Water Rule. 3526. EPA Number 815–B–98–002. PB 99– Proposed Rules. FR 65:91:30194–30274. USEPA 1991b. Guidelines for Developmental 111304. (May 10, 2000). Toxicity Risk Assessment. FR USEPA 1998g. National-Level Affordability USEPA 2000i. Quantitative Cancer 56:234:63798–63826. Criteria Under the 1996 Ammendments to Assessment for MX and USEPA 1992. EPA Method 552.1. In Methods the Safe Drinking Water Act (Final Draft Chlorohydroxyfuranones. Contract NO. 68– for the Determination of Organic Report). Contact 68–C6–0039. (August 19, C–98–195. August 11, 2000, Office of Compounds in Drinking Water— 1998). Water, Office of Science and Technology, Supplement II. EPA 600/R–92/129. NTIS, USEPA 1998h. Variance Technology Health and Ecological Criteria Division, PB92–207703. Findings for Contaminants Regulated Washington, DC.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00115 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49662 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

USEPA 2000j. Drinking Water Baseline USEPA 2003c. Addendum to the Criteria Sufrin, G. 1993. Drinking water, fluid Handbook, Second Edition. Prepared by Document for Dichloroacetic Acid: intake, and bladder cancer in western New International Consultants, Inc. under External Review Draft. York. Archives of Environmental Health, contract with EPA OGWDW, Standards USEPA 2003d. Drinking Water Criteria 48(3):191–8. and Risk Management Division. March 17, Document for Brominated Ventura, S.J., W.D. Mosher, S.C. Curtin, J.C. 2000. Trihalomethanes: External Review Draft. Abma, and S. Henshaw. 2000. ‘‘Trends in USEPA 2000k. Geometries and USEPA 2003e. Drinking Water Criteria Pregnancies and Pregnancy Rates by Characteristics of Public Water Systems. Document for Brominated Haloacetic Outcome: Estimates for the United States, Final Report. EPA 815–R–00–024. Acids: External Review Draft. 1976–96.’’ National Center for Health December 2000. USEPA 2003f. Drinking Water Criteria Statistics. Vital Health Stat 21(56). USEPA 20001. EPA Method 300.1. In Document for Cyanogen Chloride, External Villanueva, C.M., F. Fernandez, N. Malats, Methods for the Determination of Organic Review Draft. J.O. Grimalt, M. Kogevinas. 2003. Meta- and Inorganic Compounds in Drinking USEPA 2003g. Drinking Water Criteria analysis of Studies on Individual Water, Volume 1. OW–OGWDW–TSC, Document for Glyoxal and Methylglyoxal: Consumption of Chlorinated Drinking External Review Draft. Water and Bladder Cancer. J Epidemiol Cincinnati, OH. EPA 815–R–00–014. USEPA 2003h. Drinking Water Criteria Community Health, 57:166–173. Available on the OGWDW Web site at Document for Haloacetonitriles: External Wagner, H.P., Pepich, B.V., Frebis, C., http://www.epa.gov/safewater/methods/ Review Draft. Hautman, D.P., Munch, D.J., and Jackson, sourcalt.html. USEPA 2003i. Economic Analysis for the P.E. 2001. A Collaborative Study of EPA USEPA 2000m. Information Collection Rule Proposed Stage 2 DBPR. Washington, DC. Method 317.0 for the Determination of Treatment Study Database CD–ROM, EPA 815–D–03–001. Inorganic Oxyhalide Disinfection By- Version 1.0. USEPA 2003j. Draft Initial Distribution Products in Drinking Water using Ion USEPA 2000n. Science Advisory Board Final System Evaluation Guidance Manual. Chromatography with the Addition of a Report. Prepared for Environmental Washington, DC. EPA 815–D–03–002. Postcolumn Reagent for Trace Bromate Economics Advisory Committee. July 27, USEPA 2003k. Technologies and Costs for Analysis. Journal of Chromatographic 2000. EPA–SAB–EEAC–00–013. Control of Microbial Pathogens and Science. Vol 39 (255–259), June 2001. USEPA 2000o. Draft Dioxin Reassessment. Disinfection Byproducts. Prepared by the Wagner, H.P., Pepich, B.V., Frebis, C., EPA/600/P–00/001B http://cfpub.epa.gov/ Cadmus Group and Malcolm Pirnie. Hautman, D.P. and Munch, D.J. 2002. U.S. ncea/cfm/part1and2.cfm?ActType=default. USEPA 2003l. Toxicologcal Review for Environmental Protection Agency Method USEPA 2001a. Relative Source Contribution Dichloroacetic Acid: Consensus Review 326.0, a new method for monitoring for Chloroform. EPA–822-R–01–006. Draft. http://www.epa.gov/iris/subst/ inorganic oxyhalides and optimization of USEPA 2001b. Toxicological Review of 0654.htm the postcolumn derivatization for the Chloroform. In support of Integrated Risk USEPA 2003m. Information Collection selective determination of trace levels of Information System (IRIS). Washington, Request. Washington, DC. EPA 815–D–03– bromate. Journal of Chromatography. A. DC. Draft. EPA/635/R–01/001. 003. Vol. 956 (93–101), May 2002. USEPA 2001c. National Primary Drinking USEPA 2003n. Draft Significant Excursion Wallace, L.A. 1997. Human exposure and Water Regulations: Filter Backwash Guidance Manual. Washington, DC. EPA Body Burden for Chloroform and Other Recycling Rule. Final Rule. FR 815–D–03–004. Trihalomethanes., Crit. Rev. Environ. Sci. 66:111:31086–31105. (June 8, 2001). USEPA 2003o. Stage 2 Occurrence Technol. 27:113–94. USEPA 2001d. Method 317.0, Revision 2.0. Assessment for Disinfectants and Waller, K., S.H. Swan, G. DeLorenze, B. Determination of Inorganic Oxyhalide Disinfection Byproducts (D/DBPs). EPA Hopkins. 1998. Trihalomethanes in Disinfection By-Products in Drinking 68–C–99–206. Drinking Water and Spontaneous Abortion. Water Using Ion Chromatography with the USEPA 2003p. Method 552.3. Determination Epidemiology. 9(2):134–140. Addition of a Postcolumn Reagent for of Haloacetic Acids and Dalapon in Waller, K., S.H. Swan, G.C. Windham, L. Trace Bromate Analysis. Revision 2.0. EPA Drinking Water by Liquid-liquid Fenster. 2001. Influence of Exposure 815–B–01–001. (Available on the OGWDW Extraction, Derivatization, and Gas Assessment Methods on Risk Estimates in Web site at http://www.epa.gov/safewater/ Chromatography with Electron Capture an Epidemiologic Study of Total methods/sourcalt.html.) Detection. Revision 1.0. (Available on the Trihalomethane Exposure and USEPA 2001e. Arsenic Rule Benefits OGWDW Web site at http://www.epa.gov/ Spontaneous Abortion. Journal of Exposure safewater/methods/sourcalt.html.) Analysis and Environmental Analysis: an SAB Review. August 30, 2001. USEPA 2003q. Method 327.0. Determination Epidemiology. 11(6): 522–531. EPA–SAB–EC–01–008. of Chlorine Dioxide and Chlorite Ion in Weisel, C.P. and W.K. Jo. 1996. Ingestion, USEPA 2002a. Method 326.0. Determination Drinking water Using Lissamine Green B Inhalation, and Dermal Exposures to of Inorganic Oxyhalide Disinfection By- and Horseradish Peroxidase with Detection Chloroform and Trichloroethene from Tap Products in Drinking Water Using Ion by Visible Spectrophotometry. Revision Water. Environmental Health Perspectives. Chromatography Incorporating the 1.0. (Available on the OGWDW Web site at 104 (1): 48–51. Addition of a Suppressor Acidified http://www.epa.gov/safewater/methods/ WHO 2000. World Health Organization, Postcolumn Reagent for Trace Bromate sourcalt.html.) International Programme on Chemical Analysis. Revision 1.0. EPA 815–R–03– USEPA 2003r. Method 415.3. Determination Safety (IPCS). Environmental Health 007. (Available on the OGWDW Web site of Total Organic Carbon, and Specific UV Criteria 216: Disinfectants and Disinfectant at http://www.epa.gov/safewater/methods/ Absorbance at 254 nm in Source Water and By-products. sourcalt.html.) Drinking Water. Revision 1.0. NERL, Williams, S.L., Rindfleisch, D.F., and USEPA 2002b. Long Term 1 Enhanced Cincinnati, OH 45268. Williams, RL. 1995. Deadend on Haloacetic Surface Water Treatment Rule. January 14, USEPA 2003s. Arsenic in Drinking Water: Acids (HAA). In Proceedings of the 1994 2002. 67 FR 1812. Cessation Lag Model. Prepared by Sciences AWWA Water Quality Technology USEPA 2002c. Affordability Criteria for International. Contract No. 68–c–98–195. Conference, November 1994. Small Drinking Water Systems: an EPA January, 2003. Windham GC, Waller K, Anderson M, Science Advisory Board Report. December Veeramachaneni, D.N.R., T.T. Higuchi, J.S. Fenster L, Mendola P, Swan S. 2003. 2002. EPA–SAB–EEAC–03–004. Palmer, and C.M. Kane. 2000. Chlorination by-Products in Drinking USEPA 2003a. Health Risks to Fetuses, Dibromoacetic Acid, a Disinfection By- Water and Menstrual Cycle Function. Infants, and Children: A Review. Office of product in Drinking Water, Impairs Sexual Environ Health Perspect: doi:10.1289/ Water, Office of Science and Technology, Function and Fertility in Male Rabbits. ehp.5922. http://ehpnet1.niehs.nih.gov/ Health and Ecological Criteria Division. Paper presented at the annual meeting for docs/2003/5922/abstract.html USEPA 2003b. Addendum to the Criteria the Society for the Study of Reproduction, Yang, C.Y., H.F. Chiu, M.F. Cheng, et al. Document for Monochloroacetic Acid and Madison, Wisconsin. 1998. Chlorination of Drinking Water and Trichleoeacetic Acid: External Review Vena, JE, Graham, S, Freudenheim, J, Cancer Mortality in Taiwan. Draft. Marshall, J, Zielezny, M, Swanson, M, Environmental Research 78(1):1–6.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00116 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49663

Yang, V., B. Cheng, S. Tsai, T. Wu, M. Lin 2. Section 141.2 is amended by Finished water is water that is M. and K. Lin. 2000. Association between adding, in alphabetical order, introduced into the distribution system Chlorination of Drinking Water and definitions for ‘‘Combined distribution of a public water system and is intended Adverse Pregnancy Outcome in Taiwan. Environ. Health. Perspect. 108:765–68. system’’, ‘‘Consecutive system’’, for distribution without further Zheng, M., S. Andrews, and J. Bolton. 1999. ‘‘Consecutive system entry point’’, treatment, except that necessary to Impacts of medium-pressure UV on THM ‘‘Dual sample sets’’, ‘‘Finished water’’, maintain water quality. and HAA formation in pre-UV chlorinated ‘‘Locational running annual average’’, * * * * * drinking water. Proceedings, Water Quality and ‘‘Wholesale system’’ to read as Locational running annual average Technology Conference of the American follows: Water Works Association, Denver, CO. (LRAA) is the average of sample § 141.2 Definitions. analytical results for samples taken at a List of Subjects * * * * * particular monitoring site during the previous four calendar quarters. 40 CFR Part 141 Combined distribution system is the Chemicals, Indians-lands, interconnected distribution system * * * * * Intergovernmental relations, Radiation consisting of the distribution systems of Stage 2A is the period beginning [date protection, Reporting and recordkeeping wholesale systems and of the three years following publication of the requirements, Water supply. consecutive systems that receive final rule] until the dates specified in 40 CFR Part 142 finished water from those wholesale subpart V of this part for compliance system(s). with Stage 2B, during which systems Administrative practice and * * * * * must comply with Stage 2A MCLs in procedure, Chemicals, Indians-lands, § 141.64(b)(2). Radiation protection, Reporting and Consecutive system is a public water recordkeeping requirements, Water system that buys or otherwise receives * * * * * supply. some or all of its finished water from Wholesale system is a public water one or more wholesale systems, for at system that treats source water and then 40 CFR Part 143 least 60 days per year. sells or otherwise delivers finished Chemicals, Indians-lands, Water Consecutive system entry point is a water to another public water system for supply. location at which finished water is at least 60 days per year. Delivery may Dated: July 11, 2003. delivered at least 60 days per year from be through a direct connection or Linda J. Fisher, a wholesale system to a consecutive through the distribution system of one Acting Administrator. system. or more consecutive systems. For the reasons set forth in the * * * * * 3. In § 141.23, the table in paragraph preamble, title 40 chapter I of the Code Dual sample set is a set of two (k)(1) is amended by revising entries 13, of Federal Regulations is proposed to be samples collected at the same time and 18, 19, and 20; revising the amended as follows: same location, with one sample undesignated text after the table; and analyzed for TTHM and the other adding a new footnote 19 to read as PART 141—NATIONAL PRIMARY sample analyzed for HAA5. Dual sample follows: DRINKING WATER REGULATIONS sets are collected for the purposes of § 141.23 Inorganic chemical sampling and 1. The authority citation for part 141 conducting an IDSE under subpart U of analytical requirements. continues to read as follows: this part and determining compliance with the TTHM and HAA5 MCLs under * * * * * Authority: 42 U.S.C. 300f, 300g–1, 300g–2, 300g–3, 300g–4, 300g–5, 300g–6, 300j–4, subpart V of this part. (k) Inorganic analysis: 300j–9, and 300j–11. * * * * * * * * * *

