TZ.
                         EPA-815-00-004
p=l
Thursday;

June 22, 2000
         Part H



         Environmental

         Protection Agency

         40 CFR Parts 141 and 142
         National Primary Drinking Water
         Regulations; Arsenic and Clarifications to
         Compliance and New Source
         Contaminants Monitoring; Proposed Rule

-------
38888
Federal Register/Vol. 65, No. 121/Thursday, June  22,  2000/Proposed Rules
ENVIRONMENTAL PROTECTION
AGENCY

40 CFR Parts 141 and 142
[WH-FRL-6707-2]
RIN 2040-AB75

National Primary Drinking Water
Regulations; Arsenic and Clarifications
to Compliance and New Source
Contaminants Monitoring
AGENCY: Environmental Protection
Agency (EPA).
ACTION: Notice of proposed rulemaking.

SUMMARY: The Environmental Protection
Agency (EPA) is proposing a drinking
water regulation for arsenic, as required
by the 1996 amendments to the Safe
Drinking Water Act (SDWA). The
proposed health-based, non-enforceable
goal, or Maximum Contaminant Level
Goal (MCLG), for arsenic is zero, and the
proposed enforceable standard, or
maximum contaminant level (MCL), for
arsenic is 0.005 mg/L. EPA is also
requesting comment on 0.003 mg/L,
0.010 mg/L and 0.020 mg/L for the MCL.
EPA is listing technologies that will
meet the MCL, including affordable
compliance technologies for three
categories of small systems serving less
than 10,000 people. This proposal also
includes monitoring, reporting, public
notification, and consumer confidence
report requirements and State primacy
revisions for public drinking water
programs affected by the arsenic
regulation.
   In addition, in this proposal the
Agency is clarifying compliance for
 State-determined monitoring after
 exceedances for inorganic, volatile
 organic, and synthetic organic
                     contaminants. Finally, EPA is proposing
                     that States will specify the time period
                     and sampling frequency for new public
                     water systems and systems using a new
                     source of water to demonstrate
                     compliance with the MCLs. The
                     requirement for new systems and new
                     source monitoring will be effective for
                     inorganic, volatile organic, and
                     synthetic organic contaminants.
                     DATES: EPA must receive public
                     comments, in writing, on the proposed
                     regulations by September 20, 2000. EPA
                     will hold a public meeting on this
                     proposed regulation this summer. EPA
                     will publish a notice of the meeting,
                     providing date and location, in the
                     Federal Register, as well as post it on
                     EPA's Office of Ground Water and
                     Drinking Water web site at http://
                      www.epa.gov/safewater.
                     ADDRESSES: You may send written
                      comments to the W-99-16 Arsenic
                      Comments Clerk, Water Docket (MC-
                      4101); U.S. Environmental Protection
                      Agency; 1200 Pennsylvania Ave., NW,
                      Washington, DC 20460. Comments may
                      be hand-delivered to the Water Docket,
                      U.S. Environmental Protection Agency;
                      401 M Street, SW; EB-57; Washington,
                      DC 20460; (202)  260-3027 between 9
                      a.m. and 3:30 p.m. Eastern Time,
                      Monday through Friday. Comments may
                      be submitted electronically to ow-
                      docket@epamail.epa.gov. See
                      SUPPLEMENTARY INFORMATION for file
                      formats and other information about
                      electronic filing  and docket review. The
                      proposed rule and supporting
                      documents, including public comments,
                      are available for review in the Water
                      Docket at the above address.
                      FOR FURTHER INFORMATION CONTACT:
                      Regulatory information: Irene Dooley,

                          TABLE OF REGULATED  ENTITIES
(202) 260-9531, email:
dooley.irene@epa.gov. Benefits: Dr. John
B. Bennett, (202) 260-0446, email:
bennett.johnb@epa.gov General
information about the regulation: Safe
Drinking Water Hotline, phone: (800)
426-4791, or (703) 285-1093, email:
hotline.sdwa@epa.gov.

SUPPLEMENTARY INFORMATION:

Regulated Entities

  A public water system, as defined in
40 CFR 141.2, provides water to the
public for human consumption through
pipes or other constructed  conveyances,
if such system has "at least fifteen
service connections or regularly serves
an average of at least twenty-five
individuals daily at least 60 days out of
the year." A public water system is
either a community water system (CWS)
or a non-community water system
(NCWS). A community water system, as
defined in § 141.2, is "a public water
system which serves at least fifteen
service connections used by year-round
residents or regularly serves at least
twenty-five year-round residents." T,he
definition in § 141.2 for a non-transient,
non-community water system
[NTNCWS] is "a public water system
that is not a [CWS] and that regularly
serves at least 25 of the same persons
over 6 months per year." EPA has an
inventory totaling over 54,000
community water systems and
approximately 20,000 non-transient,
non-community water systems
nationwide. Entities potentially
regulated by this action are community
water systems and non-transient, non-
community water systems. The
following table provides examples of the
regulated entities under this rule.
                Category
                                                              Examples of potentially regulated entities
 Industry 	

 State, Tribal, and Local Government

 Federal Government	
                        Privately owned/operated community water supply systems using ground water or mixed
                         ground water and surface water.
                        State, Tribal, or local government-owned/operated water supply systems using ground water
                         or mixed ground water and surface water.
                        Federally owned/operated community water supply systems using ground water or mixed
                         ground water and surface water.                                    	
   The table is not intended to be
 exhaustive, but rather provides a guide
 for readers regarding entities likely to be
 regulated by this action. This table lists
 the types of entities that EPA is now
 aware could potentially be regulated by
 this action. Other types of entities not
 listed in this table could also be
 regulated. To determine whether your
 facility is regulated by this action, you
 should carefully examine the
                       applicability criteria in §§ 141.11 and
                       141.62 of the rule. If you have any
                       questions regarding the applicability of
                       this action to a particular entity, consult
                       Irene Dooley, the regulatory information
                       person listed in the FOR FURTHER
                       INFORMATION CONTACT section.

                       Additional Information for Commenters

                         Please submit an original and three
                       copies of your comments and enclosures
 (including references). To ensure that
 EPA can read, understand, and therefore
 properly respond to comments, the
 Agency would prefer that comments
 cite, where possible, the paragraph(s) or
 sections in the notice or supporting
 documents to which each comment
 refers. Commenters should use a
 separate paragraph for each issue
 discussed. Electronic comments must be
 submitted as a WordPerfect 5.1, WP6.1

-------
                  Federal  Register/Vol. 65, No.  121/Thursday, June 22,  2000/Proposed Rules
                                                                     38889
 or WPS file or as an ASCII file avoiding
 the use of special characters. Comments
 and data will also be accepted on disks
 in WP5.1, WP6.1 or WP8, or ASCII file
 format. Electronic comments on this
 Notice may be filed online at many
 Federal Depository Libraries.
 Commenters who want EPA to
 acknowledge receipt of their comments
 should include a self-addressed,
 stamped envelope. No facsimiles (faxes)
 will be accepted.

 Availability of Docket
   The docket for this rulemaking has
 been established under number W-99-
 16, and includes supporting
 documentation as well as printed, paper
 versions of electronic comments. The
 docket is available for inspection from
 9 a.m. to 4 p.m., Monday through
 Friday, excluding legal holidays, at the
 Water Docket; EB 57; U.S. EPA; 401 M
 Street, SW; Washington, D.C. For access
 to docket materials, please call (202)
 260-3027 to schedule an appointment.

 Abbreviations Used in This Proposed
 Rule
 >—greater than
 >—greater than or equal to
 <—less than
 5—less than or equal to
 § —Section
 ACWA—Association of California Water
   Agencies
 AA—activated alumina
 As (III)—trivalent arsenic. Common
   inorganic  form in water is arsenite
 As (V)—pentavalent arsenic. Common
   inorganic  form in water is arsenate
 ATSDR—Agency for Toxic Substances
   and Disease Registry, U.S. Department
   of Health & Human Services
 ASTM—American Society for Testing
   and Materials
 ASV—anodic stripping voltammetry
 AWQC—Ambient Water Quality
   Criterion
 AWWA—American Water Works
   Association
 BAT—best available technology
 BFD—Blackfoot disease
 BOD—biochemical oxygen demand
 BOSC—Board of Scientific Counselors,
   ORD
 CASRN—Chemical Abstracts Service
  registration number
 CCA—chromated copper arsenate
 CCR—consumer confidence report
 CDC—Centers for Disease Control and
  Prevention
 CFR—Code of Federal Regulations
 CPI—Consumer Price Index
CSFII—Continuing Survey of Food
  Intakes by  Individuals
CV—coefficient of variation=standard
  deviation divided by the mean x 100
CWS—community water system
 CWSS—Community Water System
   Survey                        :
 DBFs—disinfection byproducts
 DBPR—Disinfectants/Disinfection By-
   products Rule
 DMA—Di-methyl arsinic acid, cacodylic
   acid, (CH3)2HAsO2
 DSMA—Disodium methanearsonate
 DWSRF-—Drinking Water State
   Revolving Fund
 DNA—Deoxyribonucleic acid
 EB—East Tower Basement
 EDL—Estimated Detection Limit
 EDR—Electrodialysis Reversal
 e.g.—such as
 EJ—Environmental Justice         !
 EO—Executive Order
 EPA—U.S. Environmental Protection
   Agency                        |
 FDA—Food and Drug Administration
 FR—Federal Register
 FTE—full-time equivalents  (employees)
 GDP—Gross Domestic Product
 GFAA—Graphite Furnace Atomic  :
   Absorption
 GHAA—Gaseous Hydride Atomic  '
   Absorption
 GI—gastrointestinal
 gw—ground water
 HRRCA—Health Risk Reduction and
   Cost Analysis                  .
 IARC—International Agency for
   Research on Cancer
 ICP-MS—Inductively Coupled. Plasma
   Mass Spectroscopy
 i.e.—that is
 ICP-AES—Inductively Coupled Plasma-
   Atomic Emission Spectroscopy   :
 IESWTR—-Interim Enhanced Surface;
   Water Treatment Rule           :
 lOCs—inorganic contaminants      :
 IRFA—Initial Regulatory Flexibility '
   Analysis                       i
 IRIS—Integrated Risk Information
   System                        ',
 IX—Ion exchange
 K—thousands
 kg—kilogram, which is one thousand
   grams
 L—Liter, also referred to as lower case
   "1" in older citations
 LCso—The concentration of a chemical
   in air or water which is expected to
   cause death in 50% of test animals
   living in that air or water
 LCP—laboratory certification program
 LDso—The dose of a chemical taken by
   mouth or absorbed by the skin which
   is expected to cause death in 50% of
  the test animals so treated
 LOAEL—Lowest-observed-adverse-  ,
  effect level                      ;
 LS—lime softening                '
LT2ESWTR—Long-Term 2 Enhanced
  Surface Water Treatment Rule
M—millions
m3—Cubic meters
MCL—maximum contaminant level  :
MCLG—maximum contaminant level
  goal                            :
 MDL—method detection limit
 Metro—Metropolitan Water District of
   Southern California
 mg—Milligrams—one thousandth of
   gram, 1 milligram = 1,000 micrograms
 mg/kg—milligrams per kilogram
 mg/m3—Milligrams per cubic meter
 microgram (ug)—One-millionth of gram
   (3.5 x 10~8 oz., 0.000000035 oz.)
 ug/L—micrograms per liter
 M/DBP—Microbial/Disinfection By-
   product
 MMA—Mono-methyl arsenic, arsonic
   acid, CH3H2AsO3
 MOS—margin of safety
 MSMA—Monosodium methanearsonate
 NAOS—National Arsenic Occurrence
   Survey
 NAS—-National Academy of Sciences
 NAWQA—National Ambient Water
   Quality Assessment, USGS
 NCI—National Cancer Institute
 NCWS—non-community water system
 NOW AC—National Drinking Water
   Advisory Council
 NELAC—National Environmental
   Laboratory Accreditation Council
 NIRS—National Inorganic and
   Radionuclide Survey
 NIST—National Institute of Standards
   and Technology
 NOAEL—No-observed-adverse-effect
   level
 NODA—notice of data availability
 NOEL—No-observed-effect level
 NPDWR—National Primary Drinking
   Water Regulation, OGWDW
 NRC—National Research Council, the
   operating arm of NAS
 NTNCWS—non-transient non-
   community water system
 NTTAA—National Technology Transfer
   and Advancement Act of 1995
 NWIS—National Water Information
   System
 O&M—operational and maintenance
 OGWDW—Office of Ground Water and
   Drinking Water
 PBMS—Performance-Based
   Measurement System
 PE—performance evaluation, studies to
   certify laboratories for EPA drinking
   water testing
 P.L.—Public Law
 PNR—Public notification rule
 POD—point of departure
 POE—Point-of-entry treatment devices
 POU—Point-of-use treatment devices
 ppb—Parts per billion. Also, ug/L or
   micrograms per liter
 ppm—Parts per million. Also, mg/L or
   milligrams per liter
 PQL—Practical quantitation level
 PRA—Paperwork Reduction Act
 PT—performance testing
 PWS—Public water systems
PWSS—Public Water Systems
  Supervision
RCRA—Resource Conservation and
  Recovery Act

-------
38890
Federal Register/Vol. 65, No. 121/Thursday, June 22,  2000/Proposed Rules
REFs—relative exposure factors
RFA—Regulatory Flexibility Act
RfD—Reference dose
RIA—Regulatory Impact Analysis
RMCL—Recommended Maximum
  Contaminant Level
RO—reverse osmosis
RWS—Rural Water Survey
SAB—Science Advisory Board
SBA—Small Business Administration
SBREFA—Small Business Regulatory
  and Enforcement Flexibility Act, SBA
SDWA—Safe Drinking Water Act of
  1974, as amended
SOWS—Safe Drinking Water
  Information System
SER—Small Entity Representative for
  SBREFA
SISNOSE—Substantial impact on a
  significant number of small entities,
  SBREFA
SM—Standard Methods for the
  Examination of Water and Wastewater
SMRs—Standardized mortality ratios,
  comparing deaths in test areas to
  deaths in unexposed areas
SSCTs—Small System Compliance
  Technologies
STP-GFAA—Stabilized Temperature
  Platform Graphite Furnace Atomic
  Absorption
SW—Office of Solid Waste publication
   or test method
SW-846—Solid Waste publication #846,
  Test Methods for Solid and Hazardous
  Waste
TC—toxicity characteristic
TDS—total dissolved solids
TNC—transient, non-community
TOG—total organic carbon
ug—Microgram, 1000 micrograms = 1
   milligram
UMRA—Unfunded Mandates Reform
   Act
U.S.—United States
USDA—U.S. Department of Agriculture
USGS—U.S. Geological Survey
USPHS—U.S. Public Health Service
VSL—Value of Statistical Life
WESTCAS—Western Coalition of Arid
   States
WHO—World Health Organization
WITAF—Water Industry Technical
   Action Fund
WS—water supply
WTP—Willingness to  pay
Table of Contents
I. Summary of Regulation
II. Background
   A. What is the Statutory Authority for the
    Arsenic Drinking Water Regulation?
   B. What is arsenic?
   C. What are the sources of arsenic
    exposure?
   1. Natural Sources of Arsenic
   2. Industrial Sources of Arsenic
   3. Dietary Sources
   4. Environmental Sources
   D. What is the regulatory history for
    arsenic?
                         1. Earliest U.S. Arsenic Drinking Water
                           Standards
                         2. EPA's 1980 Guidelines
                         3. Research and Regulatory Work
                         E. EPA's Arsenic Research Plan
                       III. Toxic Forms and Health Effects of Arsenic
                         A. What are the toxic forms of arsenic?
                         B. What are the effects of acute toxicity?
                         C. What cancers are associated with
                           arsenic?
                         1. Skin Cancer
                         2. Internal Cancers
                         D. What non-cancer effects are associated
                           with arsenic?
                         E. What are the recent developments in
                           health effects research?
                         1. Funding of Health Effects Research
                         2. Expert Panel on Arsenic Carcinogenicity
                         3. NAS Review of EPA's Risk Assessment
                         4. May 1999 Utah Mortality Study
                         5.1999 Review of health effects
                         6. Study of Bladder and Kidney Cancer in
                           Finland
                         F. What did the National Academy of
                           Sciences/National Research Council
                           report?
                         1. The National Research Council and its
                           Charge
                         2. Exposure
                         3. Essentiality
                         4. Metabolism and Disposition
                         5. Human Health Effects and Variations in
                           Sensitivity
                         6. Modes of Action
                         7. Risk Considerations
                         8. Risk Characterization
                       IV. Setting the MCLG
                         A. How did EPA approach it?
                         B. What is the MCLG?
                         C. How will a health advisory protect
                           potentially sensitive subpopulations?
                         D. How will the Clean Water Act criterion
                           be affected by this regulation?
                       V. EPA's Estimates of Arsenic Occurrence
                         A. What data did EPA evaluate?
                         B. What databases did EPA use?
                         C. How did EPA estimate national
                           occurrence of arsenic  in drinking water?
                         D. What are the national occurrence
                           estimates of arsenic in drinking water for
                           community water systems?
                         E. How do EPA's estimates compare with
                           other recent national occurrence
                           estimates?
                         F. What are the national occurrence
                           estimates of arsenic in drinking water for
                           non-transient, non-community water
                           systems?
                         G. How do arsenic levels vary from source
                           to source and over time?
                         H. How did EPA evaluate co-occurrence?
                          1. Data
                         2. Results of the Co-occurrence Analysis
                           (US EPA, 1999f)
                        VI. Analytical Methods
                         A. What section of SDWA requires the
                           Agency to specify analytical methods?
                         B. What factors does the Agency consider
                           in approving analytical methods?
                         C. What analytical methods and method
                           updates are currently approved for the
                           analysis of arsenic in  drinking water?
                         D. Will any of the approved methods for
                           arsenic analysis be withdrawn?
                         E. Will EPA propose any new analytical
                           methods for arsenic analysis?
  F. Other Method-Related Items
  1. The Use of Ultrasonic Nebulization with
    ICP-MS
  2. Performance-Based Measurement
    System                        :
  G. What are the estimated costs of analysis?
  H. What is the practical quantitation limit?
  1. PQL determination
  2. PQL for arsenic
  I. What are the sample collection, handling
    and preservation requirements for
    arsenic?
  J. Laboratory Certification
  1. Background
  2. What Are the Performance Testing
    criteria for arsenic?
  3. How often is a laboratory required to
    demonstrate acceptable PT performance?
  4. Externalization of the PT Program
    (formerly known as the PE Program)
VII. Monitoring and Reporting Requirements
  A. What are the .existing monitoring and
    compliance requirements?
  B. How does the Agency plan to revise the
    monitoring requirements?
  C. Can States grant monitoring waivers?
  D. How can I determine if I have an MCL
    violation?
  E. When will systems have to complete
    initial monitoring?
  F. Can I use grandfathered data to satisfy
    the initial monitoring requirement?
  G. What are the monitoring requirements
    for new systems and sources?
  H. How does the Consumer Confidence
    Report change?                 \
  I. How will public notification change?
VIII. Treatment Technologies
  A. What are the Best Available      ;
    Technologies (BATs) for arsenic? What
    are the issues associated with these
    technologies?
  B. What are the likely treatment trains?
    How much will they cost?        ,
  C. How are variance and compliance
    technologies identified for small
    systems?
  D. When are exemptions available?  '-.
  E. What are the small systems compliance
    technologies?
  F. How does the Arsenic Regulation
    overlap with other regulations?   ;
IX. Costs
  A. Why does EPA analyze the regulatory
    burden?
  B. How did EPA prepare the baseline
    study?
  1. Use of baseline data
  2. Key data sources used in the baseline
    analysis for the RIA?            :
  C. How were very large system cost ;
    derived?                      ;
  D. How did EPA develop cost estimates?
  E. What are the national treatment costs of
    different MCL options?
  1. Assumptions affecting the development
    of the decision tree
  2. Assumptions affecting unit cost curves
X. Benefits of Arsenic Reduction
  A. Monetized Benefits of Avoiding Bladder
    Cancer
  1. Risk reductions: The Analytic Approach
  2. Water Consumption
  3. Monte Carlo Analysis
  4. Relative Exposure Factors
  5. NRG Risk Distributions

-------
                    Federal Register/Vol.  65, No. 121/Thursday, June 22, 2000/Proposed  Rules
                                                                             38891
   6. Estimated Risk Reductions
   B. "What if?" scenario for lung cancer risks
   C. Evaluation of Benefits
   1. Fatal Risks and Value of a Statistical Life
     (VSL)
   2. Nonfatal Risks and Willingness to Pay
     (WTP)
   D. Estimates of Quantifiable Benefits of
    Arsenic Reduction
   F. NDWAC Working Group  (NOWAC,
    1988) on Benefits
 XI. Risk Management Decisions: MCL and
    NTNCWSs
   A. What is the Proposed MCL?
   1. Feasible MCL
   2, Principal Considerations  in Analysis of
    MCL Options
   3. Findings of NRC and Consideration of
    Risk Levels
   4. Non-monetized Health Effects
   5. Sources of Uncertainty
   6. Comparison of Benefits and Costs
   7. Conclusion and Request for Comment
   B. Why is EPA proposing a total arsenic
    MCL?
   C. Why is EPA proposing to require only
    monitoring and notification for
    NTNCWSs?
   1. Methodology for analyzing NTNCWS
    risks
   2. Results
 XII. State Programs
   A. How does arsenic affect a State's
    primacy program?
   B. When does a State have to apply?
   C. How are Tribes affected?
 XIII. HRRCA
   A. What are the requirements for the
    HRRCA?
   B. What are the quantifiable and non-
    quantifiable health risk reduction
    benefits?
   C. What are the Quantifiable and Non-
    Quantifiable Costs?
   D. What are the Incremental Benefits and
    Costs?
   E. What are the  Risks of Arsenic Exposure
    to the General Population  and Sensitive
    Subpopulations?
   F. What are the Risks Associated with Co-
    Occurring Contaminants?
   G. What are the Uncertainties in the
    Analysis?
XIV. Administrative Requirements
   A. Executive Order 12866: Regulatory
    Planning and  Review
   B. Regulatory Flexibility Act (RFA), as
    amended by the Small Business
    Regulatory Enforcement Fairness Act of
    1996 (SBREFA), 5 U.S.C. 601 et seq.
   1. Overview
   2. Use of Alternative Small Entity
    Definition
   3. Initial Regulatory Flexibility Analysis
  a. Number of Small Entities Affected
  b. Reporting, Recordkeeping  and Other
    Requirements for Small Systems
  4. Small Business Advocacy Review
    (SBAR) Panel  Recommendations
C. Unfunded Mandates Reform Act (UMRA)
  1. Summary of UMRA Requirements
  a. Authorizing legislation
  b. Cost-benefit analysis
  c. Financial Assistance
  d. Estimates of future compliance costs and
    disproportionate budgetary effects
   e. Macroeconomic effects
   f. Summary of EPA's consultation with
     State, local, and tribal governments and
     their concerns
   g. Nature of State, local, and Tribal
     government concerns and how EPA '
     addressed these concerns
   h. Regulatory Alternatives Considered !
   2. Impacts on Small Governments
   D. Paperwork Reduction Act (PRA)
   E. National Technology Transfer and
     Advancement Act (NTTAA)
   F. Executive Order 12898: Environmental
     Justice                           i
   G. Executive Order 13045: Protection of
     Children from Environmental Health
     Risks and Safety Risks             ;
   H. Executive Order 13132: Federalism
   I. Executive Order 13084: Consultation and
     Coordination with Indian Tribal    \
     Governments
   J. Request for Comments on Use of Plain
    Language
 XV. References
 List of Tables
 Table V-l. Summary of Arsenic Data Sources
 Table V-2. Regional Exceedance Probability
    Distribution Estimates
 Table V-3. Statistical Estimates of Number of
    Ground Water CWSs with Average
    Arsenic Concentrations in Specified
    Ranges                           \
 Table V-4. Statistical Estimates of Number of
    Surface Water CWSs with Average
    Arsenic Concentrations in Specified
    Ranges
 Table V-5. Comparison of CWSs from EPA,
    NAOS,  and USGS Estimates Exceeding
    Arsenic Concentrations            :
 Table V-6. Statistical Estimates of Number of
    Ground Water NTNCWSs with Average
    Arsenic Concentrations in Specified ;
    Ranges
 Table V-7. Statistical Estimates of Number of
    Surface Water NTNCWSs with Average
    Arsenic Concentrations in Specified
    Ranges
 Table V-8. Correlation of Arsenic with  !
    Sulfate and Iron (surface and ground
    waters)
 Table V-9. Correlation of Arsenic with Radon
    (ground water)                    :
 Table VI-1. Approved Analytical Methods
    (and Method Updates) for Arsenic (CFR
    141.23)
 Table VI-2. Estimated Costs for the Analysis
    of Arsenic in Drinking Water
 Table VI-3. Acceptance Limits and PQLs for
    Other Metals (in order of decreasing  '
    PQL)
 Table VII-1. Comparison of Sampling,
    Monitoring, and Reporting Requirements
 Table VII-2.  Treatment in-place at small  :
    water systems (US EPA, 1999e and US
    EPA, 1999m)
 Table VII-3.  Table Identifying Regulatory I
    Changes
 Table VII-4.  Table Listing Deleted Sections
 Table VIH-1. Best Available Technologies ;
    and Removal Rates
 Table VIII-2. Treatment Technology Trains
 Table VIII-3. Annual Costs of Treatment
    Trains (Per household)
Table VIII-4. Affordable Compliance
    Technology Trains for Small Systems
Table VIH-5. Affordable Compliance     |
    Technology Trains for Small Systems
 Table IX-1. Summary of General Baseline
     Categories of Affected Entities
 Table IX-2. List of Large Water Systems that
     Serve More Than 1 Million People
 Table IX-3. Total Annual Costs for Large
     Systems for (serving more than 1 million
     people)
 Table IX-4. Systems Needing to Add Pre-
     Oxidation
 Table IX-5. Percent of Systems with
     Coagulation-Filtration and Lime-
     Softening in Place
 Table IX-6. Waste Disposal Options
 Table IX-7. Ground Water: Arsenic and
     Sulfate
 Table IX-8. Surface Water: Arsenic and
     Sulfate
 Table IX-9. Ground Water: Arsenic and Iron
 Table IX-10. Surface Water: Arsenic and Iron
 Table IX-11. National Annual Treatment
     Costs (Dollars in Millions)
 Table IX-12. Total Annual Costs per
     Household (Dollars)
 Table IX-13. Incremental National Annual
     Costs (Dollars in Millions)
 Table IX-14. Incremental Annual Costs per
     Household (Dollars)
 Table X-l. Source of Water Consumed
 Table X-2a. Bladder Cancer Incidence Risks1
     for High Percentile U.S. Populations
     Exposed At or Above MCL Options,
     After Treatment2 (Community Water
     Consumption Data 3)
 Table X-2b. Bladder Cancer Incidence Risksl
     for High Percentile U.S. Populations
     Exposed At or Above MCL Options,
     After Treatment2 (Total Water
     Consumption Data 3)
 Table X-3a. Percent of Exposed Population
     At 10 nd"sh;4 Rjgjj or Higher for
     Bladder Cancer Incidence1 After
     Treatment2 (Community Water
     Consumption Data 3)
 Table X-3b. Percent of Exposed Population
     At 10 ndash;4 Rjg^ or Higher for
     Bladder Cancer Incidence1 After
     Treatment2 (Total Water Consumption
     Data3)
 Table X-4a. Mean Bladder Cancer Incidence
     Risks' for U.S. Populations Exposed At
     or Above MCL Options, after Treatment2
     (Community Water Consumption Data 3)
 Table X-4b. Mean Bladder Cancer Incidence
     Risks1 for U.S. Populations Exposed At
     or Above MCL Options, after Treatment2
     (Total Water Consumption Data 3)
 Table X-5. Lifetime Avoided Medical Costs
     For Survivors (preliminary estimates,
     1996 dollars1)
 Table X-6. Mean Bladder Cancer Incidence
    Risks i for U.S. Populations Exposed At
    or Above MCL Options, after Treatment2
    (Composite of Tables X-5a and X-5b)
 Table X-7. Estimated Costs and Benefits from
    Reducing Arsenic in Drinking Water
    ($millions, 1999)
 Table XI-1. Estimated Costs and Benefits
    from Reducing Arsenic in Drinking
    Water (In 1999 $ millions)
Table XI-2. Exposure Factors Used in the
    NTNC Risk Assessment
Table XI-3. Composition of Non-Transient,
    Non-Community Water Systems
    (Percentage of Total NTNC Population
    Served by Sector)

-------
38892
Federal Register/Vol.  65, No.  121/Thursday,  June 22, 2000/Proposed Rules
Table XI-4. Upper Bound School Children
    Risk Associated with Current Arsenic
    Exposure in NTNC Water Systems
Table XI-5. Non-Transient Non-Community
    Benefit Cost Analysis
Table XI-0. Sensitive Group Evaluation
    Lifetime Risks
Table XHI-1. Risk Reduction from Reducing
    Arsenic in Drinking Water
Table XIII—2. Mean Bladder Cancer Risks and
    Exposed Population
Table XIFI-3. Estimated Costs and Benefits
    from Reducing Arsenic in Drinking
    Water (in 1999 S millions)
Table XIII-4. Estimated Annualized National
    Costs of Reducing Arsenic Exposures (in
    1999 S millions)
Table XIII-5. Estimated Annual Costs per
    Household1 (in 1999$)
Table XIII-6. Summary of the Total Annual
    National Costs of Compliance with the
    Proposed Arsenic Rule Across MCL
    Options (in 1997 S millions)
Table XIII-7. Estimates of the Annual
    Incremental Risk Reduction, Benefits,
    and Costs of Reducing Arsenic in
    Drinking Water ($millions, 1999)
Table XIV-1.  Profile of the Universe of Small
    Water Systems Regulated Under the
    Arsenic Rule
Table XIV-2.  Average Annual Cost per CWS
    by Ownership
Table XlV-3.  Average Compliance Costs per
    Household for CWSs Exceeding MCLs
Table XIV-4.  Average Compliance Costs per
    Household for CWSs Exceeding MCLs as
    a Percent of Median Household Income
Table XIV-5.  Hour Burden per Activity for
    Public Water Systems
Table XIV-6.  Hour Burden per Activity for
    States and Tribes
I. Summary of Regulation
  EPA is proposing an arsenic
regulation for community water
systems, which are systems that provide
piped water to at least fifteen service
connections used by year-round
residents or regularly serves at least
twenty-five year-round residents. This
proposal will require non-transient,
non-community water systems
(NTNCWS) to monitor for arsenic and
report exceedances of the MCL. The
proposed health-based, non-enforceable
goal, or Maximum Contaminant Level
Goal (MCLG), is zero, hased on EPA's
revised risk characterization.
  EPA evaluated the analytical
capability and laboratory capacity,
likelihood of water systems choosing
treatment technologies for several sizes
of systems based on source water
properties, and the national occurrence
of arsenic in water supplies to
determine the proposed Maximum
Contaminant Level (MCL). Furthermore,
the Agency analyzed the quantifiable
and nonquantifiable costs and health
risk reduction benefits likely to occur at
the treatment levels considered, and  the
effects on sensitive subpopulations.
Based on the determination that the
                      costs for the feasible MCL do not justify
                      the benefits, EPA is proposing an MCL
                      of 0.005 mg/L and requesting comment
                      on 0.003 mg/L, 0.010 mg/L, and 0.020
                      mg/L. The treatment technologies for
                      large systems are primarily coagulation/
                      filtration and lime softening, while EPA
                      expects that small systems (serving less
                      than 10,000 people) will be able to use
                      ion exchange, activated alumina, reverse
                      osmosis, nanofiltration, and
                      electrodialysis reversal. The effective
                      date will be five years after the final rule
                      comes out for community water systems
                      serving 10,000 people or less, and three
                      years after promulgation for all other
                      community water systems. EPA is
                      proposing that States applying to adopt
                      the revised arsenic MCL may use their
                      most recently approved monitoring and
                      waiver plans or note in their primacy
                      application any revisions to those plans.
                        The Agency is clarifying the
                      procedure used for determining
                      compliance after exceedances for
                      inorganic, volatile organic, and
                      synthetic organic contaminants in this
                      proposal. Finally, EPA is proposing in
                      this proposal that States will specify the
                      time frame which new systems and
                      systems using a new source of water
                      have to demonstrate compliance with
                      the MCL's including initial sampling
                      frequencies and compliance periods for
                      new systems and systems that use a new
                      source of water for inorganic, volatile
                      organic, and synthetic organic
                      contaminants.

                      II. Background

                      A. What Is the Statutory Authority for
                      the Arsenic Drinking Water Regulation?
                        Section 1401 of the Safe Drinking
                      Water Act (SDWA) requires a "primary
                      drinking water regulation" to specify a
                      maximum contaminant level (MCL) if it
                      is economically and technically feasible
                      to measure the contaminant and include
                      testing procedures to insure compliance
                      with the MCL and proper operation and
                      maintenance. In addition, section
                      1401(lKD)(i) requires EPA to establish
                      the minimum quality of untreated, or
                      raw, water taken into a public water
                      system. A national primary drinking
                      water regulation (NPDWR) that
                      establishes an MCL also lists the
                      technologies that are feasible to meet the
                      MCL, but systems are not required to
                      use the listed technologies (section
                      1412(b)(3)(E)(i)). As a result of the 1996
                      amendments to SDWA, when issuing a
                      NPDWR, EPA will also list affordable
                      technologies for small systems serving
                      10,000 to 3301, 3300 to 501, and 500 to
                      25 that achieve compliance with the
                      MCL or treatment technique. EPA can
                      list modular (packaged) and point-of-
entry and point-of-use treatment units
for the three small system sizes, as Ipng
as the units are maintained by the
public water system or its contractors.
Home units must contain mechanical
warnings to notify customers of
problems (section 1412(b)(4)(E)(ii)). In
section 1412(b)(12)(A) of SDWA, as
amended August 6, 1996, Congress
directed EPA to propose a national
primary drinking water regulation for
arsenic by January 1, 2000 and issue the
final regulation by January 1, 2001. At
the same time, Congress directed EPA to
develop a research plan by February 2,
1997 to reduce the uncertainty in
assessing health  risks from low levels of
arsenic and conduct the research in
consultation with the National Academy
of Sciences,  other Federal agencies, and
interested public and private entities.
The amendments allowed EPA to enter
into cooperative agreements for
research.
   Section 1412(a)(3) requires EPA to
propose a maximum contaminant level
goal (MCLG) simultaneously with the
national primary drinking water
regulation. The MCLG is defined in
section!412(b)(4)(A) as "the level at
which no known or anticipated adverse
effects on the health of persons occur
and which allows an adequate margin of
safety." Section  1412(b)(4)(B) specifies
that each national primary drinking
water regulation will specify a
maximum contaminant level (MCL) as
close to the MCLG as is feasible, with
two exceptions added in the 1996 ,
amendments. First, the Administrator
may establish an MCL at a level other
than the  feasible level if the treatment
to meet the feasible MCL would increase
the risk from other contaminants or the
technology would interfere with the
treatment of other contaminants
(section!412(b)(5)). Second, if benefits
at the feasible level would not justify
the costs, EPA may propose and
promulgate an MCL "that maximizes
health risk reduction benefits at a cost
that is justified by the benefits (section
  When proposing an MCL, EPA must
publish, and seek public comment on,
the health risk reduction and cost
analyses (HRRCA) of each alternative
maximum contaminant level considered
(section 1412(b)(3)(C)(i)). This includes
the quantifiable and nonquantifiable
benefits from reductions in health risk,
including those from removing co-
occurring contaminants (not counting
benefits resulting from compliance with
other proposed or final regulations),
costs of compliance (not counting costs
resulting from other regulations), any
increased health risks (including those
from co-occurring contaminants) that

-------
                  Federal Register/Vol. 65, No.  121/Thursday, June 22,  2000/Proposed Rules
                                                                      38893
 may result from compliance,
 incremental costs and benefits of each
 alternative MCL considered, and the
 effects on sensitive subpopulations (e.g.,
 infants, children, pregnant women,
 elderly, seriously ill, or other groups at
 greater risk). EPA must analyze the
 quality and extent of the information,
 the uncertainties in the analysis, and the
 degree and nature of the risk.
   The 1996 amendments also require
 EPA to base its action on the best
 available, peer-reviewed science and
 supporting studies and to present health
 effects information to the public in an
 understandable fashion. To meet the
 latter obligation, EPA must specify,
 among other things, the methodology
 used to reconcile inconsistencies in the
 scientific data for the final regulation
 (section 1412(b)(3)(B)(v)).
   Section 1451(a) allows EPA to
 delegate primary enforcement
 responsibility to federally recognized
 Indian Tribes, providing grant and
 contract assistance, using the
 procedures applied to States. Section
 1413(a)(l) allows EPA to grant States
 primary enforcement responsibility for
 NPDWRs when EPA has determined
 that the State has adopted regulations
 that are no less stringent than EPA's.
 States must adopt comparable
 regulations within two years of EPA's
 promulgation of the final rule, unless a
 two-year extension is justified. State
 primacy also requires, among other
 things, adequate enforcement (including
 monitoring and inspections) and
 reporting. EPA must approve or deny
 State applications within 90 days of
 submission (section 1413(b)(2)). In some
 cases, a State submitting revisions to
 adopt a national primary drinking water
 regulation has enforcement authority for
 the new regulation while EPA action on
 the revision is pending (section
B. What Is Arsenic?
  Arsenic is an element that occurs
naturally in rocks, soil, water, air,
plants, and animals. Arsenic is a
metalloid, which exhibits both metallic
and nonmetallic chemical and physical
properties. The primary valence states
for arsenic are 0, — 3, +3 and +5.
Although arsenic is found in nature to
a small extent in its elemental form (0
valence), it occurs most often as
inorganic and organic compounds in
either the As (III) (+3)  or As (V) (+5)
valence states. The trivalent forms of
inorganic arsenic [As (III) (e.g., arsenite,
H3AsO3)] and the pentavalent forms [As
(V) (e.g., arsenate, H2AsO4-, HAsO42-)]
are inorganic species which tend to be
more prevalent in water than the
organic arsenic species (Irgolic, 1994;
 Clifford and Zhang, 1994). The
 dominant inorganic species present in
 water is largely a function of the pH and
 the oxidizing/reducing conditions
 which affects the need for pretreatment
 and removal effects. Arsenates are more
 likely to occur in aerobic surface waters
 and arsenites are more likely to occur in
 anaerobic ground waters.         '
 C. What Are the Sources of Arsenic
 Exposure?

 1. Natural Sources of Arsenic
   There are numerous natural sources
 as well as human activities that may
 introduce arsenic into food and drinking
 water. The primary natural sources
 include geologic formations (e.g., rocks,
 soil, and sedimentary deposits),
 geothermal activity, and volcanic ;
 activity. Arsenic and its compounds
 comprise 1.5-2% of the earth's crust
 (Welch, personal communication).
 While concentrations of arsenic in the
 earth's crust vary, the average
 concentrations are generally reported to
 range from 1.5 to 5 mg/kg. Arsenic is a
 major constituent of many mineral,
 species in igneous and sedimentary
 rocks. It is commonly present in the
 sulfide ores of metals including copper,
 lead, silver, and gold. There axe over
 100 arsenic-containing minerals,
 including arsenic pyrites (e.g., FeAsS),
 realgar (AsS), lollingite (FeAs2, Fe2As3,
 FeaAss), and orpiment (AszSs).
 Geothermal water can be a source of
 inorganic arsenic in surface water and
 ground water. Welch et al. (1988)
 identified fourteen areas in the Western
 United States where dissolved arsenic
 concentrations ranged from 80 to 15,000
 jig/L. In addition, natural emissions of
 arsenic are associated with forest fires
 and grass fires. Volcanic activity  .
 appears to be the largest natural source
 of arsenic emissions to the atmosphere
 (ATSDR, 1998). Arsenic compounds,
 both inorganic and organic, are also
 found in food.

 2. Industrial Sources of Arsenic   i
   Major present and past sources of
 arsenic include wood preservatives;
 agricultural uses, industrial uses,  :
 mining and smelting. The human
 impact on arsenic levels in water
 depends on the level of human activity,
the distance from the pollution sources,
and the dispersion and fate of the
arsenic that is released. The production
of chromated copper arsenate (CCA), an
inorganic arsenic compound and wood
preservative, accounts for         '
approximately 90% of the  arsenic used
annually by industry in the United
States (USGS, 1998; USGS, 1999). CCA
is used to pressure treat lumber, which
 is typically used for the construction of
 decks, fences, and other outdoor
 applications. In addition to wood
 preservatives, the other EPA-registered
 use of inorganic arsenic is for sealed ant
 bait. In the past, agricultural uses of
 arsenic included pesticides, herbicides,
 insecticides, defoliants, and soil
 sterilants. Inorganic arsenic pesticides
 are no longer used for agricultural
 purposes; the last agricultural
 application was voluntarily canceled in
 1993 (58 FR 64579, US EPA, 1993b).
   Organic forms of arsenic are
 constituents of some agricultural
 pesticides that are currently used in the
 U.S. Monosodium methanearsonate
 (MSMA) is the most widely applied
 organoarsenical pesticide, which is used
 to control broadleaf weeds  and is
 applied to cotton (Jordan et al., 1997).
 Small amounts of disodium
 methanearsonate (DSMA, or cacodylic
 acid) are also applied to cotton fields as
 herbicides. The Food and Drug
 Administration regulates other organic
 arsenicals (e.g., roxarsone and arsanilic
 acid) used as feed additives for poultry
 and swine for increased rate of weight
 gain, improved feed  efficiencies,
 improved pigmentation, and disease
 treatment and prevention. These
 additives undergo little or no
 metabolism before excretion (NAS,
 1977; Moody and Williams, 1964;
 Aschbacher and Feil, 1991).
  Arsenic and arsenic compounds
 (arsenicals) are used for a variety of
 industrial purposes,  including:
 electrophotography,  catalysts,
 pyrotechnics, antifouling paints,
 pharmaceutical substances, dye and
 soaps, ceramics, alloys (automotive
 solder and radiators), battery plates,
 optoelectronic devices, semiconductors,
 and light emitting diodes in digital
 watches  (Azcue and Nriagu, 1994). In
 addition, burning of fossil fuels,
 combustion of wastes, mining and
 smelting, pulp and paper production,
 glass manufacturing, and cement
 manufacturing can result in emissions
 of arsenic to the environment (US EPA,
 1998). Arsenic has been identified as a
 contaminant of concern at 916 of the
 1,467 National Priorities List
 (Superfund) hazardous waste sites
 (ATSDR, 1998).
 3. Dietary Sources
  Because arsenic is naturally
 occurring, the entire population is
 exposed to low levels of arsenic through
 food, water, air, and contact with soil.
 The National Research Council report
 (NRG, 1999) described in sections III.C.
 and III.E.3. provides Food and Drug
Administration (FDA) "market basket"
 data for total arsenic intake by age

-------
38894
Federal Register/Vol.  65,  No. 121/Thursday, June  22, 2000/Proposed Rules
group. NRC assumed that, for fish and
seafood, inorganic arsenic is 10% of the
total arsenic and that other food
contains entirely inorganic arsenic.
These assumptions are probably high
and conservative for public health
protection to avoid underestimating the
contributions from food. Table 3-5 in
the 1999 NRC report characterizes
inorganic arsenic intake from food in
the U.S. as being 1.3 ug/day for infants
under one year old, 4.4 ug/day for 2-year
olds, almost 10 ug/day for 25-30 year-
old males, with a maximum of 12.5 ug/
day for 60-65 year-old males (females
had lower arsenic intake in every age
group). Macintosh et al. (1997)
estimated that 785 adults had a mean
inorganic arsenic consumption of 10.22
(lg/day, with a standard deviation of
6.54 ug/day and a range of 0.36-123.84
ug/day based on semi-quantitative food
surveys.
  Likewise, the 2 L/day assumption of
adult drinking water intake used to
develop the MCLG does not represent
intake by the average person; rather it
represents intake of a person in the 90th
  Sercentile. (See Section X.B.l.a. for a
  ascription of water consumption for
the general population.)
4. Environmental Sources
  Internal exposure after skin contact
with water or soil containing arsenic or
inhalation of arsenic from air is believed
to be low. Studies of inorganic arsenic
absorption from skin from cadavers
estimated 0.8% uptake from soil and
1.9% uptake from water over a 24-hour
period (Wester et al., 1993). EPA's
arsenic health assessment document for
the Clean Air Act (US EPA, 1984) cited
respiratory arsenic as being about 0.12
ug/day from a daily ventilation rate of
20 m3 using a 1981 national average
arsenic air concentration of 0.006 ug/m3.
Assuming 30 percent absorption, the
daily amount of arsenic from breathing
would be 0.03 ug, so air is a minor
source of arsenic (50 FR 46936 at 46960;
US EPA, 1985b). At this time, EPA is
basing health risks on estimates of
arsenic exposure from food and water.
The Centers for Disease Control and
Prevention (CDC) is initiating a study of
arsenic intake from bathing. EPA
requests comment on whether available
data on skin absorption and inhalation
indicate that these are significant
exposure routes that should be
considered in the risk assessment.
D. What is the Regulatory History for
Arsenic?
  Regulation of arsenic has been the
subject of scientific debate that has
lasted for decades despite research and
scientific review. The controversy has
                      affected policy and regulatory decisions
                      for arsenic in drinking water from low,
                      environmental exposure.

                      1. Earliest U.S. Arsenic Drinking Water
                      Standards
                        In 1942 the U.S. Public Health Service
                      first established an arsenic drinking
                      water standard for interstate water
                      carriers at 0.05 mg arsenic per liter (mg/
                      L, or 50 Ug/L), as measured with a
                      colorimetric method. The report did not
                      cite any reason for choosing that level,
                      but it defined "safety of water supplies"
                      as "the danger, if any, is so small that
                      it cannot be discovered by available
                      means of observation (US Public Health
                      Service 1943)." In 1946, the Surgeon
                      General of the U.S. Public Health
                      Service noted that the American Water
                      Works Association had accepted the
                      1942 drinking water standards,
                      including the arsenic standard (U.S.
                      Public Health Service 1946). In 1962
                      (U.S. Public Health Service 1962)  the
                      U.S. Public Health Service issued more
                      stringent drinking water standards for
                      arsenic of 0.01 mg/L (10 ug/L) for  a
                      water supply in 42 CFR 72.205(b)(l) and
                      0.05 mg/L in 42 CFR 72.205(b)(2)  as
                      grounds for rejection of a water supply,
                      as measured by the current edition of
                      Standard Methods for the Examination
                      of Water and Wastewater per 42 CFR
                      72.207(a).
                        The Safe Drinking Water Act of 1974
                      amended the Public Health Service Act
                      and specified that EPA set primary and
                      secondary drinking water standards. On
                      December 24, 1975 (40 FR 59566 at
                      59570; US EPA, 1975), EPA issued a
                      National Interim Primary Drinking
                      Water Regulation for arsenic in
                      § 141.23(b) of 0.05 mg/L (50 Ug/L),
                      effective 18 months later (§ 141.6).
                      Commenters recommended an MCL of
                      100 ug/L, saying there were no observed
                      adverse health effects (40 FR 59566 at
                      59576; US EPA, 1975). EPA noted long-
                      term chronic effects at 300-2,750  ug/L,
                      but observed no illnesses in a California
                      study at 120 ug/L. Drinking 2 liters of
                      water a day containing arsenic at  50 ug/
                      L would provide approximately 10% of
                      total ingested arsenic from food and
                      water, estimated to be 900 Ug/day. The
                      section on arsenic noted that arsenic has
                      been believed to be a carcinogen
                      "[sjince the early nineteenth century
                       * *; however evidence from animal
                      experiments and human experience has
                      accumulated to strongly suggest that
                      arsenicals do not produce cancer. One
                      exception is a report from Taiwan
                      *  *  *. The text goes on to note
                      occupational skin and lung cancer from
                      arsenic dust and skin cancer in England
                      from drinking water with 12 mg/L. (US
                      EPA, 1976 Appendix A).
2. EPA's 1980 Guidelines
  Scientific data at the time the 1980
Ambient Water Quality Guidelines were
formulated did not support a safe or
"threshold" concentration for
carcinogens, so EPA's public health
policy was
"that the recommended concentration for
maximum protection of human health is
zero. In addition, the Agency presented a
range of concentrations corresponding to
incremental cancer risks of 10 - 7 to 10 - 5
(one additional case of cancer in populations
ranging from ten million to 100,000,
respectively) * * * [that did not necessarily
represent] an Agency judgement on an
'acceptable' risk level (45 FR 79318 at 79323,
US EPA, 1980)."
  In the November 28,1980 Federal
Register document, using its then
current risk assessment approach
(assumed toxicity increased as a natural
logarithm linear function across
species), EPA set the Clean Water Act
surface water quality criterion for
arsenic at  2.2 nanograms (ng/L) (0.0022
ug/L) at an increased cancer risk of
10 —6. The criterion was to prevent skin
cancer in humans drinking
contaminated water and eating aquatic
organisms from those water bodies (45
FR  79318  at 79326). The 1980 Federal
Register notice indicated that drinking
water standards consider a range of
factors, including health effects,
technological and economic feasibility
of removal, and monitoring capability.
On the other hand the Clean Water Act
criteria of section 304(a)(l) "have no
regulatory significance under the
SDWA." The Clean Water Act section
304(a)(l) criteria are more similar to the
health-based goals of the recommended
maximum contaminant levels (now
referred to as MCLGs), than to MCLs;
and differences in mandates "may result
in differences between the two
numbers." (45 FR 79318 at 79320; US
EPA, 1980). In 1992, the Clean Water
Act criterion was recalculated based on
the updated cancer risk assessment in
EPA's Integrated Risk Information
System (IRIS) database, to a level of
0.018 ug/L for arsenic at a 10 - 6 cancer
risk (57 FR 60848; US EPA, 1992c).
3. Research and Regulatory Work
  The 1980 National Academy of
Science (NAS) Volume III of "Drinking
Water and Health" report encouraged
EPA to research whether arsenic is
essential for humans, as demonstrated
in four studies of mammalian species.
The 1983  NAS Volume V report
projected that 0.05 mg/kg of total
arsenic may be a desirable level for
people, and 25 to 50 ug a day may be
required (as cited in 50 FR 46936 at
46960; US EPA, 1985b).

-------
                  Federal  Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed Rules
                                                                     38895
   In 1983, EPA requested comment on
 whether the arsenic MCL should
 consider carcinogenicity, other health
 effects, and nutritional requirements,
 and whether MCLs are necessary for
 separate valence states {e.g., arsenite vs.
 arsenate) (48 FR 45502 at 45512; US
 EPA, 1983). On November 13,1985,
 EPA proposed (50 FR 46936; US EPA,
 1985b) a recommended maximum
 contaminant level (RMCL), a non-
 enforceable health goal now known as
 an MCLG,  of 50 ug/L based on the 1983
 NAS conclusion that 50 Hg/L balanced
 toxicity and possible essentiality and
 provided "a sufficient margin of safety"
 (50 FR 46936 at 46960). EPA also
 requested comment on alternate RMCLs
 of 100 Hg/L based on noncarcinogenic
 effects (calculated from an animal study
 and an uncertainty factor of 1000) and
 0 ug/L based on carcinogenicity (50 FR
 46936 at 46961). EPA chose not to base
 the proposed RMCL on the animal study
 because each dose group had only four
 Rhesus monkeys. Also, at that time;
 studies had "not detected increased
 risks via drinking water in the USA" (50
 FR 46936 at 46960). The 1985 proposed
 drinking water regulation preamble
 noted the 1980 excess cancer risk values
 derived from the ambient water quality
 criteria were based on skin cancer using
 the 1968 Tseng et al. study (50  FR 46936
 at 46961).
   The June 19,1986 amendments to the
 Safe Drinking Water Act (SDWA; Public
 Law 99-339) converted the 1975 interim
 arsenic standard to a National Primary
 Drinking Water Regulation  (section
 1412(a)(l)), subject to  revision by 1989
 (section 1412(b)(l)). Review of the
 arsenic risk assessment issues caused
 the Agency to miss the 1989 deadline
 for proposing a revised NPDWR. As a
 result of a citizen suit to enforce the
 deadline, EPA entered into  a consent
 decree providing deadlines for  issuing
 the arsenic rule.
   In 1988, EPA's Risk Assessment
 Forum issued the Special Report on
 Ingested Inorganic Arsenic: Skin Cancer;
 Nutritional Essentiality (EPA/625/3-87/
 013), in part, to evaluate the validity of
 applying skin cancer data from
 Taiwanese  studies (published in 1968
 and 1977) in dose-response assessments
 in the U.S.  As described in the report,
 the maximum likelihood estimate of risk
 ranged from 3 x 10 - 5  to 7 x 10 - 5 for
 a 70-kilogram person consuming 2 liters
 of water per day contaminated with 1  ug
 of arsenic per liter. Calculated at the 50
 |jg/L standard, the U.S. lifetime risk of
 skin cancer ranged from Ixl0-3to3
 x 10 —3, which means  one to three skin
 cancers would occur in a group of one
thousand people drinking water
containing arsenic at 50 |J.g/L. Existing
 studies could not determine whether
 arsenic was an essential nutrient.  '
   After reviewing the scientific
 evidence for carcinogenicity, EPA's
 Science Advisory Board (US EPA, !l989a
 and b) stated in its August 1989 and
 September 1989 reports that (1) the
 animal studies suggesting arsenic is an
 essential nutrient are not definitive; (2)
 the skin changes seen in hyperkeratosis
 may not always result in skin cancer; (3)
 the 1968 Taiwan data demonstrate, that
 high doses of ingested arsenic can 'cause
 skin cancer; (4) the Taiwan study is
 inconclusive to determine cancer risk at
 levels ingested in the United States
 (U.S.); and (5) As (III) levels below:200-
 250 ug per day may be detoxified. SAB
 recommended that EPA set the MCL
 using a non-linear dose-response (at
 some low dose, arsenic would not be
 toxic). The SAB report recommended
 that EPA revise the risk assessment
 based on dose of arsenic to target tissues
 (the concentration of arsenic that  :
 damages tissues, rather than the
 concentration in water) and consider
 detoxification.                   j
   The SAB also reviewed EPA's April
 12, 1991 Arsenic Research        .
 Recommendations (US EPA, 1991c).
 The final report provided SAB's
 recommendations (US EPA, 1992a) and
 "identified research needed to resolve
 major uncertainties about inorganic
 arsenic cancer risk" to evaluate if work
 could be done in three to five years. It
 noted that "important work can bejdone
 within the time available. Although the
 results from this work will not     '.
 completely resolve any issue, *  *  * the
 results will likely significantly improve
 the Agency's ability to evaluate the risk.
 *  *  * through improved knowledge of
 arsenic metabolism and * *  * as a i
 carcinogen." The report reflected
 uncertainty as to whether or not EPA
 could obtain enough  data to regulate
 arsenic using a non-linear model, which
 needed more information on how  !
 arsenic induces cancer. The group noted
 that it would take longer than five years
 to develop an animal model to help
 understand the toxicity of arsenic. SAB
 recommended four short-term studies:
 (a) Investigation of chromosome   ;
 damage, arsenic metabolites, and the
 times cells are most susceptible to  '•
 arsenic, (b) study of human liver   '
 capacity to add methyl groups to   •
 arsenic, (c) identifying the species in
urine in several populations to look for
 evidence of saturation of methylatiqn
 enzymes, and (d) comparing methylated
arsenic excreted in the U.S., Taiwan,
Mexico, and Argentina to consider the
effect of nutritional or genetic
differences on methylation capacity.
However, if time were not a factor, SAB
 ranked developing an animal model of
 arsenic-induced cancer as the first
 priority.
   In 1993 SAB reviewed EPA's draft
 "Drinking Water Criteria Document on
 Inorganic Arsenic (US EPA, 1993a)." In
 1995, SAB reviewed the analytical
 methods, occurrence estimate, treatment
 technologies, and approach for
 assigning costs in the regulatory impact
 analysis (US EPA, 1995). Besides
 highlighting previous SAB reviews of
 1989,1992, and 1994 on health effects,
 the 1995 report recommended changes
 to the practical quantitation limit
 approach, use of occurrence data,
 review of technologies, and support for
 the decision tree, with some
 reservations.
   EPA held internal  workgroup
 meetings throughout 1994, addressing
 risk assessment, treatment, analytical
 methods, arsenic occurrence,  exposure,
 costs, implementation issues, and
 regulatory options. EPA decided in early
 1995 to defer the arsenic regulation in
 order to better characterize health
 effects and assess cost-effective removal
 technologies for small utilities.
  The 1996 amendments to SDWA
 included a new statutory deadline for
 the arsenic regulations, as  discussed in
 section II.A.

 E. EPA's Arsenic Research Plan
  EPA held a workshop in March 1994
 entitled "Workshop on Developing an
 Epidemiology Research Strategy for
 Arsenic in Drinking Water." The cover
 letter to the final report (US EPA,
 1997b), dated April 14, 1997, notes that
 EPA has been using the
 recommendations to  direct its research
 directions. The report listed ten projects
 and seventeen conclusions on exposure,
 endpoints, study design and statistical
 power, population selection, feasibility
 of conducting a study in the U.S.,
 international studies, importance of
 developing biomarkers to measure
 health effects of arsenic, and animal
 studies.
  In 1995, the Water  Industry  Technical
 Action Fund (WITAF) ( funded by the
 American Water Works Association,
 National Association of Water
 Companies, Association of Metropolitan
 Water Agencies, National Rural Water
 Association, and National Water
 Resources Association), the AWWA
 Research Foundation, and the
 Association of California Water
Agencies (ACWA) sponsored an Expert
 Workshop on Arsenic Research Needs
 in Ellicott City, MD, May 31-June 2,
 1995. The final report (AWWA ef al.,
 1995) identified research projects in
mechanisms, epidemiology, toxicology,
and treatment. It identified ten high

-------
38896
Federal  Register/Vol. 65, No. 121/Thursday, June  22,  2000/Proposed Rules
priority projects which would need over
S3 million to fund, eleven medium
priority projects needing over S6
million, and ten low priority projects
costing over S9 million, that totaled over
S19 million in research needs.
  Congress recognized the importance
of health effects research in regulating
arsenic, as demonstrated by the 1996
statutory requirement to develop a
research plan within  180 days "in
support of drinking water rulemaking to
reduce the uncertainty in assessing
health risks associated with exposure to
low levels of arsenic  * * * (section
1412(b)(12)(A)(ii)). In the research plan
EPA recognized that "[t]he research
needs are broader than those tiiat EPA
can address alone, and it is anticipated
that other entities will be involved in
conducting some of the needed research
(US EPA, 1998a)." (See section m.E.l.
on industry-funded research and the
arsenic research plan (at wmv.epa.gov/
ORD/WebPubs/final/arsenic.pdf) for
EPA-funded projects.) In December
1996, EPA submitted its draft research
plan for peer review by its Board of
Scientific Counselors' (BOSC) Ad Hoc
Committee, and the committee met in
January 1997. The February 1998
Arsenic Research  Plan addressed the
June 1997 comments from BOSC.
  Major areas covered in the research
plan included studies to:
  • Improve our qualitative and
qxiantitative assessment of the human
toxicity of arsenic;
  • Understand mechanisms of arsenic
toxicity that may aid in extension of the
observed human findings when
extrapolation is required;           ^
  • Measure exposures of the US     '
population to arsenic from various
sources (particularly diet) to allow
bolter definition of cumulative
exposures to arsenic;
  • Refine treatment technologies that
may better remove arsenic from water
supplies;
  • Improve methods for analyzing and
monitoring arsenic in drinking water.
  EPA also set priorities in the plan and
identified projects that met the short
term and long term criteria:
Short Term Criteria
  1. Will the research improve the
scientific basis for risk assessments
needed for proposing a revised arsenic
MCL by January 1, 2000?
  2. Will the research improve the
scientific basis for risk management
decisons needed for proposinig a
revised arsenic MCL by January 1, 2000?
Long Term Criteria
  1. Will the research improve the
scientific basis for risk assessment and
                      risk management decisions needed to
                      review and develop future MCLs
                      beyond the year 2000?
                        2. Is the research essential to
                      improving our scientific understanding
                      of the health risks of arsenic?
                        The research plan included the
                      following priority topics for research
                      under the five major areas of
                      investigation supporting drinking water
                      rulemaking:
                      Exposure Analysis
                        • Arsenic speciation and
                      preservation: Improvements in
                      analytical methods to support water
                      treatment decisions.
                        • Measurement of background
                      exposures to arsenic in U.S. population,
                      particularly inorganic arsenic intake in
                      the U.S. diet.
                        • Development and evaluation of
                      biomarkers (e.g., species of arsenic in
                      urine) of exposures.
                        • Development of standard reference
                      material for arsenic in water, food,
                      urine, tissues.
                      Cancer Effects
                        • Further study of internal cancers
                      associated with arsenic exposures.
                        • Dose response data on
                      hyperkeratosis as a likely precursor to
                      skin cancer.
                        • Research  on factors influencing
                      human susceptibility including age,
                      genetic characteristics and dietary
                      patterns.
                        • Metabolic and pharmacokinetic
                      studies that can identify dose dependent
                      metabolism.
                        • Mechanistic studies for arsenic-
                      induced genotoxicity and
                      carcinogenicity.
                      Noncancer Effects
                        • Development of human dose-
                      response data for hyperkeratosis,
                      cardiovascular disease, neurotoxicity
                      and developmental effects.
                        • Development of additional health
                      effects and hazard identification data on
                      other non-cancer endpoints such as
                      diabetes and hematologic effects.
                      Risk  Management Research

                        • Identification of limitations of
                      treatment technologies and impacts on
                      water quality.
                        • Development of treatment
                      technologies for small water systems.
                        • Development of data on cost and
                      performance capabilities of various
                      treatment options.
                        • Consideration of residuals
                      management issues, including disposal
                      options and costs.
Risk Assessment/Characterization
  • Development of risk
characterizations to provide interim
support to States and local
communities.
  • Development of predictive tools
and statistical models for assessing
bioavailability, interactions and dose-
response as better mass balance data
become available.
  • Comprehensive assessment of
exposure levels and incorporation of
data into risk estimates for better
characterization of actual risks
associated with arsenic exposure.
  • Comprehensive assessment of
arsenic mode of action provide a greater
understanding of biological mechanisms
and factors that may impact the shape
of a dose response curve.
  • Comprehensive assessment of non-
cancer risks and consideration of
appropriate modeling tools for
quantitative estimation of non-cancer
risks.
  • Comprehensive assessment of
human dose-response data for
hyperkeratosis, cardiovascular disease,
neurotoxicity and developmental
effects.                            :
in. Toxic Forms and Health Effects of
Arsenic
A. What Are the toxic Forms of Arsenic?
  Arsenic exists in several forms which
vary in toxicity and occurrence.
Accordingly, for this proposed
regulation, it is important to consider
those forms that can exert toxic effects
and to which people may be exposed.
For example, the metallic form of
arsenic (0 valence) is not absorbed from
the stomach and intestines and does not
exert adverse effects. On the other hand,
a volatile compound such as arsine
(AsHs) is toxic, but is not present in
water or food. Moreover, the primary
organic forms (arsenobetaine and
arsenocholine) found in fish and
shellfish seem to have little or no
toxicity (Sabbioni et al, 1991).
Arsenobetaine quickly passes  out of the
body in urine without being
metabolized to other compounds
(Vahter, 1994). Arsenite (+3) and
arsenate (+5) are the most prevalent
toxic forms of inorganic arsenic that are
found in drinking water. However,
recovery of identified arsenic species in
vegetables, grains and oils  has been
limited and difficult, so little is known
about types of species in these foods
(NRG, 1999).
  In animals and humans, inorganic
pentavalent arsenic is converted to
trivalent arsenic that can be methylated
(i.e., chemically bonded to a methyl
group, which is a carbon atom linked to

-------
                 Federal Register/ Vol. 65, No. 121/Thursday, June. 22, 2000/Proposed  Rules
                                                                      38897
three hydrogen atoms) to mono-methyl
arsenic (MMA) and di-methyl arsinic
acid (DMA), which are organic
arsenicals. The primary route of
excretion for arsenic metabolites is in
the urine. Studies indicate that the
organic arsenicals MMA and DMA were
hundreds of times less likely to produce
genetic changes in animal cells than
inorganic arsenicals. Moreover, many
studies reported organic arsenicals to be
less reactive in tissues, to kill less cells,
and to be more easily excreted in urine
(NRG, 1999).

B. What Are the Effects of Acute
Toxicity?
  Inorganic arsenic can exert toxic
effects after acute (short-term) or
chronic (long-term) exposure. From
human acute poisoning incidents, the
LDso of arsenic has been estimated to
range from 1 to 4 mg/kg (Vallee et al.,
1960, Winship, 1984). This dose would
correspond to a lethal dose range of 70
to 280 mg for 50% of adults weighing
70 kg. At nonlethal, but high acute
doses, inorganic arsenic can cause
gastroenterological effects, shock,
neuritis (continuous pain)  and vascular
effects in humans (Buchanan, 1962).
Such incidents usually occur after
accidental exposures. However,
sometimes high dose acute exposures
may be self-administered. For example,
inorganic arsenic is a component of
some herbal medicines and adverse
effects have been reported  after use. In
one report of 74 cases (Tay and Seah,
1975), the primary signs were skin
lesions (92%), neurological (i.e., nerve)
involvement (51%), and
gastroenterological, hematological (i.e.,
blood) and renal (i.e., kidney) effects (19
to 23%). Although acute or short-term
exposures to high doses  of inorganic
arsenic can cause adverse effects, such
exposures do not occur from public
water supplies in the U.S. at the current
MCL of 50 Ug/L. EPA's proposed
drinking water regulation addresses the
long-term, chronic effects of exposure to
low concentrations of inorganic arsenic
in drinking water.

C. What Cancers Are Associated With
Arsenic?
  Inorganic arsenic is a multi-site
human carcinogen by the drinking water
route. Asian, Mexican and South
American populations with exposures
to arsenic in drinking water generally at
or above several hundred micrograms
per liter are reported to have increased
risks of skin, bladder, and lung cancer.
The current evidence also suggests that
the risks of liver and kidney cancer may
also be increased following exposures to
inorganic forms of arsenic. The weight
of evidence for ingested arsenic as a
causal factor of carcinogenicity is much
greater now than a decade ago, and the
types of cancer occurring as a result of
ingesting inorganic arsenic have even
greater health implications for U.Si and
other populations than the occurrence
of skin cancer alone. (Until the late
1980s skin cancer had been the cancer
classically associated with arsenic in
drinking water.) Epidemiologic studies
(e.g., of people) provide direct data on
arsenic risks from drinking water at
exposure levels much closer to those of
regulatory concern than environmental
risk assessments based on animal  •-
toxicity studies.

1. Skin Cancer
  Early reports linking inorganic arsenic
contamination of drinking water to skin
cancer came from Argentina (Neubauer,
1947, reviewing studies published :as
early as 1925) and Poland (Geyer, 1898,
as reported in Tseng et al., 1968).
However, the first studies that observed
dose-dependent effects of arsenic  i
associated with skin cancer came from
Taiwan (Tseng et al., 1968; Tseng,
1977). These studies focused EPA's
attention on the health effects of
ingested arsenic. Physicians physically
examined over 40,000 residents from 37
villages and 7500  residents exposed to
0.017 mg/L arsenic ( reference group).
The study population was divided!into
three groups based on exposure to
inorganic arsenic (0 to 0.29, 0.30 to 0.59
and >0.60 mg of inorganic As/Liter)
measured at the village level. A dose-
and age-related increase of arsenic-
induced skin cancer among the villagers
was noted. No skin cancers were
observed in the low arsenic reference
areas. The 1999 NRG report rioted that
the "primary limitation of this study
* *  * was the lack of detail" reported,
such as grouping individuals into  :
"broad exposure groups" (rather than
grouping into 37 village exposures).
This limits the usefulness of these
studies. However, these Tseng reports
and other corroborating studies such as
those by Albores et al. (1979) and  !
Cebrian et al. (1983) on drinking wpter
exposure and exposures to inorganic
arsenic in medicines (Cuzick et al.',
1982) and in pesticides (Roth, 1956) led
the EPA, using skin cancer as the
endpoint, to classify inorganic arsenic
as a human carcinogen (Group A) by the
oral route (US EPA, 1984).

2. Internal Cancers
  Exposure to  inorganic arsenic in
drinking water has also been associated
with the development of internal  '
cancers. "No human studies of    ;
sufficient statistical power or scope
have examined whether consumption of
arsenic in drinking water at the current
MCL results in an increased incidence
of cancer or noncancer effects (NRG,
1999, pg. 7)."
  Chen et al. (1985) used standardized
mortality ratios (SMRs) to evaluate the
association between ingested arsenic
and cancer risk in Taiwan. (SMRs, ratios
of observed to  expected deaths from
specific causes, are standardized to
adjust for differences in the age
distributions of the exposed and
reference populations.) The authors
found statistically significant increased
risks of mortality for bladder, kidney,
lung,  liver and colon cancers. A
subsequent mortality study in the same
area of Taiwan found significant dose-
response relationships for deaths from
bladder, kidney,  skin, and lung cancers
in both sexes and from liver and
prostate cancer for males. They also
found increases in peripheral and
cardiovascular diseases but not in
cerebrovascular accidents (Wu et al.,
1989). There are  several corroborating
reports of the increased risk of cancers
of internal organs from ingested arsenic
including two  from two South American
countries. In Argentina, significantly
increased risks of death from bladder,
lung and kidney  cancer were reported
(Hopenhayn-Rich et al, 1996; 1998). In
a population of approximately  400,000
in northern Chile, Smith et al. (1998)
found significantly increased risks of
bladder and lung cancer mortality.
  There have only been a few studies of
inorganic arsenic exposure via drinking
water in the U.S., and most have not
considered cancer as an endpoint.
People have written EPA asking that the
new MCL be set considering that these
U.S. studies have not seen increases in
cancers at the low levels of arsenic
exposure in U.S. drinking water.
Optimally, low-exposure arsenic studies
involve long-term residency (20-40
years  with known drinking water
arsenic exposure), access to health
records, populations large enough to
detect statistically significant increases
in cancers and other health endpoints,
and limited use of multiple sources of
water (bottled, filtered, beverages, food
prepared outside the home).
  Recently, Lewis et al. (1999)
conducted a mortality study of a
population in Utah whose drinking
water contained relatively low
concentrations of arsenic (averaged 18-
191 M-g/L). They reported no significant
increase in bladder or lung mortality.
They  did report a statistically significant
dose-response  for an increased risk of
prostate cancer mortality. Smoking is an
established risk factor for bladder and
lung cancer, and inorganic arsenic

-------
38898
Federal  Register/Vol. 65, No. 121/Thursday, June  22,  2000/Proposed Rules
behaves as a comutagen even though it
is not mutagenic alone (NRG, 1999, pg.
200). It is possible that inorganic arsenic
potentiates other risk factors for these
cancers. This potential role is consistent
with the NRC, 1999 view that arsenic's
mode of action may be to interfere with
cell "housekeeping" functions that
normally repair genetic damage and
ensure that damaged cells die
(programmed cell death) rather than
reproduce (see section IILD.2. below).
D. What Non-Cancer Effects Are
Associated With Arsenic?
  A large number of adverse
noncarcinogenic effects have been
reported in humans after exposure to
drinking water highly contaminated
with inorganic arsenic. The earliest and
most prominent changes are in the skin,
e.g., hyperpigmentation and keratoses
(calus-like growths). Other effects that
have been reported include alterations
in gastrointestinal, cardiovascular,
hematological (e.g., anemia),
pulmonary, neurological,
immunological and reproductive/
developmental function (ATSDR, 1998).
  The most common symptoms of
inorganic arsenic exposure appear on
the skin and occurr after 5-15 years of
exposure equivalent to 700 fig/day for a
70 kg adult, or within 6 months to 3
years at exposures equivalent to 2,800
ug/day for a 70 kg adult (pg. 131 NRC,
1999). They include alterations in
pigmentation and the development of
keratoses which are localized primarily
on the palms of the hands, the soles of
the feet and the torso. The presence of
hyperpigmentation and keratoses on
parts of the body not exposed to the sun
is characteristic of arsenic exposure
(Yeh, 1973, Tseng, 1977). The same
alterations have been reported in
patients treated with Fowler's solution
(l% potassium arsenite; Cuzick et al.,
1982), used for asthma, psoriasis,
rheumatic fever, leukemia, fever, pain,
and as a tonic  (WHO 1981 and NRC
1999).
  Chronic exposure to inorganic arsenic
is often associated with alterations in
gastroenterological (GI) function. For
example, noncirrhotic hypertension is a
relatively specific, but not commonly
found manifestation in inorganic
arsenic-exposed individuals and may
not become a clinical observation until
the patient demonstrates GI bleeding
(Morris et al, 1974; Nevens et al., 1990).
Physical examination may reveal spleen
and liver enlargement, and
histopathological examination of tissue
specimens may demonstrate periportal
fibrosis (Morris ef al., 1974; Nevens et
al,, 1990; Guha Mazumder et al., 1997).
There have been a few reports of
                     cirrhosis after inorganic arsenic
                     exposure, but the authors of these
                     studies did not determine the subjects'
                     alcohol consumption (NRC 1999).
                       Development of peripheral vascular
                     disease (hardening of the arteries to the
                     arms and legs, that can cause pain,
                     numbness, tingling, infection, gangrene,
                     and clots) after inorganic arsenic
                     exposure has also been reported. In
                     Taiwan, blackfoot disease (BFD, a severe
                     peripheral vascular insufficiency which
                     may result in gangrene of the feet and
                     other extremities) has been the most
                     severe manifestation of this effect. Tseng
                     (1977) reported over 1,000 cases of BFD
                     in the arsenic study areas of Taiwan.
                     Less severe cases of peripheral vascular
                     disease have been described in Chile
                     (Zaldivar et al., 1974) and Mexico
                     (Cebrian, 1987). In a Utah study,
                     increased SMRs for hypertensive heart
                     disease were noted in both males and
                     females after exposure to inorganic
                     arsenic-contaminated drinking water
                     (Lewis et al., 1999). These reports link
                     exposure to  inorganic arsenic effects on
                     the cardiovascular system.
                       Studies in Taiwan (Lai et al., 1994)
                     and Bangladesh (Rahman et al., 1998)
                     found an increased risk of diabetes
                     among people consuming arsenic-
                     contaminated water. Two Swedish
                     studies found an increased risk of
                     mortality from diabetes among those
                     occupationally exposed to arsenic
                     (Rahman and Axelson, 1995; Rahman et
                     al., 1998).
                       Although  peripheral neuropathy
                     (numbness, muscle weakness, tremors,
                     ATSDR 1998) may be present after
                     exposure to  short-term, high doses of
                     inorganic arsenic (Buchanan, 1962; Tay
                     and Seah, 1975), there are no studies
                     that definitely document this effect after
                     exposure to  levels of less than levels
                     (<50 ug/L) of inorganic arsenic in
                     drinking  water. Hindmarsh et al. (1977)
                     and Southwick et al. (1983) have
                     reported  limited evidence of peripheral
                     neuropathy in Canada and the U.S.,
                     respectively, but it was not reported in
                     studies from Taiwan, Argentina or Chile
                     (Hotta, 1989, as cited by NRC 1999).
                       There have been a few, scattered
                     reports in the literature that inorganic
                     arsenic can affect reproduction and
                     development in humans (Borzysonyi et
                     al., 1992; Desi et al., 1992; Tabacova ef
                     al., 1994). After reviewing the available
                     literature on arsenic and reproductive
                     effects, the National Research Council
                     panel (NRC 1999) wrote that "nothing
                     conclusive can be stated from these
                     studies."
                       Based on the studies  mentioned in
                     this section, it is evident that inorganic
                     arsenic contamination of drinking water
                     can cause dermal and internal cancers,
affect the GI system, alter cardiovascular
function, and increase risk of diabetes*
based on studies of people exposed to
drinking water well above the current
arsenic MCL. EPA's MCL is chosen to be
protective of the general population
within an acceptable risk range, not at
levels at which adverse health effects
are routinely seen (see section III.F.7. on
risk considerations).

E. What Are the Recent Developments in
Health Effects Research?

1. Funding of Health Effects Research
  As mentioned earlier in section II.A.,
Congress recognized that we needed
more research to determine the health
effects at low levels of arsenic (below
the observed health effects and below 50
ug/L). On December 6, 1996, EPA issued
a Federal Register notice (61 FR 64739;
US EPA, 1996e) asking for public
comment on four arsenic health
research topics to fund research projects
with $2 million from EPA
appropriations and $1 million in funds
raised by water industry groups (US
EPA, 1996d). In addition, the Office of
Research and Development's (ORD's)
Board of Scientific Counselors (BOSC)
peer reviewed the draft research topics
and the arsenic research plan. In the fall
of 1997, EPA and the industry partners
funded their respective choices for
arsenic research, after having the
applications peer reviewed. EPA issued
three grants for the following research;
Dose Response of Skin Keratoses and
Hyper-Pigmentation, Arsenic
Glutathione Interactions and Skin
Cancer, and Cellular Redox Status. The
water industry groups awarded two
contracts, studying Contribution of
Arsenic From Dietary Sources and
Tumor Studies in Mice.

2. Expert Panel on Arsenic
Carcinogenicity  '
  As part of the Integrated Risk
Information System (IRIS) update effort,
EPA sponsored an "Expert Panel on
Arsenic Carcinogenicity: Review and
Workshop" in May 1997 (US EPA,
1997d). The panel evaluated existing
data to comment on arsenic's
carcinogenic mode of action and the
effect  on dose-response extrapolations.
The panel noted that arsenic
compounds have not formed DNA
adducts (i.e., bound to DNA) nor caused
point mutations. Trivalent inorganic
forms inhibit enzymes, but arsenite and
arsenate do not affect DNA replication.
The panel discussed several modes of
action, concluding that arsenic
indirectly affects DNA, inducing
chromosomal changes. The panel
thought that arsenic-induced

-------
                 Federal Register/Vol.  65,  No. 121/Thursday, Juna 22, 2000/Proposed Rules
                                                                     38899
chromosomal abnormalities could
possibly come from errors in DNA
repair and replication that affect gene
expression; that arsenic may increase
DNA hypermethylation and oxidative
stress; that arsenic may affect cell
proliferation (cell death appears to be
nonlinear); and that arsenic may act as
a co-carcinogen. Arsenite causes cell
transformation but not mutation of cells
in culture. It also induces gene
amplification (multiple copies of DNA
sequences) in a way which suggests
interference with DNA repair or cell
control instead of direct DNA damage.
The panel noted that all identified
modes of action support a nonlinear
dose-response curve, that few data
supports any one mode as most
important, and that more than one mode
of action may be operating. At low doses
the slope of the dose response would
decrease, and at very low doses "might
effectively be linear but with a very
shallow slope, probably
indistinguishable from a threshold."
  In terms of implications for the risk
assessment, the panel noted that risk per
unit dose estimates from human studies
can be biased either way. For the
Taiwanese study, the "* *  * biases
associated with the use of average doses
and with the attribution of all increased
risk to arsenic would both lead to an
overestimation of risk (US EPA, 1997d,
page 31)." While health effects are most
likely observed in people getting high
doses, the effects are assigned to the
average dose of the exposure group.
Thus, risk per unit dose estimated from
the average doses would lead to an
overestimation of risk (US EPA, 1997d,
page 31).
3. NAS Review of EPA's Risk
Assessment
  In 1997, at EPA's request, the National
Academy of Sciences' (NAS)
Subcommittee on Arsenic of the
Committee on Toxicology of the
National Research Council (NRC) met.
Their charge was to review EPA's
assessments of arsenic. The NAS/NRC
Subcommittee finished their work in
March 1999 (The report can be viewed
from the National Academy Press
website: www.nap.edu/books/
0309063337/html/index.html). The
detailed discussion of their work is in
section III.F. In general, the  NRC report
confirms and extends concerns about
human carcinogenicity of drinking
water containing arsenic and offers
perspective on dose-response issues and
needed research. For the decisions in
this regulation, the EPA has relied upon
the NRC report as presenting the best
available, peer reviewed science as of its
completion and has augmented it with
more recently published, peer reviewed
information. Further work on the risk
assessment will also be done before the
final rule is issued to analyze the risks
of internal cancers. The NRC provided
risk numbers for bladder cancer using
the Agency's approach. The NRC report
noted that "some studies have shown
that excess lung cancer deaths attributed
to arsenic are 2-5 fold greater than the
excess bladder cancer deaths. *  * , *
(NRC, 1999, pg. 8)." The NRC
recommended that EPA analyze risks of
internal cancers both separately arid
combined. Peer-reviewed quantitative
analysis of lung tumor risk is expected
to be available for consideration in the
final rulemaking. Meanwhile, this:
proposal, in a "what if" analysis  ,
(discussed in section X.B), estimates the
potential monetary benefits that would
result if the lung cancer and bladder
cancer risks were the same, which
would be the case if the excess lung
cancer deaths actually were 2- to 5-fold
greater than the excess bladder cancer
deaths.                         !
4. May 1999 Utah Mortality Study
  EPA scientists conducted an   '.
epidemiological study of 4,058
Mormons exposed to arsenic in drinking
water in seven communities in Millard
County, Utah (Lewis et al, 1999). The
151 samples from their public and!
private drinking water sources had
arsenic concentrations ranging from 4 to
620 ug/L with seven mean (arithmetic
average) community exposure
concentrations of 18 to 191 p.g/L arid all
seven community exposure medians
(mid-point of arsenic values) <200:|o.g/L.
Observed causes of death in the study
group (numbering 2,203) were    '•
compared to those expected from the
same causes based upon death rates for
the general white male and female'
population of Utah. Several factors
suggest that the study population may
not be representative of the rest of the
United States. The Mormon church, the
predominant religion in Utah, prohibits
smoking and consumption of alcohol
and caffeine. Utah had the lowest i
statewide smoking rates in the U.S. from
1984 to 1996, ranging from 13 to 17%.
Mormon men had about half the cancers
related to smoking (mouth, larynx,' lung,
esophagus, and bladder cancers) as the
U.S. male population from 1971 to 1985
(Lyon et al.,  1994). The Utah study
population was relatively small (-4,000
persons) and primarily English,   :
Scottish, and Scandinavian in ethnic
background.
  While the study population males had
a significantly higher risk of prostate
cancer mortality, females had no  '
significantexcess risk of cancer   j
mortality at any site. Millard County
subjects had higher mortality from
kidney cancer, but this was not
statistically significant. Both males and
females in the study group had less risk
of bladder, digestive system and lung
cancer mortality than the general Utah
population. The Mormon females had
lower death rates from breast and female
genital cancers than the State rate.
These decreased death rates were not
statistically significant.
  Although deaths due to hypertensive
heart disease were roughly twice as high
as expected in both sexes, increases in
death did not relate to increases  in dose,
calculated as the years of exposure
times the median arsenic concentration.
The Utah data indicate that heart
disease should be considered in  the
evaluation of potential benefits of U.S.
regulation. Vascular effects have also
been reported as an effect of arsenic
exposure in studies in the U.S. (Engel et
al. 1994), Taiwan (Wu et al., 1989) and
Chile (Borgono et al., 1977). The overall
evidence indicating an association of
various vascular diseases with arsenic
exposure supports consideration of this
endpoint in evaluation of potential
noncancer health benefits of arsenic
exposure reduction.
5.1999 Review of Health Effects
  Tsai et al. (1999) estimated
standardized mortality ratios (SMR's)
for 23 cancer and non-cancer causes  of
death in women and 27 causes of death
in men in an area of Taiwan with
elevated arsenic exposures (Tsai, et al.,
1999). The SMRs in this study are an
expression of the ratio between deaths
that were observed in an area with
elevated arsenic levels and those that
were expected to occur, compared to
both the mortality of populations in
nearby areas without elevated arsenic
levels and to the national population.
Drinking water (250-1,140 (J.g/L) and
soil (5.3-11.2 mg/kg) in the Tsai  (1999)
population study had high arsenic
content. There are, of course, possible
differences between the population and
health care in Taiwan and the United
States; and arsenic levels in the U.S.  are
not generally as high as they were in the
study area of Taiwan. However, the
study gives an indication of the types of
health effects that may be associated
with arsenic exposure via drinking
water. The study reports a high
mortality rate (SMR > 3) for both sexes
from bladder, kidney, skin, lung, and
nasal cavity cancers and for vascular
disease. Females also had high
mortalities for laryngeal cancer.
  The SMRs calculated by Tsai (1999)
used the single cause of death noted on
the death certificates. Many chronic

-------
38900
Federal Register/Vol.  65,  No. 121/Thursday, June 22, 2000/Proposed Rules
diseases, including some cancers, are
not generally fatal. Consequently, the
impact indicated by the SMR in this
study may underestimate the total
impact of these diseases. The causes of
death reported in this study are
consistent with what is known about the
adverse effects of arsenic. Tsai et al.
(1999) identified "bronchitis, liver
cirrhosis, nephropathy, intestinal
cancer, rectal cancer, laryngeal cancer,
and cerebrovascular disease" as
possibly "related to chronic arsenic
exposure via drinking water," which
had not been reported before. In
addition, people in the  study area were
observed to have nasal cavity and larynx
cancers not caused by occupational
exposure to inhaled arsenic.
6. Study of Bladder and Kidney Cancer
in Finland
  Kurttio et al.  (1999) conducted a case-
cohort design study of 61 bladder and
49 kidney cancer cases  and 275 controls
to evaluate the risk of these diseases
with respect to  arsenic drinking water
concentrations. In this study the median
exposure was 0.1 Hg/L,  the maximum
reported was 64 ng/L, and 1% of the
exposure was greater than 10 jag/L.  The
authors reported that very low
concentrations  of arsenic in drinking
water were significantly associated with
being a case of bladder  cancer when
exposure occurred 2-9  years prior to
diagnosis. Arsenic exposure  occurring
greater than  10  years prior to diagnosis
was not associated with bladder cancer
risk. Arsenic was not associated with
kidney cancer risk even after
consideration of a latency period.
F. What Did the National Academy of
Sciences/National Research  Council
Report?
1. The National Research Council and
Its Charge
  Due to controversy surrounding the
risk assessment of inorganic  arsenic,
EPA asked the National Research
Council (NRC) to do the following:  (1)
Review EPA's characterization of
potential human health risks from
Jngestion of inorganic arsenic in
drinking water; (2) review the available
data on the carcinogenic and
noncarcinogenic effects of inorganic
arsenic; (3) review the data on the
metabolism, kinetics and mechanism(s)/
mode(s) of action of inorganic arsenic;
and (4) identify research needs to fill
data gaps. To accomplish this task, NRC
convened a panel of scientific experts
with backgrounds in chemistry,
toxicology, genetics, epidemiology,
nutrition, medicine, statistics and risk
assessment. In addition to the general
                      expertise of the panel members, many
                      had conducted research on inorganic
                      arsenic. NRC identified the thirteen
                      scientists with "diverse perspectives
                      and technical expertise" that peer
                      reviewed the draft report. The report
                      noted that "EPA did not request, nor did
                      the subcommittee endeavor to provide,
                      a formal risk assessment for arsenic in
                      drinking water (NRC, 1999)."

                      2. Exposure
                        Arsenic is naturally occurring and
                      ubiquitously distributed in the earth's
                      surface. Because of this, the general
                      population is exposed to low levels of
                      arsenic through the food supply. The
                      NRC report provides FDA market basket
                      data for inorganic arsenic intake by age
                      group which, along with similar data for
                      water intake, will permit
                      communication of total exposure
                      estimates of the general population by
                      age group. The assumption is made in
                      the FDA data that, for fish and seafood,
                      inorganic arsenic is 10% of total arsenic.
                      This 10% assumption is acknowledged
                      to be conservative and has been adopted
                      for public health  protection so as not to
                      underestimate the contribution from
                      fish and seafood. Likewise, the 2 L/day
                      assumption of adult drinking water
                      intake does not represent intake by the
                      average person; rather it represents
                      intake of a person in the 90th percentile.
                      3. Essentiality
                        The NRC report examined the
                      question of essentiality of arsenic in the
                      human diet. It found no information on
                      essentiality in humans and only data in
                      experimental animals suggesting growth
                      promotion (arsenicals are fed to
                      livestock for this  reason). Inorganic
                      arsenic has not been found to be
                      essential for human well-being or
                      involved in any required biochemical
                      pathway. Given this and the fact that
                      arsenic occurs naturally in food,
                      consideration of essentiality is not
                      necessary for public health decisions
                      about water.

                      4. Metabolism and Disposition
                        Data from humans show that
                      inorganic arsenic is readily absorbed
                      and transported through the body.  It has
                      a half-life in the body of approximately
                      four days and is primarily excreted in
                      the urine. If a human is exposed to the
                      inorganic arsenate form (+5 valence),
                      the arsenite will be reduced to arsenite
                      (+3). Some of the arsenite will be
                      sequentially methylated to form
                      monomethylarsonic acid (MMA) and
                      dimethylarsinic acid  (DMA). This
                      methylation process decreases acute
                      toxicity and facilitates excretion from
                      the body. Individuals and populations
vary in their metabolism of arsenic.
Such variations may be due to genetic
differences, species and dose of
inorganic arsenic ingested, nutrition,
disease and possibly other factors.
Whether these methylated products
(MMA and DMA) play a role in the
development of cancer and noncancer
endpoints is unknown at the present
time (NRC, 1999). The NRC report
recommended that experiments be
conducted on the factors affecting
interspecies  differences in inorganic
arsenic toxicity including use of human
tissue when  available.
5. Human Health Effects and Variations
in Sensitivity
  The NRC panel concluded that there
is sufficient evidence that chronic
ingestion of inorganic arsenic causes
bladder, lung and skin cancers and
adverse noncancer effects on the
cardiovascular systems, mainly from
studies exposed to "several hundred
micrograms per liter. Few data address
the degree of cancer risk at lower
concentrations of ingested arsenic (NRC,
1999, pg. 130)." The Utah study (Lewis
et al., 1999), published  after the NRC
report, indicates that cardiovascular
effects can occur at lower exposures
than those seen in the studies available
for the NRC report. At the present time,
the NRC report indicates that there is
insufficient evidence to judge whether
inorganic arsenic can affect
reproduction or development in
humans. However, inorganic arsenic can
pass through the placenta (Concha et al.,
1998), and developmental toxicity needs
investigation. In animal studies,
intraperitoneal (injection into the
abdominal cavity) administration of
inorganic arsenic can cause
malformations, and oral dosing has been
reported to alter fetal growth and
viability. The NRC report recommended
additional studies to characterize the
dose-response curve for inorganic
arsenic-induced cancer and noncancer
health endpoints. They also stated that
the reported beneficial effects of
inorganic arsenic in animals should be
carefully monitored. In addition, the
potential effects of inorganic arsenic on
human reproduction should be
investigated.
  There are many factors (genetics, diet,
metabolism, health and sex) that may
affect a human's response to inorganic
arsenic exposure. For example,
reduction in methylation of inorganic
arsenic methylation can cause humans
to retain more arsenic in their tissues.
The retention of a greater arsenic load
could place a person at a greater risk.
The NRC report (1999) recommended
that various  factors that have the ability

-------
                 Federal Register/Vol. 65, No.  121 /Thursday, June  22,  2000/Proposed Rules
                                                                     38901
to alter a human's response to inorganic
arsenic exposure be carefully examined.
Specifically, these studies should focus
on the extent of human variability with
respect to metabolism, tissue deposition
and excretion under different
environmental conditions.
  Humans are variable in their
metabolic processing of inorganic
arsenic, and internal dose will vary from
person to person because of this as well
as because of diet, nutritional status,
lifestyle, and health status. Human
variability also exists in response
characteristics (susceptibility). The full
quantitative extent of this variability is
not known. For instance, men are more
susceptible than women to bladder
cancer throughout the world even
though bladder cancer rates vary from
region to region. We do not know
whether arsenic may have a greater
effect at different ages (e.g., infants v.s.
adults).
6. Modes of Action
  Knowledge of a "mode of action"
means that data are available to describe
the key events at the cellular and/or
subcellular level that lead to the
development of the cancer or noncancer
endpoint. A number of potential  modes
of carcinogenic action have been
proposed for arsenic, with varying
degrees of supporting  data. The key
events in the cancer process caused by
arsenic exposure are not known.
Nevertheless, the data are sufficient to
support the conclusion of the NRG
report and the EPA 1997 expert panel
workshop report that:  "Arsenic
exposure induces chromosomal
abnormalities without direct reaction
with DNA (US EPA, 1997d)."
  There is strong evidence against a
mode of action for inorganic arsenic
involving direct reaction with DNA.
One of the hallmarks of direct DNA
reactivity is multi-species carcinogenic
activity. For arsenic, long-term
bioassays for carcinogenic activity in
rats, mice, dogs, and monkeys have been
uniformly negative (Furst, 1983). The
kinds of genetic alterations seen in both
in vivo and in vitro studies of arsenic
effects are at the level  of loss and
rearrangement of chromosomes; these
are results of errors of "cellular
housekeeping" either in DNA repair or
in chromosome replication. The NRG
and EPA expert panel  (US EPA, 1997d)
reports examined several lines of
evidence for various modes of action
that might be operative. These included
changes in DNA methylation patterns
that could change gene expression and
repair, oxidative stress, potentiation of
effects of mutations caused by other
agents, cell proliferative effects, and
interference with normal DNA repair
processes. Further examination in both
of these reports of dose-response shapes
associated with these effects led to the
conclusion that they involve processes
that have either thresholds of dose at
which there would be no response or
sublinearity of the dose response
relationship (response decreasing   ;
disproportionately as dose decreases).
  The NRG report concluded: "For :
arsenic carcinogenicity, the mode  of
action has not been established, but the
several modes of action that are     ;
considered plausible (namely, indirect
mechanisms of mutagenicity)  would
lead to a sublinear dose-response curve
at some point below the point at which
a significant increase in tumors is
observed. *  *  * However, because a
specific mode (or modes) of action has
not yet been identified, it is prudent not
to rule out the possibility of a linear
response."
  The NRG report noted that in certain
in vitro studies of human and  animal
cells,  genotoxic effects have been shown
to occur at submicromolar
concentrations of arsenite that are  :
similar to concentrations found in urine
of humans ingesting water at the current
MCL. This emphasizes the potentially
low margin of exposure (health effects
observed at concentrations eight times
above the MCL) for arsenic in water,at
the current MCL.
  For noncancer effects, inhibition of
cellular respiration in mitochondria1 by
arsenic may be the focal point of its;
toxicity. In addition, inorganic arsenic
causes oxidative stress that  could play
a role in the development of adverse
health effects. The NRG report (19991)
recommended that biomarkers of   :
inorganic arsenic exposure and cancer
appearance be thoroughly studied. Such
data might better characterize the dose-
response effects of inorganic arsenic at
lower exposure levels. For noncancer
effects, a greater understanding of
arsenic's effects on cellular respiration
and subsequent effects of methylation
and oxidative stress are needed (NRG,
1999).                           '
  NRC recommended several  mode of
action studies, using biomarkers, to help
predict the shape of the dose-response
curve for cancer and non-cancer
endpoints. NRC concluded that "   '
* * *Additional epidemiological   ,
evaluations are needed to characterise
the dose-response relationship for  ',
arsenic-associated cancer and non-
cancer endpoints, especially at low
doses."                      •-.-.,..
7. Risk Considerations
  The NRC study used the results  of
epidemiological, (i.e., human) studies;
research on the mode of action, and
information about factors affecting
sensitivity to arsenic to project to risks
to the U.S. population. The numerical
estimation of risk in the NRC report has
several features to consider. The range
of drinking water levels associated with
health endpoints in the available studies
is generally hundreds of ppb which is,
however, within a factor of 10 of the
existing standard of 50 ppb. Because of
uncertainty about the shape of the dose-
response relationship below this range
of observed responses, the NRC report
used the approach of the 1996 EPA
proposed carcinogen risk assessment
guidelines (US EPA,  1996b). For the
male bladder cancer deaths which were
emphasized in the report, NRC used a
lower limit on the dose associated with
a 1% (1 in 100) cancer response, and the
LEDoi is estimated to be -400 ppb. This
is a point of departure for extrapolating
to exposure levels outside the range of
observed data based on inference.
Consistent with the proposed revisions
to the Guidelines for  Cancer Risk
Assessment, the report shows both a
linear extrapolation and a margin of
exposure extrapolation (difference
between the point of departure and
selected exposure). Because current data
on potential modes of action are
supportive of sub-linear extrapolations,
the linear approach could overestimate
risk at low doses. However, EPA
believes that within the several-fold
range (lOx) just below the point of
departure, this should make little
difference. EPA's scientists note that it
makes an increasing difference as dose
decreases, and the difference results in
an overestimate of risk at lower
exposures. With a straight-line
extrapolation from the point of
departure, the report  estimated risk to
be 1.0 to 1.5 x 10-3 at the current MCL
of 50 ppb and the margin of exposure
to be less than 8.
  As described further in section X.A.,
EPA used parts of NRC's risk analysis
and applied U.S. water consumption,
weights, and estimate of population
exposed to arsenic to model the U.S.
population risk. In selecting the
proposed MCL, EPA considered the
uncertainties of the quantitative dose-
response assessment  for inorganic
arsenic's health effects, particularly the
possible nonlinearity of the dose-
response and multiple cancer risks.
Given the current outstanding questions
about human risk at low levels of
exposure,  decisions about safe levels are
public health policy judgments.
8. Risk Characterization
  In 1983 the National Academy of
Sciences (NAS, 1983) defined risk

-------
 38902
Federal Register/Vol. 65,  No. 121/Thursday, June 22, 2000/Proposed Rules
assessment as containing four steps:
hazard identification, dose-response
assessment, exposure assessment, and
risk characterization. Risk
characterization is the process of
estimating the health effects based on
evaluating the available research,
extrapolating to estimate health effects
at exposure levels, and characterizing
uncertainties. In risk management,
regulatory agencies such as EPA
evaluate alternatives and select the
regulatory action. Risk management
considers "political, social, economic,
and engineering information" using
value judgments to consider "the
acceptability of risk and the
reasonableness of the costs of control
(MAS, 1983)."
   Unlike most chemicals, there is a
large data base on the effects of arsenic
on humans. Inorganic arsenic is a
human poison, and oral or inhalation
exposure to the chemical can induce
many adverse health conditions in
humans. Specifically oral exposure to
inorganic arsenic in drinking water has
been reported to cause many different
human illnesses, including cancer and
noncancer effects, as described in
Section HI. The NRG panel (1999)
reviewed the inorganic arsenic health
effects data base. The panel members
concluded that the studies from Taiwan
provided the current best available data
for the risk assessment of inorganic
arsenic-induced cancer. (There are
corroborating studies from Argentina
and Chile.) They obtained more detailed
Taiwanese internal cancer data and
modeled the data using the multistage
Weibull model and a Poisson regression
model. Three Poisson data analyses
showed a 1% response level of male
bladder cancer at approximately 400 ug
of inorganic arsenic/L. The 1% level
was used as a Point of Departure (POD)
for extrapolating to exposure levels
outside the range of observed data.
  For an agent mat is either acting by
reacting directly with DNA or whose
mode of action has not been sufficiently
characterized, EPA's public health
policy is to assume that dose and
response will be proportionate as dose
decreases (linearity of the extrapolated
dose-response curve). This is a science
policy approach to provide a public
health conservative assessment of risk.
The dose-response relationship is
extrapolated oy taking a straight line
from the POD rather than by attempting
to extend the model used for the
observed range. This approach was
adopted by the NRG report which
additionally noted that using this
approach for arsenic data provides
results with alternative models that are
consistent at doses below the observed
                     range whereas extending the alternative
                     models below the observed range gives
                     inconsistent results. Drawing a straight
                     line from the POD to zero gives a risk
                     of 1 to 1.5 per 1,000 at the current MCL
                     of 50 ug/L. Since some studies show
                     that lung cancer deaths may be 2- to 5-
                     fold higher than bladder cancer deaths,
                     the combined cancer risk could be even
                     greater. The NRG panel also rioted that
                     the MCL of 50 ug/L is less than 10-fold
                     lower than the 1% response level for
                     male bladder cancer. Based on its
                     review, the consensus opinion of the
                     NRG panel was that the current MCL of
                     50 ug/L does not meet the  EPA's goal of
                     public-health protection. Their report
                     recommended that EPA lower the MCL
                     as soon as possible.
                     IV. Setting the MCLG

                     A. How Did EPA Approach It?
                       For the decisions in this regulation,
                     the EPA has relied upon the NRG report
                     as presenting the best available, peer
                     reviewed science as of its completion
                     and has augmented it with more
                     recently published, peer reviewed
                     information. EPA used the 1999 NRG
                     report and  other published scientific
                     papers to characterize the potential
                     health hazards of ingested inorganic
                     arsenic. As NRG (1999) noted, DMA
                     may enhance the carcinogenicity of
                     other chemicals, but more data are
                     needed. Based on current knowledge,
                     the organic forms of arsenic in fish and
                     shellfish do not appear to present a
                     significant  risk to humans. The overall
                     weight of evidence indicates that the
                     inorganic arsenate and arsenite forms
                     found in drinking water are responsible
                     for the adverse health effects of ingested
                     arsenic. EPA focused its risk assessment
                     on the carcinogenic effects of inorganic
                     arsenic (the forms found in drinking
                     water sources).
                       A factor that could modify the degree
                     of individual response to inorganic
                     arsenic is its metabolism. There is
                     ample evidence (NRG, 1999) that the
                     quantitative patterns of inorganic
                     arsenic methylation vary considerably
                     and that the extent of this variation is
                     unknown. It is certainly possible that
                     the metabolic patterns of people affect
                     their response to inorganic arsenic.
                       There are studies underway in
                     humans and experimental  animals
                     under the EPA research plan and other
                     sponsorships. Over the next several
                     years these will provide better
                     understanding of the mode(s) of
                     carcinogenic action of arsenic,
                     metabolic processes that are important
                     to its toxicity, human variability in
                     metabolic processes, and the specific
                     contributions of various food and other
sources to arsenic exposure in the U.S.
These are important issues in projecting
risk from the observed data range in the
epidemiologic studies to lower
environmental exposures experienced
from U.S. drinking water.
  Until further research is completed,
questions will remain regarding the
dose-response relationship at low
environmental levels. The several
Taiwan studies have strengths in their
long-term observation of exposed
persons and coverage of very large
populations (>40,000 persons).
Additionally, the collection of
pathology data was unusually thorough.
Moreover, the populations were quite
homogeneous in terms of lifestyle.
Limitations in exposure information
exist that are not unusual in such
studies. In ecological epidemiology
studies of this kind, the exposure of
individuals is difficult to measure
because their exposure from water and
food is not known. This results in
uncertainties in defining a dose-
response relationship. The studies in
Chile and Argentina are more limited in
extent, (e.g., years of coverage, number
of persons, or number of arsenic
exposure categories analyzed), but
provide important findings which
corroborate one another and those of the
Taiwan studies.
  These epidemiological studies
provide the basis for assessing potential
risk from lower concentrations of
inorganic arsenic in drinking water,
without having to adjust for cross-
species toxicity interpretation.
Ordinarily, the characteristics of human
carcinogens can be explored and
experimentally defined in test animals.
Dose-response can be measured, and
animal studies may identify internal
transport, metabolism, elimination, and
subcellular events that explain the
carcinogenic process. Arsenic presents
unique problems for quantitative  risk
assessment because there is no test
animal species in which to study its
carcinogenicity. While such studies
have been undertaken, it appears  that
test animals, unlike humans, do not
respond to inorganic arsenic exposure
by developing cancer. Their metabolism
of inorganic arsenic is also
quantitatively different than humans.
Inorganic arsenic does not react directly
with DNA. If it did, it would be
expected to cause similar effects across
species and to cause response in a
proportionate relationship to dose.
Moreover, its metabolism, internal
disposition, and excretion are different
and vary across animal and plant
species and humans—in test studies and
in nature.

-------
                 Federal Register/Vol. 65, No.  121/Thursday,  June 22, 2000/Proposed  Rules
                                                                     38903
  Until more is known, EPA will take a
traditional, public health conservative
approach to considering the potential
risks of drinking water containing
inorganic arsenic. EPA recognizes that
the traditional approach may
overestimate risk, as explained in the
next section.

B. What Is the MCLG?
  EPA concludes that exposure to
inorganic arsenic induces cancer in
humans. It also is associated with
adverse noncancer effects such as
hypertension (NRG, 1999). The NRG
report stated that "Data on the modes of
action for carcinogenicity can help to
predict the shape of cancer dose-
response curves below the level of
direct observation of tumors. *  * * For
arsenic carcinogenicity * * * modes of
action that are considered most
plausible (namely, indirect mechanisms
of mutagenicity) lead to a sublinear
dose-response at some point below the
level at which a significant increase in
tumors is observed. However, because a
specific mode (or modes) of action has
not been identified at this time, it is
prudent not to rule out the possibility of
a linear response (NRG 1999, pgs. 213-
214)." The expert panel report fUS EPA,
1997d, pg. 31) stated: "* *  * for each of
the modes of action regarded as
plausible, the dose-response would
either show a threshold or would be
nonlinear. * *  * [HJowever, "the dose
response for arsenic at low doses would
likely be truly nonlinear—i.e., with a
decreasing slope as the dose decreased.
However, at very low doses such a curve
might effectively be linear but with a
very shallow slope, probably
indistinguishable from a threshold." In
the absence of a known mode of
action(s), EPA has no basis for
determining the shape of a sublinear
dose-response curve for inorganic
arsenic. As a result, consistent with EPA
public health policy, EPA will continue
to use a linear dose-response curve for
inorganic arsenic effects. Using a linear
type of a dose-response curve, EPA is
proposing an MCLG of zero. The Agency
welcomes comments on setting a
nonzero MCLG and submission of data
supporting a nonzero MCLG.
C. How Will a Health Advisory Protect
Potentially Sensitive Subpopulations?
  The NRG report was inconclusive
about the health risks to pregnant
woman, developing fetus, infants,
lactating women, and children. When
the Agency completes this mlemaking,
it intends to issue a health advisory on
arsenic in drinking water, in order to
decrease  risk to sensitive
subpopulations prior to the
implementation of the new MCL. The
effective date of a revised MCL will ,be
three to five years after the final rule is
issued (2004-2006).               ;
  A health advisory is a non-regulatory
document that supports water providers
in their independent decisions on
actions to take regarding water
contaminants and their communication
with the general public. In the health
advisory on arsenic the Agency intepds
to address a precautionary step to
protect infants. This step would be ^o
avoid using water containing high levels
of arsenic to make up infant formula.
The reason for this precaution is that
epidemiologic studies indicate that
arsenic in drinking water (Lewis et al.,
1999) affects the cardiovascular system.
While there are no studies of effects of
arsenic on human infants, both the '
cardiovascular system and brain (and its
vascular system) continue to develop
after birth (Thompson, P.M et al. 2000);
thus, the effects discussed in this notice
on the cardiovascular system raise a
concern about potential  effects of  '
arsenic on infant development. In large
part, causes of cerebrovascular incidents
(stroke) in children are not understood
except for certain, known associations
with organic diseases and genetic  ;
diseases. Congenital and acquired heart
disease are the most common cause of
stroke in children. The current toxicity
data on arsenic do not contradict this
precautionary view.

D. How Will the Clean Water Act
Criterion Be Affected by This
Regulation?
  EPA is also working to harmonize, the
human health arsenic criteria for the
Clean Water Act (CWA) and the SDWA.
The major reason for the present
difference (discussed in  section II.Di)
between the MCL and the Ambient ,
Water Quality Criterion (AWQC) was
the result of using separate bases for
determining the two standards. The:
AWQC for arsenic was derived from the
risk .assessment for arsenic-induced ;skin
cancer, while the current SDWA MGL,
adopted in  1975 as a National Interim
Primary Drinking Water Regulation,1
evolved from the U.S. Public Health.
Service standard dating back to the
1940s. The Agency will use the
conclusions of the NRG (1999) report to
form the human health basis for both
the AWQC  and the MCL. However, the
CWA and SDWA statutes require thit
the Agency consider different factor?
during the derivation of a standard. For
example, SDWA requires that the
Agency consider: (1) Cost/benefit
analyses, including sizes of the public
water systems, (2) the level of arsenic
that can be  analyzed by laboratories 'on
a routine basis, [i.e., the practical
quantitation limit (PQL)] and (3)
treatment techniques for removing the
chemical from the water. On the other
hand, the CWA requires the EPA to
consider water and fish consumption
(including amount of fish eaten, percent
lipid in the fish and the
bioaccumulation factor for the chemical
in the fish), but not cost/benefits,
analytical or treatment techniques.
Accordingly, developing a AWQC under
the CWA may produce a standard that
differs from the MCL derived under the
SDWA even though both standards are
based on the same health endpoint. The
Agency will begin work on a new
AWQC for arsenic after promulgating
the MCL for arsenic.

V. EPA's Estimates of Arsenic
Occurrence
  One of the key components in the
development of the proposed arsenic
rule is the analysis of arsenic occurrence
in public water supplies, both
community water systems  (CWS) and
non-transient, non-community, water
systems (NTNCWS). EPA's national
occurrence assessment of arsenic
provides a basis for estimating:
  (1) The number of systems expected
to exceed various arsenic levels;
  (2) the number of people exposed to
the different levels of arsenic; and
  (3) the variability in arsenic levels in
water systems among the wells and/or
entry points to the distribution system.
EPA uses the estimate of the total
number of systems and populations
affected in the United States in its cost-
benefit analysis. EPA is seeking
comment on its analysis of arsenic
occurrence in the U.S., as well as
requesting additional data.

A. What Data Did EPA Evaluate?
  For previous occurrence analyses EPA
used four older national arsenic
databases: (1) The National Inorganic
and Radionuclide Survey (MRS),
conducted from 1984 to 1986, for
ground water CWSs; (2) a 1976-1977
National Organic Monitoring Survey
(NOMS); (3) a 1978-1980 Rural Water
Survey (RWS); and (4) the 1978
Community Water System Survey
(CWSS) for surface water CWSs.
However, these older databases have
several limitations. First, the  surveys of
surface water systems will not reflect
changes in raw water sources which
occurred in the last twenty years.
Second, filtration treatment added to
comply with the Surface Water
Treatment Rule (110 54 FR 27486, June
29, 1989) would tend to decrease
arsenic exposure, through incidental
arsenic removal. Finally, most of the

-------
38904
Federal Register/Vol. 65, No. 121/Thursday, June 22, 2000/Proposed Rules
data were censored (reported as less
than the analytical test method
detection level or reporting limit, e.g.,
"not detected" or "<5 ng/L"). MRS,
CWSS, and RVVS, respectively, had
93%, 97%, and 90% censored data. This
limits the estimation of low level
occurrence of arsenic and makes it
statistically difficult to extrapolate
occurrence with the limited amount of
non-censored data. The EPA Science
Advisory Board recommended that EPA
abandon the older data when sufficient
new data become available because of
                      the high percentage of censored data in
                      the older surveys and the difficulty of
                      using highly censored data sets to
                      estimate occurrence (US EPA, 1995).
                      Therefore, with improved analytical
                      techniques for detecting arsenic at lower
                      levels, as low as 0.5 u.g/L, and the lower
                      reporting limits in the new data
                      received by EPA, the Agency focused
                      the data evaluation on post-1980 data
                      sources for estimating national
                      occurrence.
                        Since 1992, EPA OGWDW has
                      received arsenic databases from other
                      EPA offices, States, public water
utilities, and associations. EPA
combined the compliance monitoring
data obtained from States into the "25
States" database. The Agency evaluated
the databases listed in Table V—1. (Note
that EPA's database, the Safe Drinking
Water Information System (SDWIS),
only records violations of the current
arsenic MCL, so it is censored at 50 jig/
L.) A more detailed description of the
databases and evaluations are presented
in the EPA document titled "Arsenic
Occurrence  in Public Drinking Water
Supplies," (US  EPA, 2000b).
                                 TABLE V-1.—SUMMARY OF ARSENIC DATA SOURCES
Data source
25 States1 	
Metro2 	 	 	
NAOS3 	 	 	
USGS* 	 	 	
ACWA8 	 	
WESTCAS8 	
Reporting level
(MJ/L)
<1 to 10 	
1 	
0.5 	
1 	
0.1 to 1 	
not available 	
Number of CWSs
>1 9,000
140
<517
not available
(20,000 sites).
180 (1 500 sam-
ples).
not available 	
Source water
Surface Ground
Surface Ground
Surface Ground
Ground
Surface Ground
Ground 	
Water type
finished
raw & finished
raw & predicted finished
raw
finished
finished. '
  1 Arsenic compliance monitoring data from community water systems (CWSs) from Alabama, Alaska, Arizona, Arkansas, California, Illinois, In-
diana, Kentucky, Kansas, Maine, Michigan, Minnesota, Missouri, Montana, Nevada, New Hampshire, New Jersey, New Mexico, North Carolina,
North Dakota, Ohio, Oklahoma, Oregon, Texas, and Utah.
  2 Metropolitan Water District of Southern California (MWDSC, or Metro) 1992-1993 national survey of 140 CWSs serving more than 10,000
people.
  31996 National Arsenic Occurrence Survey (NAOS) funded by the Water Industry Technical Action Fund (WITAF), which includes the following
organizations: American  Water Works Association, National Association of Water Companies, Association of Metropolitan Water Agencies, Na-
tional Rural Water Association, and National Water Resources Association.
  * U.S. Geological Survey (USGS) ambient (raw water) ground water from approximately 20,000 wells throughout the U.S. used for various pur-
poses, Including public supply, research, agriculture, industry and domestic supply.
  61993 survey from  180 water agencies, utilities, and cities in southern  California, conducted by the Association of California Water Agencies
(ACWA).
  "1997 Western Coalition of Arid States (WESTCAS) Research Committee Arsenic Occurrence Study which aggregated arsenic data (e.g., me-
dian arsenic value for  county, city, or provider) from Arizona, New Mexico,  and Nevada.
0, What Databases Did EPA Use?
  EPA evaluated the databases for
representativeness, accuracy and
coverage of community water systems in
the U.S. EPA determined that the
compliance monitoring data from the 25
States ("25-States database") would
establish the most accurate and
scientifically defensible national
occurrence and exposure distributions
of arsenic in public ground water and
surface water supplies. Figure V.l
shows the coverage of these States in the
U.S. The 25-States database provides
more finished water arsenic data, from
over 19,000 ground and surface water
CWSs, than the other national
databases. EPA is interested in finished
water data, rather than raw water data,
because it indicates the current arsenic
levels in water systems after treatment
and reflects their customers' level of
exposure to arsenic. The 25-States
database provides system and
individual arsenic data for a significant
number of CWSs in each State. The
arsenic data can be linked directly to
                      specific water systems by their
                      identification code to obtain additional
                      information in SDWIS, such as
                      population served, system type (e.g.,
                      CWS, NTNCWS), source type (e.g.,
                      ground water, surface water, purchased
                      water, ground water under the
                      influence), and location. For this reason,
                      EPA chose to use the compliance
                      monitoring data from the States of
                      California, Nevada, New Mexico, and
                      Arizona, rather than the data about
                      these States from ACWA and
                      WESTCAS.
                        Most of the 25-States data had
                      reporting limits of less than 2 u.g/L. In
                      addition, the database includes multiple
                      samples from the water systems over
                      time and from multiple sources within
                      the systems. The multiple samples
                      provide for a more accurate estimate of
                      the arsenic levels in the systems, than
                      a survey with one sample per system.
                      The arsenic compliance monitoring data
                      provides point-of-entry or well data
                      within systems from  eight States, which
                      is used for intrasystem variability
analysis (discussed in Section V.G).
Intrasystem variability analysis provides
an understanding of the variation of
arsenic levels within a system, from
well to well or entry point to entry
point.
  EPA also received arsenic data from
Florida, Idaho, Iowa, Louisiana,
Pennsylvania, and South Dakota;
however EPA did not include these
States in the database. These States
either provided data that (1) could  not
be linked to CWSs; (2) did not indicate
if the results were censored or non-
censored; (3)  were all zero, without
providing the analytical/reporting limit;
or (4) rounded results to the nearest ten
ug/L.
  EPA used the USGS and NAOS
databases and their occurrence      !
estimates for  comparison purposes. In
addition, EPA used the NAOS approach
to partitioning of the U.S. for its
analysis.
  We combined State data sets with
different data naming conventions, and
the database development and data

-------
                 Federal  Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed Rules
                                                                    38905
conditioning process is described in
Appendix D—3 of the occurrence
support document (US EPA, 2000b).
Appendix D-l identifies who provided
the data and data provided for each
State in the 25-State database. Appendix
D—2 lists the data names we used to
develop the national database. We
assumed that the data represented
compliance sampling, and some States
have reportedly provided source water
data and compliance data. If you are
aware of errors in our data set, ple'ase let
us know. Also, additional data would
reduce the uncertainty of our national
occurrence estimate. We encourage
commenters to submit arsenic
compliance monitoring data sets either
from States not already in our data set,
more recent data that were not included
in the described data sets, or a  more
official version of compliance data. We
will use this information to obtain a
more representative national occurrence
estimate for the final rule.
BILLING CODE 6560-5O-P

-------
38906
Federal Register/Vol. 65, No. 121/Thursday, June  22, 2000/Proposed Rules
     c
     O
     DC

     N-

     (D
     S
     (0

     Q



     8


     (C
     • MB
     MHMM

     Q.



     O

     o


     .9
     (0
     0)

     ts
     I
     U)
BILLING CODE S560-50-C

-------
                 Federal Register/Vol.  65,  No. 121/Thursday,  June 22, 2000/Proposed Rules
                                                                     38907
C. How Did EPA Estimate National
Occurrence of Arsenic in Drinking
Water?
  EPA derived the national estimates of
arsenic occurrence in three steps: (1)
Estimate system means; (2) estimate
State distribution of system means; and
(3) estimate national distributions of
system means.
  As discussed in section V.B, EPA
determined that the 25-States database
would be used for estimating national
occurrence. EPA calculated a system
average for each water system in its
database. When the database provided 5
or more detected (greater than the
reporting limit) arsenic samples in a
system, we used the method  of
"regression on order statistics" (Helsel
and Conn, 1988) to extrapolate values
for the non-detected observations, then
calculated the arithmetic mean. When
there were 1 to  4 detected  values, we
substituted half the reporting limit for
each non-detected value (less than the
reporting limit) and calculated an
arithmetic average. When there were no
detected values (all samples  had  non-
detected values), we set the arsenic
system average as a non-detect at the
mode (most frequently occurring) of the
reporting limits. As a result,  each
system has a calculated system mean,
either a non-detected or detected value.
  In order to estimate the distribution of
systems means in a State, EPA
aggregated the system means into a
single distribution and derived separate
estimates of percentage of systems with
average arsenic values greater than 2,3,
5, 10,15, 20, 25, 30, 40, and  50 Ug/L
(referred to as exceedance estimates).
We  developed separate estimates for
ground water and surface water systems.
Within each State, EPA fit a  lognormal
distribution to the population of
estimated system means, and used the
fitted distribution to estimate
exceedance probabilities. However,
when fitting the lognormal distribution,
EPA excluded system means which
were estimated to be less than their
reporting limit, since these require more
extrapolation below the reporting limit
and were judged to be less reliable. EPA
also did not make exceedance estimates
below the most frequently occurring
reporting limit or censoring  point in
each of the States.
   To estimate the national distribution
of system means, EPA grouped the
States into the seven regions developed
in the NAOS (Frey and Edwards, 1997).
Frey and Edwards derived a natural
occurrence factor by weighting   !
detection, number of data points, and
higher arsenic values from data in the
USGS WATSTORE water quality
database and the Metro survey. Then
they grouped States into seven regions
based on the calculated natural   ;
occurrence factors. Figure V.I is a map
of the U.S. with the NAOS regions. With
this regional grouping of States, EPA
developed separate regional estimates
for surface water and ground water
systems. In a separate analysis, EPA
found the national result from using the
NAOS regions to be similar to grouping
States into different regions, based on a
preliminary examination of generally
related exceedance probabilities.:
  EPA derived each regional estimate by
using exceedance estimates from ithe
States with compliance monitoring data
in the region, weighted by the number
of community water systems in those
specific States. For example, we Used
the exceedance estimates from Montana
and North Dakota, weighted by the
number of community water systems in
those States, to  derive the North Central
region estimate. Within each region, we
estimated the percentages of systems
with average arsenic values greater than
2, 3, 5, 10, 15, 20, 25, 30, 40, and 50  Ug/
L. We then weighted the regional
exceedance estimates, by the total
number of community water systems in
each region (including the number of
community water systems in the States
without compliance monitoring data) to
obtain national estimates of percentages
of systems with average  arsenic values
greater than 2, 3, 5, 10, 15, 20, 25, 30,
40, and 50 Ug/L.                '.
  EPA believes that separate estimates
are not justified for different system
sizes. A graphical analysis ("box and
whisker" plots) of the occurrence
distributions suggests that in some
regions, systems in different size
categories do have different mean
concentrations. However the differences
in means are much smaller than the
variability of the observed       •
concentrations. Moreover, the
differences do not vary with system size
in a consistent way. For example > for
ground water systems, arsenic
concentrations in the New England
Region (NAOS Region 1) decrease as
system size increases, while in the Mid-
Atlantic and South Central regions
(NAOS Regions 2 and 5), arsenic
concentrations increase as system size
increases. In the four remaining regions,
no systematic patterns are evident. For
these reasons, and because additional
stratification decreases the precision of
the estimates, EPA has not developed
separate estimates for different system
sizes.
  The method of substitution that EPA
used for non-detected concentrations
(described above) is different from the
method that water systems use for
determining compliance with the MCL:
We substituted positive values for non-
detects, while for purposes of
compliance, non-detected
concentrations are treated as zero.
Therefore, our estimates of occurrence
will be higher on average than those
found by water systems monitoring for
compliance with the MCL. As a result
we might overestimate both the costs
and benefits of the proposed MCL.
However we believe that our estimate of
occurrence is justified, for two reasons.
First, it is more accurate (less biased).
Second, as the detection limits of
analytical methods continue to improve
(i.e., lower than 1 |ig/L), the difference
between the two substitution methods
will be small and will occur in the range
below the MCL.
D. What Are the National Occurrence
Estimates of Arsenic in Drinking Water
for Community Water Systems?
  Arsenic is found in both ground water
and surface water sources. Figure V.I
presents the regions of the United States
referred to in this discussion. Table V—
2 data indicate that higher levels of
arsenic tend to be found in ground
water sources (e.g., aquifers) than in
surface water sources (e.g., lakes, rivers).
The 25-States finished water data also
indicate that the North Central, Midwest
Central, and New England regions of the
United States tend to have low to
moderate (2-10 (ig/L) ground water
arsenic levels, while the Western region
tends to have higher levels of ground
water arsenic (>10 (ig/L) than the other
regions. Systems in the other regions of
the U.S. may have high levels of arsenic
(hot spots), while many systems and
portions of the States in the listed
regions may not have any detected
arsenic in their drinking water.

-------
 38008
Federal  Register/Vol. 65,  No. 121/Thursday, June 22, 2000/Proposed Rules
                      TABLE V-2.—REGIONAL EXCEEDANCE PROBABILITY DISTRIBUTION ESTIMATES
Region
Percent of systems exceeding arsenic concentrations (ng/L) of:
2
3
5
10
15
20
25
30
40
50 ',
                                                Ground Water Systems
New England ,.„.. 	 	 	 	 	 	 	 .
Mid Atlantic 	 	 	 	 	 	
South East 	 	 	 	 	 	
Midwest ...„.., 	 ....... ........ 	 	 ..
South Central 	 	 	 	 ...„.,
North Central ,.,„....„„.,.. 	 	 	 	 .
West ..,.„.».„ 	 	 	

29

2
28
27
29
42

21

•)
21
19
21
31

21
*0 3
0 5
14
10
13
25

7
*1
0 2
Q
4
3
12

4
n ^
0 1
4
2
4
7


OH
n H7

1




n oft
n n^

n ft

/i




-i

•t


















                                                Surface Water Systems
Now England „, 	 	 	 	 ..
Mid Atlantic ...,„,........ 	 	 	
South East »„».„ 	 	 	
Midwest ...„...,„.....,.... 	 	 , .
South Central 	 	 	 	 	 .
North Central 	 	 	 	 	 	 	 	
West *........................ ...

11

0 8
4
g
20
19

*8

02
3
4
10
13

*9
*0 1
0 03
•j
•|
4
•7

1 0
*0 1
n nni
0 4
0 3
0 8
0

Oc
n 01

O O
01
0 2


n A.
n nm

0-t
C\ f\Q
n 1
H

n *}




n n^







n c\o

















O.o
   Estimates at these regions and levels are inconsistent, in that the estimated % exceedances at lower values are smaller than the estimates at
 higher values. This inconsistency occurs because fewer States were used to estimate % exceedances at lower levels EPA did not attempt to re-
 solve the inconsistency, but combined the regional distribution into a national distribution which is consistent
   The estimates of the number of CWSs
 expected to exceed different arsenic
 levels is based on the distribution of
 average arsenic concentrations in water
 systems. Using the data from the 25-
 States database, EPA estimates that
 5.4% of ground water CWSs and 0.7%
 of surface CWSs have average arsenic
 levels above 10 ug/L. Similarly, 12.1%
 and 2.9% of ground water CWSs and
                     surface water CWSs, respectively, have
                     average arsenic levels above 5 Ug/L.
                     Tables V—3 and V-4 provide estimates
                     by system size category. The percentage
                     of systems that have average arsenic
                     levels within a specific range does not
                     vary across the system size categories.
                     For example, 2.3% of ground water
                     systems in each of the five system size
                     categories have average arsenic levels in
 the range of >10 ug/L to 15 ug/L.
 Therefore, the arsenic exceedance
 estimates have the same distribution in
 any system size. These estimates of
 percent (or probability) of systems that
 have average arsenic levels within a
 specific range are multiplied by the
 number of systems in each size category
 to derive the number of systems in
 Table V-3 and V-4.
         TABLE V-3.—STATISTICAL ESTIMATES OF NUMBER OF GROUND WATER CWSs WITH AVERAGE ARSENIC
                                       CONCENTRATIONS IN SPECIFIED RANGES
System size (population served)
25 to 500 	 	
501 to 3,300 	 	 	
3,301 to 10,000 	
10,001 to 50,000 	
>50,000 	
Total 	
(% of systems) 	
Number of systems with average arsenic concentrations in specified ranges (|ig/L; 43,749 systems total)
<2.0
21,325
7,616
1,811
933
154
31,840
(72.8%)
>2.0 to
3.0
2,158
771
183
94
16
3,221
(7.4%)
>3.0 to
5.0
2,268
810
193
99
16
3,386
(7.7%)
>5.0 to
10.0
1,960
700
167
86
14
2,927
(6.7%)
>10.0 to
15.0
674
241
57
29
5
1,006
(2.3%)
>15.0to
20.0
314
112
27
14
2
468
(1.1%)
>20.0 to
30.0
287
103
24
13
2
429
(1.0%)
>30.0 to
50.0
188
67
16
8
1
280
(0.6%)
>50.0
129
46
11
6
1
192
(0.4%')
  Note: Totals may not add updue to rounding of the number of systems to the nearest whole number. Systems serving fewer than 25 people
are not included in this table. The estimates in this table do not take into account most treatment in place; in particular most of the systems in
the >50,0  column will have treated for arsenic in order to reduce their concentration below 50 ug/L. See text for more details
  In Tables V-3 and V-4, the estimated
numbers of systems with mean
concentrations above 50 ug/L do not
represent the number of systems which
are believed to be out of compliance
with the current MCL of 50 ug/L; nor do
they represent actual systems at all.
Rather, they are statistical
extrapolations above 50 ug/L, based
                     primarily on data below 50 Ug/L. Since
                     most data below 50 Ug/L comes from
                     systems which have not treated for
                     arsenic, the ">50.0" columns in Tables
                     V-3 and V-4 do not take into account
                     most treatment currently in place.
                     Therefore, the ">50.0" columns
                     represent the estimated number of
                     systems which would have mean
arsenic concentrations above 50 ug/L if
they had not treated for arsenic. By
comparison with Tables V-3 and V-4,
during the three-year period from
September 1994 through August 1997,  :
EPA recorded a total of 14 samples from
10 public water systems in which
arsenic concentrations exceeded 50 ug/
J-j.

-------
                 Federal Register/Vol. 65, No. 121/Thursday,  June 22, 2000/Proposed Rules
                                                                    38909
        TABLE V-4.—STATISTICAL ESTIMATES OF NUMBER OF SURFACE WATER CWSs WITH AVERAGE ARSENIC
                                      CONCENTRATIONS IN SPECIFIED RANGES
System size (population served)
25 to 500
501 to 3,300 	
3,301 to 10,000 .. . .
10,001 to 50,000
> 50,000 ...
Total 	
(% of systems)

Number of systems with average arsenic concentrations in specified ranges ((xg/L; 10,683 systems total)
<2.0
2,794
3,308
1,656
1,384
477
9,622
(90.1%)
>2.0 to
3.0
122
144
72
60
21
419
(3.9%)
>3.0 to
5.0
94
111
56
47
16
323
(3.0%)
>5.0 to
10.0
69
82
41
34
12
239
(2.2%)
>10.0to
15.0
11
'. 13
6
5
2
37
(0.4%)
>15.0to
20.0
4
5
3
2
1
15
(0.1%)
>20.0 to
30.0
4
4
2
2
1
13
(0.1%)
>30.0 to
50.0
2
3
1
1
0
8
(0.1%)
>50.0
2
2
1
1
0
7
(0.1%)
  Note: Totals may not add up due to rounding of the number of systems to the nearest Whole number. Systems serving fewer than 25 people
are not included in this table. The estimates in this table do not take into account most treatment in place; in particular most of the systems in
the ">50.0" column will have treated for arsenic in order to reduce their concentration below 50 ng/L. See text for more details.
E. How Do EPA's Estimates Compare
With Other Recent National Occurrence
Estimates?

  In addition to EPA's national
occurrence results presented in section
V.D., two additional studies recently
developed national occurrence
estimates for arsenic in drinking water:
the NAOS study (Frey and Edwards,
1997), and the USGS study of arsenic
occurrence in ground water (USGS,
2000). The databases that supported the
NAOS and USGS estimates are briefly
described in section V.A., "What data
did EPA evaluate?" Each of these
occurrence estimates was developed in
a slightly different manner. Whereas
EPA's occurrence estimates are based on
compliance monitoring data from more
than 19,000 CWSs in 25 states, the  ;
NAOS occurrence estimates are based
on a stratified random sampling from
representative groups defined by source
type, system size, and geographic
location. The NAOS database contains
435 predicted finished water arsenic
data points (derived from raw water
arsenic concentrations and treatment
information), from more than 400 CWSs.
The USGS analysis is based on arsenic
ambient (untreated, or raw water)  ;
ground water data, providing 17,496
samples for 1,528 counties (with 5 or
more data points) in the United States
(out of a total of 3,222 counties). USGS
derived exceedance estimates for each
county by calculating the percentage of
data points in each county exceeding
specific concentrations, from 1 (ig/L to
50 (ig/L. Then USGS associated the
percentages for each county with the
number of CWSs that use ground water
in these counties, which was based on
data derived from SDWIS. This
information was aggregated for all of the
appropriate counties to derive the
national estimates for ground water
CWSs. USGS did not have estimates for
surface water CWSs.
        TABLE V-5 — COMPARISON OF CWSs FROM EPA, NAOS, AND USGS ESTIMATES EXCEEDING ARSENIC
                                                CONCENTRATIONS
% CWS exceeding
2 ng/L 	
5|x.g/L 	
10 n/L 	

EPAGW
&SW
(percent)
24 1
10 3
4 5

NAOS
GW&
SW (per-
cent)
21 7
115
4 5

EPAGW
(percent)
27 2
12 1
5 4

USGS
GW (per-
cent)
25 0
1^ R
7 6

  Table V-5 compares the EPA, NAOS,
and USGS estimates of the percent of
samples exceeding various arsenic
concentrations. At a concentration of 2
u.g/L, the EPA national exceedance
estimate for both surface water and
ground water CWSs  (24.1 percent) is
higher than the NAOS estimate (21.7
percent). At 5 ug/L, the EPA and NAOS
predicted exceedance probabilities are
relatively similar (10.3 and 11.5 percent,
respectively). These two estimates are
the same at 10  Ug/L (4.5 percent). For
ground water CWSs, the USGS and EPA
estimates are also relatively similar. At
2 (ig/L, the EPA national ground water
exceedance estimate (27.2 percent) is
slightly higher than the USGS estimate
(25.0 percent).  At 5 and 10 ug/L, the
USGS exceedance estimates  (13.6
percent and 7.6 percent, respectively)
are slightly higher than the EPA
estimates (12.1 percent and 5.4 percent).
This comparison of exceedance ,
probabilities suggests that EPA's arsenic
occurrence projections based on
compliance monitoring data are
relatively close to the NAOS and USGS
projections through the range of this
comparison. In addition, the USGS :
estimates are expected to be slightly
higher than the EPA estimates for
ground water, because they are based on
raw water arsenic levels (untreated).
F. What Are the National Occurrence
Estimates of Arsenic in Drinking Water
for Non-Transient, Non-Community
Water Systems?