4 13 3 SM (18th, 19th 4 Contaminant and methodology EPA ASTM ed.) SM (20th ed.) Other

******* 13. Fluoride: Ion Chromatography ...... 6 300.0 D4327–97 4110 B 4110 B 19 300.1 Manual Distill.; Color. SPADNS...... 4500–F B, D 4500–F B, D Manual Electrode ...... D1179–93B 4500–F C 4500–F C Automated Electrode ...... 380–75WE 11 Automated Alizarin ...... 4500–F E 4500–F E 129–71W 11

******* 18. Nitrate: Ion Chromatography ...... 6 300.0 D4327–97 4110 B 4110 B B1011 8 19 300.1 6 Automated Cadmium Reduction ...... 353.2 D3867–90A 4500–NO3 F 4500–NO3 F 7 Ion Selective Electrode ...... 4500–NO3 D 4500–NO3 D 601 Manual Cadmium Reduction ...... D3867–90B 4500–NO3 E 4500–NO3 E 19. Nitrite: Ion Chromatography ...... 6 300.0 D4327–97 4110 B 4110 B B–10118 19 300.1 6 Automated Cadmium Reduction ...... 353.2 D3867–90A 4500–NO3 F 4500–NO3 F Manual Cadmium Reduction ...... D3867–90B 4500–NO3 E 4500–NO3 E Spectrophotometric ...... 4500–NO2 B 4500–NO2 B 20. Orthophosphate: 12

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00117 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49664 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

4 13 3 SM (18th, 19th 4 Contaminant and methodology EPA ASTM ed.) SM (20th ed.) Other

Colorimetric, automated, ascorbic acid ...... 6365.1 . 4500–P F 4500–P F Colorimetric, ascorbic acid, single reagent ...... D515–88A 4500–P E 4500–P E Colorimetric, phosphomolybdate ...... I–1601–855 Automated–segmented flow ...... I–2601–905 Automated discrete ...... I–2598–855 Ion Chromatography ...... 6 300.0 D4327–97 4110 B 4110 B 19 300.1

******* Note: The procedures shall be done in accordance with the documents listed below. The incorporation by reference of the following docu- ments listed in footnotes 1–11 and 16–19 was approved by the Director of the Federal Register in accordance with 5 U.S.C. 552(a) and 1 CFR part 51. Copies of the documents may be obtained from the sources listed below. Information regarding obtaining these documents can be ob- tained from the Safe Drinking Water Hotline at 800–426–4791. Documents may be inspected at EPA’s Drinking Water Docket, EPA West, 1301 Constitution Avenue NW., Room B102, Washington, DC 20460 (Telephone: 202–566–2426); or at the Office of the Federal Register, 800 North Capitol Street, NW., Suite 700, Washington, DC. ******* 3 Annual Book of ASTM Standards, 1994, 1996, or 1999, Vols. 11.01 and 11.02, ASTM International; any year containing the cited version of the method may be used. The previous versions of D1688–95A, D1688–95C (copper), D3559–95D (lead), D1293–95 (pH), D1125–91A (conduc- tivity) and D859–94 (silica) are also approved. These previous versions D1688–90A, C; D3559–90D, D1293–84, D1125–91A and D859–88, re- spectively are located in the Annual Book of ASTM Standards, 1994, Vol. 11.01. Copies may be obtained from ASTM International, 100 Barr Harbor Drive, West Conshohocken, PA 19428. 4 Standard Methods for the Examination of Water and Wastewater, 18th edition (1992), 19th edition (1995), or 20th edition (1998). American Public Health Association, 1015 Fifteenth Street, NW, Washington, DC 20005. The cited methods published in any of these three editions may be used, except that the versions of 3111 B, 3111 D, 3113 B and 3114 B in the 20th edition may not be used. 5 Method I–2601–90, Methods for Analysis by the U.S. Geological Survey National Water Quality Laboratory—Determination of Inorganic and Organic Constituents in Water and Fluvial Sediment, Open File Report 93–125, 1993; For Methods I–1030–85; I–1601–85; I–1700–85; I–2598– 85; I–2700–85; and I–3300–85 See Techniques of Water Resources Investigation of the U.S. Geological Survey, Book 5, Chapter A–1, 3rd ed., 1989; Available from Information Services, U.S. Geological Survey, Federal Center, Box 25286, Denver, CO 80225–0425. 6 ‘‘Methods for the Determination of Inorganic Substances in Environmental Samples’’, EPA/600/R–93/100, August 1993. Available at NTIS, PB94–120821. 7 The procedure shall be done in accordance with the Technical Bulletin 601 ‘‘Standard Method of Test for Nitrate in Drinking Water’’, July 1994, PN 221890–001, Analytical Technology, Inc. Copies may be obtained from ATI Orion, 529 Main Street, Boston, MA 02129. 8 Method B–1011, ‘‘Waters Test Method for Determination of Nitrite/Nitrate in Water Using Single Column Ion Chromatography,’’ August 1987. Copies may be obtained from Waters Corporation, Technical Services Division, 34 Maple Street, Milford, MA 01757. ******* 11 Industrial Method No. 129–71W, ‘‘Fluoride in Water and Wastewater’’, December 1972, and Method No. 380–75WE, ‘‘Fluoride in Water and Wastewater’’, February 1976, Technicon Industrial Systems. Copies may be obtained from Bran & Luebbe, 1025 Busch Parkway, Buffalo Grove, IL 60089. 12 Unfiltered, no digestion or hydrolysis. 13 Because MDLs reported in EPA Methods 200.7 and 200.9 were determined using a 2X preconcentration step during sample digestion, MDLs determined when samples are analyzed by direct analysis (i.e., no sample digestion) will be higher. For direct analysis of cadmium and ar- senic by Method 200.7, and arsenic by Method 3120 B sample preconcentration using pneumatic nebulization may be required to achieve lower detection limits. Preconcentration may also be required for direct analysis of antimony, lead, and thallium by Method 200.9; antimony and lead by Method 3113 B; and lead by Method D3559–90D unless multiple in-furnace depositions are made. ******* 19 ‘‘Methods for the Determination of Organic and Inorganic Compounds in Drinking Water’’, Vol. 1, EPA 815–R–00–014, August 2000. Avail- able at NTIS, PB2000–106981.

* * * * * in Drinking Water, EPA/600/4–88–039, PB92–207703, PB95–261616 and PB95– 4. Section 141.24 is amended by December 1988, Revised, July 1991. 104774, U.S. Department of Commerce, revising paragraph (e)(1) and by revising Methods 547, 550 and 550.1 are in 5285 Port Royal Road, Springfield, entry 30 in the table in paragraph (e)(1) Methods for the Determination of Virginia 22161. The toll-free number is to read as follows: Organic Compounds in Drinking 800–553–6847. Method 6651 shall be § 141.24 Organic chemicals, sampling and Water—Supplement I, EPA/600–4–90– followed in accordance with Standard analytical requirements. 020, July 1990. Methods 548.1, 549.1, Methods for the Examination of Water 552.1 and 555 are in Methods for the and Wastewater, 18th edition (1992), * * * * * (e) * * * Determination of Organic Compounds 19th edition (1995), or 20th edition (1) The following documents are in Drinking Water—Supplement II, (1998), American Public Health incorporated by reference. This EPA/600/R–92–129, August 1992. Association (APHA); any of these three incorporation by reference was Methods 502.2, 504.1, 505, 506, 507, editions may be used. Method 6610 approved by the Director of the Federal 508, 508.1, 515.2, 524.2 525.2, 531.1, shall be followed in accordance with Register in accordance with 5 U.S.C. 551.1 and 552.2 are in Methods for the Standard Methods for the Examination 552(a) and 1 CFR Part 51. Copies may Determination of Organic Compounds of Water and Wastewater, (18th Edition be inspected at EPA’s Drinking Water in Drinking Water—Supplement III, Supplement) (1994), or with the 19th Docket, 1301 Constitution Avenue, NW., EPA/600/R–95–131, August 1995. edition (1995) or 20th edition (1998) of EPA West, Room B102, Washington, DC Method 1613 is titled ‘‘Tetra-through Standard Methods for the Examination 20460 (Telephone: 202–566–2426); or at Octa-Chlorinated Dioxins and Furans by of Water and Wastewater; any of these the Office of the Federal Register, 800 Isotope-Dilution HRGC/HRMS’’, EPA/ publications may be used. The APHA North Capitol Street, NW., Suite 700, 821–B–94–005, October 1994. These documents are available from APHA, Washington, DC. Method 508A and documents are available from the 1015 Fifteenth Street NW., Washington, 515.1 are in Methods for the National Technical Information Service, D.C. 20005. Other required analytical Determination of Organic Compounds NTIS PB91–231480, PB91–146027, test procedures germane to the conduct

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00118 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49665

of these analyses are contained in Chlorinated Acids in Drinking Water by Immunoassay’’, February 2001 is Technical Notes on Drinking Water Liquid-Liquid Microextraction, available from Syngenta Crop Methods, EPA/600/R–94–173, October Derivatization and Fast Gas Protection, Inc., 410 Swing Road, Post 1994, NTIS PB95–104766. EPA Methods Chromatography with Electron Capture Office Box 18300, Greensboro, NC 515.3 and 549.2 are available from U.S. Detection,’’ Revision 1.0, April 2000, 27419, Phone number (336) 632–6000. Environmental Protection Agency, EPA/815/B–00/001 and EPA Method Method 531.2 ‘‘Measurement of N- National Exposure Research Laboratory 552.3, ‘‘Determination of Haloacetic methylcarbamoyloximes and N- (NERL)—Cincinnati, 26 West Martin Acids and Dalapon in Drinking Water methylcarbamates in Water by Direct Luther King Drive, Cincinnati, OH by Liquid-Liquid Microextraction, Aqueous Injection HPLC with 45268. ASTM Method D 5317–93 is Derivatization, and Gas Chromatography Postcolumn Derivatization,’’ Revision available in the Annual Book of ASTM with Electron Capture Detection,’’ 1.0, September 2001, EPA 815/B/01/002 Standards, (1999), Vol. 11.02, ASTM Revision 1.0, July 2003 can be accessed can be accessed and downloaded International, 100 Barr Harbor Drive, and downloaded directly on-line at West Conshohocken, PA 19428, or in http://www.epa.gov/safewater/methods/ directly on-line at http://www.epa.gov/ any edition published after 1993. EPA sourcalt.html. The Syngenta AG–625, safewater/methods/sourcalt.html. Method 515.4, ‘‘Determination of ‘‘Atrazine in Drinking Water by

Contaminant EPA method 1 Standard methods ASTM Other

******* 30. Dalapon ...... 552.1, 515.1, 552.2, 515.3, 515.4, 552.3

******* 1 For previously approved EPA methods which remain available for compliance monitoring until June 1, 2001, see paragraph (e)(2) of this section.