  The 25-States database contains data
for non-transient, non-community water
systems (NTNCWSs) in 15 States (two
additional States only provided data
from two systems). NTNCWSs are
public water systems that regularly
serve at least 25 of the same persons '
more than 6 months a year. Most
NTNCWSs serve less  than 3,300 people
(99.5%) and use ground water (96%).
  EPA calculated basic statistics for
ground water CWSs and NTNCWSs in
each of these States. EPA compared the
data and found that arsenic
distributions in NTNCWSs are quite

-------
 38910
Federal  Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed  Rules
 similar to arsenic distributions in CWSs.
 In general, the means, standard
 deviations, and level of censoring for
 CWSs in a particular State are very close
 to the levels observed in NTNCWSs in
 that State. In some States, mean levels
 are slightly higher in CWSs than in
 NTNCWSs, whereas in others, mean
 levels are slightly lower in CWSs. There
 is no clear pattern and the differences
 are relatively minor, suggesting that any
 differences are due to random variation,
 rather than systematic underlying
 differences between NTNCWSs and
 CWSs. As a result, the occurrence
                     distributions for CWSs were used to
                     derive the occurrence distributions for
                     NTNCWS systems. If the NTNCWSs
                     data from the 15 States were used to
                     derive the estimates, there would have
                     been less spatial coverage of United
                     States, which would have resulted in
                     more uncertainty in the estimate. The
                     NTNCWSs estimates are presented in
                     Tables V-6 and V-7.
                       As in the case of Tables V—3 and V—
                     4, the estimated numbers of systems in
                     Tables V-6 and V-7 with mean
                     concentrations above 50 ug/L do not
                     represent the number of systems which
 are believed to be out of compliance
 with the current MCL of 50 u,g/L; nor do
 they represent actual systems at all.
 Rather they represent the estimated
 number of systems which would have
 mean arsenic concentrations above 50
 ug/L if they had not treated for arsenic.
 By comparison with Tables V-6 and V—
 7, during the three-year period from
 September 1994 through August 1997,
 EPA recorded a total of 14 samples from
 10 public water systems in which
 arsenic concentrations exceeded 50 Ug/
 L.
      TABLE V-6.—STATISTICAL ESTIMATES OF NUMBER OF GROUND WATER NTNCWSs WITH AVERAGE ARSENIC
                                      CONCENTRATIONS IN SPECIFIED RANGES
System size (population served)
25 to 500 	 	 	
501 to 3,300 	
3,301 to 10,000 	
10 001 to 50 000 .
> 50,000 	
Total 	
(% of systems) 	

Number of systems with average arsenic concentrations in specified ranges (ug/L; 19,293 systems total)
<2.0
12,088
1,902
43
8
0
14,041
(72.8%)
>2.0 to
3.0
1,223
192
4
1
0
1,421
(7.4%)
>3.0 to
5.0
1,285
202
5
1
0
1,493
(7.7%)
>5.0 to
10.0
1,111
175
4
1
0
1,291
(6.7%(
>10.0to
15.0
382
60
1
0
0
444
(2.3%)
>15.0 to
20.0
178
28
1
0
0
206
(1.1%)
>20.0 to
30.0
163
26
1
0
0
189
(1.0%)
>30.0 to
50.0
106
17
0
0
0
123
(0.6%)
>50.0
73
11
0
0
0
85
(0.4%)
  Note: Totals may not add up due to rounding of the number of systems to the nearest whole number. Systems serving fewer than 25 people
are nol included in this table. The estimates in this table do not take into account most treatment in place; in particular most of the systems in
tha ">50.0" column will have treated for arsenic in order to reduce their concentration below 50 ug/L. See text for more details.

      TABLE V-7.—STATISTICAL ESTIMATES OF NUMBER OF SURFACE WATER NTNCWSs WITH AVERAGE ARSENIC
                                      CONCENTRATIONS IN SPECIFIED RANGES
System size (population served)
25 to 500 	
501 to 3,300 	 	
3,301 to 10,000 	
10,001 to 50,000 	
50,000 	 	
Total 	
(% of systems) 	

Number of systems with average arsenic concentrations in specified ranges ("ug/L; 764 systems total)
<2.0
502
163
18
4
2
688
(90.1%)
>2.0 to
3.0
22
7
1
0
0
30
(3.9%)
>3.0 to
5.0
17
5
1
0
0
23
(3.0%)
>5.0 to
10.0
12
4
0
0
0
17
(2.2%)
>10.0to
15.0
2
1
0
0
0
3
(0.4%)
>15.0to
20.0
1
0
0
0
0
1
(0.1%)
>20.0 to
30.0
1
0
0
0
0
1
(0.1%)
>30.0 to
50.0
0
0
0
0
0
1
(0.1%)
>50.0
; o
0
0
1 0
' 0
0
(0.1%)
  Note: Totals may not add up due to rounding of the number of systems to the nearest whole number. Systems serving fewer than 25 people
are not included in this table. The estimates in this table do not take into account most treatment in place; in particular most of the systems in
tha ">50.0" column will have treated for arsenic in order to reduce their concentration below 50 (ig/L. See text for more details.
G. How Do Arsenic Levels Vary From
Source To Source and Over Time?

  EPA analyzed the variability of
arsenic concentrations within a system,
from well to well or entry point to entry
point (sampling point). This analysis
allows EPA to estimate the number of
sampling points in a system that may be
above the proposed MCL and to
Improve estimation of the treatment
costs for systems with multiple
sampling points. The result of the
intrasystem analysis is a constant
coefficient of variation (CV), which is
one of the inputs to the cost-benefit
                     computer modeling. EPA analyzed six
                     of the eight States that provided
                     intrasystem data: California, Utah, New
                     Mexico, Oklahoma, Illinois and Indiana.
                     Arkansas and Alabama were not
                     analyzed because these States had very
                     little occurrence of arsenic and almost
                     all of the arsenic values were below the
                     detection limit. After statistical analysis
                     of 127 systems with five or more
                     sampling points, EPA derived an
                     arithmetic average CV of 0.64 or 64%.
                     The EPA document titled "Arsenic
                     Occurrence in Public Drinking Water
Supplies," presents this statistical
analysis (US EPA, 2000b).
  USGS examined its raw water arsenic
data to assess the variability of arsenic
levels over time and to determine
whether there are temporal trends
(USGS, 2000). Data came from about 350
wells with 10 or more arsenic analyses
collected over different time periods.
These wells were used for various
purposes, such as public supply,
research, agriculture, industry, and
domestic supply, and encompassed ;
non-potable and potable water quality.
USGS conducted a regression analysis

-------
                 Federal Register/Vol.  65, No. 121/Thursday,  June 22, 2000/Proposed Rules
                                                                     38911
of arsenic concentration and time for
each well and found that most of the
wells had little or no change in
concentration over time (low "r-
squared" values when arsenic
concentrations were regressed with
time). Arsenic levels for most of the
wells probably do not consistently
increase or decrease over time. In
addition, USGS found that well depth
had no relationship to temporal
variability. To determine the extent of
the temporal variability, EPA analyzed
the CVs for the mean arsenic level in the
wells. More than 100 wells had a CV
and standard deviation of zero. Most of
these wells consistently had arsenic
concentrations below the detection limit
of 1 ug/L.  EPA examined the CVs for the
other wells in relation to the mean
arsenic level and found a relatively
constant CV on the lognormal scale. The
geometric mean of the CVs, excluding
CVs of zero, is 0.39 or 39%. The report
(USGS, 2000) listed several factors that
may contribute to this  variability,
including natural variability in
geochemistry or source of
contamination, sampling technique, and
changes in pumping over time.
H. How Did EPA Evaluate Co-
Occurrence?

  Sections 1412(b)(3)(C)(i)(II), (III) and
(VI) of the SDWA, as amended in 1996,
require EPA to take into account
activities under preceding rules which
may have impacts on each new
successive rule. To fulfill this need EPA
began the analysis of the co-occurrence
of drinking water contaminants. The
information on co-occurrence will be
used to determine the level  of overlap
in regulatory requirements. For
example, this will include cases where
treatment technologies applied for one
regulation may resolve monitoring and/
or additional treatment needs for
another regulation or where water
utilities may incur costs for installing
multiple treatments to address other co-
occurring substances. This information
may also be used to show where specific
levels of one contaminant may interfere
with the treatment technology for
another.
1. Data
  For the co-occurrence analysis, EPA
relied on data from the National Water
Information System (NWIS), a U.S. ;
Geological Survey  (USGS) database.! The
NWIS database was used for several
reasons:
  • It contains both ground and surface
water data;
  • It is national in scope, representing
raw water samples from approximately
40,000 observation stations across th.e
U.S.; and                        ;
  • It provides latitude/longitude
coordinates for monitoring stations,:
which can be used in subsequent   j
analyses to associate with Public Water
Supply Systems.
  NWIS contains a water quality data
storage retrieval system developed by
the USGS Water Resources Division.
NWIS is a distributed water database;
data can be processed over a network of
computers at USGS offices throughout
the U.S. The system comprises the
Automated Data Processing System, the
Ground Water Site Inventory System,
the Water-Quality System, and the
Water-Use Data System. NWIS does:not
represent Public Water Supply Systems
directly but can be associated with them
because it provides latitude/longitude
coordinates for monitoring stations.
  Using the NWIS  data, arsenic was;
analyzed with 18 other constituents.
The other constituents included:
Sulfate, radon, radium, uranium, nitrate,
antimony, barium,  beryllium, cadmium,
chromium, cyanide, iron, manganese,
mercury, nickel, nitrite, selenium,
thallium, hardness, and total dissolved
solids. An additional set of ancillary
parameters were selected for use as ;
indicators of the hydrogeologic and
geochemical conditions that could
influence the co-occurrence of specific
constituents. These ancillary parameters
included: turbidity, conductance,
dissolved oxygen, pH, alkalinity, well
depth, and depth below land.

2. Results of the Co-occurrence Analysis
(US EPA, 1999f)

   Dissolved arsenic was observed to
have 5442 detected counts and total
arsenic was observed to have 1273
detected counts in the database at the
minimal threshold level of 2 ug/L. The
national co-occurrence estimates
derived from the USGS NWIS data
revealed several correlations between
arsenic/sulfate and arsenic/iron at the
threshold levels chosen by EPA as likely
to affect treatment  (see section VIII.).
First, a significant correlation was
observed between dissolved arsenic and
sulfate in surface water and ground
water samples at the national level. The
analysis  of the surface and ground water
data from EPA Regions 1, 2, 4, 5,  6,  7,
8, 9 and 10 show 339 co-occurrence
frequency counts of the data above the
threshold values of dissolved arsenic >5
ug/L and sulfate >250 mg/L (Table V-8).
For total arsenic and sulfate there are
143 co-occurrence  frequency counts for
the same threshold levels. There was no
significant co-occurrence of arsenic and
sulfate in EPA Region 3. Secondly, a
correlation was observed between
dissolved arsenic and iron and total
arsenic and iron in surface and ground
waters from EPA Regions 1, 2, 4, 5,  7,
8 and 9 (Table V-8). There are 562 co-
occurrence frequency counts of the  data
above the threshold levels of dissolved
arsenic >5 Ug/L and iron >300 u,g/L.
There are 542 co-occurrence frequency
counts of the data above the threshold
values of total arsenic >5 ug/L and iron
>300 ug/L. There was no significant co-
occurrence of arsenic and iron in EPA
Regions 3, 6 and 10.
          TABLE V-8.—CORRELATION OF ARSENIC WITH SULFATE AND IRON (SURFACE AND GROUND WATERS)
EPA regions
1,2,4,5,6,7,
8, 9, 10.
1, 2,4, 5,7, 8, 9


Arsenic types (threshold levels >5 |ig/L)
Dissolved Arsenic 	
Total Arsenic 	
Dissolved Arsenic 	
Total Arsenic 	

Correlation elements and their threshold level
Sulfate (>250 mg/L)
Sulfate (>250 mg/L)
Iron (>300 (J.g/L)
Iron (>300 ug/L)

Frequency
counts
339
143
562
542

  The results also show some co-
occurring pairs of arsenic with radon.
This appears to occur in EPA Regions 5
and 6 for ground water. However, the
co-occurrence of arsenic and radon at
levels of concern is not significant
(Table V-9). At present, the analysis.
does not show significant co-occurring
pairs between arsenic and radon in
surface water in any EPA region. The
impact from the co-occurrence of
arsenic and radon is not a concern oh
a national level because there was no
significant co-occurring pairs in EPA
Regions 1, 2, 3, 4, 7, 8, 9, and 10. EPA
requests comments on whether the
NWIS database and this analysis is
appropriate to use to represent co-

-------
38912
Federal  Register/Vol. 65, No.  121/Thursday,  June 22, 2000/Proposed Rules
occurrence of arsenic with other
constituents.
                      TABLE V-9.—CORRELATION OF ARSENIC WITH RADON (GROUND WATER)
EPA
regions
5 snd 6




Arsenic types and threshold levels (|ig/L)
Dissolved 2<5 	
Dissolved 5<10 	
Total 2S5 ... 	
Total 5S10 . . . . 	

Radon and
threshold
levels(pci/l)
100<300
300<1000
10CK300
30051000
OS100
100<300
OS100
100<300
Frequency
counts
58
140
; 124
101
2
' 2
1
' 1
VI. Analytical Methods

A. What Section ofSDWA Requires the
Agency To Specify Analytical Methods?

  Section 1401 of SDWA directs EPA to
promulgate national primary drinking
water regulations (NPDWRs) which
specify either MCLs or treatment
techniques for drinking water
contaminants (42 U.S.C. 300g-l). EPA is
required to set an MCL "if, in the
judgement of the Administrator, it is
economically and technologically
feasible to ascertain the level of a
contaminant in water in public water
systems" (SDWA section 1401(l)(C)(i)).
Alternatively, "if, in the judgement of
the Administrator, it is not
economically or technologically feasible
to so ascertain the level of such
contaminant," the Administrator may
identify known treatment techniques,
which sufficiently reduce the
contaminant in drinking water, in lieu
of an MCL (SDWA section
1401(l)(C)(ii)}. In addition, the NPDWRs
are required to include "criteria and
procedures to assure a supply of
drinking water which dependably
complies with such maximum
contaminant levels; including accepted
                     methods for quality control and testing
                     procedures to insure compliance with
                     such levels * * *" (SDWA section
                     1401(1)(D)J

                     B. What Factors Does the Agency
                     Consider in Approving Analytical
                     Methods?
                       In deciding whether an analytical
                     method is economically and
                     technologically feasible to  determine the
                     level of a contaminant in drinking
                     water, the Agency considers the
                     following factors:
                       • Is the method sensitive enough to
                     address the level of concern (i.e., the
                     MCL)?
                       • Does the method give reliable
                     analytical results at the MCL? What is
                     the precision (or reproducibility) and
                     the bias (accuracy or recovery}?
                       • Is the method specific? Does the
                     method identify the contaminant of
                     concern in the presence of potential
                     interferences?
                       • Is the availability of certified
                     laboratories, equipment and trained
                     personnel sufficient to conduct
                     compliance monitoring?
                       • Is the method rapid enough to
                     permit routine use in compliance
                     monitoring?
  • What is the cost of the analysis to
water supply systems?

C. What Analytical Methods and
Method Updates Are Currently
Approved for the Analysis of Arsenic in
Drinking Water?

  EPA approved analytical methods and
method updates for the analysis of ,
arsenic in drinking water in previous
rulemakings. EPA took the factors listed
in section VLB into consideration when
it approved these methods and updates.
The methods and updates, listed in
Table VI—1, are based on atomic
absorption, atomic emission and mass
spectroscopy methodologies and have
been used for compliance monitoring of
arsenic at the 0.05 mg/L MCL by State,
Federal and private laboratories for
many years. In this section on the
discussion of analytical methods, and in
the sections discussing the consumer
confidence rule and public notification,
EPA uses the mg/L units of measure, the
units used in the regulatory language.
Note that EPA's drinking water
analytical methods refer to mg/L instead
of jig/L, and milligrams are 1,000 times
larger than micrograms.
         TABLE VI-1 .—APPROVED ANALYTICAL METHODS (AND METHOD UPDATES) FOR ARSENIC (CFR 141.23)
Methodology
Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP— AES) 	
Inductively Coupled Plasma Mass Spectroscopy (ICP— MS) ICP— MS with Selective Ion Monitoring 	
Stabilized Temperature Platform Graphite Furnace Atomic Absorption (STP-GFAA) STP-GFAA with Mul-
tiple Depositions.
Graphite Furnace Atomic Absorption (GFAA) . . 	
Gaseous Hydride Atomic Absorption (GHAA) 	

Reference method '
200.7 (EPA)
3120B (SM)
200.8 (EPA)
200.9 (EPA)
3113B (SM)
D-2972-93C (ASTM)
3114B (SM)
D-2972-93B (ASTM)
MDL^ or
EDL!3
(mg/L)
6.008
39.050
0.0014
"(0.0001)
0.0005
5(0.0001)
3 6.001
3 0.005
3 0.0005
30.001
  "The reference methods approved for measuring arsenic in drinking water are cited in 40 CFR 141.23. The reference methods include:
  EPA » "Methods for the Determination of Metals in Environmental Samples—Supplement I", EPA/600/R-94-111, US EPA, May 1994. (US
EPA, 1994b)

-------
                  Federal Register/Vol. 65, No.  121/Thursday, June  22,  2000/Proposed Rules
                                                                      38913
   ^™* tandard Methods for the Examination of Water and Wastewater, 18th and 19th eds., Washington, D.C., 1992 and 1995 (APHA 1992
   AOT., respectively). The 19th edition of SM was approved in the December 1, 1999 final methods rule (64 FR 67450 US EPA 1999i)  '
   ASTM = Annual Book of ASTM Standards: Waster and Environmental Technology," Vol. 11.01 and 11 02  American' Society for Testina and
                      (AS™' 1"4 and 1996)" The 1"6 editi°n °f AS™ W3S aPProved in tne December 1, 1999 fina? methods rule (64 FR
 67450 US EPA
 «,2MD7.T Metnod Detection Limit = "the minimum concentration of a substance that can be measured and reported with 99% confidence that
 the analyte concentration is greater than zero." (40 CFR Part 136 Appendix B).     •      '.                             v-uimuenwj nidi
 r 3 F"; = Est.',mated De,t? ctio" Limit- (EDL) is defined as eitner the MDL or a concentration of a compound in a sample yielding a peak in the
 final extract with a signal to noise ratio of 5 whichever value is greater. Although the ASTM GFAA method (D-2972-93C) has a reported EDL of
 0 005 mg/L, this method is similar to other GFAA methods. EPA believes D-2972-93C is capable of detection limits similar to other GFAA meth-
 OQS.
   4 In 1994 (59 FR 62456; US EPA, 1994c), the Agency approved the use of the updated "Methods for the Determination of Metals in Environ-
 mental Samples-Supplement I," (US EPA, EPA/600/R-94/fl1, 1994). The revised manual allows the use of selective ion monSng w«h™CP-
 MS. The determined MDL for the direct analysis of arsenic in aqueous samples was 0.1 ng/L
   s In 1994 (59 FR 62456; US EPA, 1994c), the Agency approved the use of the updated "Methods for the Determination of Metals in Environ-
 mental Samples-Supplement I," (US EPA, EPA/600/R-94/1 1 1 , 1994). The revised manual allows the use of multiple depositions with STP-
 GFAA. The determined MDL for arsenic using multiple deposition with STP-GFAA is 0.1  ng/L.                        PU»IUUIK> wmi o
 D. Will Any of the Approved Methods
 for Arsenic Analysis Be Withdrawn?
   EPA believes all of the analytical
 methods listed in Table VI-1, with the
 exception of EPA Method 200.7 and SM
 3120B, are technically and economically
 feasible for compliance monitoring of
 arsenic in drinking water at the
 proposed MCL of 0.005 mg/L. EPA is
 proposing to withdraw approval for EPA
 Method 200.7 and SM 3120B because
 the detection limit for the first ICP-AES
 method, 0.008 mg/L, and the estimated
 detection limit for the second ICP-AES
 method, 0.050 mg/L, are inadequate to
 reliably  determine the presence of
 arsenic at the proposed MCL of 0.005
 mg/L. Analysis of the Water Supply
 (WS) studies used to derive the PQL
 (Analytical  Methods Support Document,
 US EPA, 19991) indicates that ICP-AES
 technology was rarely used for low level
 arsenic analysis. Therefore, the Agency
 believes the removal of the methods that
 use ICP-AES technologies will not have
 an impact on laboratory capacity.
   Even at the MCL options of 0.003,
 0.010 mg/L, and 0.020 mg/L, the Agency
 would still withdraw both EPA Method
 200.7 and SM  3120B. At these MCL
 options, these methods are still
 inadequate for compliance monitoring
 of arsenic in drinking water.
 E. Will EPA Propose Any New
 Analytical Methods for Arsenic
 Analysis?
   The Agency conducted a literature
 search to identify additional analytical
 methods which are capable of
 compliance  monitoring of arsenic at the
 proposed MCL of 0.005 mg/L
 (Analytical Methods Support Document,
 US EPA, 19991). A large majority of the
 analytical techniques identified from
 the literature search were from EPA's
 Office of Solid Waste SW-846 methods
 manual, which can be accessed online
 at www.epa.gov/epaoswer/hazwaste/
test/index.html. Of the Solid Waste
 methods, the Agency evaluated:
  • SW-846 Method 6020 (ICP-MS,
MDL = 0.0004 mg/L; US EPA, 1994d);
   • SW-846 Method 7060A (GFAA,
 MDL = 0.001 mg/L; US EPA, 1994e);
   • SW-846 Method 7062 (GFAA, MDL
 = 0.001 mg/L; US EPA, 1994f);
   • SW-846 Method 7063 (Anodic
 Stripping Voltammetry-ASV, MDL =
 0.0001 mg/L; US EPA, 1996d);
   In addition to the SW-846 method,
 the Agency also reviewed:          i
   • EPA Method 1632 (a wastewater
 GHAA method with an MDL = 0.000002
 mg/L or 0.002 ug/L; US EPA 1996a); and
   • EPA Method 200.15 (an ICP-AES
 with ultrasonic nebulization as part of
 the written method, MDL  = 0.003 mg/
 L or 0.002  mg/L; US EPA,  1994a).
   Although the SW-846 methods and
 the EPA 1632 wastewater  method are
 capable of reaching the detection limits
 needed at the proposed arsenic MCL,!
 most of these techniques (with the
 exception of the method using ASV
 technology) are similar to  methods that
 have already been approved for the
 analysis of arsenic in drinking water.
 The Agency does not believe approval
 of these methods for drinking water  '
 would provide additional  analytical
 benefits. Moreover, the addition of the
 SW-846 methods could complicate the
 laboratory  certification process because
 SW-846 methods are not mandatory
 procedures, but rather guidance.  At this
 time, laboratories are certified at
 different times for different EPA      ;
 programs. Therefore, laboratories    ;
 certified for both drinking  water
 methods and Office of Solid Waste
 methods may need to be certified
 separately under both programs to use
 SW-846 methods for drinking water.
   While SW-846 Method 7063 (using'
 ASV technology) is not similar to any'
 technique approved thus far, this
 method will not be approved for the
 measurement of arsenic in drinking
water because it only detects dissolved
arsenic as opposed to total arsenic.
Today's proposal would regulate  total;
arsenic in drinking water not dissolved
arsenic. The techniques currently
approved for drinking water measure
total arsenic (arsenic species in the
 dissolved and suspended fractions of a
 water sample). A preliminary total
 metals digestion would be necessary
 with the ASV technique in order to
 determine the total arsenic
 concentration in a drinking water
 sample.
   The Agency also reviewed but does
 not propose to approve EPA Method
 200.15, an ICP-AES method which
 requires the use of ultrasonic
 nebulization to introduce the sample
 into the plasma. To provide uniform
 signal response using EPA Method
 200.15, it is necessary for arsenic to be
 in the pentavalent state. The addition of
 hydrogen peroxide to the mixed acid
 solutions  of samples and standards prior
 to ultrasonic nebulization is necessary
 to convert all of the arsenic species to
 the pentavalent state. Although EPA
 Method 200.15 is capable achieving a
 MDL of 0.003 mg/L using direct analysis
 and a MDL of 0.002 mg/L using a total
 recoverable digestion and a 2-fold
 concentration, these levels of detection
 are still insufficient for compliance
 monitoring at the proposed MCL of
 0.005 mg/L.
   At the MCL options of 0.010 mg/L and
 0.020 mg/L, the Agency would approve
 the use of EPA Method 200.15 but only
 with the use of a total recoverable
 digestion and a 2-fold concentration
 (MDL = 0.002 mg/L). At an MCL option
 of 0.003 mg/L, EPA method 200.15
 would not be approved.

 F. Other Method-Related Items
 1. The Use of Ultrasonic Nebulization
 with ICP-MS
   In the September 3,1998 Analytical
Methods for Drinking Water
Contaminants Proposed Rule (63 FR
47907; US EPA 1998d), EPA proposed
the use of ultrasonic nebulization with
EPA Method 200.7 (ICP-AES) and EPA
Method 200.8 (ICP-MS). Because EPA
Method 200.7 and SM 3120B will be
withdraw for the analysis of arsenic in
drinking water under the proposed MCL
of 0.005 mg/L, ultrasonic nebulization
as a modification would not be allowed.

-------
38914
Federal  Register/Vol. 65, No.  121/Thursday,  June 22, 2000/Proposed Rules
Even with the modification of ultrasonic
nebulization, the ICP—AES method is
not capable of compliance monitoring
For arsenic at the proposed MCL of 0.005
mg/L. EPA Method 200.8 (ICP-MS)
would still be allowed for compliance
monitoring at the proposed MCL of
0.005 mg/L. The use of ultrasonic
nebulization can enhance transport
efficiency and lower the detection limits
for ICP-MS by approximately 5 to 10
fold. The final methods update rule was
published in the Federal Register on
December 1,1999 (64 FR 67450; US
EPA 1999J).
2. Performance-Based Measurement
System
   On October 6,1997, EPA published a
Notice of the Agency's intent to
implement a Performance Based
Measurement System (PBMS) in all of
its programs to the extent feasible (62
FR 52098; US EPA, 1997e). EPA is
currently determining how to adopt
PBMS into its drinking water program,
but has not yet made final decisions.
When PBMS is adopted into the
drinking water program, its intended
purpose will be to increase flexibility  in
laboratories in selecting suitable
                     analytical methods for compliance
                     monitoring, significantly reducing the
                     need for prior EPA approval of drinking
                     water analytical methods. Under PBMS,
                     EPA will modify the regulations that
                     require exclusive use of Agency-
                     approved methods for compliance
                     monitoring of regulated contaminants in
                     drinking water regulatory programs.
                     EPA will probably specify "performance
                     standards" for methods, which the
                     Agency would derive from the existing
                     approved methods and supporting
                     documentation. A laboratory would be
                     free to use any method or method
                     variant for compliance monitoring that
                     performed acceptably according to these
                     criteria. EPA is  currently evaluating
                     which relevant  performance
                     characteristics under PBMS should be
                     specified to ensure adequate data
                     quality for drinking water compliance
                     purposes. After PBMS is implemented,
                     EPA may continue to approve and
                     publish compliance methods for
                     laboratories that choose not to use
                     PBMS. After EPA makes final
                     determinations about the
                     implementation of PBMS in programs
                     under the Safe Drinking Water Act, the
                     Agency would then provide specific
instruction on the specified
performance criteria and how these  ;
criteria would be used by laboratories
for compliance monitoring of SDWA
analytes.

G. What Are the Estimated Costs of
Analysis?

  To obtain cost information on the  '.
analysis of arsenic in drinking water,
the Agency collected price information
from a random telephone survey of  :
seven commercial laboratories, which
were certified in drinking water
analysis, and from price lists posted on
the Internet (Analytical Methods
Support Document, US EPA, 19991).
Table VI—2 summarizes the results of
this survey, including the specific   ;
methodology and the associated cost,
range. The actual costs of performing an
analysis may vary with laboratory, the
analytical technique selected, and the
total number of samples analyzed by a
laboratory. The estimated cost range is
only for the analysis of arsenic and does
not include shipping and handling
costs. The Agency solicits comments
from the public on the cost estimates
listed in Table VI-2.
                TABLE VI-2.—ESTIMATED COSTS FOR THE ANALYSIS OF ARSENIC IN DRINKING WATER 1
Methodology



Graphite Furnace Atomic Absorption (GFAA) 	 > 	
Gaseous Hydride Atomic Absorption (GHAA) 	
Esti-
mated
cqst
range ($)
15 to 25.
10 to 15.
15 to, 50.
151050.
15tq50.
  1 Analytical Methods Support Document (US EPA, 1999I).
H. What Is the Practical Quantitation
Limit?

  Method detection limits (MDLs) and
practical quantitation levels (PQLs) are
two performance measures used by
EPA's drinking water program to
estimate the limits of performance of
analytic chemistry methods for
measuring contaminants in drinking
water. As cited in Table VI-1, EPA
defines the MDL as "the minimum
concentration of a substance that can be
measured and reported with 99%
confidence that the analyte
concentration is greater than zero (40
CFR part 136, appendix B)." MDLs can
be operator, method, laboratory, and
matrix specific. MDLs are not
necessarily reproducible within a
laboratory  or between laboratories on a
daily basis due to the day-to-day
analytical variability that can occur and
                      the difficulty of measuring an analyte at
                      very low concentrations. In an effort to
                      integrate this analytical chemistry data
                      into regulation development, EPA's
                      OGWDW uses the PQL to estimate or
                      evaluate the minimum, reliable
                      quantitation level that most laboratories
                      can be expected to meet during day-to-
                      day operations. EPA's Drinking Water
                      program defined the PQL as "the lowest
                      concentration of an analyte that can be
                      reliably measured within specified
                      limits of precision and accuracy during
                      routine laboratory operating conditions
                      (50 FR 46906, November 13, 1985)."
                      1. PQL Determination
                        A PQL is determined either through
                      the use of interlaboratory studies or, in
                      absence of sufficient information,
                      through the use of a multiplier of 5 to
                      10 times the MDL. The inter-laboratory
                      data is obtained from water supply (WS)
performance evaluation (PE) studies that
are conducted twice a year by EPA to
certify drinking water laboratories (now
referred to as the Performance Testing or
PT program). In addition to certification
of drinking water laboratories, WS
studies also provide:
  • Large-scale evaluation of analytical
methods;
  • A database for method validation;
  • Demonstration of method
utilization by a large number of
laboratories; and
  • Data for PQL determinations.
Using graphical or linear regression '
analysis  of the WS data, the Agency sets
a PQL at a concentration where at least
75% of the laboratories (generally EPA
and State laboratories) could perform
within an acceptable level of precision
and accuracy. This method of deriving
a PQL was used in the past for
inorganics such as antimony, beryllium,

-------
                  Federal Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed  Rules
                                                                     38915
 cyanide, nickel and thallium (57 FR
 31776 at 31800; US EPA, 1992b).
 2. PQL for Arsenic
   In 1994, EPA derived a preliminary
 PQL for arsenic based on data collected
 by the Agency from WS studies 20
 through 33 (WS 31 was excluded
 because the spiked samples were mixed
 incorrectly). In response to concerns
 from the water utility industry, the •
 results of this derivation and a separate
 evaluation conducted by the American
 Water Works Association (AWWA) were
 reviewed by the EPA Science Advispry
 Board (SAB) in 1995. The SAB noted
 that the acceptance limits of + 40% Used
 by EPA to derive the PQL in 1994 were
 wider than those for other SDWA metal
 contaminants. The acceptance limits
 and PQLs for several SDWA metals are
 shown in Table VI-3. The SAB
 recommended that EPA set the PQL
 using acceptance limits similar to those
 used for other inorganics.
          TABLE VI-3.—ACCEPTANCE LIMITS AND PQLs FOR OTHER METAL'S (IN ORDER OF DECREASING PQL)
Contaminant
Barium 	
Chromium 	
Selenium 	 	
Antimony 	 ;
Thallium 	
Cadmium 	
Beryllium 	
Mercury 	
Acceptance
limit1
(percent)
+1 ^
+15
+?n
4-qn
+qn
j-on
+15
±30
PQL(mg/L)2

001
Om
n nnfi
n nno
n nno
n nm
0.0005
  1 Acceptance limits for the listed inorganics are found at CFR 141.23 (k) (3)(ii).
  2 The PQL for antimony, beryllium and thallium was published in 57 FR 31776 at 31801 (July 17, 1992; US EPA, 1992b)  The PQL for barium
cadmium, chromium, mercury and selenium was published in 66 FR 3526 at 3459 (January 30, 1991; US EPA 1991a)
  Subsequent to SAB's
recommendation, EPA derived a new
PQL for arsenic (Analytical Methods
Support Document, US EPA, 19991).
The process employed by the Agency to
determine the new PQL utilized:
  • Data from six voluntary, low-level
(<0.006 mg/L of arsenic) WS studies;
  • Acceptance limits similar to other
low-level inorganics; and
  • Linear regression analysis to
determine the point at which 75% of
EPA Regional and State laboratories fell
within the chosen acceptance range.
  The derivation of the PQL for arsenic
was consistent with the process used to
determine PQLs for other metal
contaminants regulated under SDWA
and took into consideration the
recommendations from the SAB. Using
acceptance limits of + 30% and linear
regression analysis of WS studies 30
through 36 (excluding 31) yielded a PQL
of 0.00258 mg/L. The Agency rounded
up to derive a PQL for arsenic of 0.003
mg/L at the + 30% acceptance limit.
While the PQL represents a stringent
target for laboratory performance, the
Agency believes most laboratories, using
appropriate quality assurance and
quality control procedures, will achieve
this level on a routine basis.
I. What Are the Sample Collection,
Handling and Preservation
Requirements for Arsenic?
  The manner in which samples are
collected, handled and preserved is
critical to obtaining valid data.  Specific
sample collection, handling and
preservation procedures for SDWA
analytes are outlined in the "Manual for
the Certification of Laboratories
Analyzing Drinking Water" (US EPA,
1997a). For metals such as arsenic, the
certification manual specifies the
following:
   • Nitric acid (HNO3 at pH< 2) as the
preservative;
   • A maximum sample holding time of
6 months;
   • And a sample size of 1 liter,    '
collected in an appropriately cleaned
plastic or glass container, is suggested.
   Currently, arsenic does not have an
entry for preservation, collection, anji
holding time. EPA is proposing in this
rule, to revise the table following
§ 141.23(k)(2) to add "arsenic, ConcJ
HNO3 to pH < 2, P or G, and 6 months."
EPA requests comment on the
appropriateness of this revision.
   While 40 CFR 141.23(a)(4) allows :
compositing of up to 5 samples from the
same PWS, the detection limit required
for compositing must be Vs of the MCL.
Also, compositing for inorganic samples
must be done in the laboratory. Samples
should only be held if the laboratory
detection limit is adequate for the
number of samples being composited. In
any case, the composite is not to exceed
five samples. EPA is adding the test
methods and detection limits for the
approved  arsenic analytical methods to
the table following § 141.23(a)(4)(i).

/. Laboratory Certification

1.  Background                    '•
  The ultimate effectiveness of today's
regulation depends upon the ability of
laboratories to reliably analyze arsenic
at  the proposed MCL. The existing  ;
 drinking water laboratory certification
 program (LCP), which was established
 by States with guidance and
 recommendations from EPA, requires
 that only certified laboratories analyze
 compliance samples. External checks of
 a laboratory's ability to analyze samples
 of regulated contaminants within
 specific limits is the one means of
 judging laboratory performance and
 determining whether or not to grant
 certification. Under a performance
 testing (PT) program (formerly known as
 the performance evaluation or PE
 program), laboratories are required to
 successfully analyze PT samples
 (contaminant concentrations are
 unknown to the laboratory being
 reviewed) that are prepared by
 appropriate third parties. Successful
 participation in a PT program is a
 prerequisite for a laboratory to achieve
 certification and to remain certified for
 analyzing drinking water compliance
 samples. Achieving acceptable
 performance in these studies of
 unknown test samples provides some
 indication that the laboratory is
 following proper practices.
 Unacceptable performance may be
 indicative of problems that could affect
the reliability of the compliance
monitoring data.
  2. What Are the Performance Testing
Criteria for Arsenic?
  The Agency has historically identified
acceptable performance using one of
two different approaches:
  (a) Regressions from the performance
of preselected laboratories (using 95
percent confidence limits), or

-------
38916
Federal Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed Rules
  (b) Specified accuracy requirements.
  Acceptance limits based on specified
accuracy requirements are developed
from past PE study data. EPA has
traditionally preferred to use the second
("true value") approach because it is the
better indicator of performance and
provides laboratories with a fixed target.
Under this approach, each laboratory
demonstrates its ability to perform
within pre-defined limits. Laboratory
performance is evaluated using a
constant "yardstick" independent of
performance achieved by other
laboratories participating in the same
study. A fixed criterion based on a
percent error around the  "true" value
reflects the experience obtained from
numerous laboratories and includes
relationships of the accuracy and
precision of the measurement to the
concentration of the analyte. It also
assumes little or no bias in the
analytical methods that may result in
average reporting values  different from
the reference "true" value.
   In today's rulemaking, the Agency is
proposing that the laboratory
certification criteria for arsenic be set at
an acceptance limit of + 30 % at > 0.003
mg/L in §141.23(k)(3)(ii). Analysis of
water supply  data indicate that
laboratory capacity at this level should
be sufficient for compliance monitoring.
At this level, 75 % of EPA Regional and
State laboratories and  62 % of non-EPA
laboratories were capable of achieving
acceptable results. As  discussed in the
Analytical Methods Support Document,
(US EPA, 19991), setting  an acceptance
limit of ±20% would have decreased
laboratory capacity. EPA requests
comment on setting the acceptance limit
at the upper range of SAB's
recommendation.
                        3. How Often is a Laboratory Required
                      To Demonstrate Acceptable PT
                      Performance?
                        EPA requires that a PT (PE) sample for
                      chemical contaminants be successfully
                      analyzed at least once a year using each
                      method which is used to report
                      compliance monitoring results. For
                      arsenic this would require that the
                      laboratory successfully analyze a PT
                      (PE) sample using the method which is
                      used to report the results for compliance
                      monitoring. Additional guidance on the
                      minimum quality assurance
                      requirements, conditions of laboratory
                      inspections and other elements of
                      laboratory certification requirements for
                      laboratories conducting compliance
                      monitoring measurements are detailed
                      in the Manual for the Certification of
                      Laboratories  Analyzing Drinking Water,
                      Criteria and Procedures Quality
                      Assurance (US EPA, 1997a), which can
                      be downloaded via the Internet at
                      "http://www.epa.gov/ogwdwOOO/
                      certlab/labindex.html."

                      4. Externalization of the PT Program
                      (Formerly Known as the PE Program)
                        Due to resource limitations, on July
                      18,1996 EPA proposed options for the
                      externalization of the PT studies
                      program (61  FR 37464; US EPA, 1996c).
                      After evaluating public comment, in the
                      June 12, 1997 final notice EPA (62 FR
                      32112; US EPA, 1997c):
                      "decided on a program where EPA would
                      issue standards for the operation of the
                      program, the National Institute of Standards
                      and Technology (NIST) would develop
                      standards for private sector PE (PT) suppliers
                      and would evaluate and accredit PE
                      suppliers, and the private sector would
                      develop and manufacture PE (PT) materials
                      and conduct PE (PT) studies. In addition, as
                      part of the program, the PE (PT) providers
                      would report the results of the studies to the
study participants and to those organizations
that have responsibility for administering
programs supported by the studies."
EPA has addressed this topic in public
stakeholders meetings and in some
recent publications, including the
Federal Register notices mentioned in
this paragraph. More information about
laboratory certification and PT (PE);
externalization can be accessed at the
OGWDW laboratory certification
website under the drinking water
standards heading (www.epa.gov/
safewater).

VII. Monitoring and Reporting
Requirements
  The currently applicable monitoring
requirements for arsenic are different
than the other inorganic contaminants
(lOCs). First of all, arsenic's MCL and
compliance requirements are found in
§ 141.11, instead of in § 141.62(b).
Monitoring, compliance, and reporting
requirements for arsenic are also
different than the standardized    :
monitoring framework for the grouped
lOCs  (which does not include radon).
EPA is proposing to move arsenic to the
standardized monitoring framework for
lOCs  (antimony, asbestos, barium,
beryllium, cadmium, chromium,
cyanide, fluoride, mercury, nickel,
nitrate, nitrite, selenium, and thallium),
including the State reporting and
compliance requirements. Table VII—1
presents a  comparison of the existing
and proposed arsenic requirements, in
abbreviated form. For a full picture of
the regulations, you must look at the
regulatory language.
  In addition, EPA is proposing to
clarify the regulatory language for
sampling to determine compliance for
inorganics, volatiles and synthetic
organic contaminants.
                TABLE VII-1 .—COMPARISON OF SAMPLING, MONITORING, AND REPORTING REQUIREMENTS
                       [This table is not complete for compliance purposes, but provides an overview for readers.]
               Requirement
                                                       Current rule
                                                                            Proposed rule
 Compliance with § 141.11 (a)

 Compliance with § 141.11 (b)
                        MCL only applies to CWS and compliance is
                         calculated using §141.23.
                        MCL is 0.05 mg/L 	
 Monitoring frequency
                        Groundwater §141.23(a)(1)  One sample at
                         each entry point to the distribution system
                         (sampling point).
                        Surface water §141.23(a)(2) One sample at
                         every entry point to the distribution system
                         (sampling point).
 Would link  compliance with  50 jig/L  with
   §141.23(1) and would not add NTNCWS.
 MCL will  remain 50 ug/L for CWS serving
   10,000 or less until 5 years after publication
   of final rule, and be effective for larger sys-
   tems 3 years after publication  of final  rule.
   New lower MCL in §141.62.        ;
 NTNCWS will be subject to sampling, moni-
   toring and reporting  3 years after publica-
   tion of  final rule, but  not  subject to in-
   creased monitoring after exceedances, nor
   to MCL violations.
 No change to §141.23(a)(1).
                                                                                No change to § 141.23(a)(2).

-------
                   Federal Register/Vol. 65, No. 121/Thursday, June  22, 2000/Proposed Rules
                                                                            38917
          TABLE VII-1.—COMPARISON OF SAMPLING, MONITORING, AND REPORTING REQUIREMENTS—Continued
                        [This table is not complete for compliance purposes, but provides an overview for readers.]
                Requirement
               Current rule
                                                         Proposed rule
Compositing inorganics
Composite >Vs MCL .
Compositing by system size
Resampling composites
Compliance with § 141.11 CWSs have same re-
  quirements,  but arsenic monitoring  would
  move from §141.23(1) to §141.23(c).
Monitoring waivers § 141.23(c)
Minimum  data for  waivers:  Surface  water
  Ground  water All results  3,300  people.
§141.23(a)(4)(ii)    State     may    permit
  compositing  among different  systems,  5-
  sample limit, systems  serving <3,300 peo-
  ple.                             ;
§141.23(a)(4)(iii) Can use duplicates, of the
  original sample instead,  must be analyzed
  and reported to State within 14 days of col-
  lection.
§141.23(l)(1) CWS surface water yearly	
                                            §141.23(l)(2)  CWS ground water every three
                                              years..
None currently available for arsenic.
§141.23(m)  Supplier  must report  to State
  within 7 days  and initiate three additional
  samples at the same sampling point within
  a month.
Not currently specified.
§ 141.23(n) When the 4 analyses, rounded to
  the same  number of significant figures  as
  the MCL exceeds the MCL,  supplier must
  notify the State §141.31  and  give notice to
  the public §141.32.  Monitoring frequency
  determined by the State must continue until
  < MCL in two consecutive samples or until
  a variance, exemption, or enforcement ac-
  tion schedule becomes effective.
None currently specified for arsenic
Increased monitoring frequency
Adding  approved arsenic analytical  methods
  and  detection  limits to the table  following
  §141.23(a)(4)(i).
Same but §141.23(a)(4)(i) table will  list MCL
  and detection limits for arsenic.

No change to § 141.23(a)(4)(ii).
                                                                                       No change to § 141.23(a)(4)(ii).
No change to § 141.23(a)(4)(iii).
§141.23(c)(1) surface water one sample per
  compliance point annually.

§141.23(c)(1)  groundwater  one sample  at
  each  sampling  point during each  compli-
  ance period.
§ 141.23(c)(2) System may apply to the State.
§141.23(c)(3)Must take at least one  sample
  during waiver,  which  cannot exceed one
  compliance period (9 years).
§141.23(c)(4) at  least 3  years. At  least  3
  rounds of monitoring. At  least one  sample
  must be taken after January 1, 1990.
§141.23(c)(7) exceed MCL as calculated in
  (i), go to quarterly monitoring next quarter.
  §141.31(d) within 10 days of giving public
  notice,    contact    primacy    agency.
  §141.203(b) Tier 2 public notice no later
  than 30 days  after learning of violation and
  repeat every  3  months or at least  once  a
  year if allowed by primacy agency.
§141.23(0(1) for  IOCS,  §141.24(f)(15)(i) for
  VOCs, and §§141.24(h)(11)(i)  and  (ii) for
  SOCs will average based on # samples col-
  lected.
§141.23(i)(5) arsenic will  be reported to the
  nearest 0.001  rng/L.  § 141.23(i)(1)  moni-
  toring > annually, running annual average
  at sampling point. If less samples taken
  than required, compliance is based on aver-
  age of samples. Any sample below method
  detection limit is assigned zero for calcula-
  tion.

§141.23(0(2) monitoring   annually  or  less
  often if sampling point > MCL.
If State requires a confirmation sample, then
  compliance based on  average of the two
  samples. If State specifies additional moni-
  toring, compliance based on running annual
  average. If less samples taken  than re-
  quired, compliance is based on average of
  samples.
§141.23(f)(1) State may  require one within
  two weeks.
§141.23(c)(8) State can decrease monitoring
  after a minimum of 2 quarters for  ground
  water and 4 quarters  for surface  water
  
-------
38918
Federal Register/Vol. 65, No.  121/Thursday, June  22, 2000/Proposed  Rules
         TABLE VII-1.—COMPARISON OF SAMPLING, MONITORING, AND REPORTING REQUIREMENTS—Continued
                      [This table is not complete for compliance purposes, but provides an overview for readers.]
              Requirement
                                    Current rule
             Proposed rule
Now system and new sources ...
Subpart O Consumer Confidence Reports for
  CWS.
Subpart Q Public Notification for PWS
                       Only    mentions   waiver   eligibility   in
                         §141.23(c)(4).
                       >50 ng/L annual report §141.153(d)(6) length
                         of violation, potential health effects  using
                         Appendix C, actions taken. 25-50 |ig/L in-
                         formational statement per.§141.154(b).
                       >50  (ig/L  CWSs  Tier  2  annual  report
                         §141.203  required October 31, 2000 (if
                         they are in jurisdictions where the program
                         is directly implemented by EPA) or on the
                         date a primacy State adopts the new  re-
                         quirements (not to exceed May 6, 2002)..
§141.23(f)(1) If >MCL,  State can require a
  confirmation  sample  within two  weeks.
  §141.23(f)(3) Average used to  determine
  compliance with (i). States can  delete re-
  sults  with   obvious   sampling  errors.
  §141.23(g)  State  may  require  more  fre-
  quent monitoring.
§141.23(c)(9)  IOCS,  §141.24(f)(22) VOCs,
  §141.24(h)(20 SOCs, Compliance dem-
  onstrated within  State-specified  time  and
  sampling frequencies.                i
Lowers MCL & adds MCLG to Appendices A
  &  B to Subpart O-effective 30 days after
  final arsenic rule is published, before com-
  pliance with lower MCL is in place.
§ 141.203(b) Tier 2 public notice no later than
  30 days after learning of violation  and re-
  peat every 3 months or at least once a year
  if allowed by primacy agency.
                                                                               §141.31(d) within 10 days of giving public no-
                                                                                 tice, contact primacy agency.
                                                                               >5 ng/L CWSs & add NTNCWS to Table 1 of
                                                                                 §141.203 to  require Tier 2 annual report
                                                                                 §141.203 after effective date  of  arsenic
                                                                                 MCL (3-5 yrs).
A. What Are the Existing Monitoring
and Compliance Requirements?
  The arsenic monitoring requirements
appear in 40 CFR 141.23(a). Surface
water systems must collect routine
samples annually and ground water
systems must collect a routine sample
every three years. However, § 141.11(a)
currently only requires community
water systems (CWS) to monitor for
arsenic. EPA understands that some
States also require their non-transient
non-community water systems
(NTNCWS) to collect samples for the
analysis of arsenic as well. Under the
proposal, CWSs would continue to be
allowed to composite samples as
specified in § 141.23(a)(3); however, the
one-fifth arsenic MCL will no longer be
10 ug/L (It will be 1 ug/L).
  Sections 141.23(1) through (q) are
currently used to determine compliance
for arsenic. That is, if arsenic is detected
at a concentration greater than the
maximum contaminant level (MCL), the
community water system must collect 3
additional samples within one month at
the entry point to the distribution
system that exceeded the MCL
(§ 141.23(n)). If the average of the four
analyses performed, rounded to one
significant figure, exceeds the MCL, the
system must notify the State; and the
system must provide public notice
(§ 141.23(n)). After public notification,
the monitoring continues at the
                      frequency designated by the State until
                      the MCL "has not been exceeded in two
                      successive samples or until [the State
                      establishes] a monitoring schedule  as a
                      condition to a variance, exemption or
                      enforcement action (§ 141.23(n))."
                      Monitoring waivers are not permitted to
                      exclude a system from the sampling
                      requirements under § 141.23(l)-(q)
                      which currently apply to arsenic.

                      B. How Does the Agency Plan To Revise
                      the Monitoring Requirements?
                        The Agency is proposing to require
                      CWS and NTNCWSs to monitor for
                      arsenic using § 141.23(c). This will
                      make the arsenic monitoring
                      requirements consistent with the
                      inorganic contaminants (IOC's)
                      regulated under the standardized
                      monitoring framework. EPA is
                      proposing that NTNCWSs monitor  and
                      report arsenic results to the State and
                      public, as a Tier 2 notice in subpart Q,
                      Public Notification. However, the
                      Agency is proposing that NTNCWSs not
                      be required to meet the MCL, unlike the
                      other inorganics listed in § 141.62(b).
                      EPA's analysis for not requiring
                      NTNCWSs to comply with the MCL is
                      based on the cost-benefit analysis
                      discussed later in section XI.C. of this
                      preamble.
                        If arsenic exceeds the MCL, the CWS
                      will be triggered into quarterly
                      monitoring for that sampling point  "in
                      the next quarter after the violation
occurred (§ 141.23(c)(7)." The State may
allow the system to return to the routine
monitoring frequency when the State
determines that the system is reliably
and consistently below the MCL.
However, the State cannot make a    ;
determination that the system is reliably
and consistently below the MCL until a
minimum of 2 consecutive ground water
or 4 consecutive surface water samples
have been collected (§ 141.23(c)(8)). All
systems must comply with the sampling
requirements, unless a waiver has been
granted in writing by the State
(§141.23(c)(6)).
  As shown in Table VI—1, the approved
methods can measure to 0.001 mg/L or
below. In order to use the analytical
power of the methods, EPA is proposing
that arsenic data be reported to the
nearest 0.001 mg/L. Therefore, a result
of 0.0055 mg/L would be rounded to
0.006 mg/L, and 0.0145 mg/L would be
rounded to 0.014 mg/L (Figures ending
in "5" rounded down to end on an even
digit and up to an even digit.). During
the writing of this regulation, some
people had asked whether data above;
0.01 mg/L could be rounded to one
significant figure because the MCL is
being proposed with one significant
figure. EPA is issuing a clarification to
arsenic reporting in § 141.23(i) to
indicate that arsenic results will be   :
reported to the nearest 0.001 mg/L. The
significance for compliance purposes
will be that values between 0.010 mg/

-------
                 Federal Register/Vol.  65,  No. 121/Thursday, June 22, 2000/Proposed Rules
                                                                     38919
L and 0.014 mg/L will be averaged to
the nearest 0.001 mg/L, and the yearly
average will more closely reflect the
values measured. EPA requests
comment on these clarifications to
reporting requirements.

C. Can States Grant Monitoring
Waivers?
  As proposed, States will be able to
grant a 9-year monitoring waiver to a
system (§ 141.23(c)(3)). Waivers of
arsenic sampling requirements must be
based on all analytical results from
previous sampling and a vulnerability
assessment or the assessment from an
approved source water assessment
program (provided that the assessments
were designed to collect all of the
necessary information needed to
complete a vulnerability assessment for
a waiver). States issuing waivers must
consider the requirements  in 40 CFR
141.23(c)(2H6). In order to qualify  for
a waiver, there must be three previous
samples from a sampling point (annual
for surface water and three rounds for
groundwater) with analytical results
reported below the proposed MCL (i.e.,
the reporting limit must be < 0.005 mg/
L). The use of grandfathered data
collected after January 1,1990 that is
consistent with the analytical
methodology and detection limits of the
proposed regulation may be used for
issuing sampling point waivers. The
existing § 141.23(l)-(q) regulations do
not permit the use of monitoring
waivers. However, a State could now
use the analytical results from the three
previous compliance periods (1993-
1995,1996-1998, and 1999-2001) to
issue ground water sampling point
waivers. Surface water systems must
collect annual samples so a State could
use the previous 3 years sampling data
(1999, 2000, and 2001) to issue
sampling point waivers. One sample
must be collected during the nine-year
compliance cycle that the waiver is
effective, and the waiver must be
renewed every nine years. Vulnerability
assessments must be based on a
determination that the water system is
not susceptible to contamination and
arsenic is not a result of human activity
(i.e., it is naturally occurring).
  Although the approved analytical
methods can measure to 0.005 mg/L, not
all States have required systems to
report arsenic results below 50 |ig/L. In
this case, the States would  not have
adequate data to grant waivers until
enough data are available to make the
determinations. EPA has compliance
monitoring  data from 25 States at 10 |j.g/
L and below. On the other hand, one
State submitted data to EPA rounded to
tens of Hg/L, so some States may not be
able to grant waivers until the data are
reported below the proposed MCL.
  EPA believes that some States may
have been regulating arsenic under |ie
standardized inorganic framework being
proposed today. If so, those States will
have to ensure that existing monitoring
waivers have been granted using data
reported below the new proposed MCL.
Otherwise States will have to notify, the
systems of the new lower reporting
requirements that need to be met to;
qualify for a waiver for the proposed
MCL.

D. How Can I Determine if I Have an
MCL Violation?
  For this proposal, violations of the
arsenic MCL would be determined
under § 141.23(f)-(i). If a system
samples more frequently than annually
(e.g., quarterly), the system would be in
violation if the running annual average
at any sampling point exceeds the MCL
or if any one sample would cause the
annual average to be exceeded     ;
(§ 141.23(i)(l)). If a system conducts
sampling at an annual or less frequent
basis, the system would be in violation
if one sample (or the average of the
initial and State-required confirmation
sample(s)), at any sampling point  :
exceeds the MCL (§ 141.23(i)(2).    ;
However, States can require more
frequent monitoring per § 141.23(g) for
systems sampling annually or less often.
Therefore, the Agency is proposing to
clarify this section for situations for
IOCS in § 141.23(i)(2)) and the
corresponding sections for volatile and
synthetic organic contaminants
(§§ 141.24(f)(15)(ii) and
141.24(h)(ll)(ii), respectively. This
proposal clarifies compliance for
contaminants subject to §§141.23(i)(2)),
141.24(f)(15)(ii), and 141.24(h)(ll)(ii) by
pointing out that compliance will be
based on the running annual average of
the initial MCL exceedance and
subsequent State-required confirmation
samples. These confirmation samples
may be required at State-specified
frequencies (e.g., quarterly or some
other frequency depending on site-
specific conditions).
  In addition, the clarifications to  ;
§§ 141.23(i)(2)), (141.24(f)(15)(ii) and
141.24(h)(ll)(ii) address calculation of
compliance when a system fails to
collect the required number of samples.
Compliance (determined by the average
concentration) would be based on the
total number of samples collected. The
Agency expects systems will conduct all
required monitoring. However, some
 systems have purposely not collected
 the required number of quarterly
 samples, and in doing so some avoided
 reporting an MCL violation. While these
 systems all incurred monitoring and
 reporting violations for the uncollected
 samples, some systems divided the sum
 of the samples taken by four, which
 lowered the annual average reported to
 below the MCL, avoiding an MCL
 violation. The Agency requests
 comment on this clarification of
 exceedances determined under a State-
 determined monitoring frequency.
  For purposes of calculating MCL
 annual averages, § 141.23(i)(l) continues
 to set all non-detects equal to a value of
 zero. However,  the Agency realizes that
 some States use the detection limit or a
 fraction of the detection limit to
 calculate an average.

 E. When  Will Systems Have To
 Complete Initial Monitoring?
  The rule becomes effective 3 years
 after promulgation (about January 1,
 2004) for large PWS (serving  over
 10,000). This will require all  GW and
 SW systems serving over 10,000 to
 complete the  initial round of monitoring
 by December  31, 2004. However, States
 may allow systems, on a case-by-case
 basis, 2 additional years to comply with
 the MCL if capital improvements are
 necessary.
  The Agency is proposing a national
 finding that capital improvements are
 necessary for public water systems
 serving less than 10,000, on the basis
 that  existing treatments are not expected
 to be effective in arsenic removal. Table
 VII-2 shows the percentage of small
 systems with  no treatment in place as
 well as the percentage of systems which
 currently have in place technologies
 that  can remove arsenic. The  data shows
 that  capital improvements would be
 necessary for  many systems. The rule
 would be effective 5 years after
 promulgation (about January  1, 2006) for
 systems serving under 10,000. This
 would require these small GW systems
 to complete the initial round  of
 monitoring by the December 31, 2007
 ('05-'07 compliance period),  and small
 SW systems to complete the initial
round of monitoring by December 31,
 2006. EPA is requesting comment on
whether it is appropriate to make a
national finding that systems serving
less than 10,000 people will need the
two additional years to add capital
improvements in order to comply with
the proposed MCL. The alternative
would require States to issue  individual
two-year extensions for these small
systems.

-------
38920
Federal Register/Vol. 65, No. 121/Thursday,  June 22,  2000/Proposed Rules
       TABLE VI1-2.—TREATMENT IN-PLACE AT SMALL WATER SYSTEMS (US EPA, 1999E AND US EPA, 1999M)
System size
25-100 	
101-500 	
501-1K 	
1K-3.3K 	
3.3K-10K 	 	 	 	 	
Percent of sys-
tems with no
treatment in
place
GW
50
25
25
27
26
SW
7
6
0
0
0
Percent of sys-
tems with ion ex-
change in place
GW
1.7
1.4
2.9
1.6
2.1
SW
0
0
0
0
0
Percent of sys-
tems with coagu-
lation/filtration in
place
GW
1.7
4.1
2.4
2.7
8.1
SW
21.7
53.3
73.0
76.4
85.3
Percent of sys-
tems with lime
softening in
place
GW
2.6
2.7
2.4
2.7
3.3
SW
4.3
8.9
18.9
16.4
7.4
Percent of sys-
tems with re-
verse osmosis
in place
GW
0
0.5
0
0.4
0.6
s|w
0
0
0
• 0
0
  References: Geometries and Characteristics of Public Water Systems, August 1999, (US EPA, 1999e) Drinking Water Baseline Handbook,
February 24,1999, (US EPA, 1999m)
                                                                                                                  t
    The regulatory changes affected by the revised arsenic MCL are summarized in Table VII-3.

                               TABLE VII-3.—TABLE IDENTIFYING  REGULATORY CHANGES
       CFR citation
                                               Topic or subpart
§141.23(a)(4)
141.23(a)(5)
141,23(c)
141.23(g)
141.23(k)(2) ....
141.23(k)(3)(H)
141.62(b)(16) ..
141.62(c) ........
141.26(d) ........
141.154(b) ......
Appendix A to Subpart O of
  141.
Appendix B to Subpart O of
  141.
PN, Subpart Q, Table 1 to
  §141.203.
Appendix A to Subpart Q of
  141.
Appendix B to Subpart Q of
  141.
        Sample compositing allowed by the State.
        Detection limit for arsenic.
        Frequency of monitoring for arsenic determined in § 141.23(c).
        Standard inorganic monitoring framework, with State waivers possible.
        Confirmation sampling may be required by the State.
        More frequent monitoring may be required by the State.
        Compliance determination reporting.
        Approved methodology.
        Container, preservation, and holding time.
        Acceptance limit for certified laboratories.
        MCL for arsenic.
        BATs for arsenic.
        Small system compliance technologies (SSCTs).
        Requires CWS to report exceedances of new MCL in CCR before lower MCL is effective, removing 25-50 (ig/L
          informational statement requirement.
        Converting lower MCL compliance values for CCRs and listing MCLG.

        Changes MCLG and MCL values effective 30 days after MCL is final.

        Add NTNCWS exceeding MCL (not a violation) to Tier 2 reporting.

        Public notification regulatory citations revised.

        Standard Health Effects Language unchanged; revise MCLG, MCL.
  In order to prevent the arsenic MCL
of 5 Hg/L from becoming effective
immediately, EPA is proposing to delete
the reference to § 141.11(a) in § I4l.6(c),
which provides effective dates. While
examining §141.6(c) for sections that
affect arsenic, we found several sections
that do not exist. Therefore, EPA is
proposing to remove the reference to the
following sections in § 141.6(c) listed in
Table VII-4:

    TABLE  VII-4.—TABLE LISTING
          DELETED SECTIONS
   CFR section
141.11 (a)
    Topic or reason
New arsenic MCL would
  be effective imme-
  diately.
Section 141.11(e) does
  not exist
                           TABLE VII-4.—TABLE LISTING
                         DELETED SECTIONS—Continued
CFR section
141 14(a)(1)
141 .14(b)(2)(i)
141.14(d) 	
141.24(a)(3) 	

Topic or reason
Section 141.14 does not
exist.
Section 141.14 does not
exist.
Section 141.14 does not
exist.
Section 141.14 does not
exist.
Section 141.24(a) is re-
served.
  The Agency requests comment on
whether these deletions to § 141.6(c) are
necessary and appropriate.
F. Can I use Grandfathered Data To
Satisfy the Initial Monitoring
Requirement?

  Ground water systems may use
grandfathered data collected after Jan 1,
2002 to satisfy the sampling         '
requirements for the 2002—2004
compliance period. However, the
detection limit must be less than the
revised MCL. If the grandfathered data
is used to comply with the 2002-2004
compliance period and the analytical
result is between  the current MCL and
the revised MCL,  then that system will
be in violation of the revised MCL on
the effective date  of the rule. If the
system chooses not to use the
grandfathered data, then it must collect
another sample by December 31, 2004 to
demonstrate compliance with the
revised MCL.

-------
                  Federal Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed  Rules
                                                                      38921
 G. What Are the Monitoring
 Requirements for New Systems and
 Sources?
   The current regulations only address
 new systems and sources in the waiver
 provisions of § 141.23(c)(4), so the
 proposal specifically adds monitoring
 requirements for these systems for
 inorganic, volatile organic, and
 synthetic organics contaminants. All
 new systems or systems that use a new
 source of water that begin operation
 after the effective date of this rule would
 have to demonstrate compliance with
 the MCL within a period of time
 specified by the State. The State would
 also specify sampling frequencies to
 ensure a system can demonstrate
 compliance with the MCL. This
 requirement would be effective for all
 inorganic, volatile organic, and
 synthetic organic contaminants
 regulated in § 141.23 and § 141.24. The
 Agency recognizes that many States
 have established requirements for new
 systems  and new sources, and these are
 part of the approved State primacy
 programs. Therefore EPA believes that
 recognizing State-determined
 compliance will be the most effective
 way to regulate new systems and
 sources.  EPA requests comment on this
 proposed clarification.