* * * * * § 141.64 Maximum contaminant levels for Disinfection 5. Section 141.33 is amended by disinfection byproducts. byproduct Best available technology revising the first sentence of paragraph (a) Bromate and chlorite. The (a) introductory text, and adding Chlorite ...... Control of treatment processes maximum contaminant levels (MCLs) paragraph (f) to read as follows: to reduce disinfectant de- for bromate and chlorite are as follows: mand and control of dis- § 141.33 Record maintenance. infection treatment proc- * * * * * Disinfection byproduct MCL (mg/L) esses to reduce disinfectant levels. (a) Records of microbiological Bromate ...... 0.010 analyses and turbidity analyses made Chlorite ...... 1.0 pursuant to this part shall be kept for (b) TTHM and HAA5. not less than 5 years. * * * (1) Subpart L—RAA compliance. (i) * * * * * (1) Compliance dates for CWSs and Compliance dates. Subpart H systems (f) Copies of monitoring plans NTNCWSs. Subpart H systems serving serving 10,000 or more persons must developed pursuant to this part shall be 10,000 or more persons must comply comply with this paragraph (b)(1) kept for the same period of time as the with this paragraph (a) beginning beginning January 1, 2002 until the date records of analyses are required to be January 1, 2002. Subpart H systems specified for subpart V of this part kept under paragraph (a) of this section serving fewer than 10,000 persons and compliance in § 141.620(c). Subpart H or for three years after modification, systems using only ground water not systems serving fewer than 10,000 whichever is longer. under the direct influence of surface persons and systems using only ground 6. Section 141.53 is amended by water must comply with this paragraph water not under the direct influence of revising the table to read as follows: (a) beginning January 1, 2004. surface water must comply with this § 141.53 Maximum contaminant level goals (2) Best available technology. The paragraph (b)(1) beginning January 1, for disinfection byproducts. Administrator, pursuant to section 1412 2004 until the date specified for subpart * * * * * of the Act, hereby identifies the V of this part compliance in following as the best technology, § 141.620(c). Disinfection byproduct MCLG (mg/L) treatment techniques, or other means available for achieving compliance with Disinfection byproduct MCL Bromodichloromethane zero. the maximum contaminant levels for (mg/L) Bromoform zero. bromate and chlorite identified in this Bromate zero. Total trihalomethanes (TTHM) ...... 0.080 Chlorite 0.8 paragraph (a): Haloacetic acids (five) (HAA5) ...... 0.060 Chloroform 0.07 Dibromochloromethane 0.06 Disinfection Best available technology Dichloroacetic acid zero. byproduct (ii) Best available technology. The Monochloroacetic acid 0.03 Administrator, pursuant to section 1412 Trichloroacetic acid 0.02 Bromate ...... Control of ozone treatment of the Act, hereby identifies the process to reduce produc- following as the best technology, tion bromate. 7. Section 141.64 is revised to read as treatment techniques, or other means follows:

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00119 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49666 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

available for achieving compliance with following as the best technology, and V. These methods are effective for the maximum contaminant levels for treatment techniques, or other means compliance monitoring February 16, TTHM and HAA5 identified in this available for achieving compliance with 1999, unless a different effective date is paragraph (b)(1): the maximum contaminant levels for specified in this section or by the State. TTHM and HAA5 identified in this (2) The following documents are Best available paragraph (b)(3) for all systems that incorporated by reference. The Director Disinfection byproduct technology disinfect their source water: of the Federal Register approves this incorporation by reference in Total trihalomethanes Enhanced coagula- Disinfection (TTHM) and tion or enhanced byproduct Best available technology accordance with 5 U.S.C. 552(a) and 1 Halaocetic acids softening or CFR part 51. Copies may be inspected (five) (HAA5). GAC10, with chlo- Total Enhanced coagulation or at EPA’s Drinking Water Docket, 1301 rine as the primary trihalomethan- enhanced softening, plus Constitution Avenue, NW., EPA West, and residual es (TTHM) GAC10; or nanofiltration Room B102, Washington, DC 20460, or disinfectant. and with a molecular weight at the Office of the Federal Register, 800 ≤ Haloacetic and cutoff 1000 Dal- North Capitol Street, NW., Suite 700, (2) Stage 2A—LRAA compliance. (i) acids (five) tons; or GAC20. Washington, DC. EPA Method 552.1 is Compliance dates. The Stage 2A MCLs (HAA5). in Methods for the Determination of for TTHM and HAA5 must be complied Organic Compounds in Drinking Water- with as a locational running annual (iii) Best available technology for Supplement II, USEPA, August 1992, average at each subpart L of this part systems that buy disinfected water. The EPA/600/R–92/129 (available through compliance monitoring location under Administrator, pursuant to section 1412 National Information Technical Service § 141.136 beginning [date three years of the Act, hereby identifies the (NTIS), PB92–207703). EPA Methods after publication of the final rule] until following as the best technology, 502.2, 524.2, 551.1, and 552.2 are in the date specified for subpart V of this treatment techniques, or other means Methods for the Determination of part compliance in § 141.620(c). available for achieving compliance with the maximum contaminant levels for Organic Compounds in Drinking Water- TTHM and HAA5 identified in this Supplement III, USEPA, August 1995, Disinfection byproduct MCL (mg/L) paragraph (b)(3) for systems that buy EPA/600/R–95/131. (Available through disinfected water: NTIS, PB95–261616). EPA Method Total trihalomethanes (TTHM) ...... 0.120 300.0 for chlorite and bromide is in Haloacetic acids (five) (HAA5) ...... 0.100 Methods for the Determination of Disinfection Best available technology byproduct Inorganic Substances in Environmental (ii) Best available technology. The Samples, USEPA, August 1993, EPA/ Administrator, pursuant to section 1412 Total Improved distribution sys- 600/R–93/100 (available through NTIS, of the Act, hereby identifies the trihalomethan- tem and storage tank PB94–121811). EPA Methods 300.1 for es (TTHM) management to reduce following as the best technology, chlorite, bromate, and bromide and treatment techniques, or other means and detention time plus the Haloacetic use of chloramines for 321.8 for bromate are in Methods for the available for achieving compliance with acids (five) disinfectant residual Determination of Organic and Inorganic the maximum contaminant levels for (HAA5). maintenance. Compounds in Drinking Water, Volume TTHM and HAA5 identified in this 1, USEPA, August 2000, EPA 815–R– paragraph (b)(2): (c) Extensions. A system that is 00–014 (available through NTIS, installing GAC or membrane technology PB2000–106981). EPA Method 317.0, Disinfection Best available to comply with the MCLs in paragraphs byproduct technology Revision 2.0, ‘‘Determination of (a) or (b)(1) of this section may apply to Inorganic Oxyhalide Disinfection By- Total Enhanced coagulation the State for an extension of up to 24 Products in Drinking Water Using Ion trihalomethanes or enhanced soft- months past January 1, 2002, but not Chromotography with the Addition of a (TTHM) and ening or GAC10, with beyond January 1, 2004. In granting the Postcolumn Reagent for Trace Bromate Haloacetic acids chlorine as the pri- extension, States must set a schedule for Analysis,’’ USEPA, July 2001, EPA 815– (five) (HAA5). mary and residual compliance and may specify any B–01–001, EPA Method 326.0, Revision disinfectant. interim measures that the system must 1.0, ‘‘Determination of Inorganic take. Failure to meet the schedule or any Oxyhalide Disinfection By-Products in (3) Subpart V LRAA compliance. (i) interim treatment requirements Drinking Water Using Ion Compliance dates. The subpart V of this constitutes a violation of a National Chromatography Incorporating the part MCLs for TTHM and HAA5 must Primary Drinking Water Regulation. Addition of a Suppressor Acidified be complied with as a locational Postcolumn Reagent for Trace Bromate running annual average at each Subpart L—[Amended] Analysis,’’ USEPA, June 2002, EPA 815– monitoring location beginning the date 8. Section 141.131 is amended by R–03–007, EPA Method 327.0, Revision specified for Subpart V of this part 1.0, ‘‘Determination of Chlorine Dioxide compliance in § 141.620(c). revising paragraphs (a), (b), (d)(2), (d)(3), (d)(4)(i), (d)(4)(ii), and the table in and Chlorite Ion in Drinking Water Using Lissamine Green B and MCL paragraph (c)(1), and adding paragraph Disinfection byproduct (mg/L) (d)(6) to read as follows: Horseradish Peroxidase with Detection by Visible Spectrophotometry,’’ USEPA, Total trihalomethanes (TTHM) ...... 0.080 § 141.131 Analytical requirements. July 2003, and EPA Method 552.3, Haloacetic acids (five) (HAA5) ...... 0.060 (a) General. (1) Systems must use only Revision 1.0, ‘‘Determination of the analytical methods specified in this Haloacetic Acids and Dalapon in (ii) Best technology for systems that section, or their equivalent as approved Drinking Water by Liquid-liquid disinfect their source water. The by EPA, to demonstrate compliance Extraction, Derivatization, and Gas Administrator, pursuant to section 1412 with the requirements of this subpart Chromatography with Electron Capture of the Act, hereby identifies the and with the requirements of subparts U Detection,’’ USEPA, July 2003, can be

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00120 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49667

accessed and downloaded directly on- Health Association, 1995, 1998, and Association, 1015 Fifteenth Street, NW., line at www.epa.gov/safewater/ 2003, respectively. The cited methods Washington, DC 20005. ASTM Method methods/sourcalt.html. EPA Method published in any of these three editions D 1253–86 shall be followed in 415.3, Revision 1.0, ‘‘Determination of may be used. Standard Method 4500– accordance with the Annual Book of Total Organic Carbon and Specific UV ClO2 D shall be followed in accordance ASTM Standards, Volume 11.01, Absorbance at 254 nm in Source Water with Standard Methods for the American Society for Testing and and Drinking Water,’’ USEPA, June Examination of Water and Wastewater, Materials, 1996 or any year containing 2003, is available from: Chemical 19th or 20th Editions, American Public the cited version of the method may be Exposure Research Branch, Health Association, 1995 and 1998, used. ASTM D 6581–00 shall be Microbiological & Chemical Exposure respectively. Standard Methods 5310 B, followed in accordance with the Annual Assessment Research Division, National 5310 C, and 5310 D shall be followed in Book of ASTM Standards, Volume Exposure Research Laboratory, U.S. accordance with the Supplement to the 11.01, American Society for Testing and Environmental Protection Agency, 19th Edition of Standard Methods for Materials, 2001 or any year containing Cincinnati, OH 45268, Fax Number the Examination of Water and the cited version of the method may be 513–569–7757, Phone number: 513– Wastewater, or the Standard Methods used; copies may be obtained from the 569–7586. Standard Methods 4500–Cl for the Examination of Water and American Society for Testing and D, 4500–Cl E, 4500–Cl F, 4500–Cl G, Wastewater, 20th Edition, or the On- Materials, 100 Barr Harbor Drive, West 4500–Cl H, 4500–Cl I, 4500–ClO2 E, Line Version, American Public Health Conshohoken, PA 19428–2959. 6251 B, and 5910 B shall be followed in Association, 1995, 1998, and 2003, (b) Disinfection byproducts. (1) accordance with Standard Methods for respectively. The cited methods Systems must measure disinfection the Examination of Water and published in any of these editions may byproducts by the methods (as modified Wastewater, 19th or 20th Editions or the be used. Copies may be obtained from by the footnotes) listed in the following On-Line Version, American Public the American Public Health table:

APPROVED METHODS FOR DISINFECTION BYPRODUCT COMPLIANCE MONITORING

Standard ASTM Contaminant and methodology 1 EPA method Method 2 Method 3

TTHM: P&T/GC/ElCD & PID ...... 502.2 4 P&T/GC/MS ...... 524.2 LLE/GC/ECD ...... 551.1 HAA5: LLE (diazomethane)/GC/ECD ...... 6251 B 5. SPE (acidic methanol)/GC/ECD ...... 552.1 5 LLE (acidic methanol)/GC/ECD ...... 552.2, 552.3. Bromate: Ion chromatography ...... 300.1 ...... D 6581– 00 Ion chromatography & post column reaction ...... 317.0 Rev 2.0 6, 326.0 6 IC/ICP–MS ...... 321.8 6, 7 Chlorite: Amperometric titration ...... 4500–C1O2 E 8. Spectrophotometry ...... 327.0 8. . Ion chromatography ...... 300.0, 300.1, 317.0 Rev. 2.0, 326.0 ...... D 6581– 00 1 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 extraction; IC = ion chromatography; ICP-MS = in- ductively coupled plasma/mass spectrometer 2 219th or 20th editions or the On-Line Version of Standard Methods for the Examination of Water and Wastewater, 1995, 1998, and 2003, re- spectively, American Public Health Association; any of these editions may be used. 3 Annual Book of ASTM Standards, 2001 or any year containing the cited version of the method, Vol 11.01. 4 If TTHMs are the only analytes being measured in the sample, then a PID is not required. 5 The samples must be extracted within 14 days of sample collection. 6 Ion chromatography & post column reaction or IC/ICP-MS must be used for monitoring of bromate for purposes of demonstrating eligibility of reduced monitoring, as prescribed in § 141.132(b)(3)(ii). 7 Samples must be preserved at the time of sampling with 50 mg ethylenediamine (EDA)/L of sample and must be analyzed within 28 days. 8 Amperometric titration or spectrophotometry 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 chlo- rite in the distribution system, as prescribed in § 141.132(b)(2)(i)(B) and (b)(2)(ii).