 H. How Does the Consumer Confidence
 Report Change?
   On August 19,1998, EPA issued
 subpart O, the final rule requiring
 community water systems to provide
 annual reports on the quality of water
 delivered to their customers (63 FR
 44512; US EPA, 1998e). The first
 Consumer Confidence Reports (CCRs)
 were required by October 19,1999. The
 next reports are due by July 1, 2000, for
 calendar year 1999 data and every July
 1 after that (§ 141.152(a)). In general,
 reports must include information on the
 health effects of contaminants only if
 there has been a violation of an MCL or
 a treatment technique. For such
 violations specific "health effects
 language" in subpart O must be
 included verbatim in the report. The
 arsenic health effects language is
 currently required when arsenic levels
 exceed 50 ug/L.
  In addition, the Agency decided to
require more information for certain
contaminants because of concerns
raised by commenters. One of these
contaminants was arsenic. As explained
in the preamble to the final rule (63 FR
44512 at 44514; US EPA, 1998e) because
of concerns about the adequacy of the
current MCL, EPA decided that systems
that detect arsenic between 0.025mg/L
and the current MCL must include some
 information regarding arsenic
 (§ 141.154(b)). This informational   .
 statement is different from the health
 effects language required for an     I
 exceedance of the MCL. EPA noted that
 the requirement would be deleted upon
 promulgation of a revised MCL.
   Another issue which affects handling
 of arsenic in the CCR is the provisioii in
 the statute which authorized the
 Administrator to require inclusion of
 language describing health concerns for
 "not more than three regulated
 contaminants" other than those detected
 at levels which constitute a violation of
 an MCL (section 1414(c)(4)(B)(vi)).
 Based on stakeholder and commenter
 input, the Agency decided in the final
 CCR rule that it would use this authority
 in future rulemaking to require health
 effects language when certain MCLs are
 promulgated or revised. The health
 effects language of Subpart O would
 have to be included in reports of
 systems detecting a contaminant above
 the level of the new or revised MCL,.
 prior to the effective date of the MCL,
 although technically the systems are not
 in violation of the regulations. The
 Agency used this authority in the
 promulgation of the Disinfectants andl
 Disinfection Byproducts for one
 contaminant, Total Trihalomethanes 'on
 December 16,1998 (63 FR 69390). The
 Agency is now proposing to use this ;
 same authority to require inclusion of
 the health effects language in reports of
 systems which detect arsenic above the
 level of the revised MCL upon      :
 promulgation of these regulations. The
 Agency believes that it is important to
 provide this information to customers
 immediately. The systems have the
 flexibility to place this information in
 context and explain to customers that
 there is no on-going violation.
 Furthermore, the health advisory EPA is
 planning to issue in the near future will
 provide consumers with information
 about obtaining sources with lower
 arsenic prior to the effective date of the
 5 ug/L arsenic MCL. EPA asks for
 comment on whether the consumer
 confidence report should notify
 customers of arsenic health effects
 starting with the report issued by July 1,
 2002 for calendar year 2001.
  After the promulgation date of the ;
 revised arsenic MCL and before the
 effective date, community water systems
that detect arsenic above 5 Ug/L but  !
below 50 ug/L would include the
arsenic health effects language. Those!
systems that detect arsenic above 50 ug/
L would include the health effects
language and also report violations as
required by § 141.153(d)(6).
 I. How Will Public Notification Change?

   On May 4, 2000, EPA issued the final
 Public Notification Rule (PNR) for
 Subpart Q (US EPA 2000c) to revise the
 minimum requirements public water
 systems must meet for public
 notification of violations of EPA's
 drinking water standards and other
 situations that pose a risk to public
 health from the drinking water. Water
 systems must begin to comply with the
 new PNR regulations on October 31,
 2000 (if they are in jurisdictions where
 the program is directly implemented by
 EPA) or on the date a primacy State
 adopts the new requirements (not to
 exceed May 6, 2002). EPA's arsenic
 drinking  water regulation affects public
 notification requirements and amends
 the PNR as part of its rulemaking.
   The PNR divides the public notice
 requirements into three tiers, based on
 the seriousness of the violation or
 situation. Tier 1 is for violations and
 situations with significant potential to
 have serious adverse effects on human
 health as  a result of short-term
 exposure. Notice is required within 24
 hours of the violation. Tier 2 is for other
 violations and situations with potential
 to have serious adverse effects on
 human health. Notice is required within
 30 days, with extensions up to three
 months at the discretion of the State or
 primacy agency. Tier 3 is for all other
 violations and situations requiring a
 public notice not included in Tier 1 and
 Tier 2. Notice is required within 12
 months of the violation, and may be
 included in the consumer confidence
 report at the option of the water system.
  Today's proposal will require
 community water systems (CWS) to
 provide a Tier 2 public notice for
 arsenic MCL violations and to provide
 a tier 3 public notice for violations of
 the monitoring and testing procedure
 requirements. Today's proposal would
 also require NTNCWS to provide a Tier
 2 notice for exceedances of the MCL. As
 later explained in section XI.C., the
 Agency believes that overall risks from
 water ingested from NTNCWS cannot
 justify the costs of treatment. EPA
 believes that most States will, using
 their authority as described in
 § 141.203(b), require NTNCWS to issue
repeat notices on a yearly basis rather
than every three months. EPA requests
comment on the implementation of
arsenic public notification requirements
by the effective date of the arsenic MCL
and on the Tier 2 public notice
requirement for quarterly repeat notices
for continuing exceedances of the
arsenic MCL for NTNCWS.

-------
38922
Federal Register/Vol.  65,  No. 121/Thursday,  June 22, 2000/Proposed Rules
VIII. Treatment Technologies
  Section 1412(b)(4)(E) of the Safe
Drinking Water Act states that each
NPDWR which establishes an MCL shall
list the technology, treatment
techniques, and other means which the
Administrator finds to be feasible for
purposes of meeting the MCL.
Technologies are judged to be a best
available technology (BAT) when the
following criteria are satisfactorily met:
  • The capability of a high removal
efficiency;
  • A history of full scale operation;
  • General geographic applicability;
  • Reasonable cost;
  • Reasonable service life;
  • Compatibility with other water
treatment processes; and
  • The ability to bring all of the water
in a system into compliance.
  In order to fulfill this requirement set
forth by SDWA, EPA has identified
DATs in Section VIII.A. Their removal
efficiencies and a brief discussion of the
major issues surrounding the usage of
each technology are also given in
section VIII.A. Likely treatment trains,
of which the BAT will be the integral
part, are identified in section VIII. B.
The costs associated with these
treatment trains are also provided. More
details about the treatment technologies
and costs can be found in "Technologies
and Costs for the Removal of Arsenic
From Drinking Water" (US EPA,1999i).
  Section 1412(b)(4)(E)(ii) of the Act
also states that EPA shall list any
affordable small systems compliance
technologies that are feasible for the
purposes of meeting the MCL. The
general process by which EPA identifies
compliance, and if necessary, variance
technologies is described in section
VHI.C. The Agency, for the revised
arsenic regulation, is not proposing any
variance technologies. Compliance
technologies for arsenic are identified in
section VIII.E. More details about the
technologies and affordability
determinations can be found in
"Compliance Technologies for Arsenic"
(US EPA,1999g).
  Section VIII.F briefly discusses how
other rules, presently being developed
by the Agency, may impact the arsenic
rule, or how the arsenic rule may impact
these other regulations.
A. What Are the Best Available
Technologies (BATs) for Arsenic? What
Are the Issues Associated With These
Technologies?
  EPA reviewed several technologies as
BAT candidates for arsenic removal: ion
exchange, activated alumina, reverse
osmosis, nanofiltration, electrodialysis
reversal, coagulation assisted
                      microfiltration, modified coagulation/
                      filtration, modified lime softening,
                      greensand filtration, conventional iron
                      and manganese removal, and several
                      emerging technologies. The Agency
                      proposes that, of the technologies
                      capable of removing arsenic from source
                      water, only the technologies in Table
                      VIII—1 fulfill the requirements of the
                      SDWA for BAT determinations for
                      arsenic. The maximum percent removal
                      that can be reasonably obtained from
                      these technologies is also shown in the
                      table. These removal efficiencies are for
                      arsenic (V) removal.

                         TABLE VIII-1.—BEST AVAILABLE
                       TECHNOLOGIES AND REMOVAL RATES
                           Treatment technology
                      Ion Exchange 	
                      Activated Alumina	
                      Reverse Osmosis 	
                      Modified Coagulation/Filtration
                      Modified Lime Softening 	
                      Electrodialysis Reversal 	
Maximum
 percent
removal'
      95
      90
     >95
      95
      80
      85
                        1The percent removal figures are  for ar-
                      senic (V) removal.
                        In water, the most common valence
                      states of arsenic are As (V), or arsenate,
                      and As (III), or arsenite. As (V) is more
                      prevalent in aerobic surface waters and
                      As (III) is more likely to occur in
                      anaerobic ground waters. In the pH
                      range of 4 to 10, As (V) species
                      (H2AsO4 —and HzAsCu2") are negatively
                      charged, and the predominant As (III)
                      compound (HbAsOs) is neutral in
                      charge. Removal efficiencies for As (V)
                      are much better than removal of As (ill)
                      by any of the technologies evaluated,
                      because the arsenate species carry a
                      negative charge and arsenite is neutral
                      under these pH conditions. To increase
                      the removal efficiency when As (III) is
                      present, pre-oxidation to the As (V)
                      species is necessary.
                        Pre-oxidation. As (III)  may be
                      converted through pre-oxidation to As
                      (V) using one of several oxidants. Data
                      on oxidants indicate that chlorine,
                      potassium permanganate, and ozone are
                      effective in oxidizing As (III) to As (V).
                      Pre-oxidation with chlorine may create
                      undesirable concentrations of
                      disinfection by-products and membrane
                      fouling of subsequent treatments such as
                      reverse osmosis. EPA has completed
                      research on the chemical oxidants for
                      As (III) conversion, and is presently
                      investigating ultraviolet light
                      disinfection technology (UV) and solid
                      oxidizing media. For point-of-use  and
                      point-of-entry (POU/POE) devices,
                      central chlorination may be required for
                      oxidation of As (III).
  Coagulation/Filtration (C/F) is an
effective treatment process for removal
of As (V) according to laboratory and
pilot-plant tests. The type of coagulant
and dosage used affects the efficiency of
the process. Within either high or low
pH ranges, the efficiency of C/F is
significantly reduced. Below a pH of
approximately 7, removals with alum or
ferric sulfate/chloride are similar. A,bove
a pH of 7,  removals with alum decrease
dramatically (at a pH of 7.8, alum  !
removal efficiency is about 40%). Qther
coagulants are also less effective than
ferric sulfate/chloride. Disposal of the
arsenic-contaminated coagulation
sludge may be a concern especially if
nearby landfills are unwilling to accept
such a sludge.                   ~':
  Lime Softening (LS), operated within
the optimum pH range of greater than
10.5 is likely to provide a high
percentage of As removal. However, if
removals greater than 80% are required,
it may be difficult to remove
consistently at that level by LS alone.
Systems using LS may require
secondary treatment to meet that goal
(e.g., addition of an ion exchange unit
as a polishing step). As with C/F,
disposal of arsenic-contaminated sludge
from LS may be an issue.
  Coagulation/Filtration and Lime '
Softening are technologies primarily
used for large systems. Package plants
may make it more affordable for small
systems to employ these technologies.
Package plants are pre-engineered (i.e.,
the process engineering for the package
plants has been done by the
manufacturer). What remains for the
water system's engineer to design is the
specifics of the on-site application of the
equipment. However, these technologies
still require well trained operators. If it
is not possible to keep a trained operator
at the plant, an off-site contract operator
may be able to monitor the process with
a telemetry device. Because of these
complexities, these technologies are not
likely to be installed solely for arsenic
removal. However, if they are already in
place, modification of these two
technologies to achieve higher arsenic
removal efficiencies is a viable option.
  Activated Alumina (AA) is effective
in treating water with high total
dissolved solids (TDS). However, the
capacity of activated alumina to remove
arsenic is  very pH sensitive. High
removals  can be achieved  at high pHs,
but at shorter run lengths. The use of
chemicals for pH adjustment and bed
regeneration, storage of sulfuric acid
and sodium hydroxide, and process
oversight  increase operator
responsibilities and the need for
advanced training. (Decisions on the
certification of water operators will be

-------
                  Federal Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed Rules
                                                                      38923
 made at the State and local levels).
 Operators may have to add an acid to
 lower pH to an optimal range and then
 afterwards increase the pH to avoid
 corrosion. Sodium hydroxide and
 sulfuric acid are required in the
 regeneration process. Selenium,
 fluoride, chloride, sulfate, and silica, if
 present at high levels, may compete for
 adsorption sites. Suspended solids and
 precipitated iron can cause clogging of
 the AA bed. Systems containing high
 levels of these constituents may require
 pretreatment or periodic backwashing.
 AA is highly selective towards As (V),
 and this strong attraction results in
 regeneration problems, possibly
 resulting in 5 to 10 percent loss of
 adsorptive capacity after each run. As a
 result, AA may not be efficient in the
 long term. In addition, activated
 alumina produces highly concentrated
 waste streams, which can contain
 approximately 30,000 mg/L of total
 dissolved solids (TDS) content. Because
 of the high content of TDS in the waste
 stream, disposal of the brine must be
 taken into consideration.
   The safety issue of handling corrosive
 and caustic chemicals associated with
 this technology may make it
 inappropriate for small systems.
 Therefore, in estimating national costs,
 it was assumed that small systems
 would not adjust pH and would not
 regenerate on site. Costs were estimated
 assuming systems operated a non-
 optimal pH and operation on a "throw-
 away" basis. Regenerating the media off-
 site instead of disposing of spent media
 is another possibility.
   Ion Exchange (IX) can effectively
 remove arsenic as well. It is
 recommended as a BAT primarily for
 small, ground water systems with low
 sulfate and TDS, and as a polishing step
 after nitration. Sulfate, TDS, selenium,
 fluoride, and nitrate compete with
 arsenic for binding sites and can affect
 run length. Column bed regeneration
 frequency is a key factor in calculating
 costs. Recent research indicates that ion
 exchange may be practical up to
 approximately 120 mg/L of sulfate
 (Clifford 1994). Passage through a series
 of columns could improve removal and
 decrease regeneration frequency. As
 with AA, suspended solids and
 precipitated iron can cause clogging of
 the IX bed. Systems containing high
 levels of these constituents may require
 pretreatment. Suspended solids and
 precipitated iron may also be removed
 by backwashing.
  Ion exchange also produces  a highly
 concentrated waste by-product stream,
 and the disposal of this brine must be
 considered. Brine recycling can reduce
the amount of waste for disposal and
 lower the cost of operation. Recent
 research showed that the brine     :
 regeneration solution could be reused as
 many as 20 times with no impact on
 arsenic removal provided that some salt
 was added to the solution to provide
 adequate chloride levels for
 regeneration (Clifford 1998).
   Reverse Osmosis (RO) can provide
 removal efficiencies of greater than 95
 percent when operating pressure is ideal
 (e.g., pounds per square inch, psi).
 Water rejection (on the order of 20—
 25%) may be  an issue in water-scarce
 regions. If RO is used by small systems
 in the western U. S., water recovery will
 likely need to be optimized due to the
 scarcity of water resources. Water
 recovery is the volume of water     :
 produced by the process divided by the
 influent stream (product water/influent
 stream). Increased water recovery can
 lead to increased costs for arsenic
 removal. Since the ability to blend with
 an MCL of 5 ug/L would be limited, the
 entire stream may have to be treated.
 Therefore, most of the alkalinity and
 hardness would also be removed. In that
 case, to avoid corrosion problems and to
 restore minerals to the water, post-  :
 treatment corrosion control may be
 necessary. Discharge of reject water or
 brine may also be a concern.        '
   Electrodialysis Reversal (EDR)  can '
 produce effluent water quality
 comparable to reverse osmosis. EDR :
 systems are fully automated, require .
 little operator attention, and do not
 require chemical addition. EDR systems,
 however, are typically more expensive
 than nanofiltration and reverse osmosis
 systems. These systems are often used
 in treating brackish water to make it '-.
 suitable for drinking. This technology
 has also been applied in the industry ;for
 wastewater recovery. The technology
 typically operates at a recovery of 70 to
 80 percent. Few studies have been
 conducted to exclusively evaluate this
 process for the removal of arsenic, but
 a removal of approximately 85%  can be
 expected (US EPA, 19991).

 Other Technologies
  Coagulation Assisted Microfiltration.
 The coagulation process described   :
 previously can be linked with
 microfiltration to remove arsenic. The
 microfiltration step essentially takes the
 place of a conventional gravity filter. \
 The University of Houston recently  :
 completed pilot studies at Albuquerque,
 New Mexico on iron coagulation
 followed by a direct microfiltration  :
 system. The results of this study
indicated that iron coagulation followed
by microfiltration is capable of
removing arsenic (V) from water to yield
concentrations which are consistently
 below 2 ug/L. Critical operating
 parameters are iron dose, mixing energy,
 detention time, and pH (Clifford, 1997).
 However, since a full-scale operation
 history is one of the requirements to list
 a technology as a BAT, it is not
 presently being listed as one. It could be
 designated as such in the future if the
 technology meets that requirement.
   Oxidation/Filtration (including
 greensand filtration) has an advantage
 in that there is not as much competition
 with other ions. However, the process
 has not been used very much for arsenic
 removal. In addition, similar to
 activated alumina, greensand filtration
 may require pH adjustment to optimize
 removal, which may be difficult for
 small systems. This technology is not
 recommended for high removals. The
 maximum removal percentage was
 assumed to be 50% when estimating
 national costs. The presence of iron in
 the source water is  critical for arsenic
 removal. If the source water does not
 contain iron, oxidizing and filtering the
 water will not remove arsenic. In
 developing national cost estimates, it
 was assumed that systems would opt for
 this type of technology only if more
 than 300 ug/L of iron was present.
 Oxidation/Filtration is not being listed
 as a BAT because it does not meet the
 requirement of a high removal
 efficiency. However, since it is a
 relatively inexpensive technology, it
 may be appropriate for those systems
 that do not require much arsenic
 removal  and have high iron in their
 source water.

 Emerging Technologies
  There are several  emerging
 technologies for arsenic removal;
 however, these require more testing
 before they can be designated as a BAT.
 Iron-based media products include the
 following. Iron oxide coated sand
 removes  arsenic using adsorption; the
 sand also doubles as a filtration media.
 The technology has  only been tested at
 the bench-scale level and may have a
 high cost associated with it. Granular
 ferric hydroxide also employs an
 adsorption process and is being used in
 a number of full scale plants in
 Germany. Costs may be an issue with
 this technology as well. Iron filings are
 essentially a filter technology, initially
 developed for arsenic remediation.
 Though quite effective at remediation,
 this technology may have limited use as
 a drinking water treatment technology;
 the technology performs well when
treating high influent arsenic levels
typical of remediation, but needs to be
proven in treating lower influent levels
expected in raw drinking water to a
finished level at the  proposed MCL.

-------
38924
Federal Register/Vol. 65, No. 121/Thursday, June 22, 2000/Proposed Rules
Sulfur-modified iron appears to remove
total organic carbon (TOG) and
disinfection byproducts (DBFs) as well
as arsenic. However, it has only been
tested at the bench scale. ADI Group,
Inc."s proprietary process also has an
iron-based media that has been installed
in a number of locations.
  Nanofiltration is of interest because it
can be operated at lower pressures than
reverse osmosis, which translate into
lower operation and maintenance costs.
However, when nanofiltration is
operated at realistic recoveries, the
removal efficiency appears to be low.
  Electrodialysis Reversal (EDR),
although easier to operate than reverse
osmosis and nanofiltration, does not
appear to be competitive with respect to
costs and process efficiency.
Waste Disposal
   Waste disposal will be an important
issue for both large and small plants.
Costs for waste disposal have been
added to the costs of the treatment
technologies (in addition to any pre-
oxidation and corrosion control costs),
and form part of the treatment trains
that are listed in Section VIII.B. A
                       sufficient volume of receiving water
                       would be needed in order to directly
                       discharge the contaminated brine stream
                       from membrane technologies.
                       Otherwise, operators may have to pre-
                       treat to meet Clean Water Act permit
                       requirements prior to discharge. If the
                       plant is discharging to a sanitary sewer
                       because of the membranes, there may be
                       a very high salinity in the discharge as
                       well as high levels of arsenic that might,
                       without pretreatment, exceed local
                       sewer use regulations. Ion exchange and
                       activated alumina treatment brines will
                       be even more concentrated (on the order
                       of 30,000 TDS), and more than likely
                       will require pre-treatment prior to
                       discharge to either a receiving body of
                       water or the sanitary sewer.
                         Disposal of solid treatment residuals
                       would be problematic if they fail the
                       toxicity characteristic (TC) of the
                       Resource Conservation and Recovery
                       Act (RCRA). If they fail the TC, the
                       residuals are regulated as hazardous
                       waste because of the concentration of
                       arsenic. For the purposes of the national
                       cost estimate, it was assumed that solid
                       residuals would be disposed of at
                       nonhazardous landfills.
B. What Are the Likely Treatment
Trains? How Much Will They Cost?

  Likely treatment trains are shown in
Table VIII-2. These trains represent a
wide variety of solutions a facility may
consider when complying with the
proposed arsenic MCL. Not all solutions
may be viable for a given system. For
example, only those systems with   :
coagulation/filtration in-place will be
able to modify their existing treatment
system. The treatment trains include
DATs, waste disposal, and when
necessary, pre-oxidation and corrosion
control.
  Table VIII-2 also contains two "non-
treatment" options which may be
appropriate if the source water is of very
poor quality. "Regionalization" refers to
connecting with another system and
purchasing water, and "alternate
source" refers to finding a new source
of water (e.g. drilling a new well).
However, since arsenic is a naturally
occurring contaminant, it may be
ubiquitous at a particular site, so    ;
drilling another well may not improve
the situation.
                                    TABLE VIII-2.—TREATMENT TECHNOLOGY TRAINS
      Train
      No.
                                       Treatment technology trains
 1	  Regionalization.
 2	  Alternate Source.
 3	  Add pre-oxidation [if not in-place] and modify in-place Lime Softening.
 4	  Add pre-oxidation [if not in-place] and modify in-place Coagulation/Filtration.
 5	  Add pre-oxidation [if not in-pace] and add Anion Exchange and add POTW waste disposal and add corrosion control [if >90%
                   removal required]. Sulfate level at 25 mg/l.
 6	  Add pre-oxidation  [if not  in-place]  and add Anion Exchange and add POTW waste disposal and add corrosion control [if
                   >90% removal required]. Sulfate level at 150 mg/l.
 7	  Add pre-oxidation [if not in-place] and add Anion Exchange and add evaporation pond/non-hazardous landfill waste disposal
                   and add corrosion control [if >90% removal required]. Sulfate level at 25 mg/l.
 8	  Add pre-oxidation [if not in-place] and add Anion Exchange and evaporation pond/non-hazardous landfill waste disposal and
                   add corrosion control [if >90% removal required]. Sulfate level at 150 mg/l.
 9	  Add pre-oxidation  [if not  in-place] and add Activated Alumina and  add non-hazardous landfill (for spent  media) waste dis-
                   posal. pH at 7.
 10	  Add pre-oxidation [if not in-pace] and add Reverse Osmosis and add direct discharge waste disposal  and add corrosion con-
                   trol [if >90% removal required].
 11  	  Add pre-oxidation  [if not  in-place] and add Reverse Osmosis and add POTW waste disposal and add corrosion control [if
                   >90% removal required].
 12	  Add pre-oxidation  [if not  in-place] and add Reverse Osmosis and add chemical precipitation/non-hazardous landfill and add
                   corrosion control [if >90% removal required].
 13	  Add pre-oxidation [if not in-place] and add Coagulation Assisted Microfiltration and add mechanical dewatering/non-hazardous
                   landfill waste disposal.
 14	  Add pre-oxidation  [if not in-place] and add  Coagulation Assisted Microfiltration and add non-mechanical dewatering/non-haz-
                   ardous landfill waste disposal.
 15	  Add pre-oxidation [if not in-place] and add Oxidation/Filtration (Greensand) and add POTW for backwash stream.
 16	  Add pre-oxidation  [if not  in-place] and add  Anion Exchange and add chemical precipitation/non-hazardous landfill waste dis-
                   posal and add corrosion control [if >90%  removal required]. Sulfate level at 25 mg/l.                            :
 17	,	  Add pre-oxidation  [if not  in-place] and add  Anion Exchange and add chemical precipitation/non-hazardous landfill waste dis-
                   posal and add corrosion control [if >90%  removal required]. Sulfate level at 150 mg/l.
 18	  Add pre-oxidation [if not in-place] and add Activated Alumina and add POTW/non-hazardous landfill waste disposal. pH at 7.
 19	  Add pre-oxidation [if not in-place] and add POE Activated Alumina.
 20	  Add pre-oxidation [if not in-place] and add POU Reverse Osmosis.
 21  	  Add pre-oxidation [if not in-place] and add POU Activated Alumina.

-------
                  Federal Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed Rules
                                                                       38925
   Costs for each of these treatment
 trains are given in Table VIII-3. These
 costs are a function of system size. Some
 individual systems may experience
 household costs higher than those
 estimated in this table. The pre-
 oxidation costs and corrosion control
 costs are given separately for each
 system size category because they will
 only be incurred by some of the
 systems. In estimating national costs, it
 was assumed that only systems without
 pre-oxidation in-place would add the
 necessary equipment. It is expected that
 no surface water systems will need to
 install pre-oxidation for arsenic
 removal. Based on Table IX-4, it is
 expected that fewer than 50% of the
 ground water systems may need to
 install pre-oxidation for arsenic    :
 removal. Ground water systems without
 pre-oxidation should determine if pire-
 oxidation is necessary by determining if
 the arsenic is present as As (III) or As
 (V). Groundwater systems with
 predominantly As (V) will probably not
 need pre-oxidation to meet the MCL.
 Similarly, costs for corrosion control
 were only added to systems that used
 ion exchange or reverse osmosis to
 remove more than 90% of the arsenic in
 the raw water. It is expected that fewer
 than 1% of the affected systems will
 need to install corrosion control due to
 installation of arsenic treatment. For ion
 exchange, different treatment trains
 were used for two levels of sulfate. As
 sulfate affects regeneration frequency,
 the high sulfate treatment train is more
 expensive than the low sulfate treatment
 train.
                       TABLE VIII-3.—ANNUAL COSTS OF TREATMENT TRAINS (PER HOUSEHOLD)*

Treatment train
1 	
2 	
3 	
4 	
5 	
6 	
7 	
8 	
9 	
10 	
11 	
12 	
13 	
14 	
15 	
16 	
17 	
18 	
19 	
20 	
21 	
pre-ox** 	
corros** 	

25-100
(dollars)
$n47
96
750
462
R1Q
883
629
1227
384
2136
2136
2819
1282
1218
558
1008
1050
427
467
325
377
416
63

101-500
(dollars)
* ono

-too
QO
•\AK
OAR
PPfi
4fiQ
PP7
snn
ftnn
oqp
PQ°.
281
•JCA
ppo
OAK
PAQ
Ay?
QQQ
OO/1
cc
17

501-1000
(dollars)





-*cn
1 1\1
000
PD1



•me
-10-7






OH/I

11
. Si,
1001-
3300
(dollars)






















1 6
ze
3301-
10K
(dollars)

8

30
49
73
/o
108
197
loo
600
300
293


55
86
96
177
367

-------
38926
Federal Register/Vol.  65,  No. 121/Thursday, June 22, 2000/Proposed Rules
household income spent by an average
household on comparable goods and
services such items as housing (28%),
transportation (16%), food (12%),
energy and fuels (3.3%), telephone
(1.9%), water and other public services
(0.7%), entertainment (4.4%) and
alcohol and tobacco (1.5%).
  Another of the key factors that EPA
used to select an affordability threshold
was cost comparisons with other risk
reduction activities for drinking water.
Section 1412(b)(4)(E)(ii) of the SDWA
identifies both Point-of-Entry and Point-
of-Use devices as options for
compliance technologies. EPA
examined the projected costs of these
options. EPA also investigated the costs
associated with supplying bottled water
for drinking and cooking purposes. The
median income percentages that were
associated with these risk reduction
activities were: Point-of-Entry (>2.5%),
Point-of-Use (2%) and bottled water
(>2.5%). The complete rationale for
EPA's selection of 2.5% as the
affordability threshold is described in
Variance Technology Findings for
Contaminants Regulated Before  1996
(US EPA, 1998f).
  Based on the foregoing analysis, EPA
developed  an affordability criteria of
2.5% of median  household income, or
about S750, for the affordability
threshold (US EPA 1998f). The median
water bill for households in each small
system category was then subtracted
from this threshold to determine the
affordable level of household
expenditures for new treatment. This
difference is referred to as the "available
expenditure margin." Based on EPA's
1995 Community Water System Survey,
median water bills were about $250 per
year for small system customers. Thus,
an average available expenditure margin
of up to S500 per year was considered
affordable for the contaminants
regulated before 1996. However, EPA
expects the available expenditure
margin may be lower than $500 per
household per year for the arsenic rule
because EPA believes that water rates
are currently increasing faster than
median household income. Thus, the
"baseline" for annual water bills will
rise as treatment is installed for
compliance with regulations
promulgated after 1996, but before the
arsenic rule is promulgated.
   To account for this, EPA intends to
adjust its calculation of the baseline for
the affordability criteria as follows. The
national median annual household
water bills for each size category will be
adjusted by averaging the  total national
costs for the size category  over all of the
systems within the size category, hi
other words, the costs incurred by these
                      rules at the affected water systems will
                      be averaged over all of the systems in
                      that size category. A revised available
                      expenditure margin will be calculated
                      by subtracting the new baseline from the
                      affordability threshold. The affordable
                      technology determinations will be made
                      by comparing the projected costs of
                      treatment against the lower available
                      expenditure margin. If the projected
                      costs of all treatment technologies for a
                      given system size/source water quality
                      exceed the revised available
                      expenditure margin, then variance
                      technologies could be considered for
                      those systems. EPA requests comment
                      on this method of accounting for new
                      regulations in its affordability criteria.
                        Applying the affordability criterion to
                      the case of arsenic in drinking water,
                      EPA has determined that affordable
                      technologies exist for all system size
                      categories and has therefore not
                      identified a variance technology for any
                      system size or source water combination
                      at the proposed MCL. (See Table IX-12,
                      Total Annual Costs per Household.) In
                      other words, annual household costs are
                      projected to be below the available
                      affordability threshold for all system
                      size categories for the proposed MCL.
                      EPA solicits comment on its
                      determination in this case as well as its
                      affordability criteria more generally.
                        EPA recognizes that individual water
                      systems may have higher than average
                      treatment costs, fewer than average
                      households to absorb these costs, or
                      lower than average incomes, but
                      believes that the affordability criteria
                      should be based on characteristics of
                      typical systems and should not address
                      situations where costs might be
                      extremely high or low or excessively
                      burdensome. EPA believes that there are
                      other mechanisms that may address
                      these situations to a certain extent, such
                      as rates for disadvantaged communities
                      and grants. For instance, many utilities
                      extend special "lifeline" rates to
                      disadvantaged communities.
                        EPA also notes that high water costs
                      are often associated with systems that
                      have already installed treatment to
                      comply with a NPDWR. Such treatment
                      facilities may also facilitate compliance
                      with future standards. EPA's approach
                      to establishing the national-level
                      affordability criteria did not incorporate
                      a baseline for in-place treatment
                      technology. Assuming that systems with
                      high baseline water costs would need to
                      install a new treatment technology to
                      comply with a NPDWR may thus
                      overestimate the actual costs for some
                      systems.
                        To investigate this issue, EPA
                      examined a group of five small surface
                      water systems with annual water bills
above $500 per household per year
during the derivation of the national-
level affordability criteria. All of these
systems had installed disinfection and
filtration technologies to comply with
the surface water treatment rule. If these
systems exceeded the revised arsenic
standard, modification of the existing
processes would be much more cost-
effective than adding a new technology
to comply with the arsenic rule. These
systems have already made the
investment in treatment technology and
that is reflected in the current annual
household water bills.
  In addition, systems that meet criteria
established by the State could be   :
classified as disadvantaged
communities under section 1452(d) pf
the SDWA. They can receive additional
subsidization under the Drinking Water
State Revolving Fund (DWSRF)
program, including forgiveness of
principal. Under DWSRF, States must
provide a minimum of 15% of the
available funds for loans to small
communities and have the option of
providing up to 30% of the grant to
provide additional loan subsidies to .the
disadvantaged systems, as defined by
the State.
   As previously noted in today's
proposal, some technologies can
interfere with treatment in-place or
require additional treatment to address
side effects which will increase costs
over the arsenic treatment technology
base costs. (An example is corrosion
control for lead and copper, which may
need to be adjusted to accommodate
other treatment). While EPA tries to.
account for such interferences in its cost
estimates for each new compliance
technology, it is not possible to
anticipate all the  site specific issues
which may arise.  However, EPA has
included a discussion of the co-
occurrence of radon, sulfate, and iron in
this proposal. EPA will also provide
guidance identifying cost-effective  ;
treatment trains for ground water
systems that need to treat for both
arsenic and radon after the arsenic rule
is finalized.
   EPA encourages small systems to ,
discuss their infrastructure needs for
complying with the arsenic rule with
their primacy agency to determine their
eligibility for DWSRF loans, and if
eligible, to ask for assistance in applying
for the loans.

D. When Are Exemptions Available?
   Under section 1416(a), the State may
exempt a public water system from any
MCL and/or treatment technique
requirement if it finds that (1) due to
compelling factors (which may include
economic factors), the system is unable

-------
                  Federal Register/Vol.  65,  No.  121/Thursday, Juried,  2000/Proposed Rules
                                                                                 38927
to comply or develop an alternative
supply, (2) the system was in operation
on the effective date of the MCL or
treatment technique requirement, or, for
a newer system, that no reasonable
alternative source of drinking water  is
available to that system, (3) the
exemption will not result in an
unreasonable risk to health, and (4)
management or restructuring changes
cannot be made that would result in
compliance with this rule. Under
section 1416(b), at the same time it
grants an exemption the State is to
prescribe a compliance schedule and a
schedule for  implementation of any
required control measures. The final
date for compliance may not exceed
three years after the NPDWR effective
date except that the exemption can be
renewed for small systems for limited
time periods.
E. What Are the Small Systems
Compliance Technologies?
  Section 1412(b)(4)(EJ(ii) of SDWA, as
amended in 1996, requires EPA to issue
         a list of technologies that achieve  :
         compliance with MCLs established
         under the Act that are affordable arid
         applicable to typical small drinking
         water systems. These small public Water
         systems categories are: (1) Population of
         more than 25 but less than 500; (2)!
         Population of more than 500, but less
         than 3,300; and (3) Population of more
         than 3,300, but less than 10,000. Owners
         and operators may choose any
         technology or technique that best suits
         their conditions, as long as the MCL is
         met.
           Of the treatment trains identified in
         section VIII.B., the ones identified in
         Table VIII-4 are deemed to be affordable
         for systems serving 25-500 people and
         the ones identified in Table VIII-5 iare
         deemed to be affordable for systems
         serving 501-3,300 and 3,301-10,000
         people, as their annual costs are below
         the affordability threshold (US EPA,
         1999g). Because affordable compliance
         technologies are available, the Agency
         does not propose to identify any   ;
         variance technologies. EPA requests
comments on the affordable compliance
technology determinations for the three
size categories and the determination
that there will be no variance
technologies. Centralized compliance
treatment technologies include ion
exchange, activated alumina, modified
coagulation/filtration, modified lime
softening, and oxidation/filtration (e.g.
greensand filtration) for source waters
high in iron. In addition, point-of-use
(POU) and point-of-entry  (POE) devices
are also compliance technology options
for the smaller systems. EPA is aware
that very few water systems have had
experience with centrally managed POU
or POE options in the past. EPA requests
comments on implementation issues
associated with a centrally managed
POU or POE option for arsenic. The
non-treatment alternatives are especially
relevant for small systems. EPA is
proposing to add the abbreviations
"POU" and "POE" to the  definitions in
§ 141.2 and asks for comment on the
utility of adding them.
    TABLE VIII-4.—AFFORDABLE COMPLIANCE TECHNOLOGY TRAINS FOR SMALL SYSTEMS WITH POPULATION 25-500
           Train No.
                                 Treatment technology trains
4 ..
5 ..

6 ..

7 ..

8 ..

9 ..

15

16

17

18

19
20
21
Add pre-oxidation [if not in-place] and modify in-plape Lime Softening
Add pre-oxidation [if not in-place] and modify in-place Coagulation/Filtration
Add pre-oxidation [if not in-place] and add Anion Exchange and add POTW waste disposal and add corro-
  sion control [if >90% removal required]. Sulfate level at 25 mg/l.
Add pre-oxidation [if not in-place] and add Anion Exchange and add POTW waste disposal and add corro-
  sion control [if >90% removal required]. Sulfate level at 150 mg/l.
Add pre-oxidation [if not in-place] and add Anion  Exchange and  add  evaporation pond/non-hazardous
  landfill waste disposal and add corrosion control [if >90% removal  required]. Sulfate level at 25 mg/l.
Add pre-oxidation [if not in-place] and add  Anion Exchange and evaporation pond/non-hazardous landfill
  waste disposal and add corrosion control [if >90% removal required]. Sulfate level at 150 mg/l.
Add pre-oxidation [if not in-place] and add Activated Alumina and  add  non-hazardous landfill (for spent
  media) waste disposal. pH at 7.              i
Add pre-oxidation [if not in-place] and add Oxidation/Filtration (Greensand) and add POTW for backwash
  stream.
Add pre-oxidation [if  not in-place] and add Anion Exchange and add chemical precipitation/non-hazardous
  landfill waste disposal and add corrosion control [if >90% removal  required]. Sulfate level at 25 mg/l.
Add pre-oxidation [if  not in-place] and add Anion Exchange and add chemical precipitation/non-hazardous
  landfill waste disposal and add corrosion control [if >90% removal  required]. Sulfate level at 150 mg/l.
Add pre-oxidation [if not in-place] and add Activated Alumina and add POTW/non-hazardous landfill waste
  disposal. pH at 7.
Add pre-oxidation [if not in-place] and add POE Activated Alumina.
Add pre-oxidation [if not in-place] and add POU ReVerse Osmosis.
Add pre-oxidation [if not in-place] and add POU Activated Alumina.
  TABLE VIII-5.—AFFORDABLE COMPLIANCE TECHNOLOGY TRAINS FOR SMALL SYSTEMS WITH POPULATIONS 501-3,300
                                                 AND 3,301 TO 10,000   :
           Train No.
                                                                Treatment technology trains
                                Add pre-oxidation [if not in-place] and modify in-place Lime Softening
                                Add pre-oxidation [if not in-place] and modify in-place Coagulation/Filtration
                                Add pre-oxidation [if not in-place] and add Anion Exchange and add POTW waste disposal and add corro-
                                 sion control [if >90% removal required]. Sulfate level at 25 mg/l.
                                Add pre-oxidation [if not in-place] and add Anion Exchange and add POTW waste disposal and add corro-
                                 sion control [if >90% removal required]. Sulfate level at 150 mg/l.
                                Add pre-oxidation  [if not in-place] and  add Anion Exchange and add evaporation pond/non-hazardous
                                 landfill waste disposal and add corrosion control  [if >90% removal required]. Sulfate level at 25 mg/l.
                                Add pre-oxidation [if not in-place] and add Anion  "Exchange and evaporation pond/non-hazardous landfill
                                 waste disposal and add corrosion control [if >90% removal required]. Sulfate level at 150 mg/l.

-------
 38928
Federal Register/Vol. 65, No.  121/Thursday, June 22, 2000 / Proposed Rules
  TABLE VIII-5.—AFFORDABLE COMPLIANCE TECHNOLOGY TRAINS FOR SMALL SYSTEMS WITH POPULATIONS 501-3,300
                                          AND 3,301 TO 10,000—Continued
           Train No.
                                             Treatment technology trains
 10

 11

 12

 13

 14

 15

 16

 17

 18

 19
 20
 21
             Add pre-oxidation [if not in-place] and add Activated Alumina and add non-hazardous landfill (for spent
               media) waste disposal. pH at 7.
             Add pre-oxidation [if not in-place] and add Reverse Osmosis and add direct discharge waste disposal and
               add corrosion control [if >90% removal required].
             Add pre-oxidation [if not in-place] and add Reverse Osmosis and add POTW waste disposal and add cor-
               rosion control [if >90% removal required].
             Add pre-oxidation [if not in-place] and add Reverse Osmosis and add chemical precipitation/non-hazardous
               landfill and add corrosion control [if >90% removal required].
             Add  pre-oxidation [if not in-place] and add Coagulation Assisted  Microfiltration and  add  mechanical
               dewatering/non-hazardous landfill waste disposal.
             Add pre-oxidation [if not in-place] and add Coagulation Assisted Microfiltration and add non-mechanical
               dewatering/non-hazardous landfill waste disposal.
             Add pre-oxidation [if  not  in-place] and add Oxidation/Filtration (Greensand) and add POTW for backwash
               stream.
             Add pre-oxidation [if not in-place] and add Anion Exchange and add chemical precipitation/non-hazardous
               landfill waste disposal and add corrosion control [if >90% removal required]. Sulfate level at 25 mg/l.
             Add pre-oxidation [if not in-place] and add Anion Exchange and add chemical precipitation/non-hazardous
               landfill waste disposal and add corrosion control [if >90% removal required]. Sulfate level at 150 mg/l.
             Add pre-oxidation [if not in-place] and add Activated Alumina and add POTW/non-hazardous landfill waste
               disposal. pH at 7.
             Add pre-oxidation [if not in-place] and add POE Activated Alumina.
             Add pre-oxidation [if not in-place] and add POU Reverse Osmosis.
             Add pre-oxidation [if not in-place] and add POU Activated Alumina.
  Centralized treatment is not always a
feasible option. When this is the
situation, home water treatment devices
can be effective and affordable
compliance options for small systems in
meeting the proposed arsenic MCL.
Home water treatment can consist of
either whole-house (point-of-entry) or
single faucet (point-of-use) treatment.
  Whole-house, or POE treatment, is
necessary when exposure to the
contaminant by modes other than
consumption is a concern; this is not the
case with arsenic. Single faucet, or POU
treatment, is preferred when treated
water is needed only for drinking and
cooking purposes. POU devices are
especially applicable  for systems that
have a large flow and only a minor part
of that flow directed for potable use.
POE/POU options include reverse
osmosis, activated alumina,  and ion
exchange processes. POU systems are
easily installed and can be easily
operated and maintained. In addition,
tnese systems generally offer lower
capital costs and may reduce
engineering, legal, and other fees
associated with centralized treatment
options.
  Alloxving the usage of POU devices is
one of the new elements of the Safe
Drinking Water Act; on June 11,1998,
EPA issued a Federal Register notice
(US EPA, 1998i) to withdraw the
prohibition on the use of POU devices
as compliance technologies. The SDWA
stipulates that POU/POE treatment
systems "shall be owned, controlled and
maintained by the public water system,
                      or by a person under contract with the
                      public water system to ensure proper
                      operation and compliance with the MCL
                      or treatment technique and equipped
                      with mechanical warnings to ensure
                      that customers are automatically
                      notified of operational problems."
                        Using POU/POE devices introduces
                      some new issues. Adopting a POU/POE
                      treatment system in a small community
                      requires more record-keeping to monitor
                      individual devices than does central
                      treatment. POU/POE systems require
                      special regulations regarding customer
                      responsibilities and water utility
                      responsibilities. Use of POU/POE
                      systems does  not reduce the  need for a
                      well-maintained water distribution
                      system. On the contrary, increased
                      monitoring may be necessary to ensure
                      that the treatment units are operating
                      properly.
                        Water systems with high influent
                      arsenic concentrations (i.e., greater than
                      1 mg/L) may have difficulty meeting the
                      proposed MCL when POU/POE devices
                      are used. As a result, influent arsenic
                      concentration and other source water
                      characteristics must be considered when
                      evaluating POU/POE devices for arsenic
                      removal.
                        EPA assumed that systems would
                      more likely opt to use POU AA or RO
                      (and not IX), and POE AA (and not IX
                      nor RO), when developing national cost
                      estimates (refer to Table VIII-4).
                      Activated alumina and ion exchange
                      units face a breakthrough issue. If the
                      media or resin is not replaced and/or
                      regenerated on time, there is  a potential
for significantly reduced arsenic
removal. Activated alumina units have
the advantage of longer run lengths and
the option to use the media once and
throw it away. However, if POE ion
exchange units are regenerated on time,
they would also be an effective
treatment technology. Units with
automatic regeneration are thus viable
options. POE IX and RO units also have
a potential for creating corrosion control
problems. With ion exchange POE units,
a reduction in pH can be expected
initially with new resin, but the pH
reduction should subside over time. '

F. How Does the Arsenic Regulation
Overlap With Other Regulations?

  Several Federal rules are under
development regarding treatment
requirements that may relate to the
treatment of arsenic for this drinking
water rule. The following briefly
describes each rule, the impact the
Arsenic Rule may have on that rule, ;
and/or how each rule may impact the
arsenic standard. The Arsenic Rule is
expected to be promulgated in a similar
time frame as the Ground Water Rule,
the Radon Rule, and the Microbial and
Disinfection By-Product Rule (Final
December, 1998). In addition, the
disposal of residuals may be affected by
the hazardous waste regulations of the
Resource Conservation and Recovery
Act (RCRA).
  Ground Water Rule (GWR). The goals
of the GWR are to: (1) Provide a
consistent level of public health
protection; (2) prevent waterborne

-------
                 Federal Register/Vol. 65, No. 121/Thursday, June  22,  2000/Proposed Rules
                                                                    38929
microbial disease outbreaks; (3) reduce
endemic waterborne disease; and (4)
prevent fecal contamination from
reaching consumers. EPA has the
responsibility to develop a ground water
rule which not only specifies the
appropriate use of disinfection, but also
addresses other components of ground
water systems to assure public health
protection. This general provision is
supplemented with an additional
requirement that EPA develop
regulations specifying the use of
disinfectants for ground water systems
as necessary. To meet these
requirements, EPA worked with
stakeholders to develop a Ground Water
Rule proposal (US EPA, 2000d) and
plans to issue a final rule by late Fall
2000.
  The GWR will result in more systems
using disinfection. Under the GWR, a
system has options other than
disinfection (e.g., protecting source
water). However, if a system does add
a disinfection technology, it may
contribute to arsenic pre-oxidation.  This
largely depends on the type of
disinfection technology employed. If a
system chooses a technology such as
ultraviolet radiation,  it  may not affect
arsenic pre-oxidation. However, if it
chooses chlorination, it will contribute
to arsenic pre-oxidation. As discussed
previously, arsenic pre-oxidation from
As (III) to As (V) will enhance the
removal efficiencies of the technologies.
In addition, systems may use membrane
filtration for the GWR. In that case,
depending on the size of the membrane,
some arsenic removal can be achieved.
Thus, the GWR is expected to alleviate
some of the burden of the Arsenic Rule.
   Radon. In the 1996 Amendments to
the SDWA, Congress (section
1412(b)(13)) directed EPA to propose an
MCLG and NPDWR for radon by
August, 1999 (proposed on December
21,1999, US EPA 1999n) and finalize
the regulation by August, 2000 (section
1412(b)(13)). Like the Ground Water
Rule, the Radon Rule will also be
finalized before the Arsenic Rule.
Systems may employ aeration to comply
with the radon rule. Aeration alone,
however, will not likely be sufficient to
oxidize arsenic (III) to arsenic (V).
However, if systems do aerate, they may
be required by State regulations to also
disinfect. The disinfection process may
oxidize the arsenic, depending on the
type of disinfection employed.
Ultraviolet disinfection may not assist
in arsenic oxidation (still under
investigation by US EPA), whereas
chemical disinfection or oxidation is
likely to. Thus, the Radon Rule is
expected to alleviate some of the burden
of the Arsenic Rule.
  Microbial and Disinfection By-product
Regulations. To control disinfection and
disinfection byproducts and to   :
strengthen control of microbial
pathogens in drinking water, EPA js
developing a group of interrelated
regulations, as required by the SDWA.
These regulations, referred to
collectively as the Microbial
Disinfection By-product (M/DBP) Rules,
are intended to address risk trade-offs
between the two different types of
contaminants.                  '
  EPA proposed a Stage 1 Disinfectants/
Disinfection By-products Rule (DBPR)
and Interim Enhanced Surface Water
Treatment Rule (IESWTR) in July 1994.
EPA issued the final Stage 1 DBPB: and
IESWTR in November, 1998,
  The Agency has finalized and is
currently implementing a third rule, the
Information Collection Rule, that will
provide data to support development of
subsequent M/DBP regulations. These
subsequent rules include a Stage 2
DBPR and a companion Long-Term 2
Enhanced Surface Water Treatment Rule
(LT2ESWTR).                  :
  The IESWTR will primarily affect
large surface water systems, so EPA
does not expect much overlap with
small systems treating for arsenic.
However, the Stage 1 DBPR will affect
both large and small sized systems and
may overlap with small system's treating
for arsenic. In addition, the Stage 2
DBPR and possibly the LT2ESWTR
would have significance as far as arsenic
removal is concerned. For systems
removing DBF precursors, systems may
use nanofiltration. The use of
nanofiltration would also be relevant for
removing arsenic, and as a result, Would
ease some burden when systems  >
implement these later rules.
  Hazardous Waste. The current
toxicity characteristic (TC) regulatory
level for designating arsenic  as a
hazardous waste under the Resource
Conservation and Recovery Act (RCRA)
is 5 mg/L and is listed in 40 CFR '
261.24(a). It is important to differentiate
between the toxicity characteristic and
the toxicity characteristic leaching
procedure (TCLP). The TCLP is the
method by which a waste is evaluated
to determine if it exceeds the toxicity
characteristic. It is also important !to
note that while the toxicity
characteristic was based on multiplying
the current drinking water MCL by a
factor of 100, the TC is not directly
linked to the drinking water MCL; Thus,
lowering the drinking water  MCL;does
not mean that the toxicity characteristic
would be lowered. A separate RCRA
rulemaking would be required to lower
the toxicity characteristic regulatory
level. The drinking water standards for
several inorganic contaminants have
been lowered without any lowering of
the toxicity characteristic. For example,
the cadmium MCL was lowered from 10
|ig/L to 5 ug/L in 1991, but the TC for
cadmium still remains at 1.0 mg/L. The
drinking water standard for lead was
revised from an MCL of 50 jig/L to an
action level of 15 (ig/L. Both drinking
water standards were lowered in 1991.
The TC for lead remains at 5 mg/L. The
studies summarized below show that
arsenic residuals should be below the
current TC of 5 mg/L and could be
disposed in a non-hazardous landfill.
  In one study, sludges from four
different water treatment plants were
evaluated. (Bartley et al 1992). There
are data from two lime softening plants,
one plant with both lime softening and
coagulation/filtration processes, and one
arsenic removal plant utilizing
coagulation/filtration. The raw water
arsenic in the tow lime softening plants
and the one plant using both lime
softening and coagulation/filtration
were below 0.001 mg/L. The arsenic
removal plant was removing arsenic
from 1.1 mg/L to 0.42 mg/L using ferric
sulfate coagulation.  The product water
was blended with water from another
source to comply with the MCL. The
TCLP extracts ranged from 0.007 to
0.039 mg/L, which is considerably
below the current criterion for being
designated a hazardous waste under
RCRA.
  In another study, TCLP tests were
performed using the activated alumina
from two activated alumina plants
(Wang et al, 2000). Both plants had
similar setups (one is referred to as CS,
the other is referred to as BES). Both
systems consist of four tanks of
activated alumina with two parallel sets
of two tanks in series. The first set of
tanks are used as roughing filters and
the second set of tanks are used as
polishing filters. The units were not
regenerated, but replaced. For the CS
system, the influent arsenic
concentration ranged from 0.053 to
0.087 mg/L with an average of 0.062 mg/
L. The effluent arsenic concentration
was consistently below 0.005 mg/L.
When the activated alumina media was
removed from the roughing filters, three
samples were taken. All three samples
had  arsenic TCLP test results of less
than 0.05 mg/L. Again, these results
were well below the regulatory limit.
  The influent arsenic concentration of
the activated alumina plant referred to
as BES ranged from 0.021 to 0.076 mg/
L, with an average of 0.049 mg/L.
Effluent levels were also less than 0.005
mg/L. When the media was removed
from the two roughing filters, TCLP tests
were taken. The results were <0.05 mg/

-------
 38930
Federal Register/Vol. 65, No.  121/Thursday,  June 22, 2000/Proposed Rules
 L and 0.066 mg/L. Again, the results
 were below the regulatory limit.
   Another study examined residuals
 produced by anion exchange and
 coagulation-microfiltration (Clifford,
 1997). Experiments were performed at
 the University of Houston-US EPA
 Drinking Water Research Facility, a 10
 ft x 40 ft customized trailer containing
 various  unit processes, including ion
 exchange and coagulation-
 microfiltration, and a small analytical
 lab. The mobile research facility was set
 up at the West Mesa Pump Station in
 Albuquerque, NM.  The mean arsenic
 concentration in  the source water was
 0.021 mg/L.
   Ion exchange was field tested, and the
 media was regenerated. This initial
 waste stream was a brine from the
 regeneration process. The brine in the
 ion exchange process was reused 15
 times. The average arsenic
 concentration in  the product was below
 0.002 mg/L during the 15 cycles. The
 process  produced a highly concentrated
 spent brine, with arsenic concentrations
                     reaching 26.6 mg/L. It should be noted
                     that the arsenic concentration in the
                     brine would be lower if the brine was
                     not used as many times. After 6 months
                     of storage, the arsenic concentration
                     reduced to 11.3 mg/L. The arsenic was
                     then precipitated out of the brine using
                     iron, resulting in a brine with
                     approximately 0.037 mg/L of arsenic.
                     The precipitated sludge was then
                     subjected to the TCLP extraction
                     procedure. The TCLP extract had an
                     average arsenic concentration of 0.270
                     mg/L. This is below the current
                     threshold for being designated a
                     hazardous waste.
                       Coagulation-microfiltration was also
                     field tested. Arsenic removal to below
                     0.002 mg/L could be achieved; 12,000
                     gallons of water were filtered over 3
                     days. The backwash water, which is the
                     process waste, had less than 0.5%
                     solids. According to the TCLP Method
                     1311, for a liquid waste containing less
                     than 0.5%  solids, the liquid portion of
                     the waste after filtration, is defined as
                     the TCLP extract. About 20 backwash
                                       samples were collected, filtered, and
                                       analyzed for arsenic. The average
                                       concentration in the backwash water
                                       after filtration was 0.0026 mg/L and
                                       thus could be disposed as a
                                       nonhazardous waste. Additionally, the
                                       simulated sludge was subjected to the
                                       TCLP leaching procedure. The arsenic
                                       concentration in the TCLP extract was
                                       0.0218 mg/L, which is also considerably
                                       lower than the regulatory limit.
                                         The University of Colorado performed
                                       a series of tests of various arsenic
                                       treatment solid residuals using the
                                       TCLP test (Amy et al, 1999). The arsenic
                                       treatment processes included
                                       conventional plants utilizing lime
                                       softening, alum and ferric chloride
                                       coagulation, activated alumina, and
                                       membranes. The results of this analysis
                                       for the conventional plant residuals are
                                       presented in Table VIII-6. The data
                                       indicates that all the plants would pass
                                       the current TCLP test although the data
                                       from the iron coagulation plant do
                                       approach the limit.
                    TABLE VIII-6.—TCLP RESULTS FOR CONVENTIONAL PLANT ARSENIC RESIDUALS
Utility ID
F, coagulation sludge 	
F, softening sludge 	
F, filter sludge 	 	 	
G 	 	
J 	 	
L 	
C 	
O 	 	 	
Type of utility
Lime softening
Lime softening
Lime softening
Lime softening
Lime softening
Alum coagulation
Fe/Mn removal
Iron coagulation 	
TCLP extract
Arsenic (mg/L)







1.5596
   Table VIII-7 is a summary of TCLP
 data on liquid residuals prepared by the
 University of Colorado for activated
 alumina regenerant and a reverse
 osmosis reject water precipitated with
 ferric chloride. The activated alumina
 regenerant solution was neutralized to a
 pH of 6, which caused the aluminum to
 precipitate and adsorb the arsenic. The
 membrane reject water was treated with
 ferric chloride to remove the arsenic and
 the resulting ferric hydroxide residual
 was tested. The data indicates that solid
 residuals generated from the alumina
 regenerant and membrane residuals
 would pass the TCLP test.

 TABLE VIII-7.—TCLP TEST RESULTS
   FOR ACTIVATED ALUMINA AND  MEM-
   BRANE RESIDUALS
          Sample
Activated Alumina Column
  Regenerant 	
           TCLP
           extract
          as (mg/L)
                     TABLE  Vlll-7.—TCLP  TEST  RESULTS
                       FOR  ACTIVATED ALUMINA AND MEM-
                       BRANE RESIDUALS—Continued
Sample
Membrane Filter Reject Residu-
als 	

TCLP
extract
as (mg/L)
00179

             0.0242
  All of the previous data is from
residuals produced by central treatment.
There is no TCLP data on spent
activated alumina from POU or POE
devices. The TCLP results of spent
activated alumina media from POU and
POE devices were simulated by
assuming a worst-case scenario for 6-
month and one year replacement
frequencies (Kempic, 2000). To
determine the amount of arsenic that
could potentially leach into the
extraction fluid during the toxicity
characteristic leaching procedure, it was
assumed that the influent arsenic
concentration was 0.050 mg/L and that
 the activated alumina column adsorbed
 all of the arsenic. The first assumption
 represents the upper bound for influent
 concentrations since it is the current
 maximum contaminant level (MCL) for
 arsenic. The second assumption means
 that there would be no leakage or any
 breakthrough of arsenic through the
 column, which is not realistic. To
 calculate the total adsorbed arsenic
 mass, it was assumed that the POU unit
 treated 24 liters per day. This is the
 upper bound consumption used in the
 replacement frequency calculations.
  Two other assumptions were made to
 simulate the worst-case scenarios. In the
 TCLP, the solid phase is extracted with
 an amount of extraction fluid equal to
 20 times the weight of the solid phase.
 The dry media mass was used for the
 solid phase for this calculation rather
 the wet media mass. It was also
 assumed that all of the adsorbed arsenic
 would leach into the extraction fluid,
 which is not realistic. The estimates for
the worst-case scenarios are provided in
Table VIII-8.

-------
                 Federal Register/Vol. 65, No. 121 /Thursday, June 22, 2000/Proposed Rules
                                                                    38931
TABLE  VI11-8.—TCLP  PROJECTIONS
  FOR  ACTIVATED  ALUMINA  WORST-
  CASE SIMULATIONS
Replacement frequency
POU & 6-months 	
POE & 6-months
POU & Annual 	
POE & Annual 	

Max TCLP
Cone.
(mg/L)
2.6
0.8
10.4
3.2

  The projections for three of the worst-
case scenarios were below the TC of 5
mg/L. The worst-case maximum TCLP
concentration for annual replacement
for a POU activated alumina device was
above the TC. However, despite this
projection, activated alumina waste
should be non-hazardous. The most
unrealistic assumption was that all of
the arsenic adsorbed onto the alumina
would leach into the extraction fluid.
The TCLP uses weak acetic acid (0.57%)
at pH 5 for the extraction fluid. The
optimal pH for arsenic adsorption onto
activated alumina is between pH 5.5
and 6.0. Therefore, arsenic should be
retained on the activated alumina at this
pH. In fact, adsorbed arsenic is
extremely difficult to remove under any
conditions. A strong base (4% NaOH) is
typically used to regenerate activated
alumina. Arsenic is so strongly adsorbed
to the activated alumina that only 50 to
70% of the arsenic is eluted during
regeneration. Therefore, it is  extremely
unlikely that the spent activated
alumina from POU and POE units
would be considered hazardous.
   All of the TCLP data from solid
residuals were below the current TC of
5 mg/L. The arsenic concentrations in
TCLP extracts from alum coagulation,
activated alumina, lime softening, iron/
manganese removal, and coagulation-
microfiltration residuals were below
0.05 mg/L, which is two orders of
magnitude lower than the current TC
regulatory level. The TCLP data for iron
coagulation was mixed—the residuals
from the arsenic removal plant were
below 0.05 mg/L, but the residuals from
another iron coagulation plant were
above 1 mg/L. For anion exchange, the
TCLP data on the precipitated brine
stream was 0.27 mg/L. As was noted,
this was a highly concentrated brine
stream which had been used for fifteen
regenerations. Arsenic concentrations in
the precipitate would be lower if the
brine was used for fewer regeneration
cycles. Based on this data, EPA does not
believe that drinking water treatment
plant residuals would be classified as
hazardous waste. The TCLP data also
indicate that most residuals could meet
a much lower TC regulatory level. EPA
requests comment on whether it is
appropriate to assume that all residuals
can be disposed at a non-hazardous
landfill.                         ;

IX. Costs                        i

A. Why Does EPA Analyze the
Regulatory Burden?
  EPA is responsible for issuing
regulations that improve the quality of
the nation's drinking water and reduce
the risk of illness from exposure to!
harmful contaminants via drinking
water supplied by public water systems
(PWSs). As part of the regulatory   .
development process, the Agency is
required to analyze the regulatory cost
and burden imposed on all regulated
and affected entities and the benefits
associated with the regulation. The;
Regulatory Impact Analysis (RIA)  .
document is the principal summary of
these analyses. Assessing the impacts of
proposed SDWA regulations is a
complex process, involving many
analyses specified by various federal
mandates. In particular, EPA must
conduct analyses for the following
mandates:
• 1996 Safe Water Drinking Act   .
  (SDWA) Amendments          :
• Paperwork Reduction Act (PRA),
• Regulatory Flexibility Act (RFA)i
• Small Business Regulatory
  Enforcement Fairness Act (SBREFA)
• Unfunded Mandates Reform Act
  (UMRA)
• Executive Order (EO) 12866,
  "Regulatory Planning and Review"
• EO 12989, "Federal Actions to
  Address Environmental Justice in
  Minority Populations and Low-
  Income Populations"           !
• EO 13045, "Protection of Children
  from Environmental Health Risks and
  Safety Risks."
  Executive Order 12866 describes the
requirements for and content of the
national cost-benefit analyses. Section
1412(b)(3)(C) of SDWA, as amended in
1996, directs  EPA to seek comment on
a health risk reduction and cost analysis
(HRRCA) that will be issued with
proposed MCLs. The HRRCA must
identify quantifiable and          :
nonquantifiable costs and health
benefits of each MCL considered,
including the incremental costs and
benefits of each MCL considered. In
addition, the HRRCA must identify
benefits resulting from reducing co-
occurring contaminants and exclude
costs that will result from other
proposed or final regulations. The
Paperwork Reduction Act (PRA)
requires federal agencies to document
the cost and labor burden associat4d
with data collection, recordkeeping, and
reporting requirements of proposed
regulations. The Regulatory Flexibility
Act (RFA), as amended by the Small
Business Regulatory Enforcement
Fairness Act (SBREFA), mandates that
federal agencies consider the impact
imposed on small businesses,
governments, and non-profit
organizations. The objective of these
mandates is to provide regulatory relief
to small entities affected by SDWA
regulations by identifying alternative or
lower-cost compliance options. Finally,
the Unfunded Mandates Reform Act
(UMRA) seeks to assess the burden and
costs of federal  regulations to local and
State governments, while Executive
Order 12989 on environmental justice
instructs federal agencies to evaluate the
impact of proposed regulations on
minority and low-income populations.
Executive Order 13045 requires EPA to
state how the regulation addresses risks
for children.
  An RIA attempts to estimate the
possible outcomes in terms of costs and
benefits of various levels of regulation.
At the most basic level, an RIA is built
on estimates of the distribution of
arsenic occurrence among the various
water systems, the costs of treatment
technologies, and predictions of
responses by systems above the
regulatory level under consideration.
Because actual  compliance monitoring
at the proposed MCL has not been
required of all systems at the time of
proposal development, projections are
based on statistical estimates. EPA
believes that the current estimates
include appropriate conservative
assumptions and on average actual costs
are not likely to exceed the estimates.
One conservative assumption is that
equipment useful life is identical to
financing life. The Agency has a long
term effort in progress to better
characterize how much this issue will
affect cost estimations.
  To be complete, accurate, and
consistent, these analyses should be
based on a single, integrated set of data
and information that defines the
baseline characteristics or conditions of
the regulated community prior to
implementation of the regulation. The
regulated community is primarily the
water supply industry and State, local,
and tribal governments. However, it is
the customers of public water systems,
especially community water systems,
that ultimately  incur the cost burden
and realize the  intended health benefits
of these regulations. Therefore, the
baseline study identifies and, where
possible, quantifies the universe (e.g.,
characteristics of water suppliers, their
customers, and governmental entities) to

-------
 38932
Federal Register/Vol.  65,  No. 121/Thursday, June 22, 2000/Proposed Rules
 be used in the regulatory impact
 analysis (RIA).
   The current RIA applied national
 occurrence information in the modeling
 effort as described earlier in section V.G.
 EPA requests comment on its analyses
 for developing cost projections,
 including household costs, as well as
 additional cost information. Most
 previous RIAs conducted for the
 drinking water program assumed that all
 the water going into a system was the
 same concentration. Actually, many
 water systems (especially those serving
 more than 500 people) have multiple
 points where water enters the
 distribution system. Each of these entry
 points generally will have a different
 level ofarsenic. Consequently, water
 systems tend to be impacted by
 regulations in stages that increase with
 decreasing regulatory level. Because
 costs are spread across the entire
 system, individual household
 expenditures will vary according to
 regulatory level. Past RIAs were unable
 to incorporate this information, and for
 costing purposes, all entry points to the
 distribution system required treatment.
 The arsenic RIA is the first drinking
 water chemical RIA to incorporate
 monte carlo simulation of intra-system
 occurrence variability into the cost and
 benefits estimation. This simulation
 permits more accurate characterization
 of the relative household impacts of
 various alternatives. Several other
 changes have also been incorporated
 into the cost and benefit estimates for
 the arsenic RIA:
  Very Large Systems—Very large water
 systems, those serving more than a
 million people, can contribute a
 significant portion to estimates of
 overall costs and benefits at select
 regulatory levels. On the other hand,
 because there are so few of these
 systems and given that they are of
 complex configuration, statistically
 based estimates of arsenic occurrence
 (especially at low levels of arsenic
 incidence) introduce very large
 uncertainty  into the RIA. EPA addressed
 this issue by developing individually
 tailored estimates through the use of
 generally available occurrence
 information and Information Collection
 Rule data. Estimates were provided to
 the utilities and they were offered the
 opportunity to correct errors in the
 Agency assessment. While these
 estimates are a considerable
 improvement over past ones, it is
 important to keep in mind that they are
 merely projections and that individual
compliance costs could actually still
vary by a wide margin depending upon
rule timing, interactions with other
treatment or capital budget priorities,
                      regulatory commission decisions, or
                      actual compliance sampling results.
                        Inventory Based Modeling—Past RIAs
                      have generally developed benefit and
                      cost estimates by estimating impacts for
                      single representative community water
                      systems within a limited number of
                      size-based classes. Such an approach
                      introduces a slight positive bias to total
                      national cost estimates. This RIA has
                      gone beyond the past approach in the
                      modeling of community water system
                      and non-transient non-community water
                      system impacts. This RIA uses a monte
                      carlo approach to simulate application
                      of occurrence information to the actual
                      SDWIS inventory. Through repeated
                      simulations and assignments, the model
                      is able to develop the most robust
                      [statistically defensible] estimates of
                      actual exposure levels and to better
                      characterize the spread in household
                      costs.