(2) Analysis under this section for 141.135, and subparts U and V of this (ii) Achieve quantitative results on the disinfection byproducts must be part, the laboratory must: PE sample analyses that are within the conducted by laboratories that have (i) Analyze Performance Evaluation following acceptance limits which received certification by EPA or the (PE) samples that are acceptable to EPA become effective [date 60 days after date State, except as specified under or the State at least once during each of final rule publication] for purposes of paragraph (b)(3)of this section. To consecutive 12 month period by each certification: receive certification to conduct analyses method for which the laboratory desires for the DBP contaminants in §§ 141.64, certification.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00121 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49668 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

Acceptance DBP limits Comments (percent)

TTHM: Chloroform ...... ±20 Laboratory must meet all 4 individual THM acceptance limits in Bromodichloromethane ...... ±20 order to successfully pass a PE sample for TTHM. Dibromochloromethane ...... ±20 Bromoform ...... ±20 HAA5: Monochloroacetic Acid ...... ±40 Laboratory must meet the acceptance limits for 4 out of 5 of Dichloroacetic Acid ...... ±40 the HAAS compounds in order to successfully pass a PE Trichloroacetic Acid ...... ±40 sample for HAA5. Monobromacetic Acid ...... ±40 Dibromoacetic Acid ...... ±40 Chlorite ...... ±30 Bromate ...... ±30

(iii) Report quantitative data for ones listed in the following table for all with §§ 141.64, 141.135, 141.136, and concentrations at least as low as the DBP samples analyzed for compliance subparts U and V of this part:

Minimum re- DBP porting level Comments (ug/L) 7

TTHM 2: Chloroform ...... 1.0 Bromodichloromethane ...... 1.0 Dibromochloromethane ...... 1.0 Bromoform ...... 1.0 HAA5: 2 Monochloroacetic Acid ...... 2.0 Dichloroacetic Acid ...... 1.0 Trichloroacetic Acid ...... 1.0 Monobromoacetic Acid ...... 1.0 Dibromoacetic Acid ...... 1.0 Chlorite ...... 200. Bromate ...... 5.0 or 1.0 Laboratories that use EPA Methods 317.0 Revision 2.0, 326.0 or 321.8 must meet a 1.0 µg/L MRL for bromate. 1 The calibration curve must encompass the minimum reporting level (MRL) concentration and the laboratory must verify the accuracy of the calibration curve at the lowest concentration for which quantitative data are reported by analyzing a calibration check standard at that concentra- tion at the beginning of each batch of samples. The measured concentration for the check standard must be within ±50% of the expected value. Data may be reported for concentrations lower than the MRL as long as the precision and accuracy criteria are met by analyzing a standard at the lowest reporting limit chosen by the laboratory. 2 When adding the individual trihalomethane or haloacetic acid concentrations to calculate the TTHM or HAA5 concentrations, respectively, a zero is used for any analytical result that is less than the MRL concentration for that DBP.

(3) A party approved by EPA or the samples at the entrance to the (c) * * * State must measure daily chlorite distribution system. (1) * * *

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

Amperometric Titration ...... 4500–Cl D D 1253–86 X X X Low Level Amperometric Titration ...... 4500–Cl E X DPD Ferrous Titrimetric ...... 4500–Cl F X X X DPD Colorimetric ...... 4500–Cl G X X X Syringaldazine (FACTS) ...... 4500–Cl X Iodometric Electrode ...... 4500–Cl X DPD ...... 4500–ClO2 X Amperometric Method II ...... 4500–ClO2 X E Lissamine Green Spectrophotometric ... 327.0 X 1 X indicates method is approved for measuring specified disinfectant residual. Free chlorine or total chlorine may be measured for dem- onstrating compliance with the chlorine MRDL and combined chlorine or total chlorine may be measured for demonstrating compliance with the chloramine MRDL.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00122 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49669

* * * * * § 141.132 Monitoring requirements. most recent four quarters of monitoring) (d) * * * * * * * * on a continuing basis at each treatment (2) Bromide. EPA Methods 300.0, (b) * * * plant to reduce or remain on reduced 300.1, 317.0 Revision 2.0, 326.0, or (i) * * * monitoring for TTHM and HAA5. ASTM D 6581–00. (ii) Reduced monitoring. * * * * * (3) Total Organic Carbon (TOC). (A) Until [date three years from final 10. Section 141.134 is amended by Standard Method 5310 B (High- rule publication], systems required to revising paragraph (b) introductory text Temperature Combustion Method) or analyze for bromate may reduce to read as follows: Standard Method 5310 C (Persulfate- monitoring from monthly to quarterly, if Ultraviolet or Heated-Persulfate the system’s average source water § 141.134 Reporting and recordkeeping Oxidation Method) or Standard Method bromide concentration is less than 0.05 requirements. 5310 D (Wet-Oxidation Method) or EPA mg/L based on representative monthly * * * * * Method 415.3. Inorganic carbon must be bromide measurements for one year. (b) Disinfection byproducts. In removed from the samples prior to The system may remain on reduced addition to reporting required under analysis. TOC samples may not be bromate monitoring until the running § 141.136(e), systems must report the filtered prior to analysis. TOC samples annual average source water bromide information specified in the following must be acidified at the time of sample concentration, computed quarterly, is table: collection to achieve pH less than or equal to or greater than 0.05 mg/L based * * * * * equal to 2 with minimal addition of the on representative monthly 11. Section 141.135 is amended by acid specified in the method or by the measurements. If the running annual revising paragraph (a)(3)(ii) to read as instrument manufacturer. Acidified average source water bromide follows: TOC samples must be analyzed within concentration is ≥0.05 mg/L, the system § 141.135 Treatment technique for control 28 days. must resume routine monitoring (4) * * * of disinfection byproduct (DBP) precursors. (i) Dissolved Organic Carbon (DOC). required by paragraph (b)(3)(i) of this section. (a) * * * Standard Method 5310 B (High- (3) * * * (B) Beginning [date three years from Temperature Combustion Method) or (ii) Softening that results in removing final rule publication], systems may no Standard Method 5310 C (Persulfate- at least 10 mg/L of magnesium hardness longer use the provisions of paragraph Ultraviolet or Heated-Persulfate (as CaCO ), measured monthly (b)(3)(ii)(A) of this section to qualify for 3 Oxidation Method) or Standard Method according to § 141.131(d)(6) and reduced monitoring. A system required 5310 D (Wet-Oxidation Method) or EPA calculated quarterly as a running annual to analyze for bromate may reduce Method 415.3. DOC samples must be average. µ monitoring from monthly to quarterly, if filtered through the 0.45 m pore- * * * * * diameter filter as soon as practical after the system’s running annual average bromate concentration is less than 12. Section 141.136 is added to sampling, not to exceed 48 hours. After subpart L to read as follows: filtration, DOC samples must be 0.0025 mg/L based on monthly bromate acidified to achieve pH less than or measurements under paragraph (b)(3)(i) § 141.136 Additional compliance equal to 2 with minimal addition of the of this section for the most recent four requirements for Stage 2A. acid specified in the method or by the quarters, with samples analyzed using (a) Applicability. Any system that instrument manufacturer. Acidified Method 317.0 Revision 2.0, 325.0 or takes TTHM and HAA5 compliance DOC samples must be analyzed within 321.8. If a system has qualified for samples under this subpart at more than 28 days. Inorganic carbon must be reduced bromate monitoring under one location in its distribution system is removed from the samples prior to paragraph (b)(3)(ii)(A) of this section, subject to additional MCL requirements analysis. Water passed through the filter that system may remain on reduced beginning [date 3 years after publication prior to filtration of the sample must monitoring as long as the running of final rule] until the dates identified serve as the filtered blank. This filtered annual average of quarterly bromate for compliance with subpart V in blank must be analyzed using samples does not exceed 0.0025 mg/L § 141.620(c). Any system that takes procedures identical to those used for based on samples analyzed using samples at more than one location must analysis of the samples and must meet Method 317.0 Revision 2.0, 325.0, or calculate a locational running annual the following criteria: DOC < 0.5 mg/L. 321.8. If the running annual average average (LRAA) for each sampling point (ii) Ultraviolet Absorption at 254 nm bromate concentration is >0.0025 mg/L, and comply with the MCLs of 0.120 mg/ (UV254). Standard Method 5910 B the system must resume routine L for TTHM and 0.100 mg/L for HAA5 (Ultraviolet Absorption Method) or EPA monitoring required by paragraph listed in § 141.64(b)(2), except as Method 415.3. UV absorption must be (b)(3)(i) of this section. provided for under paragraph (c) of this measured at 253.7 nm (may be rounded * * * * * section. off to 254 nm). Prior to analysis, UV254 (e) Monitoring requirements for source (b) Compliance. (1) Systems must samples must be filtered through a 0.45 water TOC. In order to qualify for calculate a locational running annual µm pore-diameter filter. The pH of reduced monitoring for TTHM and average each quarter for each UV254 samples may not be adjusted. HAA5 under paragraph (b)(1)(ii) of this monitoring location at which they took Samples must be analyzed as soon as section, subpart H systems not TTHM and HAA5 samples under their practical after sampling, not to exceed monitoring under the provisions of monitoring plan developed under 48 hours. paragraph (d) of this section must take § 141.132(f) by averaging the results of * * * * * monthly TOC samples approximately TTHM or HAA5 monitoring at that (6) Magnesium. All methods allowed every 30 days at a location prior to any sample location during the four most in § 141.23(k)(1) for measuring treatment. In addition to meeting other recent quarters. magnesium. criteria for reduced monitoring in (2) Systems required to conduct 9. Section 141.132 is amended by paragraph (b)(1)(ii) of this section, the quarterly monitoring under this subpart revising paragraphs (b)(3)(ii) and (e) to source water TOC running annual must begin to make compliance read as follows: average must be ≤4.0 mg/L (based on the calculations under paragraph (b) of this

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00123 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49670 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

section at the end of the fourth calendar § 141.151 Purpose and applicability of this in the same units as the MCL. For the quarter that follows the compliance date subpart. MCLs for TTHM and HAA5 in in paragraph (a) of this section and at * * * * * § 141.64(b)(2) and (3), systems must the end of each subsequent quarter. (d) For the purpose of this subpart, include the highest locational running Systems required to conduct monitoring detected means: At or above the levels annual average for TTHM and HAA5 at a frequency that is less than quarterly prescribed by § 141.23(a)(4) for and the range of individual sample under this subpart must make inorganic contaminants, at or above the results for all sampling points expressed compliance calculations under levels prescribed by § 141.24(f)(7) for in the same units as the MCL. If more paragraph (b) of this section beginning the contaminants listed in § 141.61(a), at than one site exceeds the MCL, the with the first compliance sample taken or above the levels prescribed by system must include the locational after the compliance date in paragraph § 141.24(h)(18) for the contaminants running annual averages for all sites that (a) of this section. listed in § 141.61(c), at or above the exceed the MCL. (3) Failure to monitor will be treated levels prescribed by § 141.131(b)(2)(iii) as a monitoring violation for each for the contaminants or contaminant (C) When compliance with the MCL is quarter that a monitoring result would groups listed in § 141.64 and determined on a system-wide basis by be used in a locational running annual § 141.153(d)(iv), and at or above the calculating a running annual average of average compliance calculation. levels prescribed by § 141.25(c) for all samples at all sampling points: the (c) Consecutive systems. A radioactive contaminants. average and range of detection consecutive system must comply with * * * * * expressed in the same units as the MCL. the TTHM and HAA5 MCLs in 14. Section 141.153 is amended by The system is not required to include § 141.64(b)(2) at each monitoring revising paragraphs (d)(4)(iv)(B) and the range of individual sample results location in its distribution system (d)(4)(iv)(C) to read as follows: for the IDSE conducted under subpart U identified in its monitoring plan of this part. developed under § 141.132(f). § 141.153 Content of the reports. * * * * * (d) Reporting. Systems must submit * * * * * the compliance calculations and (d) * * * Subpart Q—[Amended] locational running annual averages (4) * * * under this section as part of the reports (iv) * * * required under § 141.134. 15. In Appendix A, the table is (B) When compliance with the MCL is amended by revising entries 1.G.1 and Subpart O—[Amended] determined by calculating a running 1.G.2, and endnotes 12 and 20, to read annual average of all samples taken at as follows: 13. Section 141.151 is amended by a sampling point: the highest average of revising paragraph (d) to read as any of the sampling points and the follows: range of all sampling points expressed

APPENDIX A TO SUBPART Q OF PART 141.—NPDWR VIOLATIONS AND OTHER SITUATIONS REQUIRING PUBLIC NOTICE 1

MCL/MRDL/TT violations2 Monitoring and testing procedure violations Contaminant Tier of pub- Tier of pub- lic notice Citation lic notice Citation required required

I. Violations of National Primary Drinking Water Regula- tions (NPDWR):3

******* G. Disinfection Byproducts, * * * 1. Total trihalomethanes (TTHM) ...... 2 141.1212, 3 141.3012, 141.64(b)20 141.132(a)–(b)20, 141.620–.630 2. Haloacetic acids (HAA5) ...... 2 141.64(b)20 3 141.132(a)–(b)20, 141.620–.630

* * * * * effect under the schedule in § 141.620(c). 16. In Appendix B the table is § 141.64(b)(2) takes effect on [date three years Appendix A—Endnotes amended by revising entries H.79, H.80, following final rule publication] and remains and endnote 17, and adding endnote 23, 12. §§ 141.12 and 141.30 will no longer in effect until the effective dates for subpart to read as follows: apply after December 31, 2003. V of this part compliance in the table in * * * * * § 141.620(c). 20. §§ 141.64(b)(1) and 141.132(a)-(b) apply * * * * * until §§ 141.64(b)(3) and 141.620–.630 take

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00124 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49671

APPENDIX B TO SUBPART Q OF PART 141—STANDARD HEALTH EFFECTS LANGUAGE FOR PUBLIC NOTIFICATION

Standard health 1 MCLG mg/ 2 effects language Contaminant L MCL mg/L for public notification

******* H. Disinfection Byproducts (DBPs), * * * 17: 79. Total trihalomethanes (TTHLM) ...... N/A 0.10/0.120/0.080 18, 19, 23 *** 80. Haloacetic acids (HAA5)...... N/A 0.060/0.10020, 23 ***