                      B. How Did EPA Prepare the Baseline
                      Study?
                        EPA identified baseline
                      characteristics as the first step in
                      standardizing baseline profiles and
                      information for use across all Agency
                      drinking water RIAs and related
                      analyses. The Agency has several efforts
                      underway to develop improved
                      technical approaches for cost and
                      benefit analyses, including developing
                      characteristic engineering unit costs of
                      treatment plants, assessing financial and
                      operational capacity, and considering
                      the low-cost best available treatment
                      (BAT) options for small systems. Then,
                     EPA reviewed the analytical procedures,
                     and data requirements needed to
                     conduct the analyses.
                       Table IX—1 provides an overview of
                     the overall approach for identifying and
                     classifying specific baseline
                     characteristics. This matrix organizes
                     the baseline characteristics according to
                     the various entities likely to be affected
                     by SDWA regulations and the different
                     categories  of data analysis inputs. The
                     affected entities include:
                       • State and Tribal Governments:
                     Agencies at the State or local level
                     (including certain Tribes and Alaskan
                     Native Villages) responsible for
                     implementing, administering, and
                     enforcing drinking water programs, and
                     other programs potentially affected by
                     Federal drinking water mandates.
                       • Public Water Suppliers: Utilities
                     and other entities that provide potable
                     water to 25 or more persons, 15 or more
                     service connections (includes
                     community and transient/non-transient
                     non-community water systems).
                       • Customers: All entities that
                     purchase drinking water from public
                     water systems (including residential,
 commercial, industrial, wholesale,
 governmental, agricultural, and other
 users).
   The corresponding categories of data
 analysis inputs shown in Table IX-1
 include:
   1. Technical/Operational:
 Characteristics relating to capital assets
 and operational processes, labor skills
 and training, and other variable inputs.
   2. Managerial/Organizational:
 Characteristics relating to ownership,
 control and authority, organizational
 structure and management approach.
   3. Financial/Economic:
 Characteristics relating to monetary
 factors, opportunity costs, and benefits.
   4. Socio-Economic/Demographic:
 Composition and characteristics of
 affected entities (who, where, how
 much) and demographic trends.
   Data to describe all the baseline
 conditions shown in Table IX-1 are
 contained in a comprehensive EPA
 document designed to be applicable to
 all drinking water regulatory impact;
 analyses, "The Baseline Handbook." It
 is data from this document which is
 used in Chapter 4 of the RIA for
 Arsenic.

 1. Use of Baseline Data
   Uses of baseline data include the
 following analyses:
 National and Sub-National Benefits,
    Costs, and Economic Impact
    Analyses:
   • Occurrence  Analysis
   • Exposure/Risk Assessment
   • Model Plants/System Configuration
   • Unit Engineering Cost Analysis
   • Compliance Decision Tree Analysis
   • Financial Analysis
   • Government Implementation
   • Reporting, Recordkeeping, and
    Monitoring Costs
   • Valuation of Health Benefits
   • Non-Health  Benefits Assessment
   • Economic Impact Assessment
 Small Entity Impact Analyses:
   • Small Entity Definition
   • Reporting and Recordkeeping
    Requirements for Small Entities
   • Financial Analysis for Small
    Entities
   • Socio-Economic Analysis for Small
    Entities
   • Regulatory Alternatives Analysis
Other Special Analyses:
   • Health Risks to Sensitive
    Subpopulations
  • Affordability Analyses
  • Government Budgetary Effects
  These broad analytical requirements
reflect the overlapping nature of the  .
required analyses pursuant to the
relevant statutory and administrative
mandates. For example, various
mandates, including EO 12866, SDWA,

-------
                   Federal Register/Vol.  65, No. 121/Thursday, June 22,  2000/Proposed Rules
                                                                                          38933
UMRA, and PRA, require national cost
and benefit analyses.

2. Key Data Sources Used in the
Baseline Analysis for the RIA?

  A number of different data sources
were employed in the development of
the tables included Chapter 4 of the
arsenic RIA. The key data sources used
included:
  1995 Community Water System Survey
(CWSS). This database was compiled by EPA
from a survey conducted in 1995 to profile
the operational and financial characteristics
of community water systems of all source,
size, and ownership types.
  WATER STATS, The Water Utility
Database. This  database was compiled by the
                American Water Works Association from a
                1996 survey of its member utilities. Data on
                water system operations and finances were
                collected in two stages. The first stage  ;
                involved a comprehensive census of the i
                largest water utilities (i.e., those serving
                50,000 or more persons). A second stageidata
                collection involved a statistical sample of
                smaller water utilities.               :
                  Safe Drinking Water Information System
                (SDW!S). This database serves as the U.S.
                EPA's comprehensive database of public
                water system regulatory compliance and!
                violation information. SDWIS contains the
                Agency's inventory of all public water
                supplies, both community and
                noncommunity systems and the populations
                they serve.
                                   Survey on State Program Staffing/Funding
                                 for FY-97. The Association of State Drinking
                                 Water Administrators (ASDWA) conducted a
                                 survey of State drinking water programs to
                                 solicit estimates on the number of staff (i.e.,
                                 full-time equivalents, FTEs) involved in
                                 drinking water regulatory implementation
                                 and enforcement activities by program area,
                                 as well as estimates of drinking water
                                 program revenues/funding and expenditures
                                 by major account categories.
                                   2990 Census of Population. Data from the
                                 1990 Census of Population was  used in
                                 conjunction with water system data to
                                 develop estimates for various demographic
                                 characteristics of households and
                                 communities served by public water systems.
                  TABLE IX-1.—SUMMARY OF GENERAL BASELINE CATEGORIES OF AFFECTED ENTITIES
      Affected entity
                                                                Baseline characteristics
                          1: Technical & operational
                         2: Managerial & organiza-
                                  tional
                          3: Economic & financial
                         4: Socioeconomic & demo-
                                 graphic
A: State Government.
B: Public Water Suppliers
C: Customers
                         A1.1  PWS Inspections &
                           Sanitary Surveys.
B1.1   Water Sources/In-
  takes.
B1.2  Source Contamina-
  tion/Protection.
B1.3  Physical Configura-
  tion.
B1.4  Plant Condition	
B1.5  Plant Flow/Capacity
B1.6  Treatment/Waste
  Processes In-Place.
B1.7  Storage Capacity ...
B1.8  Distribution System
B1.9  Residence Time 	
B1.1   Monitoring/Labora-
  tory.
C1.1   POU/POE Systems
  In Use.
A2.1   Program Staffing ....
A2.2  Laboratory Capac-
  ity/Facilities.
A2.3  Division of Author-
  ity/Jurisdiction.
B2.1   Ownership/Organi-
  zational Structure.
B2.2  Plant Operation/Op-
  erators.
A3.1  Program Expendi-
  tures.
A3.2  Program Funding/
  .Revenues.

B3.1  Operating Expenses
B3.2  Operating Reve-
  'nues.
B3.3  Non-Operating Ex-
  penses.
B3.4  Assets & Liabilities
B3.5  Rate Structures/
  jUser Burden.
B3.6  Capital Investment
  Expenditure.
                                                                          A4.1  State PWS Profile.
B4.1  PWS Type.
B4.2  PWS Size/Cus-
  tomer Base.
B4.3  PWS Source Water.
B4.4  Geographic Loca-
  tion.
                                                  C2.1  Alternative Water
                                                    Use.
                                                  C2.2  Public Attitudes/
                                                    Perceptions.
                         C3.1
                         C3.2
      Residential Income
      Nonresidential In-
                                                   come.
                                                  C3.3  Residential Water
                                                   'Costs.
                                                  C3.4  Nonresidential
                                                   iWater Costs.
                                                  C3.5  Cost of Drinking
                                                   Water Alternatives.
                                                  C3.6  Medical Costs 	
                                                  C3.7  Non-Medical Costs
                                                  C3.8  Community Finan-
                                                   cial Information.
C4.1  Population Profile.
C4.2  Customer Water
  Use.
C. How Were Very Large System Costs
Derived?

  EPA must conduct a thorough cost-
benefit analysis, and provide
comprehensive, informative, and
understandable information to the
public about its regulatory efforts. As
part of these analyses, EPA evaluated
the regulatory costs of compliance for
very large systems, who would be
subject to the new arsenic drinking
                water regulation. The nation's 25 largest
                drinking water systems (i.e., those  ;
                serving a million people or more)
                supply approximately 38 million pebple
                and generally account for about 15 to 20
                percent of all compliance-related costs.
                Accurately determining these costs for
                future regulations is critical. As a result,
                EPA has developed compliance cost
                estimates for the arsenic and radon .
                regulations for each individual system
                that serves greater than 1 million   >
                                 persons. These cost estimates help EPA
                                 to more accurately assess the cost
                                 impacts and benefits of the arsenic
                                 regulation. The estimates also help the
                                 Agency identify lower cost regulatory
                                 options and better understand current
                                 water systems' capabilities and
                                 constraints.
                                   The system costs were calculated for
                                 the 24 public water systems that serve
                                 a retail population greater than 1
                                 million persons and one public water

-------
38934
Federal  Register/Vol. 65, No. 121/Thursday, June 22, 2000/Proposed  Rules
system that serves a wholesale
population of 16 million persons. Table
IX—2 lists these 25 public water systems.
The distinguishing characteristics of
these very large systems include:
                       (1) A large number of entry points
                     from diverse sources;
                       (2) mixed (i.e., ground and surface)
                     sources;
                       (3) Occurrence not conducive to
                     mathematical modeling;
                       (4) Significant levels of wholesaling;
  (5) Sophisticated in-place treatment;
  (6) Retrofit costs dramatically
influenced by site-specific factors; and
  (7) Large amounts of waste
management and disposal which can
contribute substantial costs.
             TABLE IX-2— LIST OF LARGE WATER SYSTEMS THAT SERVE MORE THAN 1 MILLION PEOPLE

1 	 	 	
2 	
3 	 	
4 	
5 	 	 	
6 	 	 	
7 	
8 	
9 	 	 	
10 	
1 1 	 	
12 	
13 	 	
14 	
15 	 	
16 	 	 	 	
17 	 	 	 	 	
18 	
19 	 	 	 	 	 	
20 	 , 	
21 	
22 	
23 	 	
24 	
25 	

PWS ID #
AZ0407025
CA01 10005
CA1910067
CA1910087
CA3710020
CA3810001
CA4310011
CO01 1 6001
FL41 30871
GA1210001
IL0316000
MA6000000
MD01 50005
MD0300002
MI0001800
MO6010716
NY5110526
NY7003493
OH1 800311
PA1510001
PR0002591
TX0570004
TX1010013
TX150018
WA5377050

Utility name
Phoenix Municipal Water System.
East Bay Municipal Utility District.
Los Angeles — City Dept. of Water and Power.
Metropolitan Water District of Southern California.
San Diego — City of.
San Francisco Water Department.
San Jose Water Company.
Denver Water Board.
Miami-Dade Water And Sewer Authority — Main System.
City of Atlanta.
City of Chicago.
Massachusetts Water Resource Authority.
Washington Suburban Sanitation Commission.
Baltimore City.
City of Detroit.
St. Louis County Water County.
Suffolk County Water Authority.
New York City Aqueduct System.
City of Cleveland.
Philadelphia Water Department.
San Juan Metropoiitano.
Dallas Water Utility.
City of Houston — Public Works Department.
San Antonio Water System.
Seattle Public Utilities.

  Generic models cannot incorporate all
of these considerations; therefore, in-
depth characterizations and cost
analyses were developed utilizing
several existing databases and surveys.
  The profile for each system contains
information such as design and average
daily flows, treatment facility diagrams,
chemical feed processes, water quality
parameters, system layouts, and intake
and aquifer locations. System and
treatment data were obtained from the
following sources:
  (1) The Information Collection Rule
(1997);
  (2) The Community Water Supply
Survey (1995);
  (3) The Association of Metropolitan
Water Agencies Survey (1998);
  (4) The Safe Drinking Water
Information System (SDWIS); and
  (5) The American Water Works
Association WATERSTATS Survey
(1997).
  While these sources contained much
of the information necessary to perform
cost analyses, the Agency was still
missing some of the detailed arsenic
occurrence data in these large water
systems. Where major gaps existed,
especially in groundwater systems,
occurrence data obtained from the
                      States of Texas, California, and Arizona,
                      the Metropolitan Water District of
                      Southern California Arsenic Study
                      (1993), the National Inorganic and
                      Radionuclides Study (EPA, 1984), and
                      utilities were used. Based on data from
                      the studies, detailed costs estimates
                      •were derived for each of the very large
                      water systems.
                        Cost estimates were generated for
                      each system at several MCL options.
                      The total capital costs and operational
                      and maintenance (O & M) costs were
                      calculated using the profile information
                      gathered on each system, conceptual
                      designs (i.e., vendor estimates and RS
                      Means), and modified EPA cost models
                      (i.e., Water and WaterCost models). The
                      models were modified based on the
                      general cost assumptions developed in
                      the Phase I Water Treatment Cost
                      Upgrades (EPA, 1998).
                        Preliminary cost estimates were sent
                      to all of the systems for their review.
                      Approximately 30% of the systems
                      responded by submitting revised
                      estimates and/or detailed arsenic
                      occurrence data. Based on the
                      information received, EPA revised the
                      cost estimates for those systems. Based
                      on the results, the majority of the very
                      large systems will not have capital or
O&M expenditures for complying with a
MCL of 5 ng/L (Table IX-3). More
detailed costs estimates for each very
large water system can be found in the
water docket.

TABLE  IX-3.—TOTAL  ANNUAL COSTS
  FOR  LARGE SYSTEMS FOR  (SERVING
  MORE THAN 1 MILLION PEOPLE)
MCL option
(W3/L)
3 	
5 	
10 	
20
Number
systems
treating
3
3
3
3
Cost
[$millions] 1
$16-18
11-12
6.6-7.47
2.6-2.7
  1 The lower number shows costs annualized
at 3%;  the  higher number  shows  costs
annualized at 7% capital costs. The 7% rate
represents the standard discount rate pre-
ferred  by OMB  for benefit-cost analyses  of
government programs and regulations.

D. How Did EPA Develop Cost
Estimates?

  EPA developed national cost
estimates by using the occurrence data,
unit cost curves, and a decision tree,
The occurrence data provides a measure
of the number of systems that would
need to install treatment in each size

-------
                 Federal Register/Vol.  65,  No. 121/Thursday, June 22, 2000/Proposed  Rules
                                                                     38935
category (the occurrence data was
described in Section V). The unit cost
curves provide a measure of how much
a technology will cost to install. Unit
cost curves are continuous functions;
they are a function of system size and
provide an estimated cost for all design
and average flows. The costs for a
treatment train for the average flow in
each size category were given
previously in Table VIII-3. The unit cost
curves can be found in "Technologies
and Costs for the Removal of Arsenic
From Drinking Water" (US EPA, 1999i).
  EPA then developed a decision tree,
which is a prediction of what treatment
technology trains facilities would likely
install to comply with options
considered for the revised arsenic   •
standard. A brief discussion of this ,
decision tree follows. A copy of the full
300+ page flowchart and supporting
documentation can be found in
"Decision Tree for the Arsenic     !
Rulemaking Process" (US EPA, 1999d).
The following figure is a brief
representation of this flowchart. As
shown in the flowchart, EPA considered
the impact of (1) MCL option and
influent arsenic concentration; (2)   •
system size; (3) regional effects (water
scarcity); (4) source water type (that;is,
ground water or surface water); (5)
existing treatment in-place; (6) waste
disposal issues and costs; and (7) co-
occurrence of iron and sulfate, to
estimate what systems are likely to
install.
  Ultimately, the decision tree was
expressed in decision matrices, in
which EPA assigned probabilities as to
how often each of the treatment trains
in Table VIH-2 will likely be used. EPA
developed a different decision matrix
for the eight system size categories,  for
three different removal efficiencies
(<50%, 50-90% and >90%), and for two
source waters (ground and surface). In
general, to the extent possible (e.g.,
based on source water quality), EPA
assumed that systems would employ the
least-cost technology that can meet the
MCL option.
BILLING CODE 8560-50-P

-------
 38936
Federal Register/Vol. 65,  No. 121/Thursday, June 22, 2000/Proposed Rules
                                                      MCL Target
                                                     2, 5,10,15. 20
                                                      System Size
                                                       <1Kor>1K
                                                        Region
                                                     (NW, SW, East)
                                                      Water Type
                                                      (GWorSW)
     Low Sulfate
      High Iron
        High
       Sulfata
       Low Iron
        High
       Sulfate
      High Iron
                                                                              '-Furthe>V^adet'Uate
                                                                              ..Treatment^,
                                                                                                          Low Sulfate
                                                                                                           High
                                                                                                          Sulfate
BILLING CODE 6S60-50-C

-------
                 Federal Register/Vol. 65, No. 121/Thursday, June  22,  2000/Proposed Rules
                                                                    38937
  MCL option. EPA developed a
decision tree that accounted for
treatment technology limitations, and
only assigned non-zero probabilities in
the matrices to those technologies
capable of reaching each MCL option.
The maximum removal percentages are
given in Table VIII—1. For instance,
since greensand filtration is only
assumed capable of removing 50% of
the influent arsenic, for an influent level
of 20 |ig/L, the technology is assumed to
be capable of only producing product
water with 10 (Xg/L of arsenic. Therefore,
for an MCL option of 5 ug/L, no usage
was assumed for greensand filtration at
a 20 Hg/L level of influent arsenic.
  System size. The decision tree also
depends on system size.  For instance,
small systems are assumed to operate
activated alumina on a throw-away
basis, and thus the probability of using
a treatment train that employs on-site
regeneration is assumed  to be zero. The
converse is true for large systems; non-
zero probabilities are assumed only for
those trains that employ regeneration
on-site.
   Water scarcity. Water scarcity was
also taken under consideration when
developing the decision  tree. It was
assumed that this issue would adversely
affect the selection of reverse osmosis,
since the technology rejects a significant
portion of the influent water. However,
the costs for reverse osmosis treatment
trains are much higher than others (refer
to Table VIII-3), and systems would
likely opt for other, less expensive,
treatment options. For the range of MCL
options considered, it was assumed that
ion exchange would be capable of;
delivering the required removal
efficiencies. Thus, water scarcity,
though considered in the decision tree,
did not affect percentages assigned to
reverse osmosis.                :
  Source water type. Source water type
is also a factor in the decision tree. It
affects the unit cost curves;  one set of
curves were developed for surface
water, and another was developed for
ground water. The treatment-in-place
data and co-occurrence data (as shown
below) are sorted by source water :type.
Also, certain technologies are
considered appropriate for one source
water type, but not the other. For ;
instance, greensand filtration is
considered relevant only for ground
waters.
  Existing treatment in-place.
Treatments that may already exist at
facilities were taken into account in the
decision tree. It was assumed that
systems would need to pre-oxidize, if
they weren't doing so already. Table IX—
4 shows the number of systems that
were assumed to require addition of pre-
oxidation (Source: US EPA,1999e).

 TABLE IX-4.—SYSTEMS NEEDING To
        ADD PRE-OXIDATION

System size

25-100 	
101-500 .. ..
501-1 K
1.001-3.3K 	
3.301-10K 	
10,001-SOK 	
50.001-100K ....
100,001-1 M 	
Percent of
ground
water sys-
tems
54
30
24
24
27
13
41
16
Percent of
surface
water sys-
tems
9
4
0
0
3
1
2
0
  It was also assumed that those
systems that had coagulation/filtration
in place, or lime softening in place,
would modify those treatments to
optimize for arsenic removal, since it is
a relatively inexpensive option. The
percent of systems with these treatments
in place is given in Table IX—5 (Source:
US EPA,1999e). However, for higher
removals (>90%), it was assumed that
only half of the systems would be able
to achieve the desired removal with a
modification. For those systems, an
additional cost of a polishing step, such
as ion exchange, was added.
          TABLE IX-5.—PERCENT OF SYSTEMS WITH COAGULATION-FILTRATION AND LIME-SOFTENING IN PLACE
System size
25-100 	
101 500 	
501 1K 	
1 001 33K 	
3301 10K 	
10 001 50K 	
50001 100K 	
100001 1 M 	

Percent of
ground water
systems with
CF in place
2
; 4
2
3
8
4
: 4
5

Percent of sur-
face water
systems with
CF in place
22
53
73
76
85
92
85
94

Percent of
ground water
systems with
LS in place
3
3
2
3
3
5
3
10

Percent of sur-
face water
systems with
LS in place
4
9
19
16
7
8
5
5

   Waste disposal issues and costs.
 Waste disposal of arsenic contaminated
 sludges and brines was also factored
 into the decision tree, and waste costs
 were added to the treatment trains. The
 waste disposal options for each of the
 technologies considered are given in
 Table IX-6. For ion exchange and
 activated alumina, it was assumed that
 the waste streams would be too
 concentrated to discharge directly.  For
 these technologies, it was assumed that
 some of the smallest systems would be
 able to take advantage of evaporation
 ponds, but that this option would be
 cost prohibitive in medium and large
 systems. It was assumed that most
 systems would opt for either chemical
 precipitation or discharge to a sanitary
 sewer. EPA also assumed that systems
 would dispose of spent activated'
 alumina media in non-hazardous
 landfills. Costs for reverse osmosis are
 prohibitive (In Table VIII-3, Annual
 Costs of Treatment Trains, compare
 lines 11,12, and 13 against other
 technologies), but if used, EPA assumed
 the relatively large amount  of reject
water would be discharged directly
(because it would not be as concentrated
as ion exchange and activated alumina
waste streams), to a sanitary sewer or by
chemical precipitation. For coagulation
assisted microfiltration, modified
coagulation filtration, and modified
lime softening, EPA assumed the waste
would be discharged to non-hazardous
landfills after the sludge is mechanically
or non-mechanically dewatered. For
greensand filtration, it was assumed that
the spent media would be disposed of
in a non-hazardous landfill.

-------
38938
Federal Register/Vol.  65,  No. 121/Thursday, June 22, 2000/Proposed Rules
                                    TABLE IX-6.—WASTE DISPOSAL OPTIONS
Treatment tech
Ion Exchange 	 , 	 	 	
Activated Alumina 	
Reverse Osmosis 	 	
Coag Assisted Micro-filtration 	
Groonsand 	
Modify CF 	
Modify LS 	 	 	 	

POTW
waste dis-
posal
•
•
,/





Evap pond
i/
i/






Non-haz
landfill

^/

i^

^/


Direct dis-
charge








Chemical
precip








Mech
dewater




•

•
•
Non-mec(i
dewater

t


•

•
•
  Co-occurrence of iron and sulfate.
EPA also factored into the decision tree
co-occurrence data on iron and sulfate
(shown in Tables IX-7 to IX-10, Source:
US EPA,1999f). Co-occurrence of
sulfate in water adversely affects the
performance of ion exchange, and
increases operation and maintenance
costs. Three sulfate-level treatment
trains were costed for ion exchange: one
                     low-level, one mid-level and one high-
                     level. The percentages in Tables IX-7 to
                     IX-8 were used as ceilings in national
                     cost estimates and limited the number
                     of systems that could be placed in the
                     decision matrices in the low-level and
                     mid-level sulfate ranges. For example,
                     the co-occurrence data shows that the
                     maximum number of systems that can
                     be costed at the low-level sulfate
 treatment train for an influent level of
 arsenic between 10 and 20 )ig/L is 35%.
 If more systems were to be placed in the
 decision matrices under ion exchange,
 no more than 39% were assumed to face
 a sulfate level between 25 and 120 mg/
 L. Any more systems assigned to ion   :
 exchange in the decision matrices were
 assumed to face high sulfate levels.
                              TABLE IX-7.—GROUND WATER: ARSENIC AND SULFATE
Influent arsenic
<10 ng/L 	
10-20 ug/L 	
>20npjl 	 	
Likelihood of sulfate
(percent)
<25 mg/L
48
35
33
25-1 20 mg/
33
39
38
I
>120 mg/L
i
1.9
26
30
                              TABLE IX-8.—ARSENIC WATER: ARSENIC AND SULFATE
Influent arsenic
<10 ng/L 	 	
10-20 ug/L 	 	 	 	
>20 (ig/L 	 	
Likelihood of sulfate
(percent)
<25 mg/L
28
20
12
25-1 20 mg/
32
30
28
>1 20 mg/L
40
5?
60
                               TABLE IX-9.—GROUND WATER: ARSENIC AND IRON
Influent arsenic
<10ug/L 	 	
10-20 ug/L 	 	
>20ug/L 	
Likelihood of sulfate
(percent)
<300 ug/L
82
81
71
>300 ug/L!
18
19
29
  TABLE IX-10.—SURFACE WATER:
        ARSENIC AND IRON
Influent arsenic
<10 uo/L 	 	
10-20 ug/L ........
>20 H8/L 	

Likelihood of sulfate
(percent)
<300 ng/L
91
92
90
>300 ug/L
9
8
10
                      Co-occurrence of iron in water
                    improves the performance of greensand
                    filtration. Greensand is relatively
                    inexpensive for small systems to use,
                    but not as effective as other treatment
                    technologies. It was assumed that
                    systems would opt for greensand
                    filtration only if the level of iron was
                    greater than 300 ug/L. EPA used the co-
                    occurrence data in Tables IX-9 to IX-10
to determine the ceiling on the number
of systems that could use greensand
filtration in the decision matrices.

E. What Are the National Treatment
Costs of Different MCL Options?
  Under the proposed option of 5 u,g/L,
the Agency estimates that annual
treatment costs to community water
systems will be $374 million per year.
If required to treat at the proposed level,

-------
                Federal  Register/Vol. 65, No. 121/Thursday, June! 22, 2000/Proposed Rules
                                                                                                        38939
                                     MCL options considered (3, 5,10, and
                                     20 Ug/L) are provided in Table IX-11.
treatment costs to non-community non-
transient systems would be $15 million
per year. National annual costs for the                                  ',
                               TABLE IX-11 .—NATIONAL ANNUAL TREATMENT COSTS
                                                [Dollars in millions]
MCL option
(W/L) ;
3 	 	 	
c 	
•in 	 • 	
20 	 ; 	
Community
water systems
$639
374
160
59
Non-commu-
nity Non-tran-
sient systems
$25
15
6
2
Total treat-
ment costs
$664
389
166
6.1
  Total annual costs per household are
given in Table IX-12. Due to economies
of scale, costs per household are higher
in the smaller size categories, and lower
in the larger size categories. For the
proposed option of 0.005 ug/L, costs are
expected to be $364 per household for
systems serving 25-100 people, and
$254 per household for systems serving
                                      101-500 people. Costs per households
                                      in systems larger than those are  '
                                      substantially lower: from $104 to $21
                                      per household. Costs per household do
                                      not vary dramatically across MCL!
                                      options. This is because of the fact that
                                      once a system installs a treatment
                                      technology to meet an MCL target; costs
                                      do not vary significantly based upon the
removal efficiency it will be operated
under. Costs are, however, somewhat
lower at less stringent MCL options.
This is because it was assumed that
some systems would blend water at
these options, and treat only a portion
of the flow.
                              TABLE IX-12.—TOTAL ANNUAL COSTS PER HOUSEHOLD
                                                    [Dollars]
System size
25 100 	 - 	 • 	
101 500 	
501 1K 	
1K 33K 	
3 3K 10K 	
10K 50K 	
50K 100K 	
100K-1M 	
3 ug/L
$368
259
, 106
: 64
44
36
30
23
5ng/L
$364
254
104
60
41
33
27
21
10ng/L
$357
246
98
57
37
29
23
18
20fig/L
$349
238
93
52
33
25
19
15
  Incremental costs are given in Tables
IX-13 and IX-14. Incremental costs
refer to the dollars that must be spent to
obtain the next, more stringent, level of
control. The national and household
                                      costs under 20 ug/L refer to the amount
                                      that must be spent to reach 20 ug/L
                                      starting from the baseline of 50 ug/L.
                                      The dollar value under 10 ug/L
                                      represents the cost differential between
20 Ug/L and 10 Ug/L. The values under
5 ug/L and 3 ug/L were derived
similarly.
                              TABLE IX-13.—INCREMENTAL NATIONAL ANNUAL COSTS
                                                [Dollars in millions]
MCL option
(W/L)
Of) 	 	
10 	 	
5 	
3 	 	
Community
water systems
$59
101
214
265
Non-commu-
nity non-tran-
sient water
systems
$2
4
9
19
Total
$61
105
223
275
                           TABLE IX-14.—INCREMENTAL ANNUAL COSTS PER HOUSEHOLD
                                                     [Dollars]        i
System size
25 100 	
101 500 	
501 1K 	
1K 3 3K 	
3.3K-10K 	
20 ng/L
$349
238
93
52
33
10 fig/L
$8
8
5
5
4
5|ig/L
$7
8
6
•3
4
3ng/L
$4
5
2
4
3

-------
 38940
Federal  Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed Rules
                      TABLE IX-14.—INCREMENTAL ANNUAL COSTS PER HOUSEHOLD—Continued
                                                      [Dollars]
System size
10K-50K 	
50K-100K 	
100K-1M 	
20 ng/L
0^
1Q
15
10ng/L


3
Sug/L


3
3|ig/L





3
3
2
   In the process of analyzing treatment
 technologies and developing cost
 estimates, EPA held several meetings
 with stakeholders to obtain input on
 assumptions made. Several of the key
 assumptions agreed to hy stakeholders
 are given below.
 1. Assumptions Affecting the
 Development of the Decision Tree
   • EPA assumed that ion exchange
 usage would be prohibited above 120
 mg/L of sulfate and 500 mg/L of TDS.
   • EPA assumed that greensand
 filtration would be used only if iron in
 the raw water was above 300 ug/L.
   • EPA assumed that systems would
 pro-oxidize, when existing chlorination
 or other oxidants are not already
 present.
   • EPA assumed that systems would
 not likely use POE-RO nor POE-IX
 because of corrosion control problems.
 Also, with IX, if the resin is not replaced
 and/or regenerated on time, there is a
 potential for arsenic peaking. EPA
 assumed that systems will most likely
 use POE-AA.
   • The breakthrough issue also exists
 with POU-IX. POU-AA has the
 advantage of a longer run length. EPA
 assumed that systems would use either
 POU-AA or POU-RO.
 2. Assumptions Affecting Unit Cost
 Curves
  • There are significant safety and
 operating efficiency risks to small
 systems when adjusting downward.
 This pM adjustment would require
 much more oversight than most small
 systems will have. EPA, in calculating
 unit costs for activate alumina assumed
 that systems would not adjust pH
 downward; thus, AA will be operated at
 a sub-optimal pH.
  • There is a danger of operating
 technologies such as ion exchange near
 breakthrough. EPA incorporated a safety
 factor, and used 80% of the MCL as the
 target when calculating costs for all
 technologies.
  • EPA assumed that small systems
 would not regenerate Activated
 Alumina on site—AA will likely be
operated on a "throw-away" basis.
  • For modifying coagulation/
 filtration, EPA considered the cost of a
new chemical feed system when
                     switching to iron. EPA costed out
                     switching coagulants for high removals.
                     For lower removals, EPA costed out
                     optimizing alum usage.
                       • EPA assumed 75% for RO recovery.
                       • For Activated Alumina, EPA
                     assumed that there will not be any
                     systems with raw water in the optimal
                     range for arsenic removal (pH between
                     5.5-6.0).
                       • For iron-coagulation-micro-
                     filtration EPA assumed systems would
                     apply a stronger iron dose rather than
                     adjusting to optimum pH.
                       • For ion exchange, one or more
                     regenerations per day is not
                     problematic. Regeneration in Ion
                     Exchange can be done automatically.
                     EPA examined cost models on
                     regeneration frequency, volume of waste
                     generated and considered computer-
                     automation for regeneration.

                     X. Benefits of Arsenic Reduction
                       The benefits associated with
                     reductions of arsenic in drinking water
                     arise from a reduction in the risk of
                     adverse human health effects, and a
                     corresponding decrease in the number
                     of expected cases and premature deaths
                     of people experiencing  these effects.
                     The various adverse health effects
                     associated with arsenic are known with
                     different levels of certainty. Presently
                     some can be quantified and some
                     cannot. The best characterized benefits
                     can be both quantified and monetized
                     (i.e., a dollar value is attached to the
                     expected decrease in number of cases),
                     while other benefits may be only known
                     well enough to describe. The latter are
                     known as qualitative benefits. The Safe
                     Drinking Water Act (SDWA)
                     amendments of 1996 require  that EPA
                     fully consider both quantifiable and
                     non-quantifiable benefits that result
                     from drinking water regulations.
                      The first step in the benefits
                     evaluation process is to consider the
                     adverse health effects that may be
                     expected to decrease with a reduction in
                     the concentrations of arsenic in drinking
                     water. Arsenic has many health effects,
                     both cancer and non-cancer. Section III.
                     discusses these health effects.
                      As discussed in section VIII.A.,
                     treatment for arsenic removal may add
                     or remove other contaminants. Using
                     chlorine or other oxidants may increase
 risk from disinfection by-products. On
 the other hand, treatments put in place
 for arsenic may incidentally reduce the
 risk from other co-occurring
 contaminants.

 A. Monetized Benefits of Avoiding
 Bladder Cancer
   Reducing arsenic levels in tap water
 will reduce the risks of suffering the
 adverse health effects described in the
 previous sections. In 1999 the National
 Research Council examined several risk
 distributions for male bladder cancer in
 42 villages in Taiwan with arsenic
 ranging from 10 to 934 ug/L, grouping
 arsenic exposure by village. Previous
 scientific studies analyzed risk using
 less specific exposure categories, which
 can obscure "the true shape of the dose
 response curve (NRC 1999, page 273)."
 Risk assessments for other adverse
 health effects have not been as
 thoroughly addressed.
   To monetize bladder cancer benefits,
 EPA calculated the number of cases
 potentially avoided based on the NRC
 bladder cancer risk analyses. The cases
 are evaluated in terms of the economic
 benefits associated with avoiding the
 cancer cases.
   In addition to the monetized benefits
 of avoiding bladder cancer, EPA has
 chosen to monetize the potential
 benefits of avoided lung cancer, using a
 "What If" analysis based on statements
 in the NRC report (see section  X.B for
 applying the "what-if" scenario to lung
 cancer).

 1. Risk Reductions: The Analytic
 Approach
  EPA applied the 1999 NRC bladder
 cancer risk assessment to U.S.  males
 and females. The following sections
 explain how we calculated risk
 reductions for populations exposed to
 MCL options of 3 ug/L and above. The
 approach for this analysis included five
 components. First, EPA used data from
 the recent EPA water consumption
 study. This study is described  in section
X.A.2. Second, Monte Carlo simulations
 (section X.A.3) were used to develop
relative exposure factors (section X.A.4).
Third, arsenic occurrence estimates
were used to identify the population
exposed to levels above 3 ug/L. Fourth,
NRC risk distributions were chosen for

-------
                 Federal Register/Vol.  65,  No. 121/Thursday, June 22, 2000/Proposed Rules
                                                                     38941
the analysis. Fifth, EPA developed
estimates of the risks faced by exposed
populations using Monte Carlo
simulations, using the relative exposure
factors, occurrence, and NRC risk
distributions mentioned above. These
components of the analysis are
described in the following sections.

2. Water Consumption
  EPA recently updated its estimates of
personal (per capita) daily average
estimates of water consumption
("Estimated per Capita Water
Consumption in the United States,"
EPA 2000a). The estimates used data
from the combined 1994, 1995, and
1996 Continuing Survey of Food Intakes
by Individuals (CSFII), conducted by the
U.S. Department of Agriculture (USDA).
The CSFII is a complex, multistage area
probability sample of the entire U.S. and
is conducted to survey the food and
beverage intake of the  U.S. Estimates of
water consumed include direct water,
indirect water and total water (Table X-
1). "Direct" water is tap water
consumed directly as a beverage.
"Indirect" water is defined as water
added to foods and beverages during
final preparation at home or by food
service establishments such as school
cafeterias and restaurants. For the
purpose of the report, indirect water did
not include "intrinsic" water which
consists of water found naturally in
foods (biological water) and water
added by commercial food and beverage
manufactures (commercial water).
"Total" water refers to combined direct
and indirect water consumption.
                                    TABLE X-1 .—SOURCE OF WATER CONSUMED
Source
Community Tap 	 	 	
Well Tap 	
Total 	 	 	

Direct
(drinking)
X
X
x

Indirect (from
food and
beverages)
X
X
x

Bottled water


X

  Per capita water consumption
estimates are reported by source.
Sources include community/tap water,
bottled water, and water from other
sources, including water from
household wells and rain cisterns, and
household and public springs. For each
source, the mean and percentiles of the
distribution of average daily per capita
consumption are reported. The
estimates are based on an average of 2
days of reported consumption by survey
respondents.
  The estimated mean daily average per
capita consumption of community/tap
water by individuals in the U.S.
population is 1 liter/person/day. For
total water, which includes bottled
water, the estimated mean daily average
per capita consumption is 1.2 liters per/
person/day. These estimates of water
consumption are based on a sample of
15,303 individuals in the 50 States and
the District of Columbia. The sample
was selected to represent the entire
population of the U.S. based on 1990
census data.
  The estimated 90th percentile of the
empirical distribution of daily average
per capita consumption of community/
tap water for the U.S. population is 2.1
liters/person/day; the corresponding
number for daily average per capita
consumption of total water is 2.3 liters/
person/day. In other words, current
consumption data indicate that 90
percent of the U.S. population
consumes up to approximately 2 liters/
person/day, which is the amount many
federal agencies use as a standard
consumption value.
  Water consumption estimates for
selected subpopulations in the U.S. are
described in the analysis, including per
capita water consumption by source for
gender, region, age categories, economic
status,  race, and residential status and
separately for pregnant women,
lactating women, and women in   :
childbearing years. The water
consumption estimates by age were
used in the computation of the relative
exposure factors discussed in the
section X.A.4.
  These water consumption numbers
differ somewhat from previous
estimates reported in earlier studies.
The mean per capita daily intake of, total
tap water, as estimated from the 1977—
78 USDA's Nationwide Food
Consumption Survey, was 1.193 liters/
person/day (reported by Ershow and
Cantor in 1989). Based on the 1977T78
study, the estimated percentile
corresponding to 2 liters per day
consumed is the 88th.

3. Monte Carlo Analysis

  Monte Carlo analysis is a technique
for analyzing problems where there are
a large  number of combinations of input
values  that are too large to calculate for
every possible result. A random number
generator is used to generate numbers
that correspond to assumptions about
the distribution or likelihood of various
input values. For each set of random
input values a single outcome is
calculated. As the simulation runs, the
outcome is recalculated for each new set
of input values and continues until ia
stopping criterion is reached. The
accuracy of this technique, like other
statistical techniques, depends on the
accuracy of the underlying assumptions
about the distribution of input values; it
does not resolve the uncertainty behind
the assumptions. For the risk
distributions calculated in this report,
the simulations were carried out 2,000
times. For each simulation, a relative
exposure factor, occurrence estimate,
and individual risk estimate were
calculated. These calculations resulted
in estimates of the risks faced by
populations exposed to arsenic
concentrations in their drinking water.
The underlying risk distribution are
described in the following sections.
4. Relative Exposure Factors

  EPA used models to integrate the new
drinking water consumption study
information into the benefits analysis.
We used distributions for both
community/tap water and total water
consumption because the community
water/tap water estimates may
underestimate actual tap water
consumption. In this analysis, we
combined the water consumption data
with data on population weight from the
U.S. Census. The weight data included
a mean and a distribution of weight for
male and females on a year-to-year basis
throughout a lifetime. Monte Carlo
analysis generated male and female
relative exposure factors (REFs) for each
of the broad age categories used in the
water consumption study. Lifetime male
and female relative exposure factors
were then estimated, where the factors
show the sensitivity of exposure to an
individual weighing 70 kilograms and
consuming 2 liters of water per day.
These life-long REFs can be directly
multiplied by the average drinking
water consumption to provide estimates
of individual lifetime consumption

-------
38942
Federal Register/Vol. 65, No. 121/Thursday, June  22,  2000/Proposed Rules
practices. The REFs provide a means to
incorporate information on various age
groups, for example children, into the
analysis, as weight and water
consumption vary among age groups.
The means and variances of the REFs
derived from this analysis were: for
community water consumption (0.60,
0.37 males; 0.64, 0.36 females), for total
water consumption (0.73, 0.39 males;
0.79, 0.37 females).
5. NRC Risk Distributions
  While the NRC's work did not
constitute a formal risk analysis, they
did examine many statistical issues
(e.g., measurement errors, age-specific
probabilities, body weight, water
consumption rate, comparison
populations, mortality rates, choice of
model) and provided a starting point for
additional EPA analyses. The report
noted that "poor nutrition, low
selenium concentrations in Taiwan,
genetic and cultural characteristics, and
arsenic intake from food" were not
accounted for in their analysis (NRC,
1999, pg. 295).
  In its 1999 report, "Arsenic in
Drinking Water,"  the NRC analyzed
bladder cancer risks using data from
Taiwan. In addition, NRC examined
evidence from human epidemiological
studies in Chile and Argentina, and
concluded that risks of bladder and lung
cancer were comparable to those "in
Taiwan at comparable levels of
exposure (NRC 1999, page 7)." The NRC
also examined the implications of
applying different mathematical
procedures to the newly available
Taiwanese data for the purpose of
characterizing bladder cancer risk.
These risk distributions are based on
bladder cancer mortality data in
Taiwan, in a section of Taiwan where
arsenic concentrations in the water are
very high by comparison to those in the
U.S. It is also an area of very low
incomes and poor diets, and the
availability and quality of medical care
is not of high quality, by U.S. standards.
In its estimate of bladder cancer risk, the
Agency assumed that within the
Taixvanese study area, the risk of
contracting bladder cancer was
relatively close to the risk of dying from
bladder cancer (that is, that the bladder
cancer incidence rate was equal to the
bladder cancer mortality rate). At the
time the study data were collected the
chances of surviving were probably poor
for individuals diagnosed with bladder
cancer. We do not have data, however,
on the rates of survival for bladder
cancer in the Taiwanese villages in the
study and at the time of data collection.
We do know that the relative survival
rates for bladder cancer in developing
                     countries overall ranged from 23.5% to
                     66.1% in 1982-1992 ("Cancer Survival
                     in Developing Countries," International
                     Agency for Research on Cancer, World
                     Health Organization, Publication No.
                     145,1998). We also have some
                     information on annual bladder cancer
                     mortality and incidence for the general
                     population of Taiwan in 1996. The age-
                     adjusted annual incidence rates of
                     bladder cancer for males and females,
                     respectively, were 7.36 and 3.09 per
                     100,000, with corresponding annual
                     mortality rates of 3.21 and 1.44 per
                     100,000 (correspondence from Chen to
                     Herman Gibb, January 3, 2000).
                       Assuming that the proportion of
                     males and females in the population is
                     equal, these numbers imply that the
                     mortality rate for bladder cancer in the
                     general population of Taiwan, at
                     present, is 45%. Since survival rates
                     have most likely improved over the
                     years since the original Taiwanese
                     study, this number represents a lower
                     bound on the survival rate for the
                     original area under study (that is, one
                     would not expect a higher rate of
                     survival in that area at that time). This
                     has implications for the bladder cancer
                     risk estimates from the Taiwan data. For
                     this estimate we have made the
                     assumption that all bladder tumors in
                     the study area in Taiwan were fatal. If
                     there were any persons with bladder
                     cancer who recovered and died from
                     some other cause, then our estimate
                     underestimated risk; that is, there were
                     more cancer cases than cancer deaths.
                     Based on the above discussion, we think
                     bladder cancer incidence could be no
                     more than 2 fold bladder cancer
                     mortality; and that an 80% mortality
                     rate would be plausible. In the benefits
                     analysis we include estimates using an
                     assumed mortality rate ranging from
                     80% to 100%.
                       In the U.S. approximately one out of
                     four individuals who is diagnosed with
                     bladder cancer actually dies from
                     bladder cancer. The mortality rate for
                     the U.S. is taken from a cost of illness
                     study recently completed by EPA (US
                     EPA, 1999a). For those diagnosed with
                     bladder cancer at the average age  of
                     diagnosis (70 years), the probability for
                     dying of that  disease during each year
                     post-diagnosis were summed over a 20-
                     year period to obtain the value of 26
                     percent. Mortality rates for U.S. bladder
                     cancer patients have  decreased overall
                     by 24 percent from 1973 to 1996.
                       In the NRC report, Table 10-11 shows
                     excess risk estimates based on the
                     Taiwanese male bladder cancer, using a
                     Poisson regression model; a risk at the
                     current MCL of 50 |ig/L is in the range
                     of 1 to 1.347 per 1,000. Table 10-12
                     presents excess lifetime risk estimates
for bladder cancer in males calculated
using EPA's 1996 proposed revision^ to
the cancer guidelines (US EPA 1996b).
EPA selected four of these distributions
as representative of the risks and
uncertainty involved (selecting     ;
relatively high and relatively low
estimates).  These distributions (mean
1.049, 95% upper confidence limit
1.347; mean 0.731, 95% upper
confidence limit 0.807; mean 1.237,
95% upper confidence limit 1.548; and
mean 1.129, 95% upper confidence
limit 1.229), were used in the EPA
Monte Carlo simulations. All of these
risk distributions are linear in the mean,
and thus may be conservative
assumptions, as the NRC report
suggested the true relationship may be
sublinear. If the true relationship is
sublinear, i.e., lower than the straight
line from 50 ug/L to zero, the true risks
at levels below 50 (ig/L are being
overestimated. Other factors which
might lower the true risk include the
use of grouped data, the high Taiwanese
dietary intake of arsenic, and the
amount of selenium in the Taiwanese
diet.
  NRC concluded that the present MCL
in drinking water of 50 fJ.g/L does not
achieve EPA's goal for public health and
requires downward revision. EPA did
not request nor did NRC recommend a
specific new MCL level.

6. Estimated Risk Reductions
  Estimated risk reductions for bladder
cancer at various MCL levels were  :
developed using Monte Carlo
simulations. The inputs to the
simulations were the distributions of
relative risk factors (described in section
X.A.4.), distributions of occurrence for
arsenic levels at 3 |ig/L and above, and
bladder cancer risk distributions from
the National Research Council report.
The relative risk factor and occurrence
distributions  represent primarily
population and occurrence variability,
while the cancer risk distributions
represent primarily uncertainty about
the true risk. Thus the  combined
distributions  reflect both variability and
uncertainty. These combined
distributions  provide our best estimate
of the actual risks faced by the exposed
population, including the percentiles of
the population facing various levels of
risk.
  Estimated risk reductions for bladder
cancer at various MCL levels are shown
in Tables X-2a and X-2b. Table X-2a
uses data on community water      .
consumption from the  new EPA study;
Table X—2b uses data on total water
consumption from the  study.
Populations at or above 10 ~4 risk levels
are shown in  Tables X-3a and X-3b.

-------
                  Federal  Register/Vol. 65,  No. 121/Thursday,  June 22, 2000/Proposed Rules
                                                                      38943
The after treatment occurrence
distributions were assumed to reflect
treatment to 80% of the MCL level. The
latter assumption is made since water
systems tend to treat below the MCL
level in order to provide a margin of
safety.
  As shown in Table X—2a, bladder
cancer risks at the 90th percentile of
water intake, for the various  MCL
options under consideration, range from
a multiple of 10 ~5 at 3 jig/L (4-6 x
10~s) to a multiple of 10~4 at 20 ug/L
(1.2-2.4 x 10~4). At 5 |ig/L , the 90th
percentile level is 6-11 x 10~5; at 10 |j.g/
L the 90th percentile is 1.0-1.7 x 10~4.
Table X—2b presents similar
information. The risk estimates in Table
X—2b are somewhat higher than those in
Table X—2a because total water
consumption is higher than community
water consumption. Since there is  ;
uncertainty about these numbers, it is
assumed that the range 1-1.5 x 10~4
represents a risk level of essentially
10 ~4. It is then assumed that risks above
1.5 x 10~4 represent risks greater than
10~4. Table X-3a gives information:
about percentages of the exposed
populations and the number of people
exposed at 10~4 risk levels and above,
and, using the stated definition for an
over 10~4 risk level, above 10~4. The
numbers in this table show that at ah
MCL of 3 (J.g/L, only a small number (not
quantifiable) face a risk level of greater
than 10 ~4. At an MCL of 5 ug/L, about
0.3 to 0.8 million face such risk levels,
at an MCL of 10 ug/L, 0.8 to 4 million,
and at an MCL of 20 |ig/L, about 2.4 to
6.4 million would be at such levels.
Table X-3b gives similar information
using total water consumption data. The
mean bladder cancer risks for the
exposed population at the various MCL
options, after treatment, are shown in
Tables X-4a and X-4b. These mean
risks are used in the computation of the
number of cases avoided, used later in
the benefits evaluation section.
 TABLE X-2A— BLADDER CANCER INCIDENCE RISKS 1 FOR HIGH PERCENTILE U.S. POPULATIONS EXPOSED AT OR ABOVE
                    MCL OPTIONS, AFTER TREATMENT2 (COMMUNITY WATER CONSUMPTION DATAS)
MCL (ug/L)
3 	
5 	
10 	
20 	

85th :
3 2 5 4 x 10~5
5 3—9 3 X'10~5
88 1 49 x 10~4
1 2—1 96 x 10~4

90th
A fi v m~5
6 11 x 10~s
1 0—1 7 x 10~4
14 2 4 x 10~4

95th

7 ^ 1*3 n v in— 5



  1 See Sections III.C. and D. for a description of other health effects, and Section X.B. for "What-if?" estimates of magnitude for lunq cancer
risks.
  2 The bladder cancer risks presented in this table provide our "best" estimates at this tijme. Actual risks could be lower, given the various un-
certainties discussed, or higher, as these estimates assume a 100% mortality rate. An 80% mortality rate is used in the computation of upper
bound benefits.
  3 Discussed in Section X.A.2.                                               !

 TABLE X-2B.—BLADDER CANCER INCIDENCE RISKS 1 FOR HIGH PERCENTILE U.S. POPULATIONS EXPOSED AT OR ABOVE
                      MCL OPTIONS, AFTER TREATMENTS (TOTAL WATER CONSUMPTION DATAS)
MCL (jj.g/L)
3 	
5 	
10 	
20 	
85th
3 8- 6 4 x 10~5
6 3-10 5 x 10~5
1 02 1 8 x *10~4
1.4-2.34x10-"
90th
4—7 y 1O — 5
7 12 x 10~5
i p p n v m— 4
1. 7-2.8 x10~4
95th

8c 1AR v 1fl — 5

2.17-3.56 x10-4
  •"See'Sections III.C. and D. for a description of other health effects, and Section X.B. for "What-if?" estimates of magnitude for luna cancer
risks.                                                                    ;                                    a
  2 The bladder cancer risks presented in this table provide our "best" estimates at this time. Actual risks could be lower given the various un-
certainties discussed, or higher, as these estimates assume a 100%  mortality rate. An 80% mortality rate is used in the computation of upper
bound benefits.
  3 Discussed in Section X.A.2.

TABLE X-3A.—PERCENT OF EXPOSED POPULATION AT 10~4 RISK OR HIGHER FOR BLADDER CANCER INCIDENCE 1 AFTER
                              TREATMENT2  (COMMUNITY WATER CONSUMPTION DATA3)
MCL (ug/L)
3 	
5 	
10 	
20 	

Percent at
10~4 risk or
higher !
<1 2 6
1 5—12
11 34
19 5-41

Population at
10~4 risk or
higher (mil-
lions)
<0 3—0 7
0 4—3 2
2991
5211

Percent over
10-4*
<1

-------
38944
Federal Register/Vol. 65,  No. 121/Thursday, June  22, 2000/Proposed Rules
TABLE X-3B— PERCENT OF EXPOSED POPULATION AT 10~4 RISK OR HIGHER FOR BLADDER CANCER INCIDENCE 1 AFTER
                                  TREATMENT2 (TOTAL WATER CONSUMPTION DATA3)                               '
MCL (ng/L)
3 . 	
5 ., 	 	 	
10 	
20 , 	 , 	
Percent at
10-" risk or
higher
<1-3
3-18
16-50
26-53
Population at
1C-4 risk or
higher (mil-
lions)
<0.3-0.8
0.8-4.8
4.3-13.4
7-14.2
Percent over
10-4*
<1
<1-4
4-23
13-33
Populati
over 10
(million:
0.3
1.1
3.5
Dn
-4
3)
i
-1.1
-6.2
-8.9
  1 See Sections III.C. and D. for a description of other health effects, and Section X.B. for "What-if?" estimates of magnitude for lung cancer

  8 The percents presented in this table provide our "best" estimates at this time. Actual percents could be lower, given the various uncertainties
discussed, or higher, as these estimates assume a 100% mortality rate. An 80% mortality rate is used in the computation of upper bound bene-
fits.
  3 Discussed in Section X.A.2.
  •Where over 10'4 means 1.5 x 10~4 or above.
  t Too low to calculate.
TABLE  X-4A.—MEAN BLADDER CAN-
   CER  INCIDENCE  RISKS'  FOR  U.S.
   POPULATIONS   EXPOSED   AT   OR
   ABOVE   MCL   OPTIONS,   AFTER
   TREATMENT2  (COMMUNITY  WATER
   CONSUMPTION DATAS)

3 .....
5 	
10 ...
?n
MCL
(P/L)




Mean exposed popu-
lation risk
2.1-3.6 x 10-s
3.6-6.1 x 10-5
5.5-9.2 x 10-'
6.9-11. 6 x 10-5
  i See Sections III.C. and D. for a description
of other health effects, and Section X.B. for
"What-if?" estimates of magnitude  for lung
cancer risks.
  1 The bladder cancer risks presented in this
table provide our "best" estimates at this time.
Actual risks could be lower,  given the various
uncertainties discussed,  or  higher, as these
estimates assume a 100% mortality rate. An
80% mortality rate is used in the computation
of upper bound benefits.
  » Discussed in Section X.A.2.

TABLE  X-4B.—MEAN BLADDER  CAN-
  CER  INCIDENCE  RISKS' FOR  U.S.
  POPULATIONS   EXPOSED   AT   OR
  ABOVE    MCL   OPTIONS,    AFTER
  TREATMENT2  (TOTAL WATER  CON-
  SUMPTION DATA3)

3 	
5 	
10 ...
?n ...
MCL
&t/L)




Mean exposed popu-
lation risk
2.6-4.5 x 10-5
4.4-7.5 x 10~s
6.7-11. 4 x 10-5
8.4-13.9x10-5
                       B. "What if?" Scenario for Lung Cancer
                       Risks
                        The NRG report "Arsenic in Drinking
                       Water" states that "some studies have
                       shown that excess lung cancer deaths
                       attributed to arsenic are 2-5 fold greater
                       than the excess bladder cancer deaths
                       (NRG, 1999, pg. 8)." Two-to-five fold
                       greater would be 3.5 fold greater on
                       average. Also in the U.S. the mortality
                       rate from bladder cancer is 26% and the
                       mortality rate of lung cancer is 88%.
                       This suggests that if the risk of
                       contracting lung cancer were identical
                       to the risk of contracting bladder cancer,
                       one would expect 3.4 times the number
                       of deaths from lung cancer as from
                       bladder cancer. Since these numbers are
                       essentially the same, it seems reasonable
                       to assume that the risk of contracting
                       lung cancer is essentially the same as
                       the rate of contracting bladder cancer,1
                       in the context of this "what-if" scenario.
                       If the risk of contracting lung cancer
                       from arsenic in drinking water is
                       approximately equal to the risk of
                       contracting bladder cancer, then the
                       combined risk estimates of contracting
                       either bladder or lung cancer would be
                       approximately double the  risk estimates
                       presented in the previous tables.
                         EPA anticipates that a peer-reviewed
                       quantification of lung cancer risk from
                       arsenic exposure may be available
                       between the time of proposal and
   i See Sections III.C. and D. for a description
 of other health effects, and Section X.B.  for
 "What-if?" estimates of magnitude for lung
 cancer risks.
   2 The bladder cancer  risks presented in this
 table provide our "best" estimates at this time.
 Actual  risks could be lower, given the various
 uncertainties discussed, or higher, as these
 estimates assume a 100% mortality rate.  An
 80% mortality rate is used  in the computation
 o( upper bound benefits.
   > Discussed In Section X.A.2.
                         11f "X" is the probability of contracting bladder
                       cancer, then 0.26X is the probability of mortality
                       from bladder cancer. If lung cancer deaths are 2 to
                       5 times as high as bladder cancer, then they are, on
                       average, 3.5 times as high and the average
                       probability of mortality from lung cancer would be
                       3.5 times 0.26X, or 0.91X, Since we also know that
                       there is a 88% mortality rate from lung cancer, then
                       if the probability of contracting lung cancer is "Y,"
                       the probability of mortality from lung cancer can
                       also  be represented as 0.88Y. Setting the two ways
                       of deriving the probability of mortality from lung
                       cancer equal, or 0.91X = 0.88Y, one can solve for
                       Y (Y= (0.91/0.88) X). Thus Y is approximately equal
                       to X, and the rate of contracting lung cancer is
                       approximately the same as the rate of contracting
                       bladder cancer.
promulgation. If so, EPA will make this
information available for public
comment through a Notice of Data  ,
Availability (NODA) and consider the
analysis and public comment for the
final rulemaking.

C. Evaluation of Benefits
  The evaluation stage in the analysis of
risk reductions involves estimating the
value of reducing the risks. Background
information on the economic concepts
that provide the foundation for benefits
valuation, and the methods that are
typically used by economists to
monetize the value of risk reductions,
such as wage-risk, cost of illness, and
contingent valuation studies are    ;
provided in the RIA. The following
sections describe the use of these  ..,
techniques to estimate the value of the
risk reductions attributable to the
regulatory options for arsenic in
drinking water. Described first  is the
approach for valuing the reductions in
fatal risks; described next is the
approach for valuing the reductions in
nonfatal risks.
  The benefits  calculated for this
proposal are assumed to begin to accrue
on the effective date of the rule and are
based on a calculation referred to as the
"value of a statistical life" (VSL),
currently estimated at $5.8 million. The
VSL is an average estimate derived from
a set of 26 studies estimating what
people are willing to pay to avoid the
risk of premature mortality. Most of
these studies examine willingness to
pay in the context of voluntary
acceptance of higher risks of immediate
accidental death in the workplace in
exchange for higher wages. This value is
sensitive to differences in population
characteristics  and perception of risks
being valued.
   For the present rulemaking analysis,
which evaluates reduction in premature
mortality due to carcinogen exposure,
some have argued that the Agency

-------
                   Federal Register/Vol. 65, No. 121/Thursday, June 22, 2000/Proposed Rules
                                                                        38945
 should consider an assumed time lag or
 latency period in these calculations.
 Latency refers to the difference between
 the time of initial exposure to
 environmental carcinogens and the
 onset of any resulting cancer. Use of
 such an approach might reduce
 significantly the present value estimate.
 EPA is interested in receiving comments
 on the extent to which the presentation
 of more detailed information on the
 timing of cancer risk reductions would
 be useful in evaluating the benefits of
 the proposed rule.
   Latency is one of a number of
 adjustments or factors that are related to
 an evaluation of potential benefits
 associated with this rule, how those
 benefits are calculated, and when those
 economic benefits occur. Other factors
 which may influence the estimate of
 economic benefits associated with
 avoided cancer fatalities include (1) a
 possible "cancer premium" (i.e., the
 additional value or sum that people may
 be willing to pay to  avoid the
 experiences of dread, pain and
 suffering, and diminished quality of life
 associated with cancer-related illness
 and ultimate fatality); (2) the
 willingness of people to pay more over
 time to avoid mortality risk as their
 income rises; (3) a possible premium for
 accepting involuntary risks as opposed
 to voluntary assumed risks; (4) the
 greater risk aversion of the general
 population compared to the workers in
 the wage-risk valuation studies; (5)
 "altruism" or the willingness of people
 to pay more to reduce risk in other
 sectors of the population; and (6) a
 consideration of health status and life
 years remaining at the time of premature
 mortality. Use of certain of these factors
 may significantly increase the present
 value estimate. EPA therefore believes
 that adjustments should be considered
 simultaneously. The Agency also
 believes that there is currently neither a
 clear consensus among economists
 about how to simultaneously analyze
 each of these adjustments nor is  there
 adequate empirical data to support
 definitive quantitative estimates for all
 potentially significant adjustment
 factors. As a result, the primary
 estimates of economic benefits
 presented in the analysis of this
 proposed rule rely on the unadjusted
 estimate. However, EPA solicits
 comment on whether and how to
 conduct these potential adjustments to
 economic benefits estimates together

 2 Some of the key sources of bias include the
characteristics of the averted risks (whether they are
voluntary or involuntary, ordinary or catastrophic,
delayed or immediate, natural or man-made, etc.);
the demographic characteristics of the group
 with any rationale or supporting data
 commenters wish to offer. Because of
 the complexity of these issues, EPA will
 ask the Science Advisory Board (SAB)
 to conduct a review of these benefits
 transfer issues associated with economic
 valuation of adjustments in mortality
 risks. Consistent with the
 recommendations of the SAB, and  ,
 subject to resolution of any technical
 problems, EPA will attempt to develop
 and present an estimate of the latency
 structure as a part of the analysis of the
 final rule, with prior solicitation of
 comment, if appropriate.           [

 I. Fatal Risks and Value of a Statistical
 Life (VSL)                         i
   To estimate the monetary value of
 reduced fatal risks (i.e., risks of
 premature death from cancer) predicted
 under different regulatory options, value
 of a statistical life (VSL) estimates are
 multiplied by the number of premature
 fatalities avoided. VSL does not refer, to
 the value of an identifiable life, but  ;
 instead to the value of small reductions
 in mortality risks in a population. A
 "statistical" life is thus the sum of small
 individual risk reductions across an :
 entire exposed population.          i
   For example, if 100,000 people would
 each experience a reduction of I/
 100,000 in their risk of premature death
 as the result of a regulation, the
 regulation can be said to "save" one
 statistical life (i.e., 100,000 x 1/100,000).
 If each member of the population of
 100,000 were willing to pay $20 for the
 stated risk reduction, the corresponding
 value of a statistical life would be $2
 million (i.e., $20 x 100,000). VSL
 estimates are appropriate only for
 valuing small changes in risk; they are
 not values for saving a particular     :
 individual's  life.
   Of the many VSL studies, the Agency
 recommends using estimates from 26
 specific studies that have been peer
 reviewed and extensively reviewed
 within the Agency (US EPA, 1997f).
 These estimates, which are derived from
 wage-risk and contingent valuation  ;
 studies, range from $0.7 million to $16.3
 million and approximate a Wiebull
 distribution with a mean of $5.8 million
 (in 1997 dollars). To value the changes
 in fatal risks  associated with the arsenic
 regulation, the mean estimate of $5.8
 million is used.                     :
  Use of these estimates to value the
 averted risks of premature death     :
 associated with the regulatory options
 for arsenic is an example of the benefit
 transfer technique, since the subject of
 most of the studies (i.e., job-related
 risks) differs from the fatal cancer risks
 averted by the regulatory options.
 Applying these studies results in several
 sources of potential bias (see latency
 discussion in section X.C.); however,
 quantitative adjustments to address
 these biases generally have not been
 developed or adequately tested and may
 be counterbalancing.2 EPA notes the
 uncertainties in the cost-benefit
 analyses, as required by section
 1412(b)(3)(C)(i)(VII) of SDWA, and
 requests comment on alternate
 approaches.

 2. Nonfatal Risks and Willingness To
 Pay (WTP)

   Estimates of the willingness to pay to
 avoid treatable, nonfatal cancers are the
 ideal economic measures used for
 evaluation of the reduction in nonfatal
 risks.  However this information is not
 available for bladder cancer.
 Willingness to pay (WTP) data to avoid
 chronic bronchitis is available, however,
 and has been used before by EPA (the
 microbial/disinfection by-product
 (MDBP) rulemaking) as a surrogate to
 estimate the WTP to avoid non-fatal
 bladder cancer. The use of such WTP
 estimates is supported in the SDWA, as
 amended, at section 1412{b)(3)(C)(iii):
 "The Administrator may identify valid
 approaches for the measurement and
 valuation of benefits under this
 subparagraph, including approaches to
 identify consumer willingness to pay for
 reductions in health risks from drinking
 water  contaminants." The WTP central
 tendency estimate of $536,000, to avoid
 chronic bronchitis, is used to monetize
 the benefits of avoiding non-fatal
 bladder cancers (Viscusi et al., 1991).
   EPA has also developed cost of illness
 estimates for bladder cancer, as reported
 in Table X-5. These estimates of direct
 medical costs are derived from a study
 conducted by Baker et al., (as cited in
 US EPA, 1997f) which uses data from a
 sample of Medicare records for 1974-
 1981. These data include the total
 charges for inpatient hospital stays,
 skilled nursing facility stays, home
 health agency charges, physician
 services, and other outpatient and
 medical services. EPA combined these
 data with estimates of survival rates and
 treatment time periods to determine the
 average costs of initial treatment and
 maintenance care for patients who do
not die of the disease.
affected (e.g., age, income); the lag between
exposure and diagnosis or incidence of the disease
(latency) as well as between incidence and death;
the baseline health status (i.e., whether a person is
currently in good health) of affected individuals;
and the presence of altruism (i.e., individual's
willingness to pay to reduce risks incurred by
others) (US EPA, 1997f).