* * * * * comply with subpart V TTHM and HAA5 system-specific monitoring as provided MCLs (with compliance calculated as a Appendix B—Endnotes in § 141.602. As an alternative, you may locational running annual average) beginning use other system-specific data that * * * * * [date 90 months following publication of provide equivalent or better information 17. Surface water systems and ground final rule]. water systems under the direct influence of on site selection for monitoring under * * * * * subpart V as provided for in surface water are regulated under subpart H 17. Part 141 is amended by adding § 141.603(a). of 40 CFR 141. Subpart H community and new subpart U to read as follows: non-transient non-community systems (b) Applicability. You are subject to ≥ serving 10,000 must comply with subpart L Subpart U—Initial Distribution System these requirements if your system is a Evaluations DBP MCLs and disinfectant maximum community water system that adds a residual disinfectant levels (MRDLs) Sec. beginning January 1, 2002. All other primary or residual disinfectant other 141.600 General requirements. than ultraviolet light or delivers water community and non-transient non- 141.601 Initial Distribution System community systems must comply with Evaluation (IDSE) requirements. that has been treated with a primary or subpart L DBP MCLs and disinfectant MRDLs 141.602 IDSE monitoring. residual disinfectant other than beginning January 1, 2004. Subpart H 141.603 Alternatives other than IDSE ultraviolet light or if your system is a transient non-community systems serving nontransient noncommunity water ≥ monitoring. 10,000 that use chlorine dioxide as a 141.604 IDSE reports. system that serves at least 10,000 people disinfectant or oxidant must comply with the 141.605 Subpart V monitoring location and adds a primary or residual chlorine dioxide MRDL beginning January 1, recommendations to the State. 2002. All other transient non-community disinfectant other than ultraviolet light or delivers water that has been treated systems that use chlorine dioxide as a Subpart U—Initial Distribution System with a primary or residual disinfectant disinfectant or oxidant must comply with the Evaluations chlorine dioxide MRDL beginning January 1, other than ultraviolet light. You must 2004. § 141.600 General requirements. conduct an Initial Distribution System * * * * * (a) The requirements of subpart U Evaluation (IDSE), unless you meet the 23. Community and non-transient non- constitute national primary drinking 40/30 certification criteria in community systems must comply with water regulations. The regulations in § 141.603(b) or the State has granted a TTHM and HAA5 MCLs of 0.120 mg/L and this subpart establish monitoring and very small system waiver for the IDSE 0.100 mg/L, respectively (with compliance or you meet the criteria defined by the calculated as a locational running annual other requirements for identifying compliance monitoring locations to be State for a very small system waiver average) beginning [date three years under § 141.603(c). If you have a very following publication of final rule] until they used for determining compliance with are required to comply with subpart V TTHM maximum contaminant levels for total small system waiver for the IDSE under and HAA5 MCLs of 0.080 mg/L and 0.060 trihalomethanes (TTHM) and haloacetic § 141.603(c), you are not required to mg/L, respectively (with compliance acids (five)(HAA5) in subpart V through submit an IDSE report. All other calculated as a locational running annual the use of an Initial Distribution System systems must submit an IDSE report, average). Community and non-transient non- even if you meet the 40/30 certification ≥ Evaluation (IDSE). IDSEs are studies, community systems serving 10,000 must used in conjunction with subpart L criteria in § 141.603(c). comply with subpart V TTHM and HAA5 MCLs (with compliance calculated as a compliance monitoring, to identify and (c) Schedule. You must comply with locational running annual average) beginning select subpart V compliance monitoring the Initial Distribution System [date six years following publication of final sites that represent high TTHM and Evaluation (IDSE) on the schedule in the rule]. Community and non-transient non- HAA5 levels throughout the distribution following table, based on your system community systems serving <10,000 must system. The studies will be based on type.

If you are this type of system You must submit your IDSE report to the state by 1

(1) Subpart H serving ≥10,000 ...... [date 24 mos. following publication of final rule] (2) Subpart H serving <10,000 ...... [date 24 mos. following publication of final rule] 2 (3) Ground water serving ≥10,000 ...... [date 24 mos. following publication of final rule] (4) Ground water serving <10,000 ...... [date 24 mos. following publication of final rule] 2 (5) Consecutive system ...... at the same time as the system with the earliest compliance date in the com- bined distribution system 3 1 Systems that meet the 40/30 certification criteria in § 141.603(b) are encouraged to submit their IDSE report as soon as the certification cri- teria are met. 2 You must comply by [date 24 mos. following publication of final rule] if you are a wholesale system and any system in the combined distribu- tion system serves at least 10,000 people. You must comply by [date 48 mos. following publication of final rule] if no system in the combined dis- tribution system serves at least 10,000 people.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00125 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49672 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

3 You must comply by [date 24 mos. following publication of final rule] if any system in the combined distribution system serves at least 10,000 people. You must comply by [date 48 mos. following publication of final rule] if no system in the combined distribution system serves at least 10,000 people.

(d) Violations. You must comply with any additional State-specified compliance samples required of a specific monitoring and reporting requirements, which may include system with your population and source requirements. You must prepare for, conducting another IDSE. water under subpart L, but are required to conduct an IDSE under this subpart, conduct, analyze, and submit your IDSE § 141.601 Initial Distribution System report no later than the date specified in Evaluation (IDSE) requirements. you are not eligible for either the 40/30 § 141.600(c). Failure to conduct a certification in § 141.603(b) or the very (a) You must conduct an IDSE that small system waiver in § 141.603(c) and required IDSE or to submit a required meets the requirements in § 141.602 or must conduct an IDSE that meets the IDSE report by the date specified in § 141.603(a) or meet the 40/30 paragraph (c) of this section is a certification criteria in § 141.603(b) or requirements in § 141.602 or monitoring violation. If you do not have received a very small system § 141.603(a). submit your IDSE report to your State, waiver for the IDSE from the State under (b) You may use any alternative listed or if you submit the report after the § 141.603(c). If you do not take the full in the table below for which you specified date, you must comply with complement of TTHM and HAA5 qualify.

IDSE ALTERNATIVES

Alternatives Eligibility Regulatory reference

(1) Monitoring ...... All systems required to conduct an IDSE ...... § 141.602 (2) System-specific study .... All systems required to conduct an IDSE ...... § 141.603(a) (3) 40/30 certification ...... Any system with all TTHM compliance samples ≤0.040 mg/L and all HAA5 compli- § 141.603(b) ance samples ≤0.030 mg/L during the period specified in § 141.603(b). (4) Very small system waiv- Any system serving <500 for which the State has granted a waiver ...... § 141.603(c) er.

(c) IDSE results will not be used for identified in this paragraph must treat disinfection) will have a minimal the purpose of determining compliance each consecutive system entry point differential effect on TTHM and HAA5 with MCLs in § 141.64. from a wholesale system as a treatment formation associated with individual (d) Additional provisions: plant for the consecutive system for the entry points. (1) You may consider multiple wells purpose of determining monitoring § 141.602 IDSE monitoring. drawing water from a single aquifer as requirements of this subpart if water is one treatment plant for determining the delivered from the wholesale system to (a) You must conduct IDSE minimum number of TTHM and HAA5 the consecutive system for at least 60 monitoring for each treatment plant as samples required, with State approval in consecutive days through any of the indicated in the table in this paragraph. accordance with criteria developed consecutive system entry points. A You must collect dual sample sets at under § 142.16(h)(5) of this chapter. consecutive system that buys all its each monitoring location. One sample State approvals made under finished water during the period in the set must be analyzed for TTHM. § 141.132(a)(2) to treat multiple wells identified in this paragraph must The other sample in the set must be drawing water from a single aquifer as monitor based on population and source analyzed for HAA5. If approved by the one treatment plant remain in effect water for the purpose of determining State under the provisions of unless withdrawn by the State. monitoring requirements of this subpart. § 141.601(d)(1), you may consider (2) If you are a consecutive system, (i) You may request that the State multiple wells drawing water from the you must comply with the IDSE allow multiple consecutive system entry same aquifer to be one treatment plant requirements in this subpart based on points from a single wholesale system to for the purpose of determining whether you buy some or all of your a single consecutive system to be monitoring requirements. You must water from another PWS during 2004 for considered one treatment plant. conduct one monitoring period during systems with an IDSE report due [date (ii) In the request to the State for the peak historical month for TTHM 24 months after publication of final approval of multiple consecutive system levels or HAA5 levels or the month of rule] or during 2006 for systems with an entry points to be considered one warmest water temperature. You must IDSE report due [date 48 months after treatment plant, you must demonstrate review available compliance, study, or publication of final rule]. A consecutive that factors such as relative locations of operational data to determine the peak system that buys some, but not all, of its entry points, detention times, sources, historical month for TTHM or HAA5 finished water during the period and the presence of treatment (such as levels or warmest water temperature. corrosion control or booster

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00126 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49673

If you are this type of system Then you must monitor At these locations for each treatment plant 1,2

(1) Subpart H serving ≥10,000 Approximately every 60 days for one year (six Eight dual sample sets per monitoring period at locations monitoring periods). other than subpart L TTHM/HAA5 monitoring locations based on conditions: If CHLORINE is used as residual disinfectant: one near dis- tribution system entry point, two at average residence time, five at points representative of highest expected TTHM (three sites) and HAA5 concentration (two sites). If CHLORAMINE is used as residual disinfectant for any part of the year: two near distribution system entry point, two at average residence time, four at points representative of highest expected TTHM (two sites) and HAA5 concentra- tion (two sites). (2) Subpart H serving 500- Approximately every 90 days for one year Two dual sample sets per monitoring period at locations other 9,999. (four monitoring periods). than the for one year subpart L TTHM/HAA5 monitoring lo- cation; one each representative of expected high periods) TTHM level and HAA5 level. (3) Subpart H serving <500 ...... Approximately every 180 days for one year Two dual sample sets per monitoring period at locations other (two monitoring periods). than the subpart L TTHM/HAA5 monitoring location; one each representative of expected high periods) TTHM level and HAA5 level. (4) Ground water serving Approximately every 90 days for one year Two dual sample sets per monitoring period at locations other ≥10,000. (four monitoring periods). than the subpart L TTHM/HAA5 monitoring location; one each representative of expected high periods) TTHM level and HAA5 level. (5) Ground water serving < Approximately every 180 days for one year Two dual sample sets per monitoring period at locations other 10,000. (two monitoring periods). than the subpart L TTHM/HAA5 monitoring location; one each representative of expected high periods) TTHM level and HAA5 level. (6) Consecutive system ...... At a frequency based on source water and —For a consecutive system that buys all its finished water, your population 3. number of samples and locations as specified in paragraph (b) of this section. —For a consecutive system that buys some, but not all, of its finished water, serves ≥10,000, and receives water from a subpart H system: at IDSE locations required of a subpart H system serving ≥10,000. —For a consecutive system that does not meet any other cri- teria in this paragraph: two dual sample sets per monitoring period at locations other than the subpart L TTHM/HAA5 compliance monitoring location; one each representative of expected high TTHM levels and HAA5 levels. 1 Including treatment plants for consecutive system entry points that operate for at least 60 consecutive days. 2 The State may require additional monitoring. 3 You must monitor at the frequency required of a subpart H system with your population if you deliver any water required to be treated under subpart H. You must monitor at the frequency required of a ground water system with your population if you deliver no water required to be treat- ed under subpart H.

(b) IDSE monitoring for consecutive systems that buy all their water.

IDSE MONITORING LOCATIONS FOR CONSECUTIVE SYSTEMS THAT BUY ALL THEIR WATER

Number of Distribution system dual sample set locations 1 dual sample Population category set locations Near entry Average Highest Highest per moni- residence TTHM HAA5 points 2 toring period time locations locations

Subpart H Consecutive Systems that buy all their water

<500 3 ...... 2 1 1 500 to 4,999 4 ...... 2 1 1 5,000 to 9,999 4 ...... 4 1 2 1 10,000 to 24,999 5 ...... 8 1 2 3 2 25,000 to 49,999 5 ...... 12 2 3 4 3 50,000 to 99,999 5 ...... 16 3 4 5 4 100,000 to 499,999 5 ...... 24 4 6 8 6 500,000 to 1,499,999 5 ...... 32 6 8 10 8 1,500,000 to 4,999,999 5 ...... 40 8 10 12 10 >=5,000,000 5 ...... 48 10 12 14 12

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00127 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49674 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

IDSE MONITORING LOCATIONS FOR CONSECUTIVE SYSTEMS THAT BUY ALL THEIR WATER—Continued

Number of Distribution system dual sample set locations 1 dual sample Average Highest Highest Population category set locations Near entry per moni- residence TTHM HAA5 points 2 toring period time locations locations

Ground Water Consecutive Systems that buy all their water

<500 3 ...... 2 1 1 500 to 9,999 4 ...... 2 1 1 10,000 to 99,999 4 ...... 6 1 1 2 2 100,000 to 499,999 4 ...... 8 1 1 3 3 ≥500,000 4 ...... 12 2 2 4 4 1 Sampling locations to be distributed through distribution system. You may not use subpart L compliance monitoring locations as IDSE sample sites. You must collect a dual sample set at each sample location. 2 If the actual number of entry points to the distribution system is fewer than the specified number of ‘‘near entry point’’ sampling sites, take ad- ditional samples equally at highest TTHM and HAA5 locations. If there is an odd extra location number, take the odd sample at highest TTHM lo- cation. If the actual number of entry points to the distribution system is more than the specified number of sampling locations, take samples first at subpart H entry points to the distribution system having the highest water flows and then at ground water entry points to the distribution sys- tem having the highest water flows. 3 You must conduct monitoring during two monitoring periods approximately 180 days apart. 4 You must conduct monitoring during four monitoring periods approximately 90 days apart. 5 You must conduct monitoring during six monitoring periods approximately 60 days apart.