-------
38946
Federal Register/Vol.  65,  No. 121/Thursday, June 22,  2000/Proposed Rules
     TABLE X-5.—LIFETIME AVOIDED MEDICAL COSTS FOR SURVIVORS (PRELIMINARY ESTIMATES, 1996 DOLLARS 1)
Type of cancer
Bladder 	

Date data
collected
1974-1981

Number of cases studied
5% of 1 974 Medicare patients
(sample from national statis-
tics).
Estimated survival rate
26 percent (after 20 years) 	

Mean value per nonfatal
case 2 ;
$179,000 (for typical individual
diagnosed at age 70)
  'These costs increase by 2.8 percent when inflated to 1997 dollars, based on the consumer price index for the costs of medical commodities
and services.
  8Undiscounted costs.
  Source: US EPA, 1999a.
D. Estimates of Quantifiable Benefits of
Arsenic Reduction

  Benefits estimates for avoided cases of
bladder cancer were calculated using
mean population risk estimates at
various MCL levels. Table X-6 gives the
mean populations risk estimates used,
which are a composite of the mean
population risk estimates discussed
earlier. Lifetime risk estimates were
converted to annual risk factors, and
applied to the exposed population to
determine the number of cases avoided.
These cases were divided into fatalities
and non-fatal cases avoided, based on
survival information. The avoided
premature fatalities were valued based
on the VSL estimates discussed earlier,
as recommended by EPA current
guidance for cost/benefit analysis. The
avoided non-fatal cases were valued
based on the willingness to pay
                      estimates for the avoidance of chronic
                      bronchitis. The upper bound estimates
                      have been adjusted upwards to reflect
                      an 80% mortality rate, which is a
                      plausible mortality rate for the area of
                      Taiwan during the Chen .study.
                        The "What if?" scenario for lung
                      cancer benefits (described in section
                      X.B.) was used to estimate potential
                      benefits for avoided cases of lung
                      cancer. This scenario is based on the
                      statement in the NRC report "Arsenic in
                      Drinking Water" that "some studies
                      have shown that excess lung cancer
                      deaths attributed to arsenic are 2-5 fold
                      greater than the excess bladder cancer
                      deaths (NRC, 1999, pg. 8)." It was
                      shown in section X.D that the statement
                      implies (if it were accurate for the  U.S.),
                      that, because of the relative U.S.
                      mortality rates for bladder and lung
                      cancer, the rate of contracting lung
                      cancer could be essentially the same as
the rate of contracting bladder cancer.
This would double the number of
cancer cases avoided, for both low and
high estimates. The potential monetized
benefits for lung cancer would be
several times higher than those for
bladder cancer, due to the higher
number of fatalities involved with lung
cancer.
  Another way of considering the
addition of lung cancer effects would be
to estimate the potential benefits from
avoided cases of lung cancer using the
2-5 times range for fatalities (that is,
taking the expected number of bladder
cancer fatalities and multiplying them
by 2 and then 5 to obtain a range of lung
cancer fatalities, and then factoring in
non-fatal cases).
   Benefits (and costs) are assumed to
accrue on the effective date of the rule.
Table X—7 displays the results.
 TABLE X-S.-MEAN BLADDER CANCER INCIDENCE RISKS 1 FOR U.S. POPULATIONS EXPOSED AT OR ABOVE MCL OPTIONS,
                             AFTER TREATMENTS (COMPOSITE OF TABLES X-SA AND X-5B)
MCL (ng/L)
3 	
	 	 	 	 	 	 	 	 	
•jn 	 	 	 • 	
20 	
Mean exposed
population risk
2.1-4.5*10-5
3.6-7.5x10-5
5.5-11.4x10-5
6.9-13.9x10 5
   1 See Sections III.C. and D. for a description of other health effects, and Section X.B. for "What-if?" estimates of magnitude for cancer risks.
   «The bladder cancer risks presented in this table provide our "best" estimates at this time. Actual risks could be lower, given the various un-
 certainties discussed, or higher, as these estimates assume a 100%  mortality rate. An 80% mortality rate is used in the computation of upper
 bound benefits.

             TABLE X-7— ESTIMATED COSTS AND BENEFITS FROM REDUCING ARSENIC IN DRINKING WATER
                                                    [Millions, 1999]
Arsenic level (u.g/1)
3 	
5 	 	 	
10 	 	 	

Total national
costs to
CWSs1
$643.1-753
377.3-441.8
163.3-191.8
Total national
costs to CWSs
and
NTNCWSs2
$644.6-756.3
378.9-444.9
164.9-194.8
Total bladder
cancer health
benefits3
$43.6-104.2
s(79)
31 .7-89.9
5(64.3)
17.9-52.1
5(37)
"What if" scenario4 and potential non-quantified benefiis
"What If" lung
cancer health
benefits esti-
mates
$47.2-448
6(213.4)
35-384
s(173.4)
19.6-224
6(100)
Potential non-quantifiable health benefits
Skin Cancer.
Kidney Cancer.
Cancer of the Nasal Passages.
Liver Cancer.
Prostate Cancer.
Cardiovascular Effects.
Pulmonary Effects.
Immunological Effects.
Neurological Effects.

-------
                   Federal Register/Vol. 65, No. 121/Thursday,  June 22,  2000/Proposed  Rules
                                                                        38947
        TABLE X-7— ESTIMATED COSTS AND BENEFITS FROM REDUCING ARSENIC IN DRINKING WATER—Continued
                                                      [Millions, 1999]
Arsenic level (jj.g/1)
20 	

Total national
costs to
CWSsi
61.6-72.9
Total national
costs to CWSs
and
NTNCWSs 2
63.2-77.1
Total bladder
cancer health
benefits 3
7.9-29.8
5(19.8)
"What if" scenario4 and potential non-quantified benefits
"What If" lung
cancer health
benefits esti-
mates
8.8^-128
6(53.4)
Potential non-quantifiable health benefits
• Endocrine Effects.
• Reproductive and Developmental Effects.
    Costs include treatment monitoring, O&M, and administrative costs to CWSs and State costs for administration of water programs The lower
 number shows costs annualized at a consumption rate of interest of 3%, EPA's preferred approach. The higher number shows costs annuaHzed
 at 7%, which represents the standard discount rate preferred by OMB for benefit-cost analyses of government programs andregu°at ons
   2 Costs include treatment, monitoring, O&M, administrative costs to CWSs; monitoring and administrative costs to NTNCWSs; and State costs
 Tor aominisirstion ot WSTGT proQrsms.                                           ,
 h ?ThD^perut?°und efjimale in9ludes an adjustment to account for a possible mortality risk of 80%. It is possible that this risk could have been
 by Chen whtehls Tknown     lncreased beneflts- The actual risk dePends on the survival rate for bladder cancer in the area of Taiwan ftudlld

                          ^ fafal"bladde'; SSftSS.""8 ""**' """ ** riSkS 6f * *** ^9 cancer ^ aSS°Ciated With arsenic are
 rhntuH,anHtrt-f   sthndic?,tes-ttiefb(ludler ca-?cer health benefits assuming an 80% mortality rate for bladder cancer in the area of the
 Chen study and starting from the midpoint of the benefits range when mortality and incidence are assumed equivalent
 risk of^tTwadd1 parentheses is the midP°int of tne range and corresponds to an assumption that the risk of  fatal lung  cancer is 3.5 times the
 F. NDWAC Working Group (NDWAC,
 1998) on Benefits
   The National Drinking Water
 Advisory Council (NDWAC)
 recommends that:
   (1) EPA should focus its benefits
 analysis efforts primarily on assessing
 effects on human health, defining these
 effects as clearly as possible and using
 the best available data to value them. It
 is also recommended that EPA should
 also consider, where appropriate, taste
 and odor improvements, reduction of
 damage to water system materials,
 commercial water treatment cost
 reductions, benefits due to source water
 protection (e.g., ecological benefits and
 non-use benefits), and benefits derived
 from the provision of information on
 drinking water quality (e.g., a
 household's improved ability to make
 informed decisions concerning the need
 to test or filter tap water);
   (2) EPA should devote substantial
 efforts to better understanding the
 health effects of drinking water
 contaminants, including the types of
 effects, their severity, and affected
 sensitive subpopulations. Better
 information is also needed on exposures
 and the effects  of different exposure
 levels, particularly for contaminants
 with threshold effects. These efforts
 should pay particular attention to
 obtaining improved information
 concerning impacts on children and
 other sensitive populations;
  (3) EPA should clearly identify and
 describe the uncertainties in the benefits
 analysis, including descriptions of
 factors that may lead the analysis to
 significantly understate or overstate
total benefits. Factors that may have
 significant but indeterminate effects on
 the benefits estimates should also be
 described;
   (4) EPA should consider both
 quantified and non-quantified benefits
 in regulatory decision-making. The
 information about quantified and nbn-
 quantified (qualitative) benefits should
 be presented together in a format, such
 as a table, to ensure that decision-
 makers consider both kinds of
 information;
   (5) EPA should  consider incremental
 benefits and costs, total benefits and
 costs, the distribution of benefits and
 costs, and cost-effectiveness in
 regulatory decision-making. This   :
 information should be presented   '
 together in a format, such as a table, to
 ensure its consideration by decision-
 makers;
   (6) Whenever EPA considers     ;
 regulation of a drinking water
 contaminant, it should evaluate and
 consider, along with water treatment
 requirements to remove a contaminant,
 source water protection options to
 prevent such a contaminant from
 occurring. The full range of benefits of
 those options should be considered.:

 XI. Risk Management Decisions: MCL
 and NTNCWSs

 A. What Is the Proposed MCL?      !
  EPA is proposing an arsenic MCL of
 5 |ig/L and soliciting comments on
 options of 3ug/L, 10 ug/L, and 20 ug/L.
EPA is also asking that commenters .
provide their rationale and any
supporting data or information for the
option they prefer.
  The SDWA generally requires that:
EPA set the MCL for each contaminant
as close as feasible to the MCLG, based
on available technology and taking costs
 to large systems into account. The 1996
 amendments to the SDWA added the
 requirement that the Administrator
 determine whether or not the
 quantifiable and nonquantifiable
 benefits of an MCL justify the
 quantifiable and nonquantifiable costs
 based on the Health Risk Reduction and
 Cost Analysis (HRRCA) required under
 section 1412(b)(3)(C). The 1996 SDWA
 amendments also provided new
 discretionary authority for the
 Administrator to set an MCL less
 stringent than the feasible level if the
 benefits of an MCL set at the feasible
 level would not justify the costs (section
 1412(b)(6)). This proposal to establish
 an MCL for arsenic of 5 ug/L is the first
 time EPA has invoked this new
 authority.
   In conducting this analysis, EPA
 considered all available scientific
 information concerning the health
 effects of arsenic, including various
 uncertainties in the interpretation of the
 results. As discussed in more detail
 below, an array of health endpoints of
 concern were considered in this
 analysis. For some of these, the risk can
 currently be quantified (i.e., expressed
 in numerical terms); and for some, it
 cannot. Similarly, there are a variety of
 health and other benefits attributable to
 reductions in levels of arsenic in
 drinking water, some of which can be
 monetized (i.e., expressed in  monetary
 terms)  and others that cannot yet be
monetized. All were considered in this
analysis. The array of factors taken into
account in making risk management
decisions for arsenic underscore the
difficulty of recommending the most
appropriate regulatory  level. A detailed

-------
38948
Federal Register/Vol. 65,  No. 121/Thursday, June  22, 2000/Proposed Rules
discussion of each of the principal
factors considered follows.
1. Feasible MCL
  Because arsenic is a carcinogen with
no established mode of action, EPA is
proposing that the MCLG be set at zero.
To establish the MCL, EPA must first
determine the level which is as close to
this level as feasible. EPA has
determined that 3 ug/L is
technologically feasible for large
systems based on peer-reviewed
treatment information and the practical
quantitation level achievable with
available analytical methods.
2. Principal Considerations in Analysis
of MCL Options
  In addition to the feasible MCL of 3
Ug/L, the Agency evaluated MCL
options of 5 ug/L, 10 ug/L, and 20 ug/
L. EPA considered the health effects
associated with arsenic, the risk levels
to the population for these health effects
that would remain after
implementation, and the costs and
benefits of the different options (both
those that could be monetized and/or
quantified now and those that could
not). The Agency's assessment centered
on the health risk posed by arsenic in
drinking water as well as on the benefits
and costs imposed by the options
evaluated. These options were then
analyzed, taking into consideration the
uncertainties involved in each of these
factors. EPA solicits public comment on
all the factors it considered in making
this decision.
  Estimates of risk levels to the
population remaining after the
regulation is in place provide a
perspective on the level of public health
protection and benefits. The SDWA
clearly places a particular focus on
public health protection afforded by
MCLs. For instance, where EPA decides
to use  its discretionary authority after a
determination that the benefits of an
MCL would not justify the costs, section
1412(b)(6)(A) requires EPA to set the
MCL at a level that "maximizes health
risk reduction benefits at a cost that is
justified by the benefits."
  The  SDWA requires the Agency to
consider both quantifiable and
nonquantifiable health risk reduction
benefits (quantifiable benefits can
include both those that are monetizable
and those that are not). Non-
monetizable benefits range from those
about which some quantitative
information is known (such as skin
cancer), and those which are more
qualitative in nature (such as some of
the non-cancer health effects associated
with arsenic). If additional potential
benefits that are presently not
                      monetized (see Table XI—1) could be
                      estimated at some future point, the
                      benefits might increase further.
                      (Important assumptions inherent in
                      EPA's benefits estimates, including the
                      value of a statistical life and willingness
                      to pay are discussed in section X.C.)
                       EPA considered the relationship of
                      the monetized benefits to the monetized
                      costs for each option. While equality of
                      monetized benefits and costs is not a
                      requirement under section
                      1412(b)(6)(A), this relationship is still a
                      useful tool in comparing costs and
                      benefits. However, EPA believes that
                      reliance on a simple arithmetic analysis
                      of whether monetized benefits outweigh
                      monetized costs is inconsistent with the
                      HRRCA's instruction to consider both
                      quantifiable and non-quantifiable costs
                      and benefits. The Agency therefore
                      believes it is necessary to also examine
                      the qualitative and non-monetized
                      benefits and consider these benefits in
                      establishing the MCL.
                      3. Findings of NRG and Consideration of
                      Risk Levels
                       The Agency based its evaluation of
                      the risk posed by arsenic at the MCL
                      options of 3 (Xg/L, 5 Ug/L, 10 Ug/L and
                      20 Ug/L on national and international
                      research, the bladder cancer risk
                      analysis provided by the National
                      Research Council (NRC) report issued
                      by the National Academy of Sciences
                      (NRC 1999), and the NRC's qualitative
                      statements of overall risk of combined
                      cancers. The Agency is relying heavily
                      on the findings of the NRC for a number
                      of reasons. In carrying out its charge, the
                      NRC assembled an independent body of
                      preeminent scientists from several
                      disciplines. This committee examined
                      and carefully analyzed more
                      information than has been available
                      before, and NRC had the draft report
                      peer reviewed by thirteen other
                      individuals with "diverse perspectives
                      and technical expertise (NRC 1999b)."
                      EPA decided, in 1996, to charge the
                      NRC with evaluating EPA's two risk
                      assessments for arsenic and considering
                      the most current national and
                      international research on arsenic. The
                      NRC determined that the current MCL
                      of 50 ug/L is not adequately protective
                      and should be revised downward as
                      soon as possible. The NRC conducted a
                      number of statistical analyses in making
                      this determination. The report also
                      recommended that EPA conduct
                      separate analyses for "bladder, lung,
                      and other internal cancers," as well as
                      consider the combined impact of these
                      various health effects.
                        Given the release date of the NRC
                      report (March 1999) relative to the
                      timing of the proposed rule and the
additional analyses needed to
definitively quantify all endpoints of
concern, EPA chose to use NRC's   -
bladder cancer analysis to quantify and
monetize the bladder cancer risk for the
proposed rule. NRC provided
quantitative risk factors for bladder  :
cancer, that, when combined with key
risk characterization scenarios by EPA
and qualitative benefits, yield risks and
benefits associated with various
possible MCL options. The NRC report
also noted that lung cancer deaths due
to arsenic could be 2 to 5 times higher
than bladder cancer deaths, considering
the frequency and incidence  of cancers
projected from international studies.
However, the report did not provided
numeric risk-based quantification
analysis for this judgment similar to that
provided for bladder cancer.  As noted in
section X.E., EPA approximated the
potential benefits of avoiding arsenic-
related lung cancer by assuming that the
probability of incidence of lung cancer
is approximately equal to that of bladder
cancer. One can then use the death rate
associated with lung cancer (88% for
lung cancer as compared to 26% for
bladder cancer) to derive benefits and to
consider the implications of this health
endpoint on risk. The risk factors
associated with various MCL options
increase under this "What If" analysis,
with 10 ug/L being on the upper end or
just outside of the Agency's 1 x 10 ~4
risk range and more stringent MCL
options being more solidly under this
risk ceiling.
  EPA anticipates that a peer reviewed
quantification of lung cancer risk from
arsenic exposure may be available
between the time of proposal and
promulgation. If so, EPA will make this
information available for public
comment through a Notice of Data  •
Availability (NODA) and consider the
analysis and public comment for the
final rulemaking.
  Individual risk varies widely
depending on susceptibility, amount of
drinking water consumption, dietary
levels of arsenic, years of exposure, and
other factors. Consequently, any single
MCL does not provide the same level of
protection to all individuals. While not
required by  statute, the Agency has
historically set protectiveness levels
within a risk range of 10 ~4 to 10 ~6. EPA
has sought to ensure that drinking water
standards were established at levels
such that less than 10% of the exposed
population faced a risk that exceeded
the chosen risk level. This conclusion is
based on a recognition of its
responsibility to protect public health,
together with its obligation to consider
a range of risk management factors
when establishing regulatory levels.

-------
                  Federal Register/Vol.  65, No. 121/Thursday, June 22, 2000/Proposed  Rules
                                                                      38949
 4. Non-Monetized Health Effects
   There are a number of important non-
 monetized benefits that EPA considered
 in its analysis. Chief among these are
 certain health impacts known to be
 caused by arsenic (such as skin cancer).
   A number of epidemiologic studies
 conducted in several countries (e.g.,
 Taiwan, Japan, England, Hungary,
 Mexico, Chile, and Argentina) report an
 association between arsenic in drinking
 water and skin cancer in exposed
 populations. Studies conducted in the
 U.S. have not demonstrated an
 association between inorganic arsenic in
 drinking water and skin cancer.
 However, these studies may not have
 included enough people in their design
 to detect these types of effects.
   There were also a large number of
 other health effects associated with
 arsenic, discussed in section III, and
 listed in Table XI. 1, which are not
 monetized, due to lack of appropriate
 quantitative data. These health effects
 include other cancers such as prostate
 cancer and cardiovascular, pulmonary,
 neurological and other non-cancer
 endpoints.
   Other benefits not monetized for this
 proposal include customer peace of
 mind from knowing drinking water has
 been treated for arsenic and reduced
 treatment costs for currently
 unregulated contaminants that may be
 co-treated with arsenic. To the extent
 that reverse osmosis is used for arsenic
 removal, these benefits could be
 substantial. Reverse osmosis is the
 primary point of use treatment, and it is
 expected that very small systems will
 use this treatment to a significant extent.
 5. Sources of Uncertainty
   Among the non-quantifiable factors
 EPA considered in choosing the
 proposed MCL was Congress' intent that
 EPA "reduce * * *  [scientific]
 uncertainty" in promulgating the
 arsenic regulation, reflected in the
 1412(b)(12) arsenic research plan
 provisions and the legislative history for
 the arsenic provision (S. Rep. 104-169,
 104th Cong., 1st Sess. at 39-40).
  All assessments of risk are
 characterized by an amount of
 uncertainty. Some of this can be
 reduced by collecting more data or data
 of a different sort; for other types of
 uncertainty, improved data or
 assessment methods can allow one to
 define the degree to which an estimate
 is likely to be above  or below the "true"
risk. For the arsenic risk assessment,
there are some definable sources of
uncertainty. These include (but aren't
limited to) the following: choice of
endpoint and population; uncertainty
 about the exact exposure of individuals
 in the study population; issues on
 applying data from rural Taiwanese to
 the heterogenous population of the U.S.;
 the inability to know precisely how a
 chemical causes cancer in humans (the
 mode of action, which affects judgments
 as to the shape of the chemical's dose
 response curve at low doses); choice of
 mathematical modeling procedures.
 Congress established a dual path for
 arsenic in SDWA: on the one hand, EPA
 is to issue a proposed MCL in 4Vz years;
 on a parallel track EPA is to develop a
 long-term research plan, complete the
 required consultations and peer
 reviews, complete the research, and
 fully consider the research results.
 While the plan has been developed and
 research is underway, not all research
 results will be available for the final
 rule. However, EPA did obtain through
 the NRG study the most authoritative
 review of existing scientific information
 available. This review examined the
 areas of uncertainty listed above.
   EPA considered uncertainty about
 arsenic's mode of action and the shape
 of the dose response curve below the
 observable range of data. EPA is
 proposing an MCLG of zero. This   .
 decision is supported by the MRC's
 findings that the dose-response    ;
 relationship at low doses is uncertain
 and that a conservative, default
 assumption of linearity is advisable. (An
 assumption of linearity in the dose-
 response relationship implies that there
 is no "safe" level that can be identified
 at which no health effects are expected
 to occur.) However, the Agency also
 notes the NRC's conclusion that "* *  *
 a sublinear dose-response curve in the
 low dose range is predicted, although
 linearity cannot be ruled out." (NRG,
 1999, pg. 6). EPA believes the NRG
 study's articulation of uncertainty about
 the shape of the dose-response curve
 below the observed health effect range
 is an important qualitative
 consideration and, given Congress'
 concern about scientific "uncertainty"
 in setting the arsenic level, guides EPA
 to a default assumption of linearity.
  The choice of one endpoint for risk
 assessment is a judgment call. While
 this choice is guided by the best
 available science, it introduces
 uncertainty. Basing the risk assessment
 on incidence  of bladder tumors will
underestimate the combined risk of all
arsenic-induced health effects. Section
XI.A.4. discusses how assessments of
other tumor types and health endpoints
would result in a higher estimate of
arsenic risk.
  Another source of uncertainty is in
the application of data from one human
population to another. EPA believes that
 the differences in dietary contributions
 of arsenic that NRG identified in the
 Taiwan study population and the U.S.
 are important to consider and a source
 of uncertainty in interpreting the
 results. NRG estimated that daily
 inorganic arsenic intake from food in
 the U.S. ranges from 1.3 ug/day for
 infants, to 4.5 ug/day for males 14-16
 years old and 5.2 ug/day for females 14-
 16 years old, to a maximum of 12.5 ug/
 day for 60-65 year-old males and 9.7 ug/
 day for 60-65 year old females. On the
 other hand, NRG cited a study (Schoof
 et al, 1998) that estimated  the
 Taiwanese obtain 31 ug/day of inorganic
 arsenic from yams and 19 ug/day from
 rice, "for a total of 50 ug/day within a
 range of estimates of 15-211 ug/day
 (NRG, 1999, pg.  51)." NRG  noted (p. 24)
 that "Limited data on dietary arsenic
 intake in the blackfoot-disease region
 now available suggest that arsenic
 intake from food is higher in Taiwan
 than in the United States."  NRG noted
 that EPA previously observed that
 arsenic intake from sources other than
 drinking water would overestimate the
 unit risk calculated from the Taiwan
 study (US EPA 1988, pg. 86). The report
 noted that improved quantification of
 arsenic in Taiwanese food might affect
 the risk assessment for arsenic in
 drinking water in the U.S. (NRG 1999,
 Pg- 6).
   In addition, the NRG report discussed
 laboratory animal studies that indicated
 that selenium reduced the toxicity of
 arsenic. While there is no direct
 evidence for humans,  NRG  noted that
 "Selenium status there [in Taiwan]
 should be considered a moderator of
 arsenic toxicity and taken into account
 when the Taiwanese data are applied to
 populations with adequate selenium
 intakes  (NRG, 1999, pg. 240)." The NRG
 report cited studies comparing urinary
 selenium concentrations and blood
 serum selenium concentrations; these
 were lower for the Taiwanese by
 comparison to other study populations
 including people in the U.S.
  NRG noted that the "model choice can
 have a major impact on estimated low-
 dose risks when the analysis is based on
 epidemiological data (NRG 1999, pg.
 294)." NRG noted that EPA's 1988 risk
 assessment used the multistage Weibull
 model to estimate a lifetime skin cancer
 risk of 1 x 10~'3 for U.S. males exposed
 to arsenic at 50 ug/L. In their report NRG
 discussed the implications (both in a
general sense and specifically for the
Tseng data) of using data from an
ecological study, and of using grouped
data. They also reported the results of
applying both a multistage Weibull  and
a Poisson model. When they re-assorted
data into varying exposure groups, there

-------
38950
Federal Register/Vol. 65,  No.  121/Thursday,  June  22, 2000/Proposed  Rules
was a strong effect on the fitted Weibull
model. NRC concluded: "Thus the fact
that grouping does have a strong effect
provides evidence of additional
measurement error in the arsenic
concentrations being assigned at the
village level (NRG, 1999, pg. 284)." NRG
used median village arsenic
concentrations to represent exposure
levels. The Expert Panel  (US EPA,
1997d) noted that biases  from using
average doses for groups leads to
overestimation of risk.
  "* * * [Dlespite a distribution of doses in
the population, those individuals exhibiting
effects would tend also to be those who
received the highest doses; because of this,
deriving an average dose based on affected
individuals would to some extent bias risk
estimates upward. Similarly attribution of the
total excess risk in the population to arsenic
exposure alone could also be expected to
Inflate the estimate of risk if the population
is also characterized by other risk factors
such as smoking, excess exposure to sunlight,
                       nutritional status, and so on (US EPA, 1997d,
                       pg- 31)."
                         The Poisson model with a quadratic term
                       for age and a linear term for exposure fit as
                       well as the multistage Weibull model, and
                       had less variability in risks from regrouping
                       the exposure intervals. Results from the NRC
                       Poisson model estimations were used in the
                       EPA analysis of bladder cancer risks.
                         NRC noted that "Ecological studies in
                       Chile and Argentina have observed risks of
                       lung and bladder cancer of the same
                       magnitude as those reported in the studies in
                       Taiwan at comparable levels of exposure."
                       This observation increases confidence in the
                       risk estimates based on the Tseng data. That
                       these populations are different in terms of
                       ethnic background, dietary patterns, and
                       potential for other exposures also decreases
                       the level of concern about generalized
                       applicability of the Taiwanese data for risk
                       assessment.
                         EPA considered these various uncertainties
                       associated with interpretation of the health
                       effects of arsenic in making risk management
                       decisions and in selecting an appropriate
                       regulatory level. The Agency requests
comment on whether we have properly
weighed the uncertainties which
overestimate and underestimate risk of the
proposed MCL.
  There is also a measure of uncertainty
about the costs associated with various  '
possible regulatory levels. EPA has provided
its best estimates of the costs, but recognizes
that a number of stakeholders have
performed independent analyses suggesting
that the costs may be higher than those
estimated by EPA. EPA requests comment on
its cost estimates and any additional     '
information commenters may have on
possible costs of the rule.
6. Comparison of Benefits and Costs
  The monetized costs and monetized
benefits of the proposed rule, and the
methodologies used to calculate them, are
discussed in detail in sections IX, X, and XIII
of this preamble and in the HRRCA. Overall
estimates of monetized costs and monetized
benefits associated with various MCL options
are provided in Table XI-1. There are also
many health effects which have not been
monetized,  as is also shown in Table XI—J.
             TABLE Xl-1 .—ESTIMATED COSTS AND BENEFITS FROM REDUCING ARSENIC IN DRINKING WATER
                                                      [In 1999 $ millions]
Arsenic level
(W3/L)
3 	
5 	 	
10 	 	
20 	

Total national
costs to
CWSs1
643.1-753
377.3-441.8
163.3-191.8
61.6-72.9
Total national
costs to CWSs
and
NTNCWSs2
644.6-756.3
378.9-444.9
164.9-194.8
63.2-77.1
Total bladder
cancer health
benefits3
43.6-104.2
5 (79)
31.7-89.9
5(64.3)
17.9-52.1
5 (37)
7.9-29.8
5(19.8)
"What if" scenario4 and potential non-quantified benefits
"What if" lung
cancer health
benefits esti-
mates
47.2-448
6(213.4)
35-384
6(173.4)
19.6-224
6(100)
8.8-128
6 (53.4)
Potential non-quantifiable health benefits
Skin Cancer.
Kidney Cancer.
Cancer of the Nasal Passages.
Liver Cancer.
Prostate Cancer.
Cardiovascular Effects.
Pulmonary Effects.
Immunological Effects.
Neurological Effects.
Endocrine Effects.
Reproductive and Developmental Effects.
   1 Costs include treatment, monitoring, O&M, and administrative costs to CWSs and State costs for administration of water programs. The lower
 number shows costs annualized at a consumption rate of interest of 3%, EPA's preferred approach. The higher number shows costs annualized
 at 7%, which represents the standard discount rate preferred by OMB for benefit-cost analyses of government programs and regulations.
   2 Costs include treatment, monitoring, O&M, administrative costs to CWSs; monitoring and administrative costs to NTNCWSs; and State costs
 lor administration of water programs.                                                                          	
   3 The upper bound estimate includes an adjustment to account for a possible mortality risk of 80%. It is possible that this risk could have been
 below 80%, which would lead to increased benefits. The actual risk depends on the survival rate for bladder cancer in the area of Taiwan studied
 by Chen, which is unknown.                                                                                 .    .  .,
   "These estimates are based on the "what if" scenario for lung cancer, where the risks of a fatal lung cancer case associated with arsenic are
 assumed to be 2-5 times that of a fatal bladder cancer case.
   8The number in parentheses indicates the bladder cancer health benefits assuming an 80% mortality rate for bladder cancer in the area of the
 Chen study, and starting from the midpoint of the benefits range when mortality and incidence are assumed equivalent.
   8The number in parentheses is the midpoint of the range and corresponds to an assumption that the risk of fatal lung cancer is 3.5 times the
 risk of fatal bladder cancer.
 7. Conclusion and Request for Comment

   In summary, based on the NRC report,
 EPA agrees that the current MCL of 50
 Hg/L is too high and must be made more
 protective of human health. Because
 EPA is proposing an MCLG for arsenic
 of 0, the MCL must be set as close as
 feasible to the MCLG, unless EPA
 invokes its discretionary authority to set
 a different MCL at a level where the
                        costs are justified by the benefits. EPA
                        believes that the feasible level for
                        arsenic is 3 p.g/L. Today, EPA is
                        proposing that the arsenic MCL be set at
                        5 ug/L.
                          EPA believes that setting the MCL at
                        3 (ig/L, the feasible level in this case,
                        may not be justified at this time, given
                        the uncertainty regarding the
                        relationship between the monetized
 benefits and the monetized costs at that
 level, the current uncertainty of the non-
 monetized benefits, and the degree of
 scientific uncertainty regarding the
 dose-response curve for an MCL at that
 level (affected by differences in
 nutrition and arsenic from food).
 Because there is a substantial possible
 imbalance between currently estimated
 monetized costs and benefits at the

-------
                  Federal Register/Vol. 65, No. 121/Thursday, June 22, 2000/Proposed  Rules
                                                                       38951
 feasible level of 3 ug/L, and a lack of
 certainty concerning the non-monetized
 costs and potential non-monetized
 benefits, EPA is proposing a standard
 other than the feasible level, using its
 discretionary authority in section
 1412(b)(6). (See Senate Rep. 104-169,
 104th Cong., 1st Sess. at 33). The statute
 requires that a level proposed or
 promulgated using this discretionary
 authority be one which maximizes
 health risk reduction at a level where
 the costs  are justified by the benefits.
 EPA believes that the 5 ug/L MCL best
 meets this statutory test. EPA solicits
 comment on this finding, as described
 in more detail below.
   As discussed earlier in section
 XI.A.4...EPA believes that there are a
 number of not yet quantified adverse
 health effects that pose a significant risk
 to public health. While the relationship
 of actual monetized benefits to
 monetized costs at 5 ug/L, $31.7-$89.9
 million for bladder cancer benefits (plus
 possible lung cancer benefits of $35-
 $384 million based on  the "What If"
 scenario) vs. $378.9-^44.9 million in
 costs, is uncertain. EPA believes the
 range of benefits supports that level,
 especially when there may potentially
 be substantial non-monetized benefits
 factored into the analysis. EPA believes
 that, given the guidance of the NRG
 report, these potential non-monetized
 benefits, including a number of non-
 cancer health effects (see Table XI-1),
 are substantial enough  to strike a
 reasonable balance between benefits and
 costs. Strict parity of monetized costs
 and monetized benefits is not required
 to find that the benefits of a particular
 MCL option are justified under the
 statutory provisions of § 1412(b)(6). In
 addition,  at 5 ug/L, the remaining risks
 (of bladder cancer) to the exposed
 population after the rule's
 implementation are well within the
 10~4 range, which is protective of
 public health. As a result, EPA finds
 that the actual risk levels (including
 risks of potential non-monetized health
 effects) at 5 ug/L are high enough to
 justify this MCL, and it is therefore the
 level which maximizes health
 protection at a level where the costs are
 justified.
  As discussed earlier,  EPA has, as a
 matter of policy typically established
 MCLs for cancer-causing contaminants
 to ensure that the risks  of excess cancer
 deaths represented by exposure to
 drinking water at the MCL over the
 course of a lifetime are  within a range
 of one in 10,000 to one  in 1,000,000.
EPA believes that this range is
reasonably protective of public health
consistent with the goals of the Safe
Drinking Water Act. In  using its
 statutory discretion under section
 1412(b)(6)(A) to set a standard less
 stringent than the feasible level that
 maximizes health risk reduction at a
 cost that is justified by the benefits, EPA
 is proposing that it should choose a
 level that falls within the afore-
 mentioned target risk range. EPA is
 proposing to stay within this risk range
 even if the monetized benefits of a
 standard set at the upper end of the
 range are below the costs, as may be the
 case with this rule. EPA believes that
 important factors in this evaluation are
 the considerable non-quantifiable
 benefits that may be attributable to the
 proposed MCL. EPA also notes, as
 discussed earlier, that Congress did not
 direct EPA to ensure strict equality of
 monetizable costs and benefits in
 applying its discretionary authorities
 under section 1412(b)(6)(A). EPA
 requests comments on its proposed use
 of the new authority under section
 1412(b)(6)(A) of the SDWA.
  The risk assessment for bladder
 cancer indicates that a standard set at 10
 ug/L would fall at the upper end of the
 target risk range, with 5 ug/L more
 solidly within that risk range. How^ver,
 there are two important sets of
 considerations when using available
 health effects information and studies to
 help determine the appropriate level for
 a proposed new standard. On the one
 hand, multiple health endpoints are of
 concern in ensuring that the standard is
 adequately protective. As noted earlier,
 the NRC expresses concern about lung
 cancer and other health endpoints and
 indicated that excess lung cancer deaths
 from arsenic in drinking water could be
 2—5 times the level of bladder cancer
 deaths.  If these other risks were fully
 quantified, the total risk at 10 ug/L
 might be well above 1 x 10~4 (the upper
 end of the risk range), given that the
 quantified risk of bladder cancer alone
 appears to be at approximately this
 level.
  On the other hand, there is
 uncertainty in the quantification of
 bladder cancer risk (as well as other
 health endpoints) and this risk estimate
 includes a number of conservative
 assumptions, as discussed previously.
 These include the assumptions of using
 a linear dose-response function; the fact
that the dose-response data from the
Taiwan epidemiologic study are based
upon grouped occurrence information
from wells used by the study
population; and the possibility that the
study population was  more susceptible
to arsenic in drinking water (as
compared to the U.S. population) due to
the relatively high dietary intake and
dietary deficiencies in other elements (e.g.
selenium) that might mitigate the results
  of arsenic. Thus, the risk of bladder
  cancer alone might be well below
  current estimates which represent EPA's
  best estimate at this time using currently
  available data and standard
  methodologies. The proposed MCL
  attempts to balance these countervailing
  considerations in establishing a level
  that is protective of public health.
    Given these competing sources of
  uncertainty, EPA believes it is
  appropriate to propose a standard at 5
  ug/L, because at this level it is more
  likely that the total risk would be within
  the target range than at a higher
  standard. However, between now and
  promulgation of the final rule, EPA will
  work to resolve as much of this
  uncertainty as possible, both in terms of
  quantifying risk of additional health
  endpoints (e.g., lung cancer) and in
  terms of reexamining conservative
  assumptions in the risk estimate. EPA
  requests comment on its proposed level
  of 5 ug/L and on its rationale for
  selecting this level. In selecting the final
  level of the standard, EPA will evaluate,
  in light of comments received and any
  new scientific information, its proposed
  way of using its discretionary authority
  under section 1412(b)(6)(A) and the
  total risk, costs, and benefits associated
  with each of the levels of the standard
  under consideration.
    EPA requests comment on other
  potential MCLs and which of the MCLs
  and rationales presented here best fits
  the statutory framework. First, EPA is
  requesting comment on setting the MCL
  at 10 ug/L. The monetized costs of
  $164.9-$194.8 million, and monetized
  benefits of $17.9-$52.1 million for
  bladder cancer (plus possible lung
  cancer benefits  of $19.6-$224 based on
  the "What If" scenario) are closer at 10
  ug/L. The risk levels (of bladder cancer)
  to the exposed population are within
  the 10~4 risk range, and the
  uncertainties already discussed in
  Section XI.A.6.  may be a basis for
  inferring lower expected possible non-
  monetized benefits than assumed for the
  MCL option of 5 ug/L.
   EPA is also requesting comment on an
 MCL option of 20 ug/L. Some
  stakeholders favor an MCL in this range
 and cite,  as justification for such a level,
 their belief that if all uncertainties are
 taken into consideration, risk estimates
 would be within the Agency's risk range
 of range of 1 xlO~6to 1 xlO~4. As can
 be seen from Table XI-1,  costs are
 considerably reduced at this level, since
 far fewer CWSs  would be impacted (i.e.,
 occurrence of arsenic,  without
 treatment, is already below this level for
 many systems). Approximately 1,200
,,CWSs would be projected to incur costs
 of approximately $63-$77 million to

-------
38952
Federal  Register/Vol. 65, No. 121/Thursday, June 22, 2000/Proposed Rules
comply with an MCL of 20 ug/L.
Benefits would also be considerably
lower than for other options, at $7.9-
S29.8 million for bladder cancer (plus
possibly S8.8-S128 million for lung
cancer, based on the "What If
scenario). EPA's principal concern with
an MCL option in this range is that it
may not be sufficiently  protective after
consideration of all health endpoints of
concern. In other words, when the
effects of bladder cancer, lung cancer,
and skin cancer are considered, together
with the various non-quantifiable
endpoints such as circulatory system
impacts, an MCL option of 20 ug/L
could result in an unacceptably high
risk, well outside of the risk range of 1
x 10~6 to 1 x 10~4. As noted above, in
using its statutory discretion to set a
standard above the feasible level, EPA is
proposing not to set a standard that
exceeds this target risk  range. However,
EPA solicits comment on an MCL
option of 20 |ig/L along with any
supporting rationale that commenters
wish to offer.
  EPA is also requesting comment on
setting the MCL at 3 ug/L. As explained
in section XI.A.l., this is the level as
close to the MCLG as is feasible. It is
also the level at which the risks are most
solidly within the 10~4 risk range of the
three MCLs considered. If EPA were to
set the MCL at this level, EPA would not
use its discretionary authority to set the
MCL at a less stringent  level based on
costs and benefits. The  Agency
estimates that the likelihood that actual
monetized benefits of $43.6-3104.2
million for bladder cancer (plus possible
lung cancer benefits of  $47.2-3448
million based  on the  "What If"
scenario), are close to monetized costs
of S644.6-S756.3 million is less certain
than at 5 ug/L. (See Table XI-1.) While
EPA believes that benefits may be
substantially less than monetized costs
for the feasible level, the feasible level
would be the most protective of the
options presented here  and would
conservatively account for the
uncertainties about the severity of
various health effects endpoints and
their potential additive impacts.
  Finally, Congress indicated interest in
assuring that EPA considered impacts of
an MCL decision on people served by
large systems who could afford
protective MCLs and an MCL of 3 would
respond to this interest. Section
1412(b)(6)(B),  however, provides that
the interests of people served by large
systems are to be considered along with
benefits and costs to systems not
expected to get small system variances.
Because this proposal does not include
small system variance technologies (i.e.,
affordable technologies for small
                     systems at the proposed MCL have been
                     identified), the interests of persons
                     served by large and small systems are
                     being considered together and the
                     provisions of section 1412(b)(6)(B) do
                     not apply in this case.

                     B. Why Is EPA Proposing a Total
                     Arsenic MCL?
                       The previous drinking water standard
                     for arsenic of 0.05 mg/L was based on
                     total arsenic. Total arsenic includes the
                     dissolved and undissolved arsenic
                     species present in drinking water and
                     makes no distinction between inorganic
                     or organic species. Consistent with the
                     previous standard for arsenic, today's
                     proposed regulation of 0.005 mg/L will
                     be based on total arsenic. From an
                     occurrence and analytical methods
                     standpoint, the Agency believes it is
                     inappropriate to make a regulatory
                     distinction between inorganic and
                     organic arsenic forms in drinking water.
                       According to Irgolic (1994) and as
                     mentioned in section II.B, the inorganic
                     arsenic species (As III and As V)  arc
                     present in drinking water, and organic
                     arsenic compounds are rarely found in
                     water supplies. Furthermore, inorganic
                     As V (arsenate) is more prevalent in
                     drinking water supplies than inorganic
                     As III (arsenite), which tends to occur in
                     anaerobic waters. If organic species are
                     present in drinking water,
                     methylarsonic acid (MMA) and
                     dimethylarsonic acid (DMA) are  the
                     predominant organic forms.  These
                     organic species, when present, can
                     result from the leaching of arsenic-
                     containing herbicides or from the
                     conversion of the inorganic forms to the
                     organic forms in the presence of
                     microbial activity. In arsenic-rich
                     ground water wells from Taiwan,
                     methylated compounds were not
                     present above concentrations of 1 ug/L.
                     No DMA  or MMA was detected in the
                     ground water samples from six districts
                     in West Bengal, India (Chatterjee et
                     al.,1995). Regarding surface water,
                     Anderson and Bruland (1991) reported
                     that organic species (DMA and MMA)
                     accounted for 1 to 59% of the total
                     arsenic concentration from fourteen lake
                     and river samples taken in California.
                     As Irgolic pointed out in his review of
                     the Anderson and Bruland study, the
                     level of the organic arsenic found in
                     these surface water samples  were in the
                     low nanomolar (nM or nm/L) range.
                     After converting the reported units from
                     nm/L to ug/L, analysis of the Anderson
                     and Bruland data indicate that only two
                     of the fourteen water samples exceeded
                     a concentration of 1 Ug/L of organic
                     arsenic (DMA and MMA combined).
                       There is currently no EPA approved
                     method for arsenic analysis in drinking
water that distinguishes inorganic
arsenic species from organic arsenic  ,
forms. The method would need to me;et
the criteria listed in section VLB. and
would require interlaboratory studies
for validation. The estimated costs of
such an analytical method could range
from $150 to 3250 per analysis. In
addition, laboratory capacity for this
type of method would most likely be
limited at this time.
  Few toxicity studies exist for organic
arsenicals. The NRC report noted that
methylated arsenic has less
developmental toxicity than inorganic
arsenic. Concentrations of DMA
administered that decreased fetal weight
produced over 50% maternal mortality
in studies with rats and mice (Rogers et
al, 1981 as reported in NRC, 1999);
hamsters had no developmental toxicity
from exposure to MMA nor DMA
(Willhite, 1981, as reported in NRC,
1999). NRC noted that EPA has two
unpublished studies of rats fed MMA
which had some increase in thyroid
tumors, but no effect on mice. In    ,
addition, MMA and DMA produced  '
mutations in cells at concentrations over
one thousand times higher than the
concentrations of inorganic arsenite and
arsenate (Moore et al., 1997 as reported
in NRC, 1999). It takes roughly ten times
more DMA than arsenite to cause
chromosome changes in a human cell
line (Oya-Ohata et al., 1996, as reported
in NRC, 1999).
  Because of the limited occurrence of
organic arsenic species in water and the
lack of a suitable and widely available
analytical method for inorganic arsenic,
the Agency believes compliance with
the proposed arsenic standard of 0.005
mg/L should be based on total arsenic.
EPA requests comments on setting the
MCL based on total arsenic and any data
or established analytical methods that
would support setting an MCL based on
inorganic arsenic.

C. Why Is EPA Proposing To Require
Only Monitoring and Notification for
NTNCWSs?
  In this rulemaking, the Agency is
soliciting comment on an approach
which would not extend coverage of the
rule to Non-Transient Non-Community
(NTNC) water systems, but would
instead create an intermediate level of
control for these systems (monitoring
and notification requirements). The
suggested approach would recognize the
lower level of risk generally posed to
individuals by these systems.
Simultaneously, it would provide a
mechanism for the public to be
adequately informed in those situations
where unusual concentrations of NTNC
systems, customer overlap, and high

-------
                   Federal Register/Vol.  65,  No. 121/Thursday,  June 22, 2000/Proposed Rules
                                                                       38953
  local arsenic water concentrations
  caused risk levels to more closely
  approach community water system
  levels.
   There are approximately 20,000
  NTNCs water systems regulated under
  the Safe Drinking Water Act. By
  definition, these systems do not serve
  over 25 people as year round residents,
  as would be the case for a community
  water system. However, they must serve
  at least 25 of the same people for over
  six months out of the year, or they
  would be classified as Transient Non-
  Community (TNG) water systems. It is
  generally an important distinction since
  the Agency has not applied regulations
  for contaminants with chronic health
  effects to TNG water systems, while it
  often has regulated NTNC systems
  similar to community water systems
  when addressing the risks posed by
  chronic contaminants.
   In the case of arsenic, the existing
 regulation does not apply to NTNC
 systems. While it is feasible to control
 arsenic in NTNC water systems,
 extending regulation to these systems
 needs to be considered in light of the
 new SDWA requirement to determine
 whether the benefits extending coverage
 to this category would justify the costs
 and whether such regulation would
 provide a reasonable opportunity for
 health risk reduction. As discussed
 elsewhere in the preamble, this analysis
 requires a balancing of both quantitative
 and non-quantitative factors. Based on
 the modeling to be discussed, the
 ninetieth percentile lifetime risk of
 contracting bladder cancer posed to an
 individual consuming water from a
 NTNC water system, even in their
 present untreated state, does not exceed
 one in 100,000. 3  As a consequence,
 costs per each bladder cancer case
 avoided at the proposed MCL would
 approach the fifty million dollar mark if
 coverage of the rule were extended to
 NTNCs. This level is well above the
 range of historical environmental risk
 management decisions.
   These much lower risk levels result
 because most individuals served by
 NTNC systems are expected to receive
 only a small portion of their lifetime
 drinking water exposure from such
 systems. For example, even with twelve
 years of perfect attendance at schools
 served by NTNC water systems, the
 water consumed by an individual
 student is estimated to represent less
than five percent of lifetime
consumption.
  3 Throughout this discussion, exposures and risks
were only considered for populations potentially
addressable by regulation, i.e., systems with present
arsenic levels in excess of 3 ug/L.
    On the other hand, there are some;
  segments of the NTNC water system
  population where exposure is a more
  significant portion of the total lifetime
  exposure. Manufacturing and other
  workers, although they represent only
  five percent of the population served by
  NTNC systems, could receive twenty to
  forty percent of their lifetime exposure
  at work. Nevertheless, as manufacturing
  workers represent a small portion of the
  NTNC population, overall risks among
  the NTNC population are small.
   Another factor of potential concern is
  the extent to which users of the different
  NTNC water systems overlap. It is
  conceivable that some areas in the
  country exist where individuals are
  subjected to arsenic exposure at a
  number of different non-community
  systems (e.g., day care center plus
  school plus factory, etc.). In such
  circumstances, individuals would be
  exposed to proportionately higher risks
 if the water systems all had elevated
 arsenic levels. For some individuals, 'the
 exposure could approach levels
 observed in corresponding community
 water systems. This concern is
 alleviated by the fact that NTNC systems
 generally serve only a very small
 portion of the total population.  For
 example, over ninety-five percent of all
 school children are served by
 community water systems. Only a small
 percentage are served by NTNC water
 systems and, of that group, only about
 twelve percent (or less than one half of
 one percent of the overall student
 population) would be expected to have
 arsenic in their water above the
 proposed regulatory level). Likewise,'
 less than 0.1 percent of the work force
 population receive water from an NTNC
 water system. With such low portions of
 the total population exposed to any
 particular type of NTNC system, the
 overall likelihood of multiple exposure
 cases in the NTNC population should
 also be small. The groups have been
 treated independently for this analysis.
 Comment and data are solicited to
 support any alternative treatments of the
 exposure data.
   Finally, although the Agency does not
 believe there is sufficient evidence to
 support unusual sensitivity on the part
 of children, they generally do consume
 more water on a weight adjusted basis.
 For this reason, NTNC systems which
 were likely to pose the greatest exposure
 risk to  children were separately
 examined and their higher relative
 doses considered in the modeling effort.
 All of these factors contributed to the
Agency's evaluation of whether or not to
 extend regulation to NTNC water
systems for arsenic and are discussed
further in the results section.
  1. Methodology for Analyzing NTNCWS
  Risks
    Determination of system and
  individual exposure factors—In the
  past, the Agency has directly used
  SDWIS population estimates for
  assessing the risks posed to users of
  NTNC water systems. In other words, it
  was assumed that the same person
  received the exposure on a year round
  basis. Under this approach it was
  generally assumed that all NTNC users
  were exposed for 270 days out of the
  year and obtained fifty percent of their
  daily consumption from these systems.
  TNC users were assumed to use the
  system for only ten days per year.
   With the recent completion of
  "Geometries and Characteristics of
  Public Water Systems (US EPA,
  1999e)," however, the Agency has
  developed a more comprehensive
  understanding of NTNC water systems.
  These systems provide water in due
  course as part of operating another line
  of business. Many systems are classified
 as NTNC, rather than TNC, water
 systems solely because they employ
 sufficient workers to trigger the "25
 persons served for over six months out
 of the year" requirement.  Client
 utilization of these systems is actually
 much less and more similar to exposure
 in TNC water systems. For instance, it
 is fairly implausible that highway rest
 areas along interstate highways serve
 the same population on a  consistent
 basis (with the exception of long
 distance truckers). Nevertheless, there
 are highway rest areas in both NTNC
 and TNC system inventories. The
 "Geometries" report suggests that
 population figures reported in SDWIS
 which have been used for past risk
 assessments generally appear to reflect
 the number of workers in the
 establishment coupled with peak day
 customer utilization.
  Under these conditions use of the
 SDWIS figures for population greatly
 overestimates the actual individual
 exposure risk for most of the exposed
 population and also significantly
 underestimates the number of people
 exposed to NTNC water.4 Adequately
 characterizing individual and
  "For example, airports constitute only about a
hundred of the NTNC water systems. Washington's
Reagan National and Dulles, Dallas/Fort Worth,
Seattle/Tacoma, and Pittsburgh airports are the five
largest of the airports. SDWIS reports that these five
airports serve about 300,000 people. In actuality,
Bureau of Transportation Statistics suggest that they
serve about eleven million passengers per year.
Examination of this information and other BTS
statistics suggests that these airports serve closer to
seven million unique individuals over the course of
a year and that exposure occurs on an average of
ten times per year per individual customer, not 270
times.

-------
38954
Federal Register/Vol. 65, No.  121/Thursday, June  22, 2000/Proposed  Rules
population risks necessitates some
adjustments to the SDWIS population
figures. For chronic contaminants, such
as arsenic, health data reflect the
consequences of a lifetime of exposure.
Consequently, risk assessment requires
the estimation of the portion of total
lifetime drinking water consumption
that any one individual would receive
from a particular type of water system.
In turn, one needs to estimate the
appropriate portions for daily, days per
year, and years per lifetime
consumption. These estimates need to
be prepared for both the workers at the
facility and the "customers" of the
facility.
  This adjustment was accomplished
through a comprehensive review of
government and trade association
statistics on entity utilization by the
U.S. Department of Commerce's
Standard Industrial Classification (SIC)
code. These figures, coupled with
SDWIS information relating to the
portion of a particular industry served
by non-community water systems, made
possible the development of two
                      estimates needed for the risk
                      assessment: customer cycles per year
                      and worker per population served per
                      day. These numbers are required to
                      distinguish the more frequent and
                      longer duration exposure of workers
                      from that of system customers.5 A more
                      detailed characterization of the
                      derivation of these numbers is
                      contained in the docket. Table XI—2
                      provides the factors used in the NTNC
                      risk assessment to account for the
                      intermittent nature of exposure.
                      Comment is solicited on the
                      appropriateness of the various factors.
                        Once the population adjustment
                      factors were derived, it was possible to
                      determine the actual population served
                      by NTNC water systems.  Table XI-3
                      provides a breakout of these figures by
                      type of establishment. Although not
                      included in Table XI-3, there are other
                      equally important characteristics to note
                      about these systems. With notable
                      exceptions (such as the airports in
                      Washington, DC and Seattle), the
                      systems generally serve a fairly small
                      population on any given day. In fact, 99
percent of the systems serve less than
3300 users on a daily basis. This means
that water production costs will be
relatively high on a per gallon basis.
  Risk calculation—Calculations of
individual risk were prepared for each
industrial sector. Even within a given
sector, however, risk varies as a function
of an individual's relative water
consumption, body weight,
vulnerability to arsenic exposure, and
the water's arsenic concentration.
Computationally, risks were estimated
by performing Monte Carlo modeling, as
was done in the community water
system risk estimation, with two   .
exceptions. First,  each realization in a
given sector was multiplied by the
portion of lifetime exposure factor
presented in Table XI-2 to reflect the
decreased consumption associated with
the NTNC system. Secondly, relative
exposure factors were limited to age
specific ratings where appropriate.6 For
example, in the case of school children,
water consumption rates and weights
for six to eighteen year olds were used.
                       TABLE XI-2.—EXPOSURE FACTORS USED IN THE NTNC RISK ASSESSMENT
NTNCWS
Water wholesalers
Nursing homes 	
Churches 	
Golf/country clubs 	
Food retailers 	
Non-food retailers 	
Restaurants 	
Hotels/motels 	
Prisons/jails 	
Service stations 	
Agricultural products/
services 	
Daycare centers 	
Schools 	
State parks 	
Medical facilities 	
Catnpgrounds/RV 	
Federal parks 	
Highway rest areas 	
Misc. recreation service
Forest Service 	
Interstate carriers 	
Amusement parks 	
Summer camps 	
Airports 	
Military bases 	
Non-water utilities 	
Office parks 	
Manufacturing: Food ....
Manufacturing: Non-
food 	
Landfills 	
Fire deoartments 	
Number of
cycles per
yr
1.00
1.00
1.00
4.50
2.00
4.50
2.00
86.00
1.33
7.00
7.00
1.00
1.00
26.00
16.40
22.50
26.00
50.70
26.00
26.00
93.00
90.00
8.50
36.50





Worker/pop/
day
0.000
0.230
0.010
0.110
0.070
0.090
0.070
0.270
0.100
0.060
0.125
0.145
0.073
0.016
0.022
0.041
0.016
0.010
0.016
0.016
0.304
0.180
0.100
0.308
1.000
1.000
1.000
1.000
1.000
1.000
1.000
Worker frac-
tion daily

0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
1.00
0.50
0.50
1.00
0.50
0.50
0.50
0.50
0.50
0.50
1.00
1.00
Worker
days/yr

250
250
250
250
250
250
250
250
250
250
250
200
250
250
180
250
250
250
250
250
250
180
250
250
250
250
250
250
250
250
Worker
exposure
years

40
40
40
40
40
40
40
40
40
40
10
40
40
40
40
40
40
40
40
40
10
10
40
40
40
40
40
40
40
40
Customer
fraction
daily
0.25
1.00
0.50
0.50
0.25
0.25
0.25
1.00
1.00
0.25
0.25
0.50
0.50
0.50
1.00
1.00
0.50
0.50
1.00
1.00
0.50
0.50
1.00
0.25





Days of use/
yr
270
365
52
52
185
52
185
3.4
270
52
52
250
200
14
6.7
5
14
7.2
14
14
2
1
7
10





Customer ex-
posure years
' 70
1 10
' 70
70
70
70
70
40
3
54
50
5
12
70
10.3
50
70
70
70
50
70
70
10
70
.




  8 For example, travel industry statistics provide
 Information on total numbers of hotel stays,
 vacancy rates, traveller age ranges, and average
 duration of stay. Those figures can be combined
 with tho SDWIS peak day population estimates to
                       allocate daily population among workers, customers
                       and vacancies. The combination of these factors
                       provides an estimate of the number of independent
                       customer cycles experienced in a year.
  "For example, school kid water consumption was
 weighted to reflect consumption between ages 6
 and 18, while factory worker consumption was
 weighted over ages 20 to 64.

-------
                  Federal Register/Vol.  65,  No.. 121/Thursday, June 22, 2000/Proposed Rules
                                                                      38955
                 TABLE XI-2— EXPOSURE FACTORS USED IN THE NTNC RISK ASSESSMENT—Continued
NTNCWS
Construction ...
Mining 	
Migrant labor camps ....
Number of
cycles per
yr



Worker/pop/
day
1 000
1 000
1.000
Worker frac-
tion daily
1 nn
1 00
1.00
Worker
days/yr

ocn
250
Worker
exposure
years


40
Customer
fraction
daily



Days of use/
yr



Customer ex-
posure years



                  TABLE XI-3.—COMPOSITION OF NON-TRANSIENT, NON-COMMUNITY WATER SYSTEMS
                                    [Percentage of total NTNC population served by sector]
Schools 	
Manufacturing 	
Airports 	
Office Parks 	
9.7
2.7
26.1
0.6
Medical Facilities 	
Restaurants 	
Non-food Retail 	
Hotels/Motels 	
8
0.9
1.6
9.2
Interstate Carriers
State Parks
Amusement Parks
H'way Rest Area 	
7.1
8.6
17.7
1.0


Other

1.3
1.8
3.5
    To illustrate the process, it was conservatively assumed that a child would  attend only NTNC served schools for
 all twelve  years. Further,  it was assumed that a child would get half of their  daily water  consumption  at  school
 (for an  average  first grader this  would  correspond  to  roughly nine ounces  of water  per  school  day). Finally,  it was
 assumed that the child would have perfect attendance and attend school  for 200  days  per year. Table XI-4 provides
 a sample output for the upper bound individual risk distribution to school children resulting from exposure  to the
 range of untreated arsenic observed in community ground water systems 7  as well as  an  estimate based on more moderate
 assumptions of four ounces per day and 150  days attendance  for four years. Upper and  lower bound risk distributions
 were prepared for both workers and "customers" at all types of NTNC water systems  and  are contained in the  docket.

    TABLE XI-4.—UPPER BOUND SCHOOL CHILDREN  RISK ASSOCIATED WITH CURRENT ARSENIC EXPOSURE  IN NTNC
                                                 WATER SYSTEMS
                                        [Risks are per 10,000 students, i.e., x  10-"]

Mean Lifetime Risk 	
90th Percentile Lifetime Risk 	
Lifetime Bladder Cancers in Student Population 	
Moderate expo-
sure scenario

n 01 Q
0.5
Upper bound
scenario

U.079
4.5
  Note: This table does not include potential non-quantified lung or skin cancers.
  The distribution of population risks
 overall was determined as part of the
 same simulation by developing sector
 weightings to reflect the total portion of
 the NTNC population served by each
 sector. Population weighted
 proportional sampling of the individual
 sectors provided an overall distribution
 of risk among those exposed at NTNC
 systems.
 2. Results

  It is important to note that the results
 presented in the discussion of NTNC
 benefits are based on the currently
 quantified health endpoint for arsenic
 related bladder cancer. As noted
 elsewhere in Section X of today's
 proposal, there are a number  of health
 end points that have not yet been
 quantified and which could provide a
rationale for extending coverage to
NTNCs—in the event that a substantial
portion of the consumers of water from
such systems fall outside the  1 in 10,000
 risk range frequently used by the
 Agency as a benchmark for such
 decisions. (Any additional data
 quantifying such endpoints would made
 available for public comment in a
 Notice of Data Availability.)
  Table XI-5 presents a summary of the
 Benefit Cost Analysis for all NTNC
 systems. As can be seen from a review
 of the Table, regulation of arsenic in
 NTNC water systems provides only very
 limited.opportunity for national risk
 reduction. Table XI-6 presents risk
 figures for three particular sets of
 individuals: children in daycare centers
 and schools, and construction workers.
 Construction and other strenuous
 activity workers comprise an extremely
 small portion of the population served
 by NTNC systems (less than 0.1%), but
 face the highest relative risks of all
 NTNC users (90th percentile risks of 0.7
 to 1.6 x 10~4 lifetime risk).
Nevertheless, there is considerable
uncertainty about these exposure
 numbers. It is quite likely that they
 overestimate consumption and may be
 revised downward by subsequent
 analysis (Any additional data
 quantifying such endpoints would made
 available for public comment in a
 Notice of Data Availability.). The risks
 for children are much lower with an
 upper bound, 90th percentile estimate
 of 1.7 x 10-5 lifetime risk.
  What is not possible to determine
 from the analysis of NTNC systems is
 the extent to which there is overlap of
 individual exposure between the
 various sectors. As mentioned earlier,
 NTNC establishments generally
 constitute a small portion of their SIC
 sectors. This fact and the observation
 that NTNC populations would only
 serve about one percent of the total
population if all of the sectors with
 significant exposure (greater than five
percent of lifetime) if they were
  7 Community ground water occurrence
information was used since NTNC systems are
almost exclusively supplied by ground water
sources. Further, as there was no depth dependence
of arsenic levels observed in the community
information, it is believed that the data are an
adequate approximation.

-------
38956
Federal  Register/Vol. 65, No. 121/Thursday, June 22,  2000/Proposed Rules
mutually exclusive,8 provide some
support for treating the SIC groups
independently. However, it is equally
plausible that there are communities
where one individual might go from an
NTNC day care center to a series of
NTNC schools and then work in an
NTNC factory.
  The Agency is concerned about the
potential for local issues to arise with
respect to combined arsenic exposures.
In the rare community where all ground
water is contaminated with the highest
levels of arsenic, risks could be outside
of the Agency's traditionally allowable
realm. Further, different levels of
protection being provided by schools
served by community water systems
versus those served by NTNC systems
could be seen  as posing equity
considerations for rural communities.
For all of these reasons, the Agency does
not believe it is appropriate to
                     completely exempt NTNC systems from
                     arsenic regulation. On the other hand, it
                     does not believe an adequate basis exists
                     to prescribe a standard.
                       The Agency is proposing to take a
                     somewhat different approach with
                     respect to NTNC water systems than
                     previously practiced. We are proposing
                     that NTNC water systems be subject to
                     arsenic monitoring requirements
                     applicable to community water systems.
                     When an individual NTNC system has
                     arsenic present in excess of the MCL for
                     community systems, it would be
                     required to post a notice to customers as
                     described in Section VII.I. of this rule.
                     The Agency believes that this approach
                     will provide localities with high arsenic
                     concentrations the opportunity to limit
                     their consumption of water from these
                     systems. Because the NTNC is not the
                     sole source of water available to these
                     consumers as would be the case with a
community water system, they would
have the ability to use bottled water, or
in the case of schools for instance, to
install voluntary treatment to reduce,
their exposure.
  The Agency requests comment on this
approach for addressing NTNC watej*
systems as well as on two other possible
approaches: exempting NTNC systems
entirely from coverage under this rule or
extending coverage to NTNC systems in
the same manner as CWSs. EPA requests
an accompanying rationale and any data
commenters wish to submit as part of
their comments on this topic. The
Agency may decide, as part of the final
rule, to incorporate any of these three
approaches without further opportunity
for comment (except where a NODA
may be issued to provide the public
with additional new information not
taken into consideration for today's
mlemaking).
                       TABLE XI-5.—NON-TRANSIENT NON-COMMUNITY BENEFIT COST ANALYSIS
                                          [All risk values are per 10,000-i.e., 10~4i
MCL option
Mean Individual Risk 	
90th Percontile Individual 	
Annual Bladder Cancers 	
Cancer Cases Avoided 	
Benefit Million Dollars 	 	 	
Cost Million Dollars 	
Untreated
Lower
bound
0.019
0.037
0.427
0
0
Upper
bound
0.042
0.08
0.95
0
0
0
10
Lower
bound
0.012
0.027
0.265
0.162
0.31
Upper
bound
0.026
0.058
0.583
0.367
0.70
6.121
5
Lower
bound
0.0077
0.017
0.16
0.267
0.51
Upper
bound
0.017
0.037
0.36
0.59
1.1
14.69
3
Lower
bound
0.0046
0.01
0.101
0.326
0.62
Upper
bound
6.01
6.022
6.215
6.735
1.4
25.21
  Note: This table does not include potential non-quantified lung cancer benefits.
                             TABLE XI-6.—SENSITIVE GROUP EVALUATION LIFETIME RISKS



Day Care Children 	
Group



Mean risk
3.2-7X10-5
3.8-7.9x10-6
3.4-6.8x1 0-6
90th percentile
risk
7.2-1 6x1 0~s
0.84-1 .7x10-s
0.74-1.5x10-5
 XII. State Programs
 A. How Does Arsenic Affect a State's
 Primacy Program?
   States must revise their programs to
 adopt any part of today's rule which is
 more stringent than the approved State
 program. Primacy revisions must be
 completed in accordance with 40 CFR
 142.12, and 142.16. States must submit
 their revised primacy application to the
 Administrator for approval. State
 requests for final approval must be
 submitted to the Administrator no later
 than 2 years after promulgation of a new
 standard unless the State requests and is
 granted an additional 2-year extension.
                        For revisions of State programs,
                      § 142.12 requires States to submit,
                      among other things, "[a]ny additional
                      materials that are listed in § 142.16 of
                      this part for a specific EPA regulation,
                      as appropriate (§ 142.12(c)(l)(ii))."
                      Based on comments from stakeholders
                      at the arsenic in drinking water
                      regulatory development meetings held
                      prior to proposal, EPA believes that the
                      information required in § 142.16(e) is
                      not required for States revising the MCL
                      for arsenic. Although that section refers
                      to applications that adopt requirements
                      of §§141.11, 141.23, 141.32, and 141.62,
                      EPA believes that existing State
                      programs which contain the
 standardized monitoring framework for
 inorganic contaminants (40 CFR 141.23)
 can ensure all CWSs monitor for
 arsenic. Therefore, EPA is proposing to
 clarify that § 141.16(e) applies only to
 new contaminants, not revisions of ,
 existing contaminants regulations. The
 Agency requests comment on whether
 this is an appropriate change.      :
   EPA believes that the requirements in
 § 142.12(c) will provide sufficient  ',
 information for EPA review of the State
 revision. The side-by-side comparison
 of requirements required in
 § 142.12(c)(l)(i) will only consist of.
 sections revised to adopt the changes
 required for the arsenic regulation and
   "This Is considerably loss than the estimated
 rural population in the U.S. which is the smallest
                      group among which users of these systems would
                      conceivably be distributed.