(c) You must prepare an IDSE (2) If you are a ground water system (a) If you complied by meeting the monitoring plan prior to starting IDSE serving ≥10,000 people, you are not provisions of §§ 141.602 or 141.603(a), monitoring and implement that plan. In required to comply with § 141.602 or your IDSE report must include the the plan, you must identify specific paragraph (a) of this section if you elements required in paragraphs (a)(1) monitoring locations and dates that certify to your State that all TTHM through (a)(3) of this section. meet the criteria in paragraphs (a) and samples taken under § 141.30 in 2003 (1) Your report must include all (b) of this section, as applicable. are ≤0.040 mg/L and that all TTHM and HAA5 compliance samples taken under TTHM and HAA5 analytical results § 141.603 Alternatives other than IDSE subpart L during 2004 are ≤0.040 mg/L from subpart L compliance monitoring monitoring. and ≤0.030 mg/L, respectively. conducted during the period of the IDSE In lieu of IDSE monitoring under (3) If you are a consecutive system presented in a tabular or spreadsheet § 141.602, you may use one of the serving <10,000 required to submit an format acceptable to the State. Your alternatives identified in paragraphs (a) IDSE report by [date 24 months report must also include a schematic of through (c) of this section for which you following publication of final rule], you your distribution system, with results, qualify to comply with this subpart. are not required to comply with location, and date of all IDSE (a) System-specific study. You may § 141.602 or paragraph (a) of this section monitoring, system-specific study perform an IDSE study based on system- if you certify to your State that all monitoring, and subpart L compliance specific monitoring or system-specific TTHM and HAA5 compliance samples samples noted. data if such a study identifies equivalent taken under subpart L during 2004 are ≤ ≤ (2) If you conducted IDSE monitoring or superior monitoring sites 0.040 mg/L and 0.030 mg/L, under § 141.602, your report must representing high TTHM and HAA5 respectively. include all IDSE TTHM and HAA5 (4) You must submit an IDSE report levels as would be identified by IDSE analytical results presented in a tabular that complies with § 141.604 and monitoring under § 141.602. You must or spreadsheet format acceptable to the contains the required certification. submit an IDSE report that complies State. Your report must also include all with § 141.604. (c) Very small system waiver. If you serve fewer than 500 people, the State additional data you relied on to justify (b) 40/30 certification. In order to may waive IDSE monitoring if the State IDSE monitoring site selection, plus qualify for the 40/30 certification, you determines that the TTHM and HAA5 your original monitoring plan must not have had any TTHM or HAA5 monitoring site for each plant under developed under § 141.602(c) and an monitoring violations during the § 141.132 is sufficient to represent both explanation of any deviations from that periods specified in paragraphs (b)(1) the highest TTHM and the highest plan. through (b)(3) of this section. HAA5 concentration in your (3) If you used the system-specific (1) You are not required to comply distribution system. If your IDSE study alternative in § 141.603(a), your with § 141.602 or paragraph (a) of this monitoring is waived, you are not report must include the basis (studies, section if you certify to your State that required to submit an IDSE report. You reports, data, analytical results, all compliance samples under subpart L must monitor under subpart V during modeling) by which you determined in 2002 and 2003 (for subpart H systems the same month and at the same that the recommended subpart V serving ≥10,000 people) or in 2004 and location as used for compliance monitoring sites representing high 2005 (for systems serving <10,000 sampling in subpart L. people that are not required to submit TTHM and HAA5 levels are comparable an IDSE report by [date 24 months § 141.604 IDSE reports. or superior to those that would following publication of final rule]) You must submit your IDSE report to otherwise have been identified by IDSE were ≤0.040 mg/L for TTHM and ≤0.030 the State according to the schedule in mg/L for HAA5. § 141.600(c).

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00128 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49675

monitoring under § 141.602. Your report required to take four dual sample sets include a rationale for selecting other must also include an analysis that per treatment plant per quarter under locations. If the State approves, you demonstrates that your system-specific routine monitoring under § 141.621, you must monitor at these other locations to study characterized expected TTHM must base your recommendations on the determine compliance under subpart V. and HAA5 levels throughout your entire locations in the distribution system If any of the criteria in this paragraph distribution system. where you expect to find the highest would cause only one location per (b) If you meet the 40/30 certification TTHM and HAA5 LRAAs. In treatment plant to be recommended, you criteria in § 141.603(b), your IDSE report determining the highest LRAA, you must identify an additional location must include all TTHM and HAA5 must evaluate both subpart L with the next highest HAA5 LRAA or analytical results from compliance compliance data and IDSE data. For request that you be allowed to monitor monitoring used to qualify for the 40/30 each plant, you must recommend only at that location. certification and a schematic of your locations with: (c) Systems that qualify for the 40/30 distribution system (with results, (1) The two highest TTHM locational certification. If you use the 40/30 location, and date of all compliance running annual averages; certification in § 141.603(b), you may samples noted). You must also include (2) The highest HAA5 locational use either subpart L compliance results of those compliance samples running annual average; and monitoring locations or you may taken after the period used to qualify for (3) An existing subpart L compliance identify monitoring locations for the 40/30 certification for State review. monitoring location identified in the (c) Your IDSE report must include § 141.132(f) monitoring plan that is the Subpart V that are different from those your recommendations and justification location of either the highest TTHM or for subpart L. You must include a for where and during what month(s) HAA5 LRAA among the three rationale for changing existing subpart L TTHM and HAA5 monitoring for compliance monitoring locations locations, choosing locations with a Subpart V should be conducted. You representative of average residence time long residence time and a detectable must base your recommendations on the (by calculating an LRAA for each residual. If you choose monitoring criteria in § 141.605. Your IDSE report compliance monitoring location using locations other than those in subpart L must also include the population the compliance monitoring results as subpart V compliance monitoring served; system type (subpart H or collected during the period of the IDSE). locations, you must retain the subpart L ground water); whether your system is (4) You may recommend locations locations with the highest TTHM and a consecutive system; and, if you other than those in paragraphs (a)(1) HAA5 LRAAs. If any of the criteria in conducted plant-based monitoring, the through (3) of this section if you include this paragraph would cause only one number of treatment plants and a rationale for selecting other locations. location per treatment plant to be consecutive system entry points. If the State approves, you must monitor recommended, you must identify an (d) Recordkeeping. You must retain a at these locations to determine additional location with the next complete copy of your IDSE report compliance under subpart V. highest HAA5 LRAA or request that you submitted under § 141.604 for 10 years (5) If any of the criteria in this be allowed to monitor only at that after the date that you submitted your paragraph (a) of this section would location. If you are required to monitor IDSE report. If the State modifies the cause fewer than four locations per at more locations under subpart V of monitoring requirements that you treatment plant to be recommended, you this part than under subpart L of this recommended in your IDSE report or if must identify an additional location(s) part, you must identify additional the State approves alternative with the next highest HAA5 LRAA. locations with a long residence time and monitoring sites, you must keep a copy (b) All groundwater systems and a detectable residual. of the State’s notification on file for 10 subpart H systems serving fewer than (d) Consecutive systems that buy years after the date of the State’s 10,000 people. If you are a system some, but not all, of their finished water. notification. You must make the IDSE required to take two dual sample sets Your recommendations must comply report and any State notification per treatment plant per quarter or per with §§ 141.601(d) and 141.605 (a) available for review by the State or the year or one TTHM and one HAA5 through (c). public. sample per plant per year for routine monitoring under § 141.621, you must (e) Consecutive systems that buy all § 141.605 Subpart V monitoring location select the locations with the highest their finished water. recommendations to the State. TTHM locational running annual (1) You must select the number of (a) Subpart H systems serving at least average and highest HAA5 locational monitoring locations specified in the 10,000 people. If you are a system running annual average, unless you following tables.

SUBPART V.—SAMPLE FREQUENCY FOR TTHM/HAA5 (AS DUAL SAMPLE SETS) FOR CONSECUTIVE SYSTEMS THAT BUY ALL THEIR WATER

Population Number of samples

Subpart H Consecutive Systems That Buy All Their Water

<500 ...... 1 TTHM and 1 HAA5 sample per year at different locations and time if the highest TTHM and HAA5 occurred at different locations and/or time or 1 dual sample set per year if the highest TTHM and HAA5 occurred at the same location and time of year, taken during the peak historical month for DBP concentrations or (if unknown) month of warmest water temperature. 500 to 4,999 ...... 1 TTHM and 1 HAA5 sample per quarter at different locations if the highest TTHM and HAA5 occurred at different locations or 1 dual sample set per quarter if the highest TTHM and HAA5 occurred at the same location. 5,000 to 9,999 ...... 2 dual sample sets per quarter. 10,000 to 24,999 ...... 4 dual sample sets per quarter. 25,000 to 49,999 ...... 6 dual sample sets per quarter. 50,000 to 99,999 ...... 8 dual sample sets per quarter.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00129 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49676 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

SUBPART V.—SAMPLE FREQUENCY FOR TTHM/HAA5 (AS DUAL SAMPLE SETS) FOR CONSECUTIVE SYSTEMS THAT BUY ALL THEIR WATER—Continued

Population Number of samples

100,000 to 499,999 ...... 12 dual sample sets per quarter. 500,000 to 1,499,999 ...... 16 dual sample sets per quarter. 1,500,000 to 4,999,999 ...... 20 dual sample sets per quarter. >=5,000,000 ...... 24 dual sample sets per quarter.

Ground Water Consecutive Systems That Buy All Their Water

<500 ...... 1 TTHM and 1 HAA5 sample per year at different locations and time if the highest TTHM and HAA5 occurred at different locations and/or time or 1 dual sample set per year if the highest TTHM and HAA5 occurred at the same location and time of year, taken during the peak historical month for DBP concentrations, or, if unknown, during month of warmest water temperature. 500 to 9,999 ...... 2 dual sample sets per year. Must be taken during the peak historical month for DBP concentrations. 10,000 to 99,999 ...... 4 dual sample sets per quarter. 100,000 to 499,999 ...... 6 dual sample sets per quarter. ≥500,000 ...... 8 dual sample sets per quarter.

(2) You must select Subpart V sufficient subpart L monitoring 141.628 Requirements for remaining on monitoring locations based on subpart L locations to identify the required increased TTHM and HAA5 monitoring compliance monitoring results collected number of Subpart V compliance based on subpart L results. during the period of the IDSE and IDSE monitoring locations, you must identify 141.629 [Reserved] 141.630 Reporting and recordkeeping monitoring results. You must follow the additional locations by selecting a site requirements. protocol in paragraphs (e)(2)(i) through representative of maximum residence (iv) of this section, unless you provide time and then a site representative of Subpart V—Stage 2B Disinfection a rationale for recommending different average residence time and repeating Byproducts Requirements locations. If required to monitor at more until the required number of than four locations, you must repeat the compliance monitoring locations have § 141.620 General requirements. protocol as necessary, alternating been identified. (a) The requirements of subpart V between sites with the highest HAA5 (f) You must schedule samples during constitute national primary drinking LRAA and the highest TTHM LRAA not the peak historical month for TTHM and water regulations. These regulations previously selected as a subpart V HAA5 concentration, unless the State establish requirements for control of monitoring location for choosing approves another month. Once you have certain disinfection byproducts that locations under paragraph (e)(2)(iii) of identified the peak historical month, supercede some requirements in subpart this section. and if you are required to conduct L and that are in addition to other (i) Location with the highest TTHM routine monitoring at least quarterly, requirements that are currently required LRAA not previously selected as a you must schedule subpart V under subpart L of this part. The subpart V monitoring location. compliance monitoring at a regular regulations in this subpart establish (ii) Location with the highest HAA5 frequency of approximately every 90 monitoring and other requirements for LRAA not previously selected as a days or fewer. achieving compliance with maximum subpart V monitoring location. 18. Part 141 is amended by adding contaminant levels for total (iii) Existing subpart L average new subpart V to read as follows: trihalomethanes (TTHM) and haloacetic residence time compliance monitoring Subpart V—Stage 2B Disinfection acids (five)(HAA5). location. Byproducts Requirements (b) Applicability. You are subject to (iv) Location with the highest TTHM Sec. these requirements if your system is a LRAA not previously selected as a 141.620 General requirements. community water system or subpart V monitoring location. 141.621 Routine monitoring. nontransient noncommunity water (3) You may recommend locations 141.622 Subpart V monitoring plan. system that adds a primary or residual other than those in paragraph (e)(2) of 141.623 Reduced monitoring. disinfectant other than ultraviolet light this section if you include a rationale for 141.624 Additional requirements for or delivers water that has been treated selecting other locations. If the State consecutive systems. with a primary or residual disinfectant 141.625 Conditions requiring increased other than ultraviolet light. approves, you must monitor at these monitoring. locations to determine compliance 141.626 Significant excursions. (c) Schedule. You must comply with under subpart V. 141.627 Requirements for remaining on the requirements in this subpart on the (4) If you used the 40/30 certification reduced TTHM and HAA5 monitoring schedule in the following table, based in § 141.603(b) and do not have based on subpart L results. on your system type.