-------
                  Federal Register/Vol. 65, No.  121/Thursday, June  22,  2000/Proposed Rules
                                                                      38957
 any other revisions requested by the
 State. In addition, the Attorney
 General's statement required in
 § 142.12(c)(l)(iii) will certify that the
 revised regulations will be effective and
 enforceable. The Agency requests
 comment on whether any other
 documentation is necessary to approve
 revisions to State programs enforcing
 the new arsenic regulation.
   The Agency is proposing to add
 § 142.16(j) to clarify primacy
 requirements relating to monitoring
 plans and waiver procedures for
 revisions of existing monitoring
 requirements such as arsenic. Section
 142.16(j) clarifies that the State simply
 needs to inform the Agency in their
 application of any changes to the
 monitoring plans and waiver
 procedures. Alternatively, a State may
 indicate in the primacy application that
 they will use the existing monitoring
 plans and waiver criteria approved for
 primacy under the National Primary
 Drinking Water Standards for other
 contaminants (for example, i.e. the
 Phase II/V rules). This information may
 be provided in the primacy application
 crosswalk which identifies revisions to
 the State primacy program.
 B. When Does a State Have To Apply?
   To maintain primacy for the Public
 Water Supply (PWS) program and to be
 eligible for interim primacy enforcement
 authority for future regulations, States
 must adopt today's rule, when final. A
 State must submit a request for approval
 of program revisions that adopt the
 revised MCL and implementing
 regulations within two years of
 promulgation unless EPA approved an
 extension per § 142.12(b).  Interim
 primacy enforcement authority allows
 States to implement and enforce
 drinking water regulations once State
 regulations are effective and the State
 has submitted a complete and final
 primacy revision application. To obtain
 interim primacy, a State must have
 primacy with respect to each existing
 NPDWR. Under interim primacy
 enforcement authority, States are
 effectively considered to have primacy
 during the period that EPA is reviewing
their primacy revision application.
 C. How Are Tribes Affected?
  Currently, no federally recognized
Indian tribes have primacy to enforce
any of the drinking water regulations.
EPA Regions implement the rules for all
Tribes under section 1451 (a)(l) of
SDWA. Tribes must submit a primacy
 application to have oversight for the
 inorganic contaminants (i.e., the Phase
 II/V rule) to obtain the authority for the
 revised arsenic MCL. Tribes with
 primacy for drinking water programs are
 eligible for grants and contract
 assistance (section 1451(a)(3)). Tribes
 are also eligible for grants under the
 Drinking Water State Revolving  Fund
 Tribal set aside grant program
 authorized by section 145 2(i) for public
 water system expenditures.
 XIII. HRRCA

 A. What Are the Requirements for the
 HRRCA?
   Section 1412(b)(3)(C) of the  1996
 Amendments requires EPA to  prepare a
 Health Risk Reduction and Cost
 Analysis (HRRCA) in support  of any
 NPDWR that includes an MCL.
 According to these requirements, EPA
 must analyze each of the following
 when proposing a NPDWR that includes
 an MCL: (1) Quantifiable and non-
 quantifiable health risk reduction
 benefits for which there is a factual
 basis in the mlemaking record to
 conclude that such benefits are likely to
 occur as the result of treatment to
 comply with each level; (2) quantifiable
 and non-quantifiable health risk
 reduction benefits for which there is a
 factual basis in the rulemaking record to
 conclude that such benefits are likely to
 occur from reductions in co-occurring
 contaminants that may be attributed
 solely to compliance with the MCL,
 excluding benefits resulting from
 compliance with other proposed or
 promulgated regulations; (3)
 quantifiable and non-quantifiable costs
 for which there is a factual basis  in the
 rulemaking record to conclude that such
 costs  are likely to occur solely  as a
 result of compliance with the MCL,
 including monitoring, treatment, and
 other costs, and excluding costs
 resulting from compliance with other
 proposed or promulgated regulations;
 (4) the incremental costs and benefits
 associated with each alternative MCL
 considered; (5) the effects of the
 contaminant on the general population
 and on groups within the general
 population, such as infants, children,
 pregnant women, the elderly,
individuals with a history of serious
 illness, or other subpopulations that are
 identified as likely to be at greater risk
 of adverse health effects due to exposure
to contaminants in drinking water than
the general population; (6) any
increased health risk that may occur as
 the result of compliance, including risks
 associated with co-occurring
 contaminants; and (7) other relevant
 factors, including the quality and extent
 of the information, the uncertainties in
 the analysis, and factors with respect to
 the degree and nature of the risk.
   This analysis summarizes EPA's
 estimates of the costs and benefits
 associated with various arsenic levels.
 Summary tables are presented that
 characterize aggregate costs and
 benefits, impacts on affected entities,
 and tradeoffs between risk reduction
 and compliance costs. This analysis also
 summarizes the effects of arsenic on the
 general population as well as any
 sensitive subpopulations and provides a
 discussion on the uncertainties in the
 analysis and any other relevant factors.

 B. What Are the Quantifiable and Non-
 Quantifiable Health Risk Reduction
 Benefits?

  Arsenic ingestion has been linked to
 a multitude of health effects, both
 cancerous and non-cancerous. These
 health effects include cancer of the
 bladder, lungs, skin, kidney, nasal
 passages, liver, and prostate. Arsenic
 ingestion has also been attributed to
 cardiovascular, pulmonary,
 immunological, neurological, endocrine,
 and reproductive and developmental
 effects. A complete list of the arsenic-
 related health effects reported in
 humans is shown in Table X-l. Current
 research on arsenic exposure has only
 been able to define scientifically
 defensible risks for bladder cancer.
 Because there is currently a lack of
 strong evidence on the risks of other
 arsenic-related health effects noted
 above, the Agency has based its
 assessment of the quantifiable health
 risk reduction benefits solely on the
 risks of arsenic induced bladder
 cancers. It is important to note that if
 the Agency were able to quantify
 additional arsenic-related health effects,
 the quantified benefits estimates may be
 significantly higher than the estimates
 presented in this analysis.
  The quantifiable health benefits of
reducing arsenic exposures in drinking
water are attributable to the reduced
number of fatal and non-fatal cancers,
primarily of the bladder. Table XIII-1
shows the health risk reductions
(number of total bladder cancers
avoided and the proportions of fatal and
non-fatal bladder cancers avoided) at
various arsenic levels.

-------
38958
Federal Register/Vol. 65, No.  121/Thursday, June  22, 2000/Proposed  Rules
                   TABLE XI11-1.—RISK REDUCTION FROM REDUCING ARSENIC IN DRINKING WATER"
Arsenic level 2(ng/L)
3 	 , 	
5 	
•)0 	
20 	 	
Risk reduction
(total bladder
cancers avoid-
ed per year)
22-42
16-36
9-21
4-12
Risk reduction
(fatal bladder
cancers avoid-
ed per year)
5.7-10.9
4.2-9.4
2.3-5.5
1-3
Risk reduction
(non-fatal
bladder can-
cers avoided
per year)
16.3-31.1
11.8-26.6
296
6.7-15.5
3-9
  1 The number of bladder cancer cases avoided provide our "best" estimates at this time. The actual number of cases could be lower, given the
various uncertainties discussed, or higher, as these estimates assume a 100% mortality rate. An 80% mortality rate is used in the computation of
upper bound benefits.
  The above ranges of total, fatal, and
non-fatal bladder cancer cases are based
on a range of mean bladder cancer risks
for exposed populations at or above
arsenic levels of 3, 5,10, and 20 jig/L
as shown in Table XHI-2. For example,
!f we multiply the risk range at 3 u.g/L
(2.1 x 10 - 5 to 4.5 x 10-5) by the
                      population exposed at 3 u.g/L (26.6
                      million), we find that the total cancers
                      avoided at this arsenic level range from
                      22 to 42 bladder cancers per year, when
                      subtracted from the number of bladder
                      cancers per year at the baseline (50 ug/
                      L). Fatal bladder cancer cases are
                      determined through the relationship
(EPA, 1999a) that approximately 26
percent of the total bladder cancer cases
avoided at each level result in fatalities.
Non-fatal bladder cancer cases are
calculated by subtracting the total
number of cancers from the number of
fatal cancer cases.
                      TABLE XIII-2.—MEAN BLADDER CANCER RISKS AND EXPOSED POPULATION'
Arsenic level (|ig/L)
Baseline (50 p.g/L):
3 	
5 	
•jO 	
20 	
Mean exposed
population risk2
2.1-4.5X10-5
3.6-7.5x10-5
5.5-11.4x10-5
6.9-13.9x10-5
Total bladder can-
cer cases avoided
per year3
22-42
16-36
!9-21
4-12
  ' The population exposed at 3 iig/L or greater is approximately 26.6 million.
  *The bladder cancer risks presented in this table provide our "best" estimates at this time. Actual risks could be lower, given the various un-
certainties discussed, or higher, as these estimates assume a 100% mortality rate. An 80% mortality rate is used in the computation of upper
bound benefits.
  »Total bladder cancer cases avoided could be higher, depending on the survival rate for bladder cancer in the study area of Taiwan for the du-
ration of the study.
  The Agency has developed monetized
 estimates of the health benefits
 associated with the risk reductions from
 arsenic exposures. The SDWA, as
 amended, requires that a cost-benefit
 analysis be conducted for each NPDWR,
 and places a high priority on better
 analysis to support rulemaking. The
 Agency is interested in refining its
 approach to both the cost and benefit
 analysis, and in particular recognizes
 that there are different approaches to
 monetizing health benefits.
  The approach used in this analysis for
 the measurement of health risk
 reduction benefits is the monetary value
 of a statistical life (VSL) applied to each
 fatal cancer avoided. Estimating the VSL
 involves inferring individuals' implicit
 tradeoffs between small changes in
 mortality risk and monetary
 compensation. In this analysis, a central
                      tendency estimate of $5.8 million
                      (1997$) is used in the monetary benefits
                      calculations. This figure is determined
                      for the VSL estimates in 26 studies
                      reviewed in EPA's recent draft guidance
                      on benefits assessment (US EPA, 1997f).
                      It is important to recognize the
                      limitations of existing VSL estimates
                      and to consider whether factors such as
                      differences in the demographic
                      characteristics of the populations and
                      differences in the nature of the risks
                      being valued have a significant impact
                      on the value of mortality risk reduction
                      benefits. Also, medical care or lost-time
                      costs are not separately included in the
                      benefits estimates for fatal cancers, since
                      it is assumed that these costs are
                      captured in the VSL for fatal cancers.
                         For non-fatal cancers, willingness to
                      pay (WTP) data to avoid chronic
                      bronchitis is used as a surrogate to
estimate the WTP to avoid non-fatal,
bladder cancers. The use of such WTP
estimates is supported in the SDWA, as
amended, at section 1412(b)(3)(C)(iii):
"The Administrator may identify valid
approaches for the measurement and
valuation of benefits under this
subparagraph, including approaches to
identify consumer willingness to pay for
reductions in health risks from drinking
water contaminants."
  A WTP central tendency estimate of
$536,000 (in 1997  $) is used to monetize
the benefits of avoiding non-fatal
cancers (Viscusi et al, 1991). The fatal,
non-fatal, and non-quantifiable health
benefits are summarized in Table XI!II—
3. As expected, the quantified bladder
cancer benefits increase as arsenic levels
decrease.

-------
                   Federal Register/Vol. 65, No. 121/Thursday, June^ 22,  2000/Proposed Rules
                                                                          38959
             TABLE XIII-3,—ESTIMATED COSTS AND BENEFITS FROM REDUCING ARSENIC IN DRINKING WATER
                                                      [In 1999 $ millions]
Arsenic level (ng/L)
3 	
5 	
10 	
20 	

Total national
costs to
CWSs'
643.1-753
377.3-441.8
163.3-191.8
61.6-72.9
Total national
costs to CWSs
and
NTNCWSs2
644.6-756.3
378.9-444.9
164.9-194.8
63.2-77.1
Total bladder
cancer health
benefits 3
43.6-104.2
5 (79)
31.7-89.9
5(64.3)
17.9-52.1
5 (37)
7.9-29.8
5(19.8)
"What if" scenario4 and potential non-quantified benefits
"What if" lung
cancer health
benefits esti-
mates
47.2-448
6(213.4)
35-384
6(173.4)
19.6-224
6 (100)
8.8-128
6 (53.4)
Potential non-quantifiable health benefits
Skin Cancer.
Kidney Cancer.
Cancer of the Nasal Passages.
Liver Cancer.
Prostate Cancer.
Cardiovascular Effects.
Pulmonary Effects.
Immunological Effects.
Neurological Effects.
Endocrine Effects.
Reproductive and Developmental Effects.
   1 Costs include treatment, monitoring, O&M, and administrative costs to CWSs and State costs for administration of water programs The lower
 number shows costs annualized at a consumption rate of interest of 3%, EPA's preferred approach. The higher number shows costs annualized
 at 7%, which represents the standard discount rate preferred by OMB for benefit-cost analyses of government programs and regulations
   2 Costs include treatment, monitoring, O&M, administrative costs to CWSs; monitoring and administrative costs to NTNCWSs- and State costs
 for administration of water programs.
   3 The upper bound estimate includes an adjustment to account for a possible mortality risk of 80%. It is possible that this risk could have been
 below 80%, which would lead to increased benefits. The actual risk depends on the survival rate for bladder cancer in the area of Taiwan studied
 by Chen, which is unknown.
   4 These estimates are based on the "what if" scenario for lung cancer, where the  risks of a fatal lung cancer case associated with arsenic are
 assumed to be 2-5 times that of a fatal bladder cancer case.
   5 The number in parentheses indicates the bladder cancer health benefits assuming an 80% mortality rate for bladder cancer in the area of the
 Chen study, and starting from the midpoint of the benefits range when mortality and incidence are assumed equivalent.
   «The number in parentheses is the midpoint of the range and corresponds to an assumption that the risk of fatal lung cancer is 3.5 times the
 nsk of fatal bladder cancer.
  Reductions in arsenic exposures may
also be associated with non-quantifiable
benefits. EPA has identified several
potential non-quantifiable benefits
associated with regulating arsenic in
drinking water. In addition to the non-
quantifiable benefits noted in Table
XIII—3, these benefits may include any
customer peace of mind from knowing
that their drinking water has been
treated for arsenic. Also, using reverse
osmosis to remove arsenic from
drinking water may also reduce other
contaminants such as sulfate, nitrate,
and iron due to the high removal
efficiency of this treatment technology.
 C. What Are the Quantifiable and Non-
 Quantifiable Costs?
  The costs of reducing arsenic to
 various levels are summarized in Table
 XIII—4, which shows that, as expected,
 aggregate arsenic mitigation costs
 increase with decreasing arsenic levels.
 Total national costs range from $646
 million per year at 3 (ig/L to $65 million
 per year at 20 jig/L.
              TABLE XIII-4.—ESTIMATED ANNUALIZED NATIONAL COSTS OF REDUCING ARSENIC EXPOSURES
                                                     [In 1999 $ millions]
Arsenic level ([ig/L)
3 	
5 	
10 	
20 	
Costs to
CWSs1
639—746 4
374—433
160 187
59-68
Total national
costs to
CWSs 2
R4<3 1 7R^
•377 T— 441 fi
163 3—191 8
61.6-72.9
Total national
costs to CWSs
and
NTNCWSs3
RAA fi 7RR ^
oyo (X-4A/1 Q
1 fi4 Q 1 CM. 8
63.2-77.1
Total cost per fatal bladder
cancer case avoided 4



20.3-60.7 (26-77.1)1.
  1 Costs include treatment and O&M costs only. The lower number shows costs annualized at 3 percent; the higher number shows costs
annualized at 7%. The 7% rate represents the standard discount rate preferred by OMB for benefit-cost analyses of government programs and
regulations.   .                                                               •
  2Costs include treatment, monitoring, O&M, and administrative costs to CWSs and State costs for administration of water proqrams Costs
annualized at 3 and 7 percent.                                                                                  v a
  3 Costs include treatment, monitoring, O&M, administrative costs to CWSs; monitoring and administrative costs to NTNCWSs; and State costs
for administration of water programs. Costs annualized at 3 and 7 percent.
  4 Range based on range of fatal bladder cancer cases avoided per year shown in Table XIII.1. The range of costs per fatal bladder
avoided could be one-half of the value presented, depending on the mortality rate for bladder cancer in the study area of Taiwan for the c
of the study. A plausible estimate for that mortality rate is 80%
                                                                           cancer
                                                                          duration
  The cost impact of reducing arsenic in
drinking water at the household level
was also assessed. Table XIII-5
examines the cost per household for
each system size category. As shown in
the table, costs per household decrease
as system size increases. Costs per
household also do not vary significantly
across arsenic levels. This is because

-------
38960
Federal Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed Rules
costs do not vary significantly with
removal efficiency; once a system
installs a treatment technology to meet
an MCL, costs based upon the removal
                     efficiency that the treatment technology
                     will be operated under remain relatively
                     flat. Per household costs are, however,
                     somewhat lower at less stringent arsenic
levels. This is due to the assumption
that some systems would blend water at
these levels and treat only a portion of
the flow.
 TABLE XI11-5.—ESTIMATED ANNUAL COSTS PER HOUSEHOLD1 (IN 1999 $) AND (NUMBER OF HOUSEHOLDS AFFECTED)
System s\ze
25-100 	
101-500 	
501-1 000 	
1001-3300 	 	
3301-10000 . 	
10,001-50,000 	
50001-100000 	
100001-1 million 	 	

3ng/L
$368
(93,900)
$259
(366,900)
$106
(356,000)
$64
2(1)
$44
2(1.6)
$36
2 (3.25)
$30
2 (1-4)
$23
2 (3.1)
SHQ/L
$364
(58,600)
$254
(229,000)
$104
(223,000)
$60
(626,000)
$41
2(1)
$33
2(2.1)
$27
2 (0.9)
$21
2(1.8)
10ng/L
$357
(27,000)
$246
(103,000)
$98
(102,000)
$57
(290,000)
$37
(478,000)
$29
(998,000
$23
(465,000)
$18
(937,000)
20 (ig/L1
$349
(10,000)
$238
(41,000)
$93
(41,000)
$52
(118,000)
'$33
(196,000)
$25
(406,000)
'$19
(189,000)
$15
(365,000)
  1 Costs include treatment and O&M costs to CWSs only.
  s Million.
  Costs per household are higher for
households served by smaller systems
than larger systems for two reasons.
First, smaller systems serve far fewer
households than larger systems and,
consequently, each household must bear
a greater percentage share of the capital
and O&M costs. Second, smaller
systems tend to have higher influent
arsenic concentrations that, on a per-
capita or per-household basis, require
                     more expensive treatment methods (e.g.,
                     a higher percentage removal efficiency)
                     to comply with an MCL.
                      Table XIII-6 summarizes the
                     estimates of total national costs of
                     compliance with the proposed MCL
                     options of 3, 5, and 10, and 20 ug/L.
                     This table is divided into two major
                     groupings; the first grouping displays
                     the estimated costs to Community Water
                     Systems (CWSs) and the second
grouping displays the estimated costs to
Non-Transient Non-Community Water
Systems (NTNCWSs). The State costs
presented in Table XIII-6 were
developed as part of the analyses to
comply with the Unfunded Mandates
Reform Act (UMRA) and also the
Paperwork Reduction Act (PRA).
Additional information on State costs is
provided in Section XIV of this
preamble.
   TABLE XIII-6.—SUMMARY OF THE TOTAL ANNUAL NATIONAL COSTS OF COMPLIANCE WITH THE PROPOSED ARSENIC
                                          RULE ACROSS MCL OPTIONS
                                               [In 1997$ millions]1
Costs
Cost of capital
CWS
3 percent
7 percent
NTNCWS
3 percent
7 percent
                                                    3|ig/L
Treatment 	
Monitoring Reporting & Recordkeeping 	 	
State & EPA Administrative Costs 	

Total Costs 	
639.2
2.2
2.2

643.6
746.4
2.9
3.7

753.0
* (25.2)
0.95
1.1

*1.2 (27.3)
*(30.5)
! 1.1
2.2

* 3.3 (3*3.8)
                                                     5(ig/L
Treatment 	
Monitoring Reporting & Recordkeeping 	
State & EPA Administrative Costs 	

Total Costs 	

373.9
1.9
1.8

377.8

436.0
2.7
3.1

441.8

*(14.7)
0.92
1.0

* 1.2 (16.6)

*(17.8)
1.1
'2.0

*3.1 (20.9)
v i '
                                                    10
Treatment 	 	
Monitoring, Reporting J
State & EPA Administr
Total Costs 	

« Recordkeeping 	
stive Costs

160.4
1.8
1.5
163.7
186.7
2.5
2.6
191.8
0.90
0.93
*1.8 (7.9)
"(7.4)
1.1
J1.9
*3.0 (10.3)

-------
                 Federal Register/Vol. 65, No. 121/Thursday, June •• 22,  2000 / Proposed Rules
                                                                    38961
  TABLE Xlli-6.—SUMMARY OF THE TOTAL ANNUAL NATIONAL COSTS OF COMPLIANCE WITH THE PROPOSED ARSENIC
                                     RULE ACROSS MCL OPTIONS—Continued
                                                 [In 1997$ millions]'      ]
Costs
Cost of capital
CWS
3 percent
7 percent
NTNCWS
3 percent
7 percent
                                                      20ug/L
Treatment 	
Monitoring Reporting & Recordkeeping 	
State & EPA Administrative Costs 	

Total Costs 	
58.9
1.7
1.3

61.9
68.3
2.4
2.3

72.9
*(2.1)
2.0
0.91

* 2.9 (5.1)
*(2.6)
2.3
1.9

* 4.2 (6.7)
  1 Totals may not add due to rounding.
  * Costs in parentheses include treatment costs if NTNCWS had to comply with the MCL.
                                   D. What Are the Incremental Benefits and Costs?
   Table XIII-7 summarizes the incremental benefits and costs associated with arsenic exposure reduction.

   TABLE XIII-7.—ESTIMATES OF THE ANNUAL INCREMENTAL RISK REDUCTION, BENEFITS, AND COSTS OF REDUCING
                                           ARSENIC IN DRINKING WATER
                                                   [$millions, 1999]       ;
Arsenic level
Incremental Risk Reduction Fatal Bladder Cancers Avoided Per Year 	
Incremental Risk Reduction, Non-Fatal Bladder Cancers Avoided Per Year
Annual Incremental Monetized Benefits1 	
Annual Incremental Costs2 	
20 ug/L
1-3
3-9
7.9-29.8
63.2
10ng/L
1 .3-2.5
3.7-6.5
10-22.3
101.7
5ng/L
1.9-3.9
5.1-11.1
13.8-37.8
214
3ng/L
1.5-1.5
4.5-4.5
11.9-14.3
265.7
  1 The incremental upper bound benefits estimates presented in this table have been adjusted upwards to reflect an 80% mortality rate, which is
a plausible mortality rate for the area of Taiwan during the Chen study.
  2 Costs include treatment, monitoring, O&M, and administrative costs to CWSs; monitoring and administrative costs to NTNCWSs and State
costs.
E. What Are the Risks of Arsenic
Exposure to the General Population and
Sensitive Subpopulations?

  The SDWA, as amended, includes
specific provisions in section
1412(b)(3)(CKiKV) to assess the effects
of the contaminant on the general
population and on groups within the
general population such as children,
pregnant women, the elderly,
individuals with a history of serious
illness, or other subpopulations that are
identified as likely to be at greater risk
of adverse health effects due to exposure
to contaminants in drinking water than
the general population. The NRG Report
concluded that there is insufficient
scientific information to permit separate
cancer risk estimates for potential
subpopulations such as pregnant
women, lactating women, and children
and that factors that influence
sensitivity to or expression of arsenic-
associated cancer and noncancer effects
need to be better characterized. Despite
the inconclusive nature of the effects on
subpopulations, EPA is planning to
issue a health advisory for arsenic in
early 2000. See section IV.C of this
preamble for further information on the
health advisory.
F. What Are the Risks Associated With
Co-Occurring Contaminants?

  The SDWA, as amended, requires
EPA to take into account the activities
under preceding rules that may have
impacts on future rules. To address this
requirement, EPA analyzed the co-:
occurrence of arsenic with other
drinking water contaminants (EPA,
1999f). The results of this analysis help
determine the level of overlap in
regulatory requirements (cost of
technology that can remove more than
one contaminant) and also indicate
where specific levels of one
contaminant may interfere with the
treatment technology for another. This
analysis indicates that there is some co-
occurrence of arsenic with sulfate,;iron,
and radon. Co-occurrence can also1
indicate the likelihood for increased, or
in this case, decreased risks due to
arsenic and selenium.
  As discussed in section XI.A.5. of the
preamble, animal studies suggest that
selenium reduces the toxicity of arsenic,
and people in Taiwan have much lower
levels of selenium in their blood and
urine than people in China, the U.S.,
and Canada. Deficient selenium intake
is linked to heart problems, and
excessive intake can lead to thick brittle
nails and changes in the nervous
system. The U.S. recommends a daily
dietary intake of 55 u,g/day for females
and 70 u,g/day for males. The WHO
lower limit of safe ranges are 30 (for
females) and 40 (for males) jig/day
(NRG, 1990). EPA's study of co-
occurrence of arsenic (at 2, 5,10, 20,
and > 20 (ig/L) and selenium above 50
(ig/L levels found no significant
correlations between arsenic and
selenium. EPA believes that, in general,
the U.S. population does not experience
selenium toxicity which would be
reduced by the presence of arsenic and
that there is sufficient selenium in the
American diet to reduce the toxicity of
arsenic. The Agency requests data and
comments on whether selenium
decreases arsenic toxicity on a regional
basis. Section V of this preamble
summarizes the results of EPA's arsenic
co-occurrence analysis.
G. What Are the Uncertainties in the
Analysis?
  The models used to estimate arsenic-
related cancer risks, risk reduction, and
monetary benefits take many inputs
which are both uncertain and highly
variable. The benefits estimates that
have been discussed in this preamble
were derived using point estimates of

-------
38962
Federal  Register/Vol. 65, No. 121/Thursday, June  22,  2000/Proposed Rules
the monetary surrogates for fatal and
non-fatal bladder cancers. The value of
statistical life (VSL) has been
approximated by a single-value estimate
of S5.8 million, and willingness-to-pay
(WTP) to avoid non-fatal bladder cancer
has been modeled as a constant with a
value of 5536,000. These are the central
tendency values derived by EPA, based
on studies from the economic literature
and previous regulatory analyses (US
EPA 1997f, Viscusi ef al, 1991).
Because the VSL is much larger than the
WTP value, the VSL value dominates
the total monetary benefits calculation.
  The studies that have been reviewed
by EPA (US EPA 1997f) have developed
a wide range of VSL values, from
§700,000 to S16.3 million. This implies
that the monetized benefits of reduced
bladder cancer risks could take a wide
range of values, depending upon the
VSL that is chosen.
  Additional sources of uncertainty in
this analysis are also found in the NRG
Report. Such uncertainties include the
shape of the dose-response curve, the
contribution of arsenic exposure from
food, and the choice of model when
conducting arsenic risk assessment.
These sources of uncertainties are
discussed in further detail in section XI.
of today's document.
XIV. Administrative Requirements
A. Executive Order 12866: Regulatory
Planning and Review
  Under Executive Order 12866,
"Regulatory Planning and Review" (58
FR 51735, October 4,1993), the Agency
must determine whether the regulatory
action is "significant" and therefore
subject to OMB review and the
requirements of the Executive Order.
The Order defines "significant
regulatory action" as one that is likely
to result in  a rule that may:
  (l) Have an annual effect on the
economy of S100 million or more or
adversely affect in a material way the
economy, a sector of the economy,
productivity, competition, jobs, the
environment, public health or safety, or
State, local, or tribal governments or
communities;
  (2) Create a serious inconsistency or
otherwise interfere with an action taken
or planned  by another agency;
  C3) Materially alter the budgetary
impact of entitlements, grants, user fees,
or loan programs or the rights and
obligations of recipients thereof; or
  (4J Raise novel legal or policy issues
arising out of legal mandates, the
President's priorities, or the principles
set forth in  the Executive Order.
                       Pursuant to the terms of Executive
                     Order 12866, it has been determined
                     that this rule is a "significant regulatory
                     action". As such, this action was
                     submitted to OMB for review. Changes
                     made in response to OMB suggestions or
                     recommendations will be documented
                     in the public record.

                     B. Regulatory Flexibility Act (RFA), as
                     Amended by the Small Business
                     Regulatory Enforcement Fairness Act of
                     1996 (SBREFA), 5 U.S.C. 601 et seq.

                     1. Overview
                       The RFA generally requires an agency
                     to prepare a regulatory flexibility
                     analysis of any rule subject to notice
                     and comment rulemaking requirements
                     under the Administrative. Procedure Act
                     or any other statute unless the Agency
                     certifies that the rule will not have a
                     significant economic impact on a
                     substantial number of small  entities.
                     Small entities include small businesses,
                     small organizations, and small
                     governmental jurisdictions.

                     2. Use of Alternative Small Entity
                     Definition
                       The RFA provides  default definitions
                     for each type of small entity. It also
                     authorizes an agency to use alternative
                     definitions for each category of small
                     entity, "which are appropriate to the
                     activities of the agency" after proposing
                     the alternative definition(s) in the
                     Federal Register and taking comment (5
                     U.S.C. 60l(3)-(5)). In addition to the
                     above, to establish an alternative small
                     business definition, agencies must
                     consult with SBA's Chief Counsel for
                     Advocacy.
                       EPA is proposing the Arsenic Rule
                     which contains provisions which apply
                     to small PWSs serving fewer than
                     10,000 persons. This is the cut-off level
                     specified by Congress in the  1996
                     Amendments to the Safe Drinking Water
                     Act for small system flexibility
                     provisions. Because this definition does
                     not correspond to the definitions of
                     "small" for small businesses,
                     governments, and non-profit
                     organizations, EPA requested comment
                     on an alternative definition of "small
                     entity" in the preamble to the proposed
                     Consumer Confidence Report (CCR)
                     regulation (63 FR 7605 at 7620,
                     February 13,  1998, US EPA 1998J).
                     Comments showed that stakeholders
                     supported the proposed alternative
                     definition. EPA also consulted with the
                     SBA Office of Advocacy on the
                     definition as it relates to small business
                     analysis. In the preamble to the final
                     CCR regulation (63 FR 44511, August
19, 1998, US EPA, 1998e), EPA stated its
intent to establish this alternative     ;
definition for regulatory flexibility
assessments under the RFA for all
drinking water regulations and has thus
used it in this proposed rulemaking.

3. Initial Regulatory Flexibility Analysis

  In accordance with section 603 of the
RFA, EPA prepared  an initial regulatory
flexibility analysis (IRFA) that examines
the impact of the proposed rule on small
entities along with regulatory
alternatives that could reduce that
impact. The IRFA is available for review
in the docket and is  summarized below.
  The RFA requires EPA to address the
following when completing an  IRFA:
  (1) Describe the reasons why action by
the Agency is being  considered;
  (2) State succinctly the objectives of,
and legal basis for, the proposed rule;
  (3) Describe, and where feasible,
estimate the types and number of small
entities to which the proposed rule will
apply;
  (4) Describe the projected reporting,
record keeping, and other compliance
requirements of the rule, including an
estimate of the classes of small entities
that will be subject to the requirements
and the type of professional skills
necessary for preparation of reports or
records;
  (5) Identify, to the extent practicable,
all relevant Federal rules that may
duplicate, overlap, or conflict with the
proposed rule; and
  (6) Describe any significant
alternatives to the proposed rule that
accomplish the stated objectives of
applicable statutes while minimizing
any significant economic impact of the
proposed rule on small entities.
  EPA has considered and addressed all
of the previously described
requirements. The following is  a
summary of the IRFA. The first and
second requirements are discussed in
section LA. of this Preamble. The third
and fourth requirements are
summarized as follows. The fifth
requirement is discussed under section
VIII.F.  of this Preamble in a subsection
addressing potential interactions
between the arsenic rule and upcoming
and existing rules affecting community
water systems. The sixth requirement,
regulatory alternatives, is detailed in
section XIII.                        ;
  a. Number of Small Entities Affected.
The number of small entities subject to
today's rule is shown in Table XIV-1
below.

-------
                 Federal Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed Rules
                                                                    38963
    TABLE XIV-1.—PROFILE OF THE UNIVERSE OF SMALL WATER SYSTEMS REGULATED UNDER THE ARSENIC RULE
Water system type
Publicly-Owned:
CWS 	
NCWS 	
Privately-Owned:
CWS 	
NCWS 	
Total Systems:
CWS
NCWS 	
Total 	
System size category
<100
1,729
1,783
13,640
8,178
15,369
9,961
25.330
101-500
5,795
3,171
11,266
4,162
17,061
7,333
24.394
501-1,000
3,785
1,182
2,124
902
5,909
2,084
7.993
1,001-3,300
6,179
361
1,955
411
8,134
772
8.906
3,301-10,000
3,649
29
654
56
4,303
85
4.388
  Source: Safe Drinking Water Information System (SDWIS), December 1998 freeze.
  b. Reporting, Recordkeeping and
Other Requirements for Small Systems.
The proposed arsenic rule continues to
require small systems to maintain
records and to report arsenic
concentration levels at the point-of-
entry to the water system's distribution
system. Small systems are also required
to provide arsenic information in the
Consumer Confidence Report or other
public notification if the system exceeds
the MCL. Arsenic monitoring and
reporting will be required annually for
surface water (and mixed surface and
ground water systems) or once every
three years for ground water systems,
unless the small system obtains a
monitoring waiver from the State,
demonstrating compliance with the
proposed MCL.  Other existing
information and reporting requirements,
such as Consumer Confidence Reports
and public notification requirements,
will be revised to include the lower
arsenic MCL (see sections VII. H. and I.).
As is the case for other contaminants,
required information on system arsenic
levels must be provided by affected
systems and is not considered to be
confidential. The professional skills
necessary for preparing the reports are
the same skill level required by small
systems for current reporting and
monitoring requirements for other
drinking water standards.
  The classes of small entities that are
subject to the proposed arsenic rule
include public water systems serving
less than 10,000 people.

4. Small Business Advocacy Review
(SBAR) Panel Recommendations

  As required by section 609(b) of the
RFA, as amended by SBREFA, EPA also
conducted outreach to small entities
and convened a Small Business
Advocacy Review Panel to obtain advice
and recommendations of representatives
of the small entities that potentially
would be subject to the rule's
requirements.
  EPA identified 22 representatives of
small entities that were most likely to be
subject to the proposal. In December,
1998, EPA prepared and distributed to
the small entity representatives (SERs)
an outreach document on the arsenic
rule titled "Information for Small Entity
Representatives Regarding the Arsenic
in Drinking Water Rule" (US EPA,
1998g).
  On December 18, 1998, EPA held a
small entity conference call from  :
Washington D.C. to provide a forum for
small entity input on key issues related
to the planned proposal of the arsenic
in drinking water rule. These issues
included, but were not limited to issues
related to the rule development, such as
arsenic health risks, treatment
technologies, analytical methods, arid
monitoring. Fifteen SERs from small
water systems participated on the call
from the following States: Alabama,
Arizona, California, Georgia,
Massachusetts, Montana, Nebraska; New
Hampshire, New Jersey, Utah, Virginia,
Washington, and Wisconsin.
  Efforts to identify and incorporate
small entity concerns into this
rulemaking culminated with the
convening of a SBAR Panel on March
30, 1999, pursuant to section 609 of
RFA/SBREFA. The four-person Panel
was headed by EPA's Small Business
Advocacy Chairperson and included the
Director of the Standards and Risk
Management Division within EPA's
Office of Ground Water and Drinking
Water, the Administrator of the Office of
Information and Regulatory Affairs with
the Office of Management and Budget,
and the Chief Counsel for Advocacy of
the SBA. For a 60-day period starting on
the convening  date, the Panel reviewed
technical background information
related to this rulemaking, reviewed
comments provided by the SERs, and
met on several occasions. The Panel also
conducted its own outreach to the SERs
and held a conference call on April 21,
1999 with the SERs to identify issues
and explore alternative approaches for
accomplishing environmental
protection goals while minimizing
impacts to small entities. Consistent
with the RFA/SBREFA requirements,
the Panel evaluated the assembled
materials and small-entity comments on
issues related to the elements of the
IRFA. A copy of the June 4,1999 Panel
report is included in the docket for this
proposed rule (US EPA, 1999c).
  Today's notice incorporates all of the
recommendations on which the Panel
reached consensus, except for a number
of recommendations on information to
include in small system guidance. The
small system guidance materials will be
provided before or soon after the final
rule is published in the Federal
Register. EPA is committed to
addressing the following Panel
recommendations regarding guidance
for small systems: highlight the various
waste disposal options and the
necessary technical and procedural
steps for small CWSs to follow in
exploring these alternatives; provide
specific recommendations and technical
information relative to the use of POU
devices; provide guidance to State and
local authorities on waste disposal
issues relative to the use of these
devices; and provide information to
assist in making treatment decisions to
address multiple contaminants in the
most cost-effective manner. The Panel
also recommended that EPA provide
guidance identifying cost-effective
treatment trains for ground water
systems that need to treat for both
arsenic and radon in the proposed rule.
However, treatment trains cannot be
accurately identified until after the
radon and arsenic standards are
finalized because these standards would
affect which treatment technologies are
appropriate. Since the co-occurrence of

-------
 38964
Federal Register/Vol. 65, No. 121/Thursday, June  22,  2000/Proposed Rules
 arsenic and radon seems to be
 statistically significant in only two EPA
 regions, the impact from this co-
 occurring pair is not significant on a
 national level. However, for the regions
 which are impacted, there is the
 potential that aeration treatment
 technology that may be used to mitigate
 radon may also help to mitigate arsenic.
 Aeration technology can oxidize the
 soluble form  of arsenic to the insoluble
 form. This would reduce the cost of
 arsenic mitigation by making it easier to
 remove arsenic. EPA will address this
 recommendation further in the small
 system guidance materials.
  The following is a summary of the rest
 of the Panel recommendations and
 EPA's response to these
 recommendations, by subject area:
  Treatment  Technologies, Waste
 Disposal, and Cost Estimates: The Panel
 recommended the following: further
 develop the preliminary treatment and
 waste disposal cost estimates; fully
 consider these costs when identifying
 affordable compliance technologies for
 all system size categories; and provide
 information to small water systems on
 possible options for complying with the
 MCL, in addition to installing any listed
 compliance technologies.
  In response to these
 recommendations, the treatment section
 of the preamble (see section VIII.A.) and
 the Treatment and Cost document (US
 EPA, 19995) describe the development
 of final cost estimates for treatment and
 waste disposal, including the request for
 comment on its projected household
 costs; how EPA identified the affordable
 compliance technologies, including the
 consideration of cost (section vm.B.);
 and information has been added to the
 treatment section about options for
 complying with the MCL other than
 installing compliance technologies,
 such as selecting to regionalize (see
 section VIII.B.).
  Regarding POU devices, the Panel
 recommended the following: continue
 to promote the use of POU devices as
 alternative treatment options for very
 small systems where appropriate;
 account for all costs, including costs
 that may not routinely be explicitly
 calculated; and consider liability issues
 from POU/POE devices when evaluating
 their appropriateness as compliance
 technologies; and investigate waste
 disposal issues with POE devices.
  In response to these
recommendations, the treatment section
of the preamble: includes an expanded
description regarding available POU
compliance treatment technologies and
conditions under which POU treatment
may be appropriate for very small
systems (see section VIII.D.); describes
                     the components which contribute to the
                     POU cost estimates (see section VIII.D.);
                     and clarifies that water systems will be
                     responsible for POU operation and
                     maintenance to prevent liability issues
                     from customers maintaining equipment
                     themselves (see section VIII.D.). In
                     addition, EPA does not recommend
                     reverse osmosis as a POE treatment
                     technology due to the evaluation of
                     corrosion control issues (see section
                     VIII.D.).
                       Relevance of Other Drinking Water
                     Regulations: The Panel recommended
                     the following: include discussion of the
                     co-occurrence of arsenic and radon in
                     the proposed rule for arsenic;  take
                     possible  interactions among treatments
                     for different contaminants into account
                     in costing compliance technologies and
                     determining whether they are nationally
                     affordable for small systems; and
                     encourage systems to be forward-
                     looking and test for the multiple
                     contaminants to determine if and how
                     they would be affected by the  upcoming
                     rules.
                       In response, the co-occurrence  section
                     of the preamble includes  a discussion
                     on the co-occurrence analysis of radon
                     and arsenic (see section V.H.), and the
                     treatment section of the preamble has
                     been expanded to describe the
                     relationship of treatment  for arsenic
                     with other drinking water rules and how
                     this issue was taken into account in cost
                     estimates (see  section VIII.F.). The
                     preamble encourages systems  to
                     consider other upcoming  rules when
                     making future plans on monitoring or
                     treatment (see section VIII.E.).
                       Small Systems Variance Technologies
                     and National Affordability Criteria: The
                     Panel recommended the following:
                     include a discussion of the issues
                     surrounding appropriate adjustment of
                     its national affordability criteria to
                     account for new regulatory
                     requirements;  consider revising its
                     approach to national affordability
                     criteria to address the concern that the
                     current cumulative approach for
                     adjusting the baseline household  water
                     bills is based on chronological order
                     rather than risk, to the extent allowed by
                     statutory and regulatory requirements;
                     and examine the data in the 1995
                     Community Water Supply Survey to
                     determine if in-place treatment
                     baselines can be linked with the current
                     annual water bill baseline in each of the
                     size categories for the proposed rule.
                       In response to these recommendation,
                     the treatment section of the preamble
                     (VIILC.) includes an expanded
                     discussion about the national
                     affordability criteria and adjusting it  to
                     account for new regulations;
                     information and rationale have been
added to explain the national
affordability approach (see section
VIILC.). The 1995 Community Water
System Survey (US EPA, 1997g) does
not provide sufficient data to link in-
place treatment baselines with annual
water bill baselines.
  Monitoring and Arsenic Species: The
Panel recommended that EPA consider
allowing States to use recent      .
compliance monitoring data to satisfy
initial sampling requirements or to
obtain a waiver and that EPA continue
to explore whether or not to make a
regulatory distinction between organic
and inorganic arsenic based on    .
compliance costs and other
considerations. In response, the    '.
monitoring section of the preamble and
the proposed regulatory language
describe the allowance of monitoring
data that meet analytical requirements
and have reporting limits sufficiently
below the revised MCL and collected
after 1990. The MCL section of the
preamble contains information and
rationale to support EPA's decision to
base the MCL on total arsenic (see
section XI).
  Considerations in setting the MCL:
The Panel recommended the following:
in performing its obligations under
SDWA, take cognizance of the scientific
findings, the large scientific       •
uncertainties, the large potential costs
(including treatment and waste disposal
costs), and the fact that this standard is
scheduled for review in the future; give
full consideration to the provisions of
the Executive Order 12866 and to the
option of exercising the new statutory
authority under SDWA sections
1412(b)(4)(C) and 1412(b)(6)(A) in the
development of the arsenic rule; and
fully consider all of the "risk
management" components of its
rulemaking effort to ensure that the
financial and  other impacts on small
entities are factored into its decision-
making processes. The Panel also
recommended that EPA take into
account both quantifiable and non-
quantifiable costs and benefits of the
standard and the needs of sensitive sub-
populations, and give due consideration
to the impact of the rule upon small
systems.
  In response to all these
recommendations, EPA describes in
detail the factors that were considered
in setting in the MCL and provides the
rationale for this selection (see section
XI).
  Applicability of proposal: The Panel
recommended that EPA carefully
consider the appropriateness of
extending the scope of the rule to Non-
Transient, Non-Community Water
Systems (NTNCWSs). In response, the

-------
                 Federal Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed  Rules
                                                                      38965
proposed MCL for arsenic does not
apply to NTNCWSs and the MCL
section of the preamble describes the
basis for this decision, including the
incremental costs and benefits
attributable to coverage of these water
systems (see section XI.C.).
  Other Issues: The Panel recommended
that EPA encourage small systems to
discuss their infrastructure needs for
complying with the arsenic rule with
their primacy agency to determine their
eligibility for DWSRF loans, and if
eligible, to ask for assistance in applying
for the loans. In response, the UMRA
section XIV.C. has been expanded to
discuss funding options for small
systems, and guidance will be written to
encourage systems to be proactive in
communicating with their primacy
agency.
  Regarding health effects, the Panel
recommended the following: Further
evaluate the Utah study and its
relationship to the studies on which the
NRG report was based and give it
appropriate weight in the risk
assessment for the proposed arsenic
standard; and examine the NRG
recommendations in the light of the
uncertainties associated with the
report's recommendations, and any new
data that may not have been considered
in the NRG report. In response to these
recommendations, the benefits and MCL
sections (sections X and XI) describe the
quantitative and non-quantitative
benefits evaluation and use of research
data.
  We invite comments on all aspects of
the proposal and its impacts on small
entities.
C. Unfunded Mandates Reform Act
(UMRA)
  Title II of the Unfunded Mandates
Reform Act of 1995 (UMRA), Public
Law 104-4, establishes requirements for
Federal agencies to assess the effects of
their regulatory actions on State, local,
and tribal governments and the private
sector. Under UMRA section 202, EPA
generally must prepare a  written
statement, including a cost-benefit
analysis, for proposed and final rules
with "Federal mandates" that may
result in expenditures to  State, local,
and tribal governments, in the aggregate,
or to the private sector, of $100 million
or more in any one year.  Before
promulgating an EPA rule, for which a
written statement is needed, section 205
of the UMRA generally requires EPA to
identify and consider a reasonable
number of regulatory alternatives and
adopt the least costly, most cost-
effective or least burdensome alternative
that achieves the objectives of the rule.
The provisions of section 205 do not
apply when they are inconsistent with
applicable law. Moreover, section 205
allows EPA to adopt an alternative other
than the least costly, most cost-effective
or least burdensome alternative if the
Administrator publishes with the final
rule an explanation on why that
alternative  was not adopted.
  Before EPA establishes any regulatory
requirements that may significantly or
uniquely affect small governments,1
including tribal governments, it must
have developed, under section 203' of
the UMRA, a small government agency
plan. The plan must provide for
notifying potentially affected small
governments, enabling officials of
affected small governments to have
meaningful and timely input in the
development of EPA regulatory
proposals with significant Federal
intergovernmental mandates and
informing,  educating, and advising
small governments on compliance with
the regulatory requirements.
1. Summary of UMRA Requirements
  EPA has  determined that this rule
contains a Federal mandate that may
result in expenditures of $100 million or
more for State, local, and Tribal
governments, in the aggregate, and the
private sector in any one year.     ;
Accordingly, EPA has prepared, under
section 202 of the UMRA, a written
statement addressing the following
areas:
   (1) Authorizing legislation;
   (2) cost-benefit analysis including an
analysis of the extent  to which the 'costs
to State, local, and tribal governments
will be paid for by the Federal
government;
   (3) estimates of future compliance
costs and disproportionate budgetary
effects;
   (4) macro-economic effects; and
   (5) a summary of EPA's consultation
with State, local, and  tribal
governments, a summary of their
concerns, and a summary of EPA's
evaluation of their concerns.      :
   A summary of this analysis follojws
and a more detailed description is;
presented in EPA's  Regulatory Impact
Analysis (RIA) of the Arsenic Rule'(US
EPA, 2000e) which is included in the
docket for this proposed rulemaking.
   a. Authorizing legislation. Today's
proposed rule is proposed pursuant to
section 1412(b)(13) of the 1996   \
amendments to the SDWA which
requires EPA to propose and promulgate
a national primary drinking water :
regulation  for arsenic, establishes a
statutory deadline of January 1, 2000, to
propose this rule, and establishes a
statutory deadline of January 1, 2001, to
promulgate this rule.
  b. Cost-benefit analysis. Section XIII.
of this Preamble, describing the
Regulatory Impact Analysis (RIA) and
Health Risk Reduction and Cost
Analysis (HRRCA) for arsenic, contains
a detailed cost-benefit analysis in
support of the arsenic rule. Today's
proposed rule is expected to have a total
annualized cost of approximately $379
to 445 million.9 This total annualized
cost includes the total annual
administrative costs of State, local, and
tribal governments, in aggregate, less
than 1% of the cost, and total annual
treatment (CWS only, as proposed),
monitoring, reporting, and record
keeping impacts on public water
systems, in aggregate, of approximately
$376.7 to 439.8 million.10 Treatment
costs estimates are presented in Sections
IX.D. and E. of this Preamble, and
administrative costs are discussed in
section 9 of the RIA (US EPA, 2000e).
  The RIA includes both qualitative and
monetized benefits for improvements in
health and safety. EPA estimates the
proposed arsenic rule will have annual
monetized benefits for bladder cancer of
approximately $43.6 to 104.2 million if
the MCL were to be set at 3 |ig/L, $31.7
to 89.9 million if set at 5  (ig/L, $17.9 to
52 million if set at 10 Hg/L, and $7.9 to
29.8 million if set at 20(ig/L (EPA also
estimates possible lung cancer benefits
based on the "What If" scenario of $47—
448 million at 3 Ug/L, $35-384 million
at 5 |ig/L, $19.6-224 million at 10 (ig/
L, and $8.8-128 million at 20 (ig/L.).11
The monetized health benefits of
reducing arsenic exposures in drinking
water are attributable to the reduced
incidence of fatal and non-fatal  bladder
cancers. Under baseline assumptions
(no control of arsenic exposure <50 jig/
L),  10—17 fatal bladder cancers and 29—
48 non-fatal bladder cancers per year are
associated with arsenic exposures
through CWSs. At a arsenic level of 3
|ig/L, an estimated 5.7 to  10.9 fatal
bladder cancers and 22 to 42 non-fatal
bladder cancers per year  are prevented.
At a level of 5 |ig/L, an estimated 4 to
9 fatal bladder cancers and 16 to 36 non-
fatal bladder cancers per year are
prevented. At a level 10 (ig/L, 2 to 6 fatal
and 9 to 21 non-fatal bladder cancers
per year are prevented. At a level 20 jig/
L, 1 to 3 fatal and 3 to 9 non-fatal
bladder cancers per year  are prevented.
  9 Costed as proposed, using the 3 percent and 7
percent discount rate cost-of-capital values in Table
X-8, in 1999 $ with NTNCWS monitoring and
reporting, but not required to comply with the MCL.
If NTNCWS were to comply with the MCL, their
treatment costs would bring the annualized cost to
$394.4 million.
  10 Source: table XII-6, in 1997 S.
  11 Source: Table X-7.

-------
38966
Federal Register/Vol.  65,  No. 121/Thursday, June  22,  2000/Proposed Rules
  In addition to quantifiable benefits,
EPA has identified several potential
non-quantifiable benefits associated
with reducing arsenic exposures in
drinking water. These potential benefits
are difficult to quantify because of the
uncertainty surrounding their
estimation. Non-quantifiable benefits
may include any peace-of-mind benefits
specific to reduction of arsenic risks  that
may not be adequately captured in the
Value of Statistical Life (VSL) estimate.
  State, local and Tribal governments
will incur a range of administrative
costs with the MCL options in
complying with the  arsenic rule.
Administrative costs associated with
water mitigation can include costs
associated with program management,
inspections, and enforcement activities.
EPA estimates the total annual costs  of
administrative activities for compliance
with the MCL to be approximately $2.8
million,
  c. Financial Assistance. Various
Federal programs exist to provide
financial assistance  to State, local, and
tribal governments to administer and
comply with this and other drinking
water rules. The Federal government
provides funding to  States that have  a
primary enforcement responsibility for
their drinking water programs through
the Public Water Systems Supervision
(PWSS) Grants program. Additional
funding is available  from other
programs administered either by EPA or
other Federal agencies. These include
the Drinking Water State Revolving
Fund (DWSRF) and  Housing and Urban
Development's Community
Development Block  Grant Program. For
example, the SDWA authorizes the
Administrator of the EPA to award
capitalization grants to States, which in
turn can provide low cost loans and
other types of assistance to eligible
public water systems. The DWSRF
                      assists public water systems with
                      financing the costs of infrastructure
                      needed to achieve or maintain
                      compliance with SDWA requirements.
                      Each State will have considerable
                      flexibility to determine the design of its
                      program and to direct funding toward
                      its most pressing compliance and public
                      health protection needs. States may
                      also, on a matching basis, use' up to ten
                      percent of their DWSRF allotments for
                      each fiscal year to assist in running the
                      State drinking water program.
                        Under PWSS Program Assistance
                      Grants, the Administrator may make
                      grants to States to carry out public water
                      system supervision programs. States
                      may use these funds to develop primacy
                      programs. States may "contract" with
                      other State agencies to assist in the
                      development or implementation of their
                      primacy program. However, States may
                      not use program assistance grant funds
                      to contract with regulated  entities (i.e.,
                      water systems). PWSS Grants  may be
                      used by States to set-up and administer
                      a State program which includes such
                      activities as: public education, testing,
                      training, technical assistance,
                      developing and administering a
                      remediation grant and loan or incentive
                      program (excludes the actual grant or
                      loan funds), or other regulatory or non-
                      regulatory measures.
                        d. Estimates of future compliance
                      costs and disproportionate budgetary
                      effects. To meet the requirement in
                      section 202 of the UMRA,  EPA analyzed
                      future compliance costs and possible
                      disproportionate budgetary effects of the
                      MCL options. The Agency believes that
                      the cost estimates, indicated previously
                      and discussed in more detail in Section
                      XIII.B of today's Preamble accurately
                      characterize future compliance costs of
                      the proposed rule.
                        With regard to the disproportionate
                      impacts, EPA considered available data
                      sources in analyzing the
disproportionate impacts upon
geographic or social segments of the
nation or industry. No rationale for
disproportionate impacts by geography
were identified. EPA will prepare a  ;
small entity compliance guide, a
monitoring/analytical manual, and a
small systems technology manual that
will assist the public and private sector.
To fully consider the potential
disproportionate impacts of this
proposed rule, this analysis also
developed three other measures:
  (1) Reviewing the impacts on small
versus large systems;
  (2) reviewing the costs to public
versus private water systems; and
  (3) reviewing the household costs for
the proposed rule.
  The first measure, the national
impacts on small versus large systems,
is shown in Section IX,  Table IX-12,
Total Annual Costs per  Household.
Small systems are defined as those
systems serving 10,000  people or less
and large systems are those systems  that
serve more than 10,000  people. The
higher compliance costs to small
systems is primarily due to the greater
number of small systems as opposed to
large systems (i.e., there are 39,420  ;
small systems versus 1,443 large
systems).
  The second measure of
disproportionate impacts evaluated  is
the relative total costs to public versus
private water systems, by size. Table
XIV—2 presents the total annualized
costs for public and private systems by
system size category for the 3 jig/L, 5 (J.g/
L, 10 (ig/L, and 20 Ug/L  arsenic levels.
The costs are comparable for public  and
private systems across system sizes for
all options. This pattern may be due in
large part to the limited number of
treatment options assumed to be
available to either public or private
systems to remove arsenic.
                          TABLE XIV-2.—AVERAGE ANNUAL COST PER CWS BY OWNERSHIP
System size
Treatment and monitoring
costs
Public
Private
Total cost
All systems
                                                    MCL = 3 |ig/L
<100 	 	 	 	
101-500 	 	
501-1,000.... 	 	 	
1,001-3,300 	 	
3,301-10,000 	
10,001-1,000,000 	
>1, 000,000 	

$9,475
25,228
34,688
60,929
135,573
578,591
3,885,713

$7,354
18,570
31 ,645
51,097
111,396
547,969


$7,559
26,588
33,474
58,189
131,197
573,423
3,885,713

                                                    MCL = 5 |j.g/L
<100 	
101-500 	
9,720
24,560
7,212
18,223
7,450
26,198

-------
                 Federal Register/Vol.  65, No. 121/Thursday, June 22, 2000/Proposed  Rules
                                                                   38967
                   TABLE XIV-2.—AVERAGE ANNUAL COST PER CWS BY OWNERSHIP—Continued

System size

501-1,000 	 	 	
1,001-3,300 	 	 	
3,301-10,000 	
10,001-1,000,000 	
>1, 000,000 	 : 	

Treatment ar

Public
34 124
57277
124,552
518647
2,669,474

d monitoring

Private
30697
48 198
102005
459 930


Total cost

All systems
32 778
54 666
120399
508 640
2 669 474

                                                  MCL = 10 [j.g/L
<100 	 '. 	
101-500 	
501-1,000 	
1,001-3,300 	
3,301-10,000 	 	 	
10,001-1,000,000 	 	 	
>1,000,000 	 .

9453
23584
32,271
53357
113338
458 340
1 395 498

7 135
17675
29 160
44785
91 244
41 5 520


7 350
19 551
31 048
50921
109278
450 835
1 395 498

                                                  MCL = 20 \ig/L
<100 	
101-500 	 	
501-1,000 	 	
1,001-3,300 	
3,301-10,000 	
10,001-1,000,000 	
>1, 000,000 	 :

9 121
22778
30493
48399
99 872
394 742
921 121

6 950
16954
27 668
41 625
79 128
334 737


7 157
18 738
29 376
46 501
95 983
384 868
921 121

  * Costs were calculated at a commercial interest rate and include system treatment, monitoring, and administrative costs; note that systems
serving over 1 million people are public surface water systems.                     ;
  The third measure, household costs,
can also be used to gauge the impact of
a regulation and to determine whether
there are disproportionately high
impacts in particular segments of the
population. A detailed analysis of
household cost impacts by system size
is presented in the RIA (US EPA 2000e).
The costs for households served by
public and private water systems are
presented in Table XIV—3. As expected,
cost per household increases as system
size decreases. Cost per household is
higher for households served by smaller
systems than larger systems for two
reasons. First, smaller systems serve far
fewer households than larger systems
and, consequently, each household
must bear a greater percentage share of
the system's costs. Second, smaller,
systems tend to have higher influent
arsenic concentrations that, on a per-
capita or per-household basis, require
more expensive treatment methods to
achieve the target arsenic level.
  There is a moderate difference in
annual cost per household for the 3 jig/
L, 5 ug/L, 10 ng/L, and 20 u.g/L levels
for each size category. However, the
costs per household are higher for
private systems than for public systems.
For public systems, the cost per
household ranges from $24.73 to
$341.78 per year at 5 u,g/L and from
$22.03 to $329.17 per year at 10 Ug/L.
For private systems, the ranges are
$21.91 to $369.21 per year, and $19.06
to $363.08 per year, respectively.
            TABLE XIV-3.—AVERAGE COMPLIANCE COSTS PER HOUSEHOLD FOR CWSs EXCEEDING MCLs
System size
Groundwater
! Public
Private
Surface water
Public
Private
                                                   MCL = 3
<100 	
101-500 	
501-1,000 	
1,001-3,300 	
3,301-10,000 	
10,001-1,000,000 	
>1, 000,000 	

$338 44
21859
10863
62.17
4467
31 29


$374 86
285 61
11260
8324
6296
31 29


32894
135 98
4544
21 13
18 34
2649
270

$385 61
18396
46 72
2791
22 94
2281


                                                   MCL = 5
<100 	
101-500 	
501-1,000 	
1,001-3,300 	
3,301-10,000 	
10,001-1,000,000 	
341 78
21311
106.00
58.31
40 60
28.12
36921
280 76
10840
7754
57 25
28.63
32348
135 22
44 86
20 07
16 89
24.73
33005
182 65
4635
2657
21 54
21.91

-------
38968
Federal Register/Vol. 65, No. 121/Thursday, June 22,  2000/Proposed Rules
      TABLE XIV-3.—AVERAGE COMPLIANCE COSTS PER HOUSEHOLD FOR CWSs EXCEEDING MCLs—Continued
System size
>1,000,000 	

Groundwater
Public


Private


Surface water
Public
1.73
Private


                                               MCL = 10}ig/L
<100 	 	 , 	
101-500 	
501-1,000 	
1,001-3,300 	
3,301-10,000 	
10,001-1,000,000 	
>1,000,000 	

329 17
203 40
, 99 45
53 70
36 30
2409


363 09
273 04
102 19
71 97
50 41
24 47


317 80
132 74
42 98
18 62
14 fia
22 03
0 89

qoic C.A
1 on op
4448
P^ 4Q
-to cc
19 06


                                               MCL = 20 p.g/L
<100 	
101-500 	
501-1.000 	 	
1,001-3,300 	
3,301-10,000 	
10,001-1,000,000 	
>1,000,000 	 	
320 13
195 99
93 27
48 03
31 38
2027

352 42
262 01
96 63
66 12
44 14
2039

310 11
132 68
42 26
18 20
1*3 "W
19 96
0.55
OO/1 Q/l
179 93
44 04
9A 87
1"7 ^


  'Costs to households were calculated at a commercial interest rate and include system treatment, monitoring, and administrative costs; note
that systems serving over 1 million people are public surface water systems.


   TABLE XIV-4.—AVERAGE COMPLIANCE COSTS PER HOUSEHOLD FOR CWSs EXCEEDING MCLs AS A PERCENT OF

                                        MEDIAN HOUSEHOLD INCOME
System size
Groundwater
Public
Private
Surface water
Public
Private
                                               MCL = 3 ng/L
<100 	
101-500 	 	 	
501-1,000 	
1,001-3,300 	
3,301-10,000 	 „
10,001-1,000,000 	
>1,000.0000 	

085
0 55
027
0 16
011
008


0 95
0 72
0 28
0 21
0 16
0 08


n FM
0 34
011
n 0=.
0 05
n n?
0 01

A OC
n AG
019
On-7
h nfi
h nfi


                                               MCL = 5
<100 	
101-500 	 	 	
501-1 ,000 	
1,001-3,300 	
3,301-10,000 	
10,001-1,000,000 	
>1,000,0000 	
0 86
0 54
027
0 15
0 10
007

0 93
071
0 27
0 20
0 14
007

0 82
0 34
n 11
n cm
n nd
0 06
0.00
n R*3
n 1 ,000,0000 	
0 83
0 51
025
0 14
0 09
0 06

0 92
0 69
0 26
0 18
0 13
0 06

0 80
n T?
n 11
0 05
n n4
n nfi
0.00
I
ri P.9
n AR
01 1
n nfi
n nc
A nc

                                              MCL = 20
<100 	
101-500 	
501-1,000 	
1,001-3,300 	
3,301-10,000 	
10.001-1,000,000 	
0.81
0 49
024
0 12
0 08
0.05
0.89
0 66
0 24
0 17
011
0.05
0.78
n TI
n 11
n n^
A AO
0.05
0.82
n 4s
01 1
n HR
n nd.
o.no

-------
                 Federal Register/Vol. 65, No. 121/Thursday, June  22,  2000/Proposed Rules
                                                                    38969
   TABLE XIV-4.—AVERAGE COMPLIANCE COSTS PER HOUSEHOLD FOR CWSs EXCEEDING MCLs AS A PERCENT OF
                                     MEDIAN HOUSEHOLD INCOME—Continued
System size
>1, 000,0000 	
Groundwater
Public

Private

Surface water
Public
0.00
Private

  * Costs to household were calculated at a commercial interest rate and include system treatment, monitoring, and administrative costs; median
household income in May 1999 was $39,648 from the 1998 annual median household income from the Census.
  To further evaluate the impacts of
these household costs, the costs per
household were compared to median
household income data for each system-
size category. The result of this
calculation, presented in Table XIV—4
for public and private systems, indicate
a household's likely share of
incremental costs in terms of its
household income. For all system sizes,
household costs as a percentage of
median household income are less than
one percent for households served by
either public or private systems. Similar
to the cost per household results on
which they are based, household
impacts exhibit little variability across
arsenic levels.
   e. Macro/economic effects. As required
under UMRA § 202, EPA is required to
estimate the potential macro-economic
effects of the regulation. These types of
effects include those on productivity,
economic growth, full employment,
creation of productive jobs, and
international competitiveness.  Macro-
economic effects tend to be measurable
in nationwide econometric models only
if the economic impact of the regulation
reaches 0.25 percent to 0.5 percent of
Gross Domestic Product (GDP). In 1998,
real GDP was $7,552 billion so a rule
would have to cost at least $18 billion
annually to have a measurable effect. A
regulation with a smaller aggregate
effect is unlikely to have any
measurable impact unless it is  highly
focused on a particular geographic
region or economic sector. The macro-
economic effects on the national
economy from the arsenic rule should
be negligible based on the fact  that,
assuming 100 percent compliance with
an MCL, the total annual costs are
approximately $756 million at the 3 ug/
L level, $445 million at the 5 Ug/L level,
about $195 million at the 10 u,g/L level,
and at the 20 |J.g/L level, about $77
million (at a 7 percent discount rate),
and the costs are not expected to be
highly focused on a particular
geographic region or industry sector.
   /. Summary of EPA's consultation
with State, local,  and tribal governments
and their concerns. Under UMRA
section 204, EPA is to provide a
summary of its consultation with
elected representatives (or their
designated authorized employees) of
affected State, local, and Tribal
governments in this rulemaking. EPA
initiated consultations with
governmental entities and the private
sector affected by this rulemaking;
through various means. This included
five stakeholder meetings announced in
the Federal Register and open to any
one interested in attending in person or
by phone, and presentations at meetings
of the American Water Works
Association (AWWA), the Association
of State Drinking Water Administrators
(ASDWA), the Association of California
Water Agencies (ACWA), and the;
Association of Metropolitan Water
Agencies (AMWA). Participants in
EPA's stakeholder meetings also
included representatives from the'
National Rural Water Association,
AMWA, ASDWA, AWWA, ACWA,
Rural Community Assistance Program,
State departments of environmental
protection, State health departments,
State drinking water programs, and a
Tribe. EPA also made presentations at
Tribal meetings in Nevada, Alaska, and
California. To address the proposed
rule's impact on small entities, the
Agency consulted with representatives
of small water systems and convened a
Small Business Advocacy Review, Panel
in accordance with the Regulatory
Flexibility Act (RFA) as amended >by the
Small Business Regulatory Enforc.ement
Fairness Act (SBREFA). Two of the
small entity representatives were '
elected officials from local governments.
EPA also invited State drinking water
program representatives to participate in
a number of workgroup meetings. In
addition to these consultations, EPA
participated in and gave presentations at
AWWA's Technical Workgroup for
Arsenic. State public health department
and drinking water program     '
representatives, drinking water districts,
and ASDWA participated in the
Technical Workgroup meetings. Finally
EPA presented the benefits analysis to
State and Tribal health and
environmental agencies.
  The public docket for this proposed
rUlemaking contains meeting summaries
for EPA's five stakeholder meetings on
arsenic in drinking water, written
comments received by the Agency, and
provides details about the nature of
State, local, and Tribal government's
concerns. A summary of State, local,
and Tribal government concerns on this
proposed rulemaking is in the next
section.
  In order to inform and involve Tribal
governments in the rulemaking process,
EPA staff attended the 16th Annual
Consumer Conference of the National
Indian Health Board on October 6-8,
1998 in Anchorage, Alaska. Over nine
hundred attendees representing Tribes
from across the country were in
attendance. During the conference, EPA
conducted two workshops for meeting
participants. The objectives of the
workshops were to present an overview
of EPA's drinking water program, solicit
comments on key issues of potential
interest in upcoming drinking water
regulations, and to solicit advice in
identifying an effective consultative
process with Tribes for the future.
  EPA, in conjunction with the Inter
Tribal Council of Arizona (ITCA), also
convened a Tribal consultation meeting
on February 24-25,1999, in Las Vegas,
Nevada to discuss ways to involve
Tribal representatives, both Tribal
council members and tribal water utility
operators, in the stakeholder process.
Approximately twenty-five
representatives from a diverse group of
Tribes attended the two-day meeting.
Meeting participants included
representatives from the following
Tribes: Cherokee Nation, Nezperce
Tribe, Jicarilla Apache Tribe, Blackfeet
Tribe, Seminole Tribe of Florida, Hopi
Tribe, Cheyenne River Sioux Tribe,
Menominee Indian Tribe, Tulalip
Tribes, Mississippi Band of Choctaw
Indians, Narragansett Indian Tribe, and
Yakama Nation.
  The major meeting objectives were to:
  (1)  identify key issues of concern to
Tribal representatives;
  (2)  solicit input on issues concerning
current OGWDW regulatory efforts;
  (3)  solicit input and information that
should be included in support of future
drinking water regulations; and

-------
  38970
Federal  Register/Vol. 65, No. 121/Thursday, June  22,  2000/Proposed Rules
   (4) provide an effective format for
 Tribal involvement in EPA's regulatory
 development process.
   EPA staff also provided an overview
 on the forthcoming arsenic rule at the
 meeting. The presentation included the
 health concerns associated with arsenic,
 EPA's current position on arsenic in
 drinking water, the definition of an
 MCL, an explanation of the difference
 between point-of-use and point-of-entry
 treatment devices, and specific issues
 for Tribes. The following questions \vere
 posed to the Tribal representatives to
 begin discussion on arsenic in drinking
 water:
   (1) What are the current arsenic levels
 in your water systems?
   (2) What are Tribal water systems
 affordability issues in regard  to arsenic?
   (3) Does your Tribe use well water,
 river water or lake water?
   (4) Purchase water from another
 drinking water utility?
   The summary for me February 24-25,
 1999 meeting was sent to all 565
 Federally recognized Tribes in the
 United States.
   EPA also conducted a series of
 workshops at the Annual Conference of
 the National Tribal Environmental
 Council which was held on May 18-20,
 1999 in Eureka,  California.
 Representatives from over 50 Tribes
 attended all, or part, of these sessions.
 The objectives of the workshops were to
 provide an overview of forthcoming
 EPA regulations affecting water systems;
 discuss changes to operator certification
 requirements; discuss funding for Tribal
 water systems; and to discuss
 innovative approaches to regulatory cost
 reduction. Meeting summaries for EPA's
 Tribal consultations are available in the
 public docket for this proposed
 rulemaking.
  g. Nature of State, local, and Tribal
 government concerns and how EPA
 addressed these concerns. State and
 local governments raised several
 concerns, including the high costs of the
 rule to small systems;  the burden of
 revising the State primacy program;  the
 high degree of uncertainty associated
 with  the benefits; the high costs of
 including Non-Transient Non-
 Community Water Systems (NTNCWSs).
 EPA modified regulations governing the
 revision of State primacy in order to
 decrease the burden of the new arsenic
 regulation in response to State concerns
 that EPA minimize paperwork and
 documentation of existing programs that
 would manage the arsenic regulation.
 Section XL asks for comment on
 alternate MCL options, based partly on
 the high costs of the rule for small
systems and uncertainty associated with
the risks.
                       Tribal representatives were generally
                     supportive of regulations which would
                     ensure a high level of water quality, but
                     raised concerns over funding for
                     regulations. With regard to the
                     forthcoming proposed arsenic rule,
                     many Tribal representatives saw the
                     health benefits as highly desirable, but
                     felt that unless additional funds were
                     made available, implementing the
                     regulation would be difficult for many
                     Tribes.
                       EPA understands the State, local, and
                     tribal government concerns with the
                     above issues. The Agency believes the
                     options  for small systems, proposed for
                     public comment in this rulemaking, will
                     address stakeholder concerns pertaining
                     to small systems and  will help to reduce
                     the financial burden to these systems.
                     Small systems compliance technologies
                     and associated costs were listed in
                     section VIII.E. Regionalization, the
                     process by which a small system can
                     connect with another system and
                     purchase water, is a non-treatment
                     option that could be considered for
                     small systems. The costs for
                     regionalization by system size are
                     presented as Treatment Train #1 in
                     Table VIII-3 of section VIII.B. Sections
                     XII.C address tribal SRF and grant
                     funding.
                       Non-Transient Non-Community Water
                     Systems (NTNCWSs)  are only required
                     to monitor and report exceedances of
                     the MCL. A detailed discussion of the
                     exposure to arsenic in NTNCWSs is
                     shown in section V.F. of this Preamble.
                     EPA has conducted a  preliminary
                     analysis on exposure and risks to
                     NTNCWSs and is soliciting public
                     comment on this preliminary analysis.
                     An analysis of the potential benefits and
                     costs of arsenic in drinking water for
                     NTNCWSs is summarized in the
                     preamble and included in the docket for
                     this proposed rulemaking (US EPA
                     2000e).
                       The Agency is basing this regulation
                     on the risks to the general population
                     and is not excluding any particular
                     segments of the population. For a more
                     complete discussion on the risks of
                     arsenic in drinking water and air, see
                     section II.C. of this Preamble.
                       h. Regulatory Alternatives
                     Considered. As required under section
                     205 of the UMRA, EPA considered
                     several regulatory alternatives in
                     developing an MCL for arsenic in
                     drinking water. In preparation for this
                     consideration, the Regulatory Impact
                     Analysis and Health Risk Reduction and
                     Cost Analysis (HRRCA) for Arsenic
                     evaluated arsenic levels of 3 ug/L, 5 jig/
                     L, 10 ug/L, and 20 ug/L.
                      The Regulatory Impact Analysis and
                    HRRCA also evaluated national costs
 and benefits of States choosing to
 reduce arsenic exposure in drinking
 water. For further discussion on the
 regulatory alternatives considered in
 this proposed rulemaking, see section
 XIII. of this Preamble. EPA examined a
 range of regulatory alternatives that
 could be employed to achieve the
 objectives of this rule and chose what it
 believes is the least burdensome such
 alternative. The regulatory approach
 embodied in this rule includes a
 proposed MCL that relies on the use of
 the Administrator's discretionary
 authority under section 1412(b)(6) of the
 SDWA to set a less stringent level than
 the feasible level. The exercise of these,
 authorities in this manner is expected to
 reduce overall burden on regulated
 entitities (as compared to the burden of
 a more stringent level) but still
 maximize health risk reduction. (See
 section XI.A for a more complete
 discussion of the rationale for the     :
 exercise of these authorities.) In terms of
 coverage of the rule, we are proposing
 that only CWSs be fully covered by the
 rule, driven, in part by consideration of
 the burden associated with not covering
 NTNCWSs in view of the minimal
 health risk reduction that would be
 achieved. The proposed approach is
 also based upon an analysis  and listing
 of least cost treatment alternatives
 (including use of point of use treatment
 devices) that are collectively expected to
 reduce regulatory burden. Finally,
 today's proposal includes an approach
 to monitoring and reporting  that
 involves a framework that provides for
 reduced regulatory burden where
 arsenic levels  are low. Also,  see EPA's
 Regulatory Impact Analysis for Arsenic
 (US EPA 2000e).

 2. Impacts on Small Governments
   In developing this rule, EPA
 consulted with small governments
 pursuant to section 203 of the UMRA to;
 address impacts of regulatory
 requirements in the rule that might
 significantly or uniquely affect small
 governments. In preparation  for the
 proposed arsenic rule, EPA conducted
 analysis on small government impacts
 and included small government officials
 or their designated representatives in
 the rule making process. EPA conducted
 stakeholder meetings on the
 development of the arsenic rule which
 gave a variety of stakeholders, including
 small governments, the opportunity for
 timely and meaningful participation  in
 the regulatory development process.
 Groups such as the National Association
 of Towns and Townships, the National
 League of Cities, and the National
Association of Counties participated  in
the proposed rulemaking process.

-------
                  Federal Register/Vol. 65, No.  121 /Thursday, June 22, 2000/Proposed Rules
                                                                     38971
 Through such participation and
 exchange, EPA notified potentially
 affected small governments of
 requirements under consideration and
 provided officials of affected small
 governments with an opportunity to
 have meaningful and timely input into
 the development of the regulatory
 proposal. See section XIV.B.B.a. for a
 summary of the Small Business Review
 Panel consultations.
   In addition, EPA will educate, inform,
 and advise small systems, including
 those run by small governments, about
 the arsenic rule requirements. One of
 the most important components of this
 process is the  Small Entity Compliance
 Guide,  required by the Small  Business
 Regulatory Enforcement Fairness Act of
 1996 shortly after the rule is
 promulgated. This plain-English guide
 will explain what actions a small entity
 must take to comply with the rule. Also,
 the Agency is  developing fact sheets
 that concisely describe various aspects
 and requirements of the arsenic rule.

 D. Paperwork Reduction Act (PRA)
   The information collection
 requirements in this proposed rule have
 been submitted for approval to the
 Office of Management and Budget
 (OMB) under the Paperwork Reduction
 Act, 44 U.S.C. 3501 et seq. An
 Information Collection Request (ICR)
 document has been prepared by EPA
 (ICR, No. 1948.01) and a copy may be
 obtained from Sandy Farmer  by mail at
 Collection Strategies Division; U.S.
 Environmental Protection Agency
 (2822); 1200 Pennsylvania Ave., NW,
 Washington, DC 20460, by email at
• farmer.sandy@epamail.epa.gov, or by
 calling (202) 260-2740. A copy may also
 be downloaded off the Internet at http:/
 /www. epa .gov/icr.
   Two types of information will be
 collected under the proposed arsenic
 rule. First, information on CWSs and
 NTNCWSs and their arsenic levels
 reported under 50 ug/L will enable the
 States and EPA to evaluate compliance
 with the lower MCL. This information,
 most of which consists of monitoring
 results, corresponds to arsenic
 information already collected from
 water systems. Arsenic monitoring and
 reporting will continue annually for
 surface water systems or once every
 three years for ground water systems,
 unless the MCL is exceeded or a State
 grants  a waiver (see section VII). Other
 existing information and reporting
 requirements, such as Consumer
 Confidence Reports (US EPA, 1998J) and
 the public notification requirements (US
 EPA, 2000c), will be amended to reflect
 the lower MCL for arsenic. As proposed,
 NTNCWSs will not be required to
comply with the MCL becau'se of the
low exposure levels as expla ined in
section XI.C. However, EPA i's requiring
NTNCWSs to report to the State and
public when it exceeds the M CL .
through public notification
requirements. As is the case for other
contaminants, required information on
system arsenic levels must be provided
by affected systems and is not
considered to be confidential. EiPA
believes the information needs
discussed previously, on compli'ance
with the MCL programs, are esse ntial to
achieving the arsenic-related hea 1th risk
reductions anticipated by EPA m ider
the proposed rule.              ;
  EPA has estimated the burden |
associated with the specific record-
keeping and reporting requirement s of
the proposed rule in an accompanying
Information Collection Request (ICR),
which is available in the public docket
for this proposed rulemaking. Burde.n
means the total time, effort, or financ ial
resources  expended by persons to:
generate, maintain, retain, or disclose or
provide information to or for a Federal'
agency. This includes the time needed
to review  instructions; develop, acquire1,
install, and utilize technology and      \
systems for the purposes of collecting,
validating, and verifying information,
processing and maintaining
information, and disclosing and
providing information; adjust the
existing procedures to comply with any
previously applicable instructions  and
requirements; train personnel to b,e able
to respond to a collection of
information; search data sources;
complete  and review the collection of
information; and transmit or otherwise
disclose the information.
  The ICR for the proposed rule covers
the information collection, reporting
and record-keeping requirements for the
three-year period following
promulgation of the Arsenic Rule.  There
are several activities that PWSs must
perform in preparation for compliance
with the revised Arsenic Rule in the
first three years. Start-up activities
include reading the final rule to become
familiar with the requirements and
training staff to perform the required
activities. The number of hours required
to perform each activity varies by
system size. The total start-up burden
per system for systems serving less than
10,000 people is estimated to be 24
hours; the total start-up burden per
system for systems serving more than
10,000 people is estimated to be 40
hours. The total hour burden for the
74,607 PWSs (including NTNCWS)
covered by this rule is estimated to be
1,847,784 hours, or an annual average of
615,928 hours. There are no monitoring,
record-keeping, reporting or equipment
costs for PWSs during the first three-
year period. EPA expects States to incur
only nominal information collection,
reporting or record-keeping costs during
the first three years. (For estimates of
the cost of information collection,
reporting and record-keeping over a 20-
year period, see ICR No. 1948.01)
  An agency may not conduct or
sponsor, and a person is not required to
respond to, a collection of information
unless it displays a currently valid OMB
control number. The OMB control
numbers for EPA's regulations are listed
in 40 CFR part 9 and 48 CFR Chapter
15.
  Comments are requested on the
Agency's need for this information, the
accuracy of the provided burden
estimates, and any suggested methods
for minimizing respondent burden,
including through the use of automated
collection techniques. Send comments
on the ICR to the Director, Collection
Strategies Division; U.S. Environmental
Protection Agency (2822); 1200
Pennsylvania Ave., NW, Washington,
DC 20460; and to the Office of
Information and Regulatory Affairs,
Office of Management and Budget, 725
17th St., NW, Washington, DC 20503,
marked "Attention: Desk Officer for
EPA." Include the ICR number in any
correspondence. Since OMB is required
to make a decision concerning the ICR
Ibetween 30 and 60 days after June 22,
2'000, a comment to OMB is best assured
o. f having its full effect if OMB receives
it by July 24, 2000. The final rule will
ret '.pond to any OMB or public
comments on the information collection
requirements contained in this proposal.

E. A Jational Technology Transfer and
Adv -ancement Act (NTTAA)
  Sesction 12(d) of the National
Technology Transfer and Advancement
Act of 1995 (NTTAA), (Public Law 104-
113, section 12(d), 15 U.S.C. 272 note),
directs EPA to use voluntary consensus
stand; irds in its regulatory activities
unless',to do so would be inconsistent
with a pplicable law or otherwise
imprac 'tical. Voluntary consensus
standai 'ds are technical standards (e.g.,
materia 1 specifications, test methods,
samplin % procedures, business
practices?, etc.) that are developed or
adopted by voluntary consensus
standard bodies. The NTTAA directs
EPA to provide to Congress, through
OMB, explanations when the Agency
decide s not to use available and
applicable voluntary consensus
standards.
  EPAMs process for selecting analytical
methods is consistent with section 12(d)
of the ] NTT A A. EPA performed a

-------
 38972
Federal Register/Vol. 65, No.  ?i21/Thursday, June  22,  2000/Proposed Rules
 literature search to identify analytical
 methods from industry, academia,
 voluntary consensus standard bodies
 and other parties that could be used to
 reliably measure total arsenic in
 drinking water at the proposed MCL of
 0.005 mg/L. Today's proposed
 rulomaking allows the use of analytical
 methods which are described in the
 "Annual Book of ASTM Standards"
 (American Society for Testing and
 Materials, 1994  and 1996) and in
 "Standards for the Examination of
 Water and Wastewater" (APHA, 1992
 and 1995). The four methods published
 by these consensus organizations
 include SM 3113B, SM 3114B, ASTM
 2972-93B and ASTM 2972-93C. These
 methods were all approved for arsenic
 analysis in previous methods-related
 rulemakings for the MCL of 0.050
 mg/L. Along with the review of other
 analytical methods, EPA also re-
 evaluated these  consensus methods for
 the new arsenic standard. The Agency
 believes these methods will still be
 reliable for compliance monitoring at
 ths proposed MCL of 0.005 mg/L.
 Additional information on these
 methods are shown in section VI. C. and
 F. of today's preamble. One consensus
 method, SM 3120B, will be withdrawn
 in today's rulemaking. As discussed in
 section VI.D.,  SM 3120B will be
 withdrawn because the detection limit
 for this method is inadequate to reliably
 determine the presence of arsenic at tho
 proposed MCL of 0.005 mg/L.
  Although no other methods were
 identified from the literature search,
 EPA welcomes comments on this asp ect
 of today's proposed rulemaking and
 specifically invites the public to ide;.itify
 potentially-applicable voluntary
 consensus standards, explain why s ucli
 standard should be considered for
 inclusion with this regulation, and ';o
 provide the  necessary information /.xom
 Inter-laboratory studies on detectio n
 limits, accuracy, recovery and precision.
 F. Executive Order 12898:
 Environmental Justice
  Executive Order 12898 "FederaJ
 Actions To Address Environment;al
 Justice in Minority Populations said
 Low-Income Populations," (59 F R 7629,
 February 16,1994) establishes a Federal
 policy for incorporating environmental
 justice into Federal agency mis sions by
 directing agencies to identify a nd
 address disproportionately hig h and
 adverse human health or environmental
 effects of its programs, policies, and
 activities on minority and low-income
populations. The Agency has
considered environmental ju;jtice-
related issues concerning thei pote ntial
impacts of this action and has consulted
                     with minority and low-income
                     stakeholders by convening a stakeholder
                     meeting via video conference
                     specificall y to address environmental
                     justice iss'aes.
                       As part of EPA's responsibilities to
                     comply with Executive Order 12898, the
                     Agency h eld a stakeholder meeting via
                     video conference on March 12, 1998, to
                     highlight components of pending
                     drinking,' water regulations and how
                     they ma y impact sensitive sub-
                     populafjons, minority populations, and
                     low-income populations. Topics
                     discussed included treatment
                     techniques, costs and benefits, data
                     quality, health effects, and the
                     regultitory process. Participants
                     inclu ded national, State, tribal,
                     municipal, and individual stakeholders.
                     EPA conducted the meeting by video
                     conference call between eleven cities.
                     This meeting was a continuation of
                     stakeholder meetings that started in
                     19'95 to obtain input on the Agency's
                     Dr inking Water programs. The major
                     objectives for the 1998 meeting were:
                       (1) Solicit ideas from Environmental
                     J ustice (EJ) stakeholders  on known
                     •'issues concerning current drinking
                     water regulatory efforts;
                       (2) Identify key issues  of concern to EJ
                     stakeholders; and
                       (3) Receive suggestions from EJ
                     stakeholders concerning ways to
                     increase representation of EJ
                     communities in OGWDW regulatory
                     efforts.
                       In addition, EPA developed a plain-
                     English guide specifically for this
                     meeting to assist stakeholders in
                     understanding the multiple and
                     sometimes complex issues surrounding
                     drinking water regulations. A meeting
                     summary for the March 12, 1998
                     Environmental Justice stakeholders
                     meeting (US EPA, 1998b) is available in
                     the public docket for this proposed
                     rulemaking.
                       During the presentation of separate
                     cities' discussions, several arsenic
                     issues came up. In Region 6 one
                     stakeholder thought that  test results for
                     arsenic (discussed in ppb and ug/L)
                     were hard to understand, and the health
                     effects appear to be complicated. Region
                     6 participants had concerns about the
                     toxic effects on mothers,  individuals
                     with different metabolisms, and
                     individuals with poor nutrition. One  of
                     the stakeholders expressed a concern
                     that the government was  not protecting
                     poorer communities against pollution.
                     In Region 7, one stakeholder lives in an
                     area that purchases water which has to
                     be monitored. The area has a shrinking
                     population that is increasing in age and
                     immune conditions. Although there are
                     pesticides in the water and air, it would
 not be economically practical to
 consolidate to a regional drinking water
 system. One member of an Indian tribe
 said Tribes tend to have more diabetes
 than the rest of the country, and
 diabetes seemed to be linked to arsenic
 exposure. In Region 8 a stakeholder
 wanted affordable or equally protective
 treatment options. A Region 8
 participant asked for disclosure of
 environmental contamination. Region 9
 reported some individual monitoring
 difficulties. Stakeholders wanted better
 access to funding sources. Stakeholders
 in Region 9 had concerns about the
 immuno-compromlsed, young children,
 and pregnant women. Some
 stakeholders wanted standard setting to
 address regional needs, include local
 governments in the standard setting,
 more technical assistance and training,
 and more stakeholder involvement.
 Tribes and large cities with low income
 families may be burdened with more of
 the risk.
   The Agency considered equity-related
 issues concerning the potential impacts
 of this action. There is no factual basis
 to indicate that minority and low
 income communities are more  (or less)
 exposed to arsenic in drinking water.
 The occurrence information suggests
 there is no difference between the
 percent of systems likely to be impacted
 in small communities versus larger
 ones. Further, arsenic in drinking water
 is primarily natural in origin (rather
 than related to contamination events)
 and a systematic bias based on
 socioeconomic factors would not be
 expected to occur. A key issue of
 concern is the potential for an uneven
 distribution of risk reduction benefits
 across water systems and society.
   The public is invited to comment on
 EPA's analysis of environmental justice
 and, specifically, to recommend
 additional methods to address
 environmental justice concerns with the
 approach for treating arsenic in drinking
 water.

 G. Executive Order 13045.-Protection of
 Children from Environmental Health
 Risks and Safety Risks
  Executive Order 13045, "Protection of
 Children from Environmental Health
 Risks and Safety Risks," (62 FR 19885
 April 23, 1997) applies to any rule that:
 (1) is determined to be "economically
 significant" as defined under Executive
 Order 12866, and (2) concerns an
 environmental health or safety risk that
 EPA has reason to believe may have a
 disproportionate effect on children. If
the regulatory action meets both criteria,
the Agency must evaluate the
 environmental health or safety effects of
the planned rule  on children, and

-------
                 Federal Register/Vol. 65, No.  121/Thursday, June  22,  2000/Proposed Rules
                                                                     38973
explain why the planned regulation is
preferable to other potentially effective
and reasonably feasible alternatives
considered by the Agency.
  This proposed rule is not subject to
the Executive Order because the Agency
does not have reason to believe the
environmental health risks or safety
risks addressed by this action present a
disproportionate risk to children.
Nonetheless, we have evaluated the
environmental health or safety effects of
arsenic in drinking water on children.
The results of this evaluation are
contained in section III.F.5. of this
Preamble. Copies of the documents used
to evaluate the environmental health or
safety effects of arsenic in drinking
water on children have been placed in
the public docket for this proposed
rulemaking.
  The public is invited to submit or
identify peer-reviewed studies and data,
of which EPA may not be aware, that
assessed results of early life exposure to
arsenic via ingestion.
H. Executive Order 13132: Federalism
  Executive Order 13132, entitled
"Federalism" (64 FR 43255, August 10,
1999), requires EPA to develop an
accountable process to ensure
"meaningful and timely input by State
and local officials in the development of
regulatory policies that have federalism
implications." "Policies that have
federalism implications" is defined in
the Executive Order to include
regulations that have "substantial direct
effects on the States, on the relationship
between the national government and
the States, or on the distribution of
power and responsibilities among the
various levels of government."
  Under section 6 of Executive Order
13132, EPA may not issue a regulation
that has federalism implications, that
imposes substantial direct compliance
costs, and that is not required by statute,
unless the Federal government provides
the funds necessary to pay the direct
compliance costs incurred by State and
local governments, or EPA consults with
State and local officials early in the
process of developing the proposed •
regulation. EPA also may not issue  a
regulation that has federalism
implications and that preempts State
law, unless the Agency consults with
State and local officials early in the
process of developing the proposed
regulation.
  If EPA complies by consulting,
Executive Order 13132 requires EPA to
provide to the Office of Management
and Budget (OMB), in a separately
identified section of the preamble to the
rule, a federalism summary impact
statement (FSIS). The FSIS must include
a description of the extent of EPA's
prior consultation with State and.'local
officials, a summary of the nature of
their concerns and the agency's position
supporting the need to issue the
regulation, and a statement of the extent
to which the concerns of State and local
officials have been met. Also, when EPA
transmits a draft final rule with  ;
federalism implications to OMB for
review pursuant to Executive Order
12866, EPA must  include a certification
from the agency's Federalism Official
stating that EPA has met the
requirements of Executive Order 13132
in a meaningful and timely manner.
  EPA has concluded that this proposed
rule  will have federalism implications.
This rule will impose substantial direct
compliance costs  on State and local
governments, and the Federal
government will not provide the funds
necessary to pay those costs.     ;
Accordingly, EPA provides the
following FSIS as required by section
6(b)  of Executive Order 13132.
  EPA consulted with State and local
officials early in the process of
developing the proposed regulation to
permit them to have meaningful and
timely input into its development.
Summaries of the meetings have been
included int public record for this
proposed rulemaking. EPA consulted
extensively with State, local, and tribal
governments. For example, we held four
public stakeholder meetings in   :
Washington, B.C.  (two meetings); San
Antonio, Texas; and Monterey,
California. Invitations to stakeholder
meetings were extended to the National
Association of Counties, The National
Governors' Association, the National
Association of Towns and Townships,
the National League of Cities, and the
National Conference of State Legislators.
In addition, several elected officials
were part of the Small Business  ;
Advocacy Review Panel convened by
EPA (as required by section 609(b) of
the Regulatory Flexibility Act).   '
Consultation has not ended, however,
but will be an on-going transactiohal
process. EPA officials presented a
summary of the rule to the National
Governor's Association in a meeting on
May 24, 2000. In addition, we
scheduled a one-day stakeholders'
meeting for the trade associations that
represent elected officials on May 30,
2000 to discuss and solicit comment on
this and other upcoming contaminant
rules. EPA will continue to seek input
from its State and local government
partners.
  Several key issues were raised by
stakeholders regarding the arsenic rule
provision, many of which were related
to reducing burden and maintaining
flexibility. The Office of Water was able
to reduce burden and increase flexibility
in a number of areas in response to
these comments. More specifically,
elected officials expressed overall
concerns about: (1) Factors considered
in setting of the MCL and (2) the
treatment technologies, their associated
costs and waste disposal costs. Specific
issues regarding the setting of the MCL
included:
  • The treatment costs associated with
a lower drinking water standard;
  • Concerns about affordability for
lower income areas;
  • Asking the Agency to delay setting
a standard below 25 ug/L until the
development of affordable technologies;
and
  • A lack of evidence for health effects
data below 50 |J.g/L.
  Specific concerns regarding the
treatment technologies, their associated
costs and waste disposal costs included:
  • The difficulty of using oxidation/
filtration for arsenic removal when
concentrations are <25 jig/L (even after
the addition of iron salts and pH
adjustment);
  • The waste disposal costs created
from the use of ion exchange;
  • The more intensive need for
operator oversight and the amount of
sludge generated using coagulation
filtration and lime softening at a high
pH;
  • The difficulty in finding and the
expense associated with activated
alumina;
  • The expense associated with
reverse osmosis, nano-filtration and pre-
oxidation.
  The Agency responded to these
concerns in several ways. We are very
sensitive to the potential costs of
treatment for a lower drinking water
standard and have examined an array of
treatment options (especially those that
are most appropriate  for small systems)
in order to identify the least cost,
affordable options that systems may use
to comply with a new standard. We
therefore do not believe that it is
necessary to delay promulgating a rule
with an MCL below 25 |ig/L pending
identification of such technologies, as
one of the comments  suggests. We have
also included higher MCL options than
the proposed MCL in the preamble for
comment, due in large part to concerns
expressed by elected  officials and other
stakeholders about the treatment costs
associated with a low MCL. These
issues are discussed in more detail in
the sections VIII. (treatment) and XI.
(regarding choice of the MCL). We also
share the concerns of elected officials in
connection with the affordability of a
new rule for lower income areas and

-------
 38974
Federal Register/Vol. 65, No. 121/Thursday, June 22, 2000/Proposed Rules
 have identified special programs and
 avenues that may be pursued to provide
 relief for such areas (see section VIII.C.).
   In response to the comment that there
 is a lack of evidence for health effects
 below 50 ug/L, we note that the National
 Academy of Sciences" National
 Research Council has categorically
 determined, based on their review of the
 most recent data and information
 concerning the health effects of arsenic,
 that the current standard of 50 ug/L is
 not protective and should be revised
 downward as soon as possible (NRG,
 1999). This topic is discussed  in more
 detail in section III.
   In response to concerns about specific
 treatment technologies, their associated
 costs and waste disposal costs, EPA
 identifies several treatment technologies
 in section VIII. Section VIII. A.
 identifies the BATs for arsenic removal
 and section VIII.B. identifies
 technologies which are considered
 affordable. The Agency agrees with the
 statement that oxidation/filtration is not
 an appropriate technology to treat
 arsenic to low levels. For this reason, it
 is not considered a BAT. The Agency
 also agrees that wastes are created using
 ion exchange. Section VIII. addresses
 the use of brine recycling in reducing
 wastes and waste disposal costs. In
 addition, regionalization or finding a
 new water source (section VIII.) are
 alternative non-treatment options to
 consider to avoid treatment and the
 costs and disposal issues associated
 with treatment. The Agency agrees with
 the concern that coagulation/filtration is
 more operator intensive but this
 technology and pH modifications are
 only considered if this treatment
 process is already in place. In regards to
 the amount of sludge produced, the
 additional amount of sludge generated
 due to the removal of arsenic is minor.
 The Agency disagrees that activated
 alumina is expensive and, difficult to
 find. As shown in Table VIII-3,
 activated alumina is one of the cheaper
 treatment technologies. The Agency
 agrees that reverse osmosis, nano-
 filtration and the need for pre-oxidation
 are expensive treatment options. In
 these cases, a PWS should consider one
 of the more affordable treatment options
 shown in section  VIII.B.
 /. Executive Order 13084: Consultation
 and Coordination with Indian  Tribal
 Governments
  Under Executive Order 13084,
 "Consultation and Coordination with
 Indian Tribal Governments," 63  FR
 27655 (May 19,1998) EPA may not
 issue a regulation that is not required by
statute, that significantly or uniquely
affects the communities of Indian Tribal
                      governments, and that imposes
                      substantial direct compliance costs on
                      those communities, unless the Federal
                      government provides the funds
                      necessary to pay the direct compliance
                      costs incurred by the Tribal
                      governments, or EPA consults with
                      those governments. If EPA complies by
                      consulting, Executive Order 13084
                      requires EPA to provide the Office of
                      Management and Budget, in a separately
                      identified section of the preamble to the
                      rule, a description  of the extent of EPA's
                      prior consultation with representatives
                      of affected Tribal governments, a
                      summary of the nature of their concerns,
                      and a statement supporting the need to
                      issue the regulation. In addition,
                      Executive Order 13084 requires EPA to
                      develop an effective process permitting
                      elected officials and other
                      representatives of Indian Tribal
                      governments "to provide meaningful
                      and timely input in the development of
                      regulatory policies on matters that
                      significantly or uniquely affect their
                      communities."
                        EPA has concluded that this rule may
                      significantly affect communities of
                      Indian Tribal governments. It will also
                      impose substantial direct compliance
                      costs on such communities, and the
                      Federal government will not provide the
                      funds necessary to  pay the direct costs
                      incurred by the Tribal governments in
                      complying with the rule. In developing
                      this rule, EPA consulted with
                      representatives of Tribal governments
                      pursuant to Executive Order 13084.
                      Summaries of the meetings have been
                      included in the public docket for this
                      proposed rulemaking. EPA's
                      consultation, the nature  of the
                      governments' concerns, and EPA's
                      position supporting the need for this
                      rule are discussed in sections XlV.C.l.f.
                      and g. of this Preamble.

                      /. Request for Comments on Use of Plain
                      Language
                       Executive Order  12866 and the
                      President's memorandum of June 1,
                      1998, require each agency to write all
                      rules in plain language. We invite your
                      comments on how to make this
                      proposed rule easier to understand. For
                      example:
                       •  Have we organized the material to
                      suit your needs?
                       •  Are the requirements in the rule
                      clearly stated?
                       •  Does the rule contain technical
                      language or jargon that isn't clear?
                       •  Would a different format (grouping
                      and order of sections, use of headings,
                      paragraphing) make the rule easier to
                      understand?
                       •  Would more (but shorter) sections
                      be better?
   •  Could we improve clarity by adding
 tables, lists, or diagrams?
   •  What else could we do to make the
 rule easier to understand?

 XV. References

   The following references are referred
 to in this notice and are included in the
 public docket together with other
 correspondence and information. The
 public docket is available as described
 at the beginning of this notice. All
 public comments received on the    !
 proposal are included in the public
 docket.
 Agency for Toxic Substances and Disease
  Registry. 1998. Draft Toxicological Profile
  for Arsenic. Prepared for the US
  Department of Health and Human Services
  by the Research Triangle Institute.
 Albores, A., M. E. Cebrian, I. Tellez and B.
  Valdez. 1979. Comparative Study of
  Chronic Hydroarsenicism in Two Rural
  Communities in the Region Lagunra of
  Mexico, [in Spanish]. Bol. Oficina Sanit.
  Panam. 86:196-205.
 Amy, G.L., M. Edwards, M. Benjamin, K.
  Carlson, J. Chwirka, P. Brandhuber, L.
  McNeill and F. Vagliasindi. 1999. Arsenic
  Treatability Options and Evaluation of
  Residuals Management Issues, Draft
  AWWARF Report.
 Anderson, L. and K.W. Bruland. 1991.
  Biogeochemistry of Arsenic in Natural
  Waters: The Importance of Methylated
  Species. Environmental Science
  Technology. 25(3):420-427.
 American Public Health Association (APHA).
  1992 and 1995. Standard Methods for the
  Examination of Water and Wastewater.
  18th Edition, American Public Health
  Association, 1015 Fifteenth Street N.W.,
  Washington, DC 20005.
 American Society for Testing and Materials
  (ASTM). 1994 and 1996. Annual Book of
  ASTM Standards. Vol. 11.01 and 11.02,
  American Society for Testing and
  Materials, 1916 Race Street, Philadelphia,
  PA 19103.
 Aschbacher, P.W. and V.J. Feil.  1991. Fate of
  [14 C] Arsanilic Acid in Pigs and Chickens.
  Journal of Agriculture and Food Chemsitry.
  38:146-148.
 AWWA Research Foundation, AWWA Water
  Industry Technical Action Fund &
  Association of California Water Agencies.
  1995. Research Needs Report: Arsenic in
  Drinking Water: Report from International
  Expert Workshop, Ellicott City, Maryland,
  May 31-June 2, 1995. Prepared by T. David
  Chen, HDR Engineering, Inc. August 1995.
Azcue, J. M. and J. O. Nriagu. 1994. Arsenic:
  Historical Perspectives. In Arsenic in the
  Environment. Part  I: Cycling and
  Characterization. Nriagu, J. O., Ed. New
  York, NY, John Wiley and Sons, Inc: pp:
  1-16.
Borzsonyi, M., A. Berecsky, P. Rudnai, M.
  Csanady and A. Horvath. 1992.
  Epidemiological Studies on Human
  Subjects Exposed to Arsenic in Drinking
  Water in Southeast Hungary. Archives of
  Toxicology. 66:77-78.

-------
                   Federal  Register/Vol.  65, No.  121/Thursday, June  22, 2000/Proposed Rules
                                                                            38975
Buchanan, W. D. 1962. Toxicity of Arsenic
  Compounds. Amsterdam, Elsevier
  Scientific Publishers, pp v-viii.
Calvert, C.C. 1975. Arsenicals in Animal
  Feeds and Waste. In Arsenical Pesticides.
  Woolson, E. A., Ed. Washington, DC,
  American Chemical Society: pp. 70-80.
Cebrian, M. 1987. Some Potential Problems
  in Assessing the Effects of Chronic Arsenic
  Exposure in North Mexico [preprint
  extended abstract]. New Orleans, LA,
  American Chemical Society.
Cebrian, M. E., A. Albores, M. Aguilar and E.
  Blakely. 1983. Chronic Arsenic Poisoning
  in the North of Mexico. Human
  Toxicology. 2:121-133.
Chatterjee, A., D. Das, B.K. Mandal, T.R.
  Chowdhury, G. Samanta, and D.
  Chakraborti. 1995. Arsenic in Ground
  Water in Six Districts of West Bengal,
  India: The Biggest Arsenic Calamity in the
  World. Part 1. Arsenic Species in Drinking
  Water and Urine of the Affected People.
  Analyst. 120: 643-650.
Chen, C. J., Y. C. Chuang, T. M. Lin and H.
  Y. Wu. 1985. Malignant Neoplasms Among
  Residents of a Blackfoot Disease Endemic
  Area in Taiwan: High Arsenic Well Water
  and Cancers. Cancer Research. 45:5895-
  5899.
Chen, C.J., H.Y. Chiou, M.H. Chiang, L.J. Lin
  and T.Y. Tai. 1996. Dose-Response
  Relationship Between Ischemic Heart
  Disease Mortality and Long-Term Arsenic
  Exposure.  Arteriosclerosis, Thrombosis,
  and Vascular Biology. 16(4):504-510.
Clifford, D. 1994. Computer Prediction of
  Arsenic Ion Exchange. Journal of the
  American Water Works Association. 86:10:
  Pgs
Clifford, D. and Z. Zhang. 1994. Arsenic
  Chemistry and Speciation. American Water
  Works Association Annual Conference.
  New York, NY. June 19-23.
Clifford, D., G. Ghurye, A. Tripp, J. Tong.
  1997. Final Report: Phases 1 and 2, City of
  Albuquerque Arsenic Study. Field Studies
  on Arsenic Removal in Albuquerque, New
  Mexico Using the University of Houston/
  EPA Mobile Drinking Water Treatment
  Research Facility. Prepared for John
  Stromp, IIII,  Water Resources Manager,
  City of Albuquerque. December 1997.
Clifford, D., G. Ghurye and A. Tripp. 1998.
  Arsenic Removal by Ion Exchange With
  and Without Brine Reuse. American Water
  Works Association Inorganic Contaminants
  Workshop. San Antonio, TX, February 22-
  24.
Concha, G., G.  Vogler, D. Lezcano, B. Nermell
  and M. Vahter. 1998. Exposure to Inorganic
  Arsenic Metabolites During Early Human
  Development. Toxicological Sciences.
  44:185-190.
Cuzick, J., S. Evans, M. Gillman, and D. A.
  Price Evans.  1982. Medicinal Arsenic and
  Internal Malignancies. British Journal of
  Cancer. 45:904-911.
Desi, 1.1992. Arsenic Contamination of
  Drinking Water in South-East Hungary.
  Geographia Medica. 22:45-53.
Ershow, A.G., and K. Cantor. 1989. Total
  Water and Tapwater Intake in the United
  States: Population-Based Estimates of
  Quantities and Sources. Prepared under
  the National Cancer Institute Order # 263—
  MD-810264.
Frey, M. M. and M. A. Edwards. 1997.   !
  Surveying Arsenic Occurrence. Journal of
  the American Water Works Association.
  89(3):105-117.
Furst, A. 1983. A New Look at Arsenic   ;
  Carcinogenesis. In Arsenic: Industrial, '
  Biomedical, Environmental Perspectives.
  Lederer, W. H. and Fensterheim, R. J., Eds.
  New York, Van Nostrand Reinhold: pp.
  151-165.                           .
Guha Mazumder, D. N., J. Das Gupta, A.
  Santra, A. Pal, A. Ghose, S. Sarkar, N.
  Chattopadhaya and D. Chakraborty. 1997.
  Non-Cancer Effects of Chronic Arsenicosis
  with Special Reference to Liver Damage. In
  Arsenic: Exposure and Health Effects.  .
  Abernathy, C. O., Calderon, R. L. and  •
  Chappell, W., Eds. London, Chapman and
  Hall: pp. 112-123.                   >
Helsel, D. R. and T. A. Cohn.  1988.
  Estimation of Descriptive Statistics for
  Multiply Censored Water Quality Data.1
  Water Resources Research.  24(12):1997r
  2004.
Hindmarsh, J. T., O. R. McLetchie, L. P. M.
  Heffernan, O. A. Hayne, H.  A. Ellenberger,
  R. F. McCurdy and H. J. Thiebaux. 1977.
  Electromyographic Abnormalities in
  Chronic Environmental Arsenicalism.
  Analytical Toxicology. 1:270-276.
Hopenhayn-Rich, C., M. L. Biggs, A. Fuch's,
  R. Bergoglio, E. E. Tello, H. Nicolli and A.
  H. Smith. 1996. Bladder Cancer Mortality
  Associated With Arsenic in Drinking Water
  in Argentina. Epidemiology. 7(2):117-124.
Hopenhayn-Rich, C., M. L. Biggs and A. H.
  Smith. 1998. Lung and Kidney Cancer ,
  Mortality Associated with Arsenic in
  Drinking Water in Cordoba, Argentina.,
  Epidemiology. 27:561-569.           '
Hotta, N. 1989. Clinical Aspects of Chronic
  Arsenic Poisoning due to Environmental
  Pollution in and around a Small Refining
  Spot. Nippon Taishitsugaku Zasshi. 53:49-
  70.
Irgolic, K. J. 1994. Determination of Total,
  Arsenic and Arsenic Compounds in
  Drinking Water. In Arsenic: Exposure and
  Health. Chappell, W. R., Abernathy, C. O.
  and Cothern, C. R., Eds. Northwood, U.K.,
  Science and Technology Letters: pp. 51-60.
Jordan, D., M. McClelland, A. Kendig and R.
  Frans. 1997. Monosodium
  Methanearsonate Influence on Broadleaf
  Weed Control with Selected          !
  Postemergence-Directed Cotton Herbicides.
  Cotton Science. 1:72-75.
Kempic, J.B. 2000. Centrally managed POU/
  POE Option for Compliance with the  !
  Arsenic Regulation. AWWA Inorganic
  Contaminants Workshop, Albuquerque,
  NM, February 27-29,  2000.
Kurttio, P, E. Pukkala, H. Kahelin, A.
  Auvinen, and J. Pekkanen. 1999. Arsenic
  Concentrations in Well Water and Risk of
  Bladder and Kidney Cancer in Finland.
  Environmental Health Perspectives
  107(9):705-710.
Lai, M.S., Y.M. Hsueh, C.J. Chien, M.P. Shyu,
  S.Y. Chen, T.L. Kuo, M.M. Wu, and T.Y.
  Tai. 1994. Ingested Inorganic Arsenic and
  Prevalence of Diabetes Mellitus. American
  Journal of Epidemiology. 139(5):484-492.
Lewis, D. R., J. W. Southwick, R. Ouellet-
  Hellstrom, J. Rench and R. L. Calderon.
  1999. Drinking Water Arsenic in Utah: A
  Cohort Mortality Study. Environmental
  Health Perspectives. 107{5):359-365.
Luchtrath, H. 1983. The Consequences of
  Chronic Arsenic Poisoning Among Moselle
  Wine Growers. Pathoanatomical
  Investigations of Post-Mortem
  Examinations Performed Between 1960
  and 1977. Journal of Cancer Research and
  Clinical Oncology. 105:173-182.
Macintosh D.L., P.L. Williams, D.J. Hunter,
  L.A. Sampson, S.C. Morris, W.C. Willett,
  and E.B. Rimm. 1997. Evaluation of a Food
  Frequency Questionnaire-Food
  Composition Approach for Estimating
  Dietary Intake of Inorganic Arsenic and
  Methylmercury. Cancer Epidemiology,
  Biomarkers & Prevention. 6:1043-1050.
Moody, J.P. and R.T. Williams. 1964. The
  Fate of Arsanilic Acid and Acetylarsanilic
  Acid in Hens. Food and Cosmetics
  Toxicology. 2:687-693.
Morris, J.S., M. Schmid, S. Newman, P. J.
  Scheuer and S. Sherlock. 1974. Arsenic
  and Noncirrhotic Portal Hypertension.
  Gastroenterology. 66:86-94.
National Academy of Sciences. 1977.
  Arsenic. Medical and Biological Effects of
  Environmental Pollutants. Washington,
  DC. National Academy Press. 332  pp.
National Drinking Water Advisory Council
  (NDWAC). 1998. Benefits Working Group
  Report to NDWAC. October 29, 1998.
National Research Council. 1983. Risk
  Assessment in the Federal Government:
  Managing the Process. National Academy
  Press.
National Research Council. 1999. Arsenic in
  Drinking Water. Washington, DC. National
  Academy Press.
Neubauer, O. 1947. Arsenical Cancer: A
  Review. British Journal of Cancer. 1:192-
  251. (as cited in US EPA.1976)
Nevens, F., J. Fevery, W. Van Steenbergen, R.
  Sciot, V. Desmet and J. De Groote. 1990.
  Arsenic and Noncirrhotic Portal
  Hypertension: A Report of Eight Cases.
  Hepatology. 11:80-85.
Oya-Ohta Y., T. Raise, and T. Ochi. 1996.
  Induction of Chromosomal Aberrations in
  Cultured Human Fibroblasts by Inorganic
  and Organic Arsenic Compounds and the
  Different Roles of Glutathione in Such
  Induction. Mutation Research.  357:123-.
  129. [as cited in NRC 1999 in XI.B]
Rahman, M. and J.O. Axelson. 1995. Diabetes
  Mellitus and Arsenic Exposure: a Second
  Look at Case-Control Data from a Swedish
  Copper Smelter. Occupational
  Environmental Medicine. 52:773—774.
Rahman, M., M. Tondel, S.A. Ahmad, and C.
  Axels. 1998. Diabetese Mellitus Associated
  with Arsenic Exposure in Bangladesh.
  American Journal of Epidemiology.
  148(2):198-203.
Rogers E.H., N. Chernoff, and B.J. Kavlock.
  1981. The Teratogenic Potential of
  Cacodylic Acid in the Rat and Mouse. Drug
  and Chemical Toxicology. 4(1):49-61.
Roth, F. 1956. Concerning Chronic Arsenic
  Poisoning of the Moselle Wine Growers
  with Special Emphasis on Arsenic
  Carcinomas. Krebsforschung. 61:287-319.
Sabbioni, E., M. Fischbach, G. Pozzi, R.
  Pietra, M. Gallorini and J. L. Piette. 1991.
  Cellular Retention, Toxicity and
  Carcinogenic Potential of Seafood  Arsenic.

-------
38976
Federal  Register/Vol.  65, No.  121/Thursday, June  22,  2000/Proposed Rules
  I. Lack of Cytotoxicity and Transforming
  Activity of Arsenobetaine in the BALB/3T3
  Cell Line. Carcinogenesis. 12:1287-1291.
Smith, A.H., M. Goycolea, R. Haque and M.
  L. Biggs. 1998. Marked Increase in Bladder
  and Lung Cancer Mortality in a Region of
  Northern Chile Due to Arsenic in Drinking
  Water. American Journal of Epidemiology.
  147(7):660~669.
Southwick, J. W., A. B. Western, M. M. Beck,
  T. Whitley, R. Isaacs, J. Petajan and C. D.
  Hanson. 1983. An Epidemiological Study
  of Arsenic in Drinking Water in Millard
  County, Utah. In Arsenic: Industrial,
  Biomedical, Environmental Perspectives.
  Lederer, W. H. and Fensterheim, R. J., Eds.
  New York, Van Nostrand Reinhold: pp.
  210-225.
Tabacova, S., D. D. Baird, L. Balabaeve, D.
  Lolova and I. Petrov. 1994. Placental
  Arsenic and Cadmium in Relation to Lipid
  Peroxides and Glutathione Levels in
  Maternal-Infant  Pairs From a Copper
  Smelter Area. Placenta. 15:873-881.
Tay, C.H. and C.S. Seah. 1975. Arsenic
  Poisoning From Anti-Asthmatic Herbal
  Preparations. Medical Journal, Australia.
  2:424-428.
Thompson, P.M., J.N. Gledd, R.P. Woods, D.
  MacDonald, A.C. Evans and A.W. Toga.
  2000.  Growth Patterns in the Developing
  Brain  Detected by Using Continuum
  Mechanical Tensor Maps. Nature. 404:190-
  193.
Tsai, S.M., T. N. Wang and Y.C. Ko. 1999.
  Mortality for Certain Diseases in Areas
  with High Levels of Arsenic in Drinking
  Water. Archives of Environmental Health.
  54(3):186-193.
Tseng, W. P., H. M. Chu, S. W. How, J. M.
  Pong,  C. S. Lin and S. Yeh. 1968.
  Prevalence of Skin Cancer in an Endemic
  Area of Chronic Arsenicism in Taiwan.
  Journal of the National Cancer Institute.
  40(3):453-463.
Tseng, W. P. 1977. Effects and Dose-Response
  Relationships of Skin Cancer and Blackfoot
  Disease with Arsenic. Environmental
  Health Perspectives. 19:109-119.
US EPA. 1975. Water Programs: National
  Interim Primary Drinking Water
  Regulations. Federal Register. Vol. 40, No.
  248, p, 59566. December 24,1975.
US EPA. 1976. National Interim Primary
  Drinking Water Regulations. Office of
  Water Supply. EPA 570/9-76-003.
US EPA. 1980. Water Quality Criteria
  Documents; Availability. Federal Register.
  Vol. 45. No. 291, p. 79318. November 28,
  1980.
US EPA. 1983. National Revised Primary
  Drinking Water Regulations; Advance
  Notice of Proposed Rulemaking. Federal
  Register. Vol. 48, No. 194, p. 45502.
  October 5,1983.
US EPA. 1984. Health Assessment Document
  for Inorganic Arsenic. Office of Health and
  Environmental Assessment, Office of
  Research and Development. EPA-600/8-
  83-021F. March, 1984.
US EPA. 1985a. National Primary Drinking
  Water Regulations;  Volatile Synthetic
  Organic Chemicals; Proposed Rule. Federal
  Register. Vol. 50, No. 219, p. 46906.
  November 13,1985.
US EPA. 1985b. National Revised Primary
  Drinking Water Regulations; Synthetic
                         Organic Chemicals, Inorganic Chemicals
                         and Microorganisms; Proposed Rule.
                         Federal Register. Vol. 50, No. 219, p.
                         46936. November 13, 1985.
                       US EPA. 1988. Special Report on Ingested
                         Inorganic Arsenic: Skin Cancer; Nutritional
                         Essentiality. Risk Assessment Forum. EPA/
                         625/3-87/013. 124 pp. July 1988.
                       US EPA. 1989a. Cover letter dated August 14,
                         1989, from SAB to EPA. Ad Hoc Work
                         Group, Drinking Water Subcommittee,
                         Environmental Health Advisory
                         Committee, Science Advisory Board
                         Report: A Critical Examination of the
                         Evidence for a Threshold For Cancer Risk
                         in Humans from Inorganic Arsenic.
                         Washington, DC. June 1989 report.
                       US EPA. 1989b. Cover letter dated September
                         28,1989 from SAB to EPA. Science
                         Advisory Board's Review of the Arsenic
                         Issues Relating to the Phase II Proposed
                         Regulations  From the Office of Drinking
                         Water. Science Advisory Board Committee:
                         Drinking Water Subcommittee of the
                         Environmental Health Committee.
                       US EPA. 1991a. National Primary Drinking
                         Water Regulations—Synthetic Organic
                         Chemicals and Inorganic Chemicals;
                         Monitoring for Unregulated Contaminants;
                         National Primary Drinking  Water
                         Regulations  Implementation; National
                         Secondary Drinking Water  Regulations.
                         Federal Register. Vol. 56, No. 20, p. 3526.
                         January 30, 1991.
                       US EPA. 1991b. Arsenic IRIS File; Arsenic,
                         Inorganic. February 1991. Used for 1992
                         National Toxics Rule, December 22, 1992,
                         57 FR 60848.
                       US EPA. 1991c. Arsenic Research
                         Recommendations memorandum dated
                         April 12, 1991 from John R. Fowle III,
                         Chair of the  Arsenic Research
                         Recommendation Workgroup, Health
                         Effects Research Laboratory.
                       US EPA. 1992a. Science Advisory Board
                         Report: Review of the Office of Research
                         and Development's Arsenic Research
                         Recommendations. Washington, DC.
                         Science Advisory Board. EPA-SAB-DWC-
                         92-018. May 1992.
                       US EPA. 1992b. National Primary Drinking
                         Water Regulations; Synthetic Organic
                         Chemicals and Inorganic Chemicals; Final
                         Rule. Federal Register. Vol. 57, No. 138, p.
                         31776. July 17,1992.
                       US EPA. 1992c. Water Quality Standards;
                         Establishment of Numeric Criteria for
                         Priority Toxic Pollutants; States'
                         Compliance; Final Rule. Federal Register.
                         Vol. 57, No.  246, p. 60848.  December 22,
                         1992.
                       US EPA. 1992d.  Bartley, C.B., P.M. Colucci,
                         and T. Stevens. The Inorganic Chemical
                         Characterization of Water Treatment Plant
                         Residuals. EPA/600/SR-92-117,
                         Cincinnati.
                       US EPA. 1993a. Science Advisory Board
                         Report: Review of the Draft Drinking Water
                         Criteria Document on Inorganic Arsenic.
                         Washington, DC. Science Advisory Board.
                         EPA-SAB-DWC-94-004. November 8,
                         1993.
                       US EPA. 1993b.  Inorganic Arsenicals;
                         Conclusion of Special Review. Federal
                         Register. Vol. 58, No. 234, p. 64579.
                         December 8, 1993.
US EPA. 1994a. EPA Method 200.15,    !
  Determination of Metals and Trace
  Elements in Water By Ultrasonic
  Nebulization Inductively Coupled Plasma-
  Atomic Emission Spectrometry. Methods
  for the Determination of Metals in
  Environmental Samples—Supplement I,
  Rev 1.2. EPA/600/R-94-111. May 1994,
US EPA. 1994b. Methods for the        ,
  Determination of Metals in Environmental
  Samples—Supplement I. EPA/600/R-94-
  111, NTIS PB 94-184942.
US EPA. 1994c. National Primary and   ,
  Secondary Drinking Water Regulations;
  Analytical Methods for Regulated Drinking
  Water Contaminants; Final Rule. Federal
  Register. Vol. 59, No. 232, p. 62456.
  December 5, 1994.
US EPA. 1994d. SW-846 Method 6020,
  Inductively Coupled Plasma-Mass
  Spectrometry. Test Methods for Evaluating
  Solid Waste: Physical/Chemical Methods.
  Third Edition, September 1994 Update II.
US EPA. 1994e. SW-846 Method 7060A,
  Arsenic (Atomic Absorption, Furnace
  Technique). Test Methods for Evaluating
  Solid Waste: Physical/Chemical Methods.
  Third Edition, September 1994 Update II.
US EPA. 1994f. SW-846 Method 7062,
  Antimony and Arsenic (Atomic
  Absorption, Borohydride Reduction). Test
  Methods for Evaluating Solid Waste,
  Physical/Chemical Methods. Third Edition,
  September 1994 Update II.
US EPA. 1995. Science Advisory Board
  Report: Review of Issues Related to the
  Regulation of Arsenic in Drinking Water.
  Washington, DC. Science Advisory Board.
  EPA-SAB-DWC-95-015. July 19, 1995.
US EPA. 1996a. EPA Method 1632, Inorganic
  Arsenic In Water by Hydride Generation
  Quartz Furnace Atomic Absorption. EPA/
  821/R-96-013. July 1996.
US EPA. 1996b. Proposed Guidelines for
  Carcinogenic Risk Assessment; Notice.
  Federal Register. Vol 61, No. 79, p. 17960.
  April 23, 1996.
US EPA. 1996c. Performance Evaluation
  Studies Supporting Administration of the
  Clean Water Act and the Safe Drinking
  Water Act. Federal Register. Vol. 61, No.
  139, p. 37464. July 18, 1996.         !
US EPA. 1996d. SW-846 Method 7063,
  Arsenic in Aqueous Samples and Extracts
  by Anodic Stripping Voltammetry (ASV).
  Test Methods for Evaluating Solid Wastes,
  Physical/Chemical Methods. Third Edition,
  December 1996, Update III.
US EPA. 1996e. Investigator-Initiated Grants
  on Health Effects of Arsenic. Federal
  Register. Vol 61, No. 236, p. 64739.
  December 6, 1996.
US EPA. 1997a. Manual for the Certification
  of Laboratories Analyzing Drinking Water.
  EPA 815/B-97/001.
US EPA. 1997b. March 1994 Workshop on
  Developing an Epidemiology Research
  Strategy for Arsenic in Drinking Water.
  Prepared for EPA's National Health and
  Environmental Effects Research Laboratory
  by SRA Technologies. April 14,1997.
US EPA' 1997c. Performance Evaluation
  Studies Supporting Administration of the
  Clean. Water Act and the Safe Drinking
  Water Act. Federal Register. Vol. 62, No.
  113, p. 32112. June 12, 1997.         :

-------
                   Federal Register/Vol. 65,  No.  121/Thursday, June 22,  2000/Proposed Rules
                                                                            38977
 US EPA. 1997d. National Center for
   Environmental Assessment. Report on the
   Expert Panel on Arsenic Carcinogenicity:
   Review and Workshop. Lexington, MA.
   Prepared by the Eastern Research Group
   under contract to US EPA. August 1997.
 US EPA. 1997e. Performance Based
   Measurement System. Federal Register.
   Vol. 62, No. 193, p. 52098. October 6, 1997.
 US EPA. 1997f. Benefits and Costs of the
   Clean Air Act. 1970-1990. Clean Air Act
   § 812. Report Prepared for U.S. Congress by
   US EPA Office of Air and Radiation.
   Chapter 6. October. EPA 410-R-97-002.
 US EPA. 1997g. Community Water System
   Survey, Volume I: Overview and Volume
   II: Detailed Survey Result Tables and
   Methodology Report. EPA 815-R-97-001a
   and EPA 815-R-97-001b. January, 1997.
 US EPA. 1998a. Research Plan for Arsenic in
   Drinking Water. Office of Research and
   Development, National Center for
   Environmental Assessment. EPA/600/R-
   98/042. www.epa.gov/ORD/WebPubs/final/
   arsenic.pdf February 1998.
 US EPA. 1998b. Environmental Justice
   Stakeholders Meeting March 12,1998
   Meeting Summary.
 US EPA. 1998c. Locating and Estimating Air
   Emissions From Sources of Arsenic and
   Arsenic Compounds. Research Triangle
   Park, NC. Office of Air Quality Planning
   and Standards. EPA-454-R-98-013. June
   1998.
 US EPA 1998d. National Primary Drinking
   Water Regulations: Analytical Methods for
   Regulated Drinking Water Contaminants;
   Final and Proposed Rule. Federal Register.
   Vol. 63, No. 171, p. 47097. September 3,
   1998.
 US EPA. 1998e. National Primary Drinking
   Water Regulations: Consumer Confidence
   Reports. Final Rule. Federal Register. Vol.
   63, No. 160, p. 44512. August 19, 1998.
 US EPA. 1998f. Variance Technology
   Findings for Contaminants Regulated
   Before 1996. Office of Water. EPA 815-R-
   98-003. September 1998.
 US EPA. 1998g. Information for Small Entity
   Representatives Regarding the Arsenic in
   Drinking Water Rule. December 3,1998.
 US EPA. 1998h. Announcement of Small
   System Compliance Technology Lists for
   Existing National Primary Drinking Water
  Regulations and Findings Concerning
   Variance Technologies. Notice of Lists of
   Technologies and Upcoming Release of
   Guidance and Supporting Documents.
  Federal Register. Vol. 63, No. 153, p.
  42032 at 43045. August 6,1998.
 US EPA. 19981. Removal of the Prohibition
  on the Use of Point of Use Devices for
  Compliance with National Primary
  Drinking Water Regulation, Federal
  Register notice (63 FR 31934). June 11,
   1998.
 US EPA. 1998J. National Primary Drinking
  Water Regulations: Consumer Confidence
  Reports. Proposed Rule. Federal Register.
  Vol. 63, No. p. 7605. February 13, 1998.
US EPA. 1999a. Cost of Illness Handbook.
  Office of Pollution Prevention and Toxics.
  Chapter 1 II.8. Cost of Bladder Cancer.
  September,  1999.
US EPA. 1999b. National Primary Drinking
  Water Regulations: Public Notification
   Rule, Proposed Rule. Federal Register. Vol.
   64, No. 92, p. 25964. May 13, 1999.
 US EPA. 1999c. Report of the Small Business
   Advocacy Review Panel on EPA's Planned
   Proposal of the National Primary Drinking
   Water Regulation for Arsenic. Cover memo
   to the Administrator and the report. June
   4, 1999.
 US EPA. 1999d. Decision Tree for the
   Arsenic Rulemaking Process. Washington,
   DC. Office of Ground Water and Drinking
   Water. 404 pp. July 1999.
 US EPA. 1999e. Geometries and
   Characteristics of Public Water Systems.
   Prepared by Science Applications
   International Corporation under contract
   with EPA OGWDW. August 15,1999.  :
 US EPA. 1999f. Co-Occurrence of Drinking
   Water Contaminants. Prepared by Science
   Applications International Corporation
   under contract 68-C6-0059 for EPA
   OGWDW. September 30, 1999.        !
 US EPA. 1999g. Small Systems Compliance
   Technology List for the Arsenic Rule.
   Prepared by ICI under contract 68-C6-
   0039. November, 1999.               :
 US EPA. 1999h. Small Systems Compliance
   Technology List for the Arsenic Rule.
   Prepared by ICI under contract 68-C6-,
   0039. November, 1999.
 US EPA. 1999L Technologies and Costs for
   the Removal of Arsenic from Drinking
   Water. Washington, DC. Office of Ground
   Water and Drinking Water. 386 pp.
   November, 1999.
 US EPA. 1999J. National Primary Drinking
   Water Regulations: Analytical Methods for
   Chemical and Microbiological
   Contaminants and Revisions to Laboratory
   Certification Requirements; Final Rule. >
   Federal Register. Vol. 64, No. 230, p.
   67450. December 1, 1999.
 US EPA. 1999k. Arsenic Risk
   Characterization, Part 1. Prepared by ISSI
   Consulting Group, Inc. for EPA Office of
   Water, Office of Standards and
   Technology. December 22, 1999.
 US EPA. 19991. Analytical Methods Support
  Document for Arsenic in Drinking Water.
  Prepared by Science Applications
  International Corporation under contract
  with EPA OGWDW, Standards and Risk
  Management Division. December, 1999.
 US EPA. 1999m. Drinking Water Baseline
  Handbook. Prepared by International  ,
  Consultants, Inc. under contract with EPA
  OGWDW, Standards and Risk Management
  Division. February 24,1999.
 US EPA. 1999n. National Primary Drinking
  Water Regulations: Radon-222, Proposed
  Rule. Federal Register. Vol. 64, No. 211, p.
  59246. EPA 815-Z-99-006. November 2,
  1999.
US EPA. 19990. Radon and Arsenic
  Regulatory Compliance Costs for the 25
  Largest Public Water Systems (With
  Treatment Plant Configurations) Prepared
  for U.S. EPA by Science Applications
  International Corporation. August 10,1999.
US EPA. 2000a. Estimated Per Capita Water
  Ingestion in the United States: Based on
  Data Collected by the United States
  Department of Agriculture's (USDA) 1994-
  1996 Continuing Survey of Food Intakes by
  Individuals. Office of Water, Office of
  Standards and Technology. EPA-822-00-
  008. April 2000.                     \
 US EPA. 2000b. Arsenic Occurrence in
   Public Drinking Water Supplies.
   Washington, DC. Office of Ground Water
   and Drinking Water. EPA 815-0-00-001.
   May 2000.
 US EPA. 2000c. National Primary Drinking
   Water Regulations: Public Notification
   Rule; Final Rule. Federal Register. Vol. 65,
   No. 87, p. 25982. May 4, 2000.
 US EPA. 2000d. National Primary Drinking
   Water Regulations: Ground Water Rule;
   Proposed Rule. Federal Register. Vol. 65,
   No. 91, p. 30193. May 10, 2000.
 US EPA. 2000e. Regulatory Impact Analysis
   (RIA) of the Arsenic Rule. May 2000.
 US GS. 1998. Reese, R.G. Jr. Arsenic. In
   United States Geological Survey Minerals
   Yearbook, Fairfax, VA, US Geological
   Survey.
 US GS. 1999. Reese, R.G. Jr. Arsenic. In
   Mineral Commodity Summaries. Fairfax,
   VA, pgs. 26-27. US Geological Survey.
   January 1999.
 US GS. 2000. Focazio, M., A. Welch, S.
   Watkins, D. Helsel & M. Horn. A
   retrospective analysis of the occurrence of
   arsenic  in ground water resources of the
   United States and limitations in drinking
   water supply characterizations. Water
   Resources Investigations Report: 99-4279.
   May 2000.
 US Public Health Service. 1943. Public
   Health Service Drinking Water Standards.
   Approved Revisions to the 1925 Drinking
   Water Standards on December 3,1942.
   Public Health Reports. 58(3):69-82.
   January 15,1943.
 US Public Health Service. 1946. Public
   Health Service Drinking Water Standards.
   Approved Revisions to the 1942 Drinking
   Water Standards by the American  Water
   Works Association. Public Health  Reports.
   61(ll):371-384. March 15, 1946.
 US Public Health Service. 1962. Chapter 1—
   Public Health Service, Department of
   Health Education and Welfare. Title  42
   Public Health, Part 72 Interstate
   Quarantine, Subpart J Drinking Water
   Standards. Federal Register, p. 2152.
  March 6, 1962.
 Vahter, M. 1994. Species differences in the
  metabolism of arsenic compounds.
  Applied Organometallic Chemistry. 8:175-
   182.
 Vallee, B. L., D. D. Ulmer and W. E. C.
  Wacker. 1960. Arsenic Toxicology and
  Biochemistry. AMA Arch. Ind. Med.
  21:56-75.
 Viscusi, W.K., W.A. Magat