If you are this type of system You must comply with subpart V by: 123

(1) Subpart H serving ≥10,000 ...... [date 72 mos following publication of final rule]. (2) Subpart H serving <10,000 ...... [date 90 mos following publication of final rule] if no Cryptosporidium monitoring is required under § 141.706(c) OR [date 102 mos following publication of final rule] if Cryptosporidium monitoring is required under § 141.706(c). (3) Ground water serving ≥10,000 ...... [date 72 mos following publication of final rule]. (4) Ground water serving <10,000 ...... [date 90 mos following publication of final rule].

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00130 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49677

If you are this type of system You must comply with subpart V by: 123

(5) Consecutive system ...... —at the same time as the system with the earliest compliance date in the combined distribu- tion system. 1 The State may grant up to an additional 24 months for compliance if you require capital improvements. 2 If you are required to conduct quarterly monitoring, you must begin monitoring in the first full calendar quarter that follows the compliance date in this table. If you are required to conduct monitoring at a frequency that is less than quarterly, you must begin monitoring in the calendar month recommended in the IDSE report prepared under § 141.604 no later than 12 months after the compliance date in this table. If you are not required to submit an IDSE report, you must begin monitoring during the calendar month identified in the monitoring plan developed under § 141.622 no later than 12 months after the compliance date. 3 If you are required to conduct quarterly monitoring, you must make compliance calculations at the end of the fourth calendar quarter that fol- lows the compliance date and at the end of each subsequent quarter (or earlier if the LRAA calculated based on fewer than four quarters of data would cause the MCL to be exceeded regardless of the monitoring results of subsequent quarters). If you are required to conduct monitoring at a frequency that is less than quarterly, you must make compliance calculations beginning with the first compliance sample taken after the compli- ance date.

(d) Monitoring and compliance. You (1) You may consider multiple wells consecutive system that buys some, but must monitor at sampling locations drawing water from a single aquifer as not all, of its finished water is identified in your monitoring plan one treatment plant for determining the considered a treatment plant for that developed under § 141.622. To minimum number of TTHM and HAA5 consecutive system. determine compliance with subpart V samples required, with State approval in (i) You may request that the State MCLs, you must calculate locational accordance with criteria developed allow multiple consecutive system entry running annual averages for TTHM and under § 142.16(h)(5) of this chapter. points from a single wholesale system to HAA5 using monitoring results Approvals made under §§ 141.132(a)(2) a single consecutive system to be collected under this subpart. If you fail and 141.601(d) remain in effect unless considered one treatment plant. to complete four consecutive quarters of withdrawn by the State. (ii) In the request to the State for monitoring, you must calculate (2) Consecutive systems. For the approval of multiple consecutive system compliance with the MCL based on an purposes of this subpart, you must entry points to be considered one average of the available data from the determine whether you buy all or some treatment plant, you must demonstrate most recent four quarters. of your water based on your that factors such as relative locations of (e) Violations. You must comply with categorization for the IDSE under entry points, detention times, sources, specific monitoring and reporting subpart U, unless otherwise directed by and the presence of treatment (such as requirements. Failure to monitor in the State. If you were not categorized corrosion control or booster accordance with the monitoring plan under subpart U, you must determine disinfection) will have a minimal required under § 141.622 is a whether you buy all or some of your differential effect on TTHM and HAA5 monitoring violation. Failure to monitor water based on your categorization formation associated with individual will also be treated as a monitoring during 2005, unless otherwise directed entry points. violation for the entire period covered by the State. by a locational running annual average (3) For the purposes of determining § 141.621 Routine monitoring. compliance calculation for the subpart monitoring requirements of this subpart, (a) You must monitor at the locations V MCLs in § 141.64(b)(3). each consecutive system entry point and frequencies listed in the following (f) Additional provisions. from a wholesale system to a table.

If you are this type of 1 system Then you must monitor At these locations for each treatment plant

(1) Subpart H serving four dual sample sets per quarter per treatment plant, —locations recommended to the State in the IDSE re- ≥10,000. taken approximately every 90 days. One quarterly set port submitted under subpart U. must be taken during the peak historical month for DBP concentrations 2. (2) Subpart H serving 500– two dual sample sets per quarter per treatment plant, —locations recommended to the State in the IDSE re- 9,999. taken approximately every 90 days. One quarterly set port submitted under subpart U.3 must be taken during the peak historical month for DBP concentrations 2. (3) Subpart H serving <500 one TTHM and one HAA5 sample per year per treat- —locations recommended to the State in the IDSE re- ment plant, taken during the peak historical month for port submitted under subpart U.4 DBP concentrations. (4) Ground water serving two dual sample sets per quarter per treatment plant, —locations recommended to the State in the IDSE re- ≥10,000. taken approximately every 90 days. One quarterly set port submitted under subpart U.3 must be taken during the peak historical month for DBP concentrations 2. (5) Ground water serving two dual sample sets per year per treatment plant, —locations recommended to the State in the IDSE re- 500–9,999. taken during the peak historical month for DBP con- port submitted under subpart U.3 centrations 2. (6) Ground water serving one TTHM and one HAA5 sample per year per treat- —locations recommended to the State in the IDSE re- <500. ment plant, taken during the peak historical month for port submitted under subpart U.4 DBP concentrations. (7) Consecutive system that based on your own population and source water, ex- —locations recommended to the State in the IDSE re- buys some, but not all, of cept that consecutive systems that receive water from port submitted under subpart U. its finished water. a subpart H system must monitor as a subpart H sys- tem.

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00131 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49678 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

If you are this type of 1 system Then you must monitor At these locations for each treatment plant

(8) Consecutive system that as specified in § 141.605(e) ...... —locations recommended to the State in the IDSE re- buys all its finished water. port submitted under subpart U. 1 Unless the State has approved or required other locations or additional locations based on the IDSE report or other information, or you have updated the monitoring plan under § 141.622. 2 A dual sample set is a set of two samples collected at the same time and same location, with one sample analyzed for TTHM and the other sample analyzed for HAA5. 3 If you have a single location that has both the highest TTHM LRAA and highest HAA5 LRAA, you may take a dual sample set only at that lo- cation after approval by the State. 4 You are required to sample for both TTHM and HAA5 at one location if that location is the highest for both TTHM and HAA5. If different loca- tions have high TTHM and HAA5 LRAAs, you may sample for TTHM only at the high TTHM location and for HAA5 only at the high HAA5 loca- tion. If you have received a very small system waiver for IDSE monitoring from the State under § 141.603(c), you must monitor for TTHM and HAA5 as a dual sample set at the subpart L monitoring location (a point representative of maximum residence time) during the month of warmest water temperature.

(b) You must begin monitoring at the under § 141.132(f) no later than the date must submit a copy of your monitoring locations you have recommended in identified in § 141.620(c) for subpart V plan to the State prior to the date you your IDSE report submitted under compliance. The monitoring plan must are required to comply with the § 141.604 following the schedule in contain the elements in paragraphs monitoring plan. § 141.620(c), unless the State requires (a)(1) through (a)(5) of this section: (d) You may modify your monitoring other locations or additional locations (1) Monitoring locations; plan to reflect changes in treatment, after its review. If you have received a (2) Monitoring dates; very small system waiver under (3) Compliance calculation distribution system operations and § 141.603(c), you must monitor at the procedures; layout (including new service areas), or location(s) identified in your monitoring (4) Monitoring plans for any other other factors that may affect TTHM or plan in § 141.132(f), updated as required systems in the combined distribution HAA5 formation. If you change by § 141.622. system if monitoring requirements have monitoring locations, you must replace (c) You must use an approved method been modified based on data from other locations with the lowest LRAA and listed in § 141.131 for TTHM and HAA5 systems; and notify the State how new sites were analyses in this subpart. Analyses must (5) Any permits, contracts, or selected as part of the next report due be conducted by laboratories that have agreements with third parties (including under § 141.630. The State may also received certification by EPA or the other PWSs, laboratories, and State require modifications in your State as specified in § 141.131. agencies) to sample, analyze, report, or monitoring plan. perform any other system requirement § 141.622 Subpart V monitoring plan. in this subpart. § 141.623 Reduced monitoring. (a) You must develop and implement (b) The monitoring plan will reflect a monitoring plan to be kept on file for the recommendations of the IDSE report (a) Systems other than consecutive State and public review. You may required under subpart U, along with systems that buy all their water. You comply by updating the monitoring plan any State-mandated modifications. The may reduce monitoring by meeting the developed under § 141.132(f) no later State must approve any monitoring sites criteria in the table in this paragraph at than the date identified in § 141.620(c) for which you are required to provide a all treatment plants in the system. You for subpart V compliance. If you have rationale in your IDSE report by may only use data collected under the received a very small system waiver § 141.605(a)(4). provisions of this subpart or subpart L under § 141.603(c), you must comply by (c) If you are a subpart H system of this part to qualify for reduced updating the monitoring plan developed serving more than 3,300 people, you monitoring.

If you are this type of Then you may reduce monitoring if you have To reduce monitoring per plant at these locations/frequency system monitoring results under § 141.621 and TTHM HAA5

(1) Subpart H serving —the LRAA is ≤0.040 mg/L for TTHM and —monitor once per quarter by taking a dual —monitor once per quarter by taking a dual ≥10,000. ≤0.030 for HAA5 at ALL monitoring loca- sample set at the location with the highest sample set at the location with the highest tions, AND TTHM LRAA or single measurement. HAA5 LRAA or single measurement. —the source water annual average TOC level, before any treatment, is ≤4.0 mg/L at each subpart H treatment plant 1. (2) Subpart H serving —the LRAA is ≤0.040 mg/L for TTHM and —monitor once per year by taking a dual —monitor once per year by taking a dual 500–9,999. ≤0.030 for HAA5 at ALL monitoring loca- sample set at the location with the highest sample set at the location with the highest tions, AND TTHM single measurement during the HHA5 single measurement during the quarter that the highest single TTHM quarter that the highest single HHA5 measurement occurred 2. measurement occurred.2 —the source water annual average TOC level, before any treatment, is ≤4.0 mg/L at each subpart H treatment plant 1. (3) Subpart H serving —monitoring may not be reduced to fewer not applicable ...... not applicable. <500. than one TTHM sample and one HAA5 sample per year. (4) Ground water serv- —the LRAA is ≤0.040 mg/L for TTHM and —monitor once per year by taking a dual —monitor once per year by taking a dual ing ≥10,000. ≤0.030 for HAA5 at ALL monitoring loca- sample set at the location with the highest sample set at the location with the highest tions. TTHM single measurement during the HHA5 single measurement during the quarter that the highest single TTHM quarter that the highest single HHA5 measurement occurred 2. measurement occurred.2

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00132 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49679

If you are this type of Then you may reduce monitoring if you have To reduce monitoring per plant at these locations/frequency system monitoring results under § 141.621 and TTHM HAA5

(5) Ground water serv- —the LRAA is ≤0.040 mg/L for TTHM and —monitor once every third year by taking a —monitor once every third year by taking a ing 500–9,999. ≤0.030 for HAA5 at ALL monitoring loca- dual sample set at the location with the dual sample set at the location with the tions. highest TTHM single measurement during highest HHA5 single measurement during the quarter that the highest single TTHM the quarter that the highest single HHA5 measurement occurred 2. measurement occurred.2 (6) Ground water serv- —the LRAA is ≤0.040 mg/L for TTHM and —monitor once every third year for TTHM at —monitor once every third year for HAA5 at ing <500. ≤0.030 for HAA5 at ALL monitoring loca- the location with the highest TTHM single the location with the highest HAA5 single tions. measurement during the quarter that the measurement during the quarter that the highest single TTHM measurement oc- highest single HAA5 measurement oc- curred 2. curred.2 (7) Consecutive sys- —the LRAA is ≤0.040 mg/L for TTHM and —monitor at the location(s) and frequency —monitor at the location(s) and frequency tem that buys some, ≤0.030 for HAA5 at ALL monitoring loca- associated with a non-consecutive system associated with a non-consecutive system but not all, of its fin- tions. with the same population and source water with the same population and source water ished water 3. type. type.2 1 TOC monitoring must comply with the provisions of either § 141.132(d) or § 141.132(e). 2 If your location for reduced monitoring for TTHM and HAA5 is the same location and if your quarter for the highest TTHM and HAA5 single measurement is the same, you may take one dual sample set at that location during that quarter. 3 Consecutive systems that buy some, but not all, of their finished water may reduce monitoring based on their own population and their wholesale system(s)’s source water type to the frequency and location(s) required in this section, unless the consecutive system treats surface water or ground water under the direct influ- ence of surface water. If the consecutive system treats surface water or ground water under the direct influence of surface water, it must base reduced monitoring on its population and classification as a subpart H system.

(b) Consecutive systems that buy all paragraph if the LRAA is ≤0.040 mg/L use data collected under the provisions their water. You may reduce monitoring for TTHM and ≤0.030 mg/L for HAA5 at of this subpart or subpart L of this part to the level specified in the table in this all monitoring locations. You may only to qualify for reduced monitoring.

REDUCED MONITORING FREQUENCY FOR CONSECUTIVE SYSTEMS THAT BUY ALL THEIR WATER.

Population Reduced monitoring frequency and location

Subpart H systems

<500 ...... Monitoring may not be reduced. 500 to 4,999 ...... 1 TTHM and 1 HAA5 sample per year at different locations or during different quarters if the highest TTHM and HAA5 measurements occurred at different locations or different quarters or 1 dual sample set per year if the highest TTHM and HAA5 measurements occurred at the same location and quarter. 5,000 to 9,999 ...... 2 dual sample sets per year; one at the location with the highest TTHM single measurement during the quarter that the highest single TTHM measurement occurred, one at the location with the highest HAA5 single meas- urement during the quarter that the highest single HAA5 measurement occurred. 10,000 to 24,999 ...... 2 dual sample sets per quarter at the locations with the highest TTHM and highest HAA5 LRAAs. 25,000 to 49,999 ...... 2 dual sample sets per quarter at the locations with the highest TTHM and highest HAA5 LRAAs. 50,000 to 99,000 ...... 4 dual sample sets per quarter—at the locations with the two highest TTHM and two highest HAA5 LRAAs. 100,000 to 499,999 ...... 4 dual sample sets per quarter—at the locations with the two highest TTHM and two highest HAA5 LRAAs. 500,000 to 1,499,999 ...... 6 dual sample sets per quarter—at the locations with the three highest TTHM and three highest HAA5 LRAAs. 1,500,000 to 4,999,999 ...... 6 dual sample sets per quarter—at the locations with the three highest TTHM and three highest HAA5 LRAAs. >=5,000,000 ...... 8 dual sample sets per quarter at the locations with the four highest TTHM and four highest HAA5 LRAAs.

Ground water systems

<500 ...... 1 TTHM and 1 HAA5 sample every third year at different locations and time if the highest TTHM and HAA5 measurements occurred at different locations and/or time or 1 dual sample set every third year if the highest TTHM and HAA5 measurements occurred at the same location and time of year. 500 to 9,999 ...... 1 TTHM and 1 HAA5 sample every year at different locations and time if the highest TTHM and HAA5 meas- urements occurred at different locations and/or time or 1 dual sample set every year if the highest TTHM and HAA5 measurements occurred at the same location and time of year. 10,000 to 99,000 ...... 2 dual sample sets per year; one at the location with the highest TTHM single measurement during the quarter that the highest single TTHM measurement occurred and one at the location with the highest HAA5 single measurement during the quarter that the highest single HAA5 measurement occurred. 100,000 to 499,999 ...... 2 dual sample sets per quarter; at the locations with the highest TTHM and highest HAA5 LRAAs. ≥500,000 ...... 4 dual sample sets per quarter; at the locations with the two highest TTHM and two highest HAA5 LRAAs.

(c) You may remain on reduced water annual average TOC level, before if the annual (or less frequent) sample monitoring as long as the TTHM LRAA any treatment, must be ≤4.0 mg/L at at any location exceeds either 0.060 mg/ ≤0.040 mg/L and the HAA5 LRAA each treatment plant treating surface L for TTHM or 0.045 mg/L for HAA5, ≤0.030 mg/L at each monitoring location water or ground water under the direct or if the source water annual average (for systems with quarterly monitoring) influence of surface water, based on TOC level, before any treatment, >4.0 or each TTHM sample ≤0.060 mg/L and monitoring conducted under either mg/L at any treatment plant treating each HAA5 sample ≤0.045 mg/L (for §§ 141.132(d) or 141.132(e). If the LRAA surface water or ground water under the systems with annual or less frequent at any location exceeds either 0.040 mg/ direct influence of surface water, the monitoring). In addition, the source L for TTHM or 0.030 mg/L for HAA5 or system must resume routine monitoring

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00133 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 49680 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules

under § 141.621 for all treatment plants contribute to TTHM and HAA5 water under the direct influence of or begin increased monitoring for all formation (such as flushing programs surface water to the State within 10 days treatment plants if § 141.625 applies. and storage tank operations and excess of the end of any quarter in which (d) The State may return your system capacity) and how these practices may monitoring is required: to routine monitoring at the State’s be modified to reduce TTHM and HAA5 (i) The number of source water TOC discretion. levels. samples taken each month during last quarter. § 141.624 Additional requirements for § 141.627 Requirements for remaining on (ii) The date and result of each sample consecutive systems. reduced TTHM and HAA5 monitoring based taken during last quarter. on subpart L results. If you are a consecutive system that (iii) The quarterly average of monthly does not add a disinfectant but delivers You may remain on reduced samples taken during last quarter. water that has been disinfected with monitoring after the dates identified in (iv) The running annual average other than ultraviolet light, you must § 141.620(c) for compliance with this (RAA) of quarterly averages from the comply with monitoring requirements subpart only if you qualify for a 40/30 past four quarters. for chlorine and chloramines in certification under § 141.603(b) or have (v) Whether the RAA exceeded 4.0 § 141.132(c)(1) and the compliance received a very small system waiver mg/L. requirements in § 141.133(c)(1) under § 141.603(c), plus you meet the (b) Recordkeeping. You must retain beginning [date three years after reduced monitoring criteria in any subpart V monitoring plans and publication of final rule] and report § 141.623(c), and you do not change or your subpart V monitoring results as monitoring results under § 141.134(c), add monitoring locations from those required by § 141.33. unless required earlier by the State. used for compliance monitoring under subpart L. If your monitoring locations PART 142— NATIONAL PRIMARY § 141.625 Conditions requiring increased under this subpart differ from your DRINKING WATER REGULATIONS monitoring. monitoring locations under subpart L, IMPLEMENTATION (a) If you are required to monitor at you may not remain on reduced a particular location yearly or less monitoring after the dates identified in 1. The authority citation for part 142 frequently than yearly under §§ 141.621 § 141.620(c) for compliance with this continues to read as follows: or 141.623, you must increase subpart. Authority: 42 U.S.C. 300f, 300g–1, 300g–2, monitoring to dual sample sets once per 300g–3, 300g–4, 300g–5, 300g–6, 300j–4, quarter (taken approximately every 90 § 141.628 Requirements for remaining on 300j–9, and 300j–11. days) at all locations if either the annual increased TTHM and HAA5 monitoring based on subpart L results. 2. Section 142.14 is amended by (or less frequent) TTHM sample >0.080 adding paragraph (a)(8) to read as If you were on increased monitoring mg/L or the annual (or less frequent) follows: HAA5 sample >0.060 mg/L at any under subpart L, you must remain on location. increased monitoring until you qualify § 142.14 Records kept by States. (b) You are not in violation of the for a return to routine monitoring under (a) * * * MCL until the LRAA calculated based § 141.625(c). You must conduct (8) Any decisions made pursuant to on four consecutive quarters of increased monitoring under § 141.625 at the provisions of 40 CFR part 141, monitoring (or the LRAA calculated the monitoring locations in the subparts U and V of this chapter. based on fewer than four quarters of monitoring plan developed under (i) Those systems for which the State data if the MCL would be exceeded § 141.622 beginning at the date has determined that the 40 CFR part regardless of the monitoring results of identified in § 141.620(c) for compliance 141, subpart L approved monitoring site subsequent quarters) exceeds the with this subpart and remain on is representative of the highest TTHM subpart V MCLs in § 141.64(b)(3). You increased monitoring until you qualify and HAA5 and therefore have been are in violation of the monitoring for a return to routine monitoring under granted a very small system waiver requirements for each quarter that a § 141.625(c). under § 141.603(c) of this chapter. The monitoring result would be used in § 141.629 [Reserved] State must provide a copy of the calculating an LRAA if you fail to decision to the system. A copy of the monitor. § 141.630 Reporting and recordkeeping decision must be kept until reversed or (c) You may return to routine requirements. revised. monitoring once you have conducted (a) Reporting. (1) You must report the (ii) System IDSE reports, plus any increased monitoring for at least four following information for each modifications required by the State. consecutive quarters and the LRAA for monitoring location to the State within Reports must be kept until reversed or every location is ≤0.060 mg/L for TTHM 10 days of the end of any quarter in revised in their entirety. and ≤0.045 mg/L for HAA5. which monitoring is required: * * * * * (i) Number of samples taken during 3. Section 142.16 is amended by § 141.626 Significant excursions. the last quarter. adding paragraph (m) to read as follows: If a significant excursion occurs, you (ii) Date and results of each sample must conduct a significant excursion taken during the last quarter. § 142.16 Special primacy conditions. evaluation and prepare a written report (iii) Arithmetic average of quarterly * * * * * of the evaluation no later than 90 days results for the last four quarters (m) Requirements for States to adopt after being notified of the analytical (LRAAs). 40 CFR part 141, subparts U and V. In result that shows the significant (iv) Whether the MCL was violated. addition to the general primacy excursion. You must discuss the (2) If you are a subpart H system requirements elsewhere in this part, evaluation with the State no later than seeking to qualify for or remain on including the requirements that State the next sanitary survey for your system. reduced TTHM/HAA5 monitoring, you regulations be at least as stringent as Your evaluation must include an must report the following source water federal requirements, an application for examination of distribution system TOC information for each treatment approval of a State program revision operational practices that may plant that treats surface water or ground that adopts 40 CFR part 141, subparts U

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00134 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2 Federal Register / Vol. 68, No. 159 / Monday, August 18, 2003 / Proposed Rules 49681

and V, must contain a description of (3) A procedure for determining that PART 143—NATIONAL SECONDARY how the State will accomplish the multiple consecutive system entry DRINKING WATER REGULATIONS following: points from a single wholesale system to (1) For PWSs serving fewer than 500 a single consecutive system should be 1. The authority citation for part 143 people, a very small system waiver treated as a single treatment plant for continues to read as follows: procedure for subpart U IDSE monitoring purposes. requirements that will apply to all Authority: 42 U.S.C. 300f et seq. (4) A procedure for addressing systems that serve fewer than 500 2. In § 143.4, the table in paragraph (b) people without the State making a consecutive systems outside the provisions of § 141.29 of this chapter or is amended by revising entries 2 and 9 system-by-system waiver determination, and footnotes 3 and 4, and by adding if the State elects to use such an part 141 subparts U and V of this footnote 6 to read as follows: authority. chapter, if the State elects to use such (2) A procedure for evaluating system- an authority. § 143.4 Monitoring. specific studies under § 141.603(a) of (5) A procedure for systems to * * * * * this chapter, if system-specific studies identify significant excursions. are conducted in the State. (b) * * *

Contaminant EPA ASTM 3 SM 4 18th and 19th ed. SM 4 20th ed. Other

******* 2. Chloride ...... 300.0 1 D4327–97 .... 4110 B ...... 4110 B. 300.1 6 ...... 4500–Cl ¥D ...... 4500–Cl ¥D ...... D512–89B .... 4500–Cl¥B ...... 4500–Cl¥B ......

******* 9. Sulfate ...... 300.0 1 D4327–97 .... 4110B ...... 4110B. 300.1 6 ...... 1 2¥ 2¥ 375.2 ...... 4500–SO4 F ...... 4500–SO4 F. 2¥ 2¥ 4500–SO4 C, D ...... 4500–SO 4 C, D. 2 2¥ D516–90 ...... 4500–SO4 ¥E ...... 4500–SO4 E.

******* ***** 1 ‘‘Methods for the Determination of Inorganic Substances in Environmental Samples’’, EPA/600/R–93–100, August 1993. Available at NTIS, PB94–120821. ***** 3 Annual Book of ASTM Standards, 1994, 1996, or 1999, Vols. 11.01 and 11.02, ASTM International; any year containing the cited version of the method may be used. Copies may be obtained from ASTM International, 100 Barr Harbor Drive, West Conshohocken, PA 19428. 4 Standard Methods for the Examination of Water and Wastewater, 18th edition (1992), 19th edition (1995), or 20th edition (1998). American Public Health Association, 1015 Fifteenth Street, NW, Washington, DC 20005. The cited methods published in any of these three editions may be used, except that the versions of 3111 B, 3111 D, and 3113 B in the 20th edition may not be used. ***** 6 ‘‘Methods for the Determination of Organic and Inorganic Compounds in Drinking Water’’, Vol. 1, EPA 815-R–00–014, August 2000. Available at NTIS, PB2000–106981.

[FR Doc. 03–18149 Filed 8–15–03; 8:45 am] BILLING CODE 6560–50–P

VerDate Jan<31>2003 21:44 Aug 15, 2003 Jkt 200001 PO 00000 Frm 00135 Fmt 4701 Sfmt 4702 E:\FR\FM\18AUP2.SGM 18AUP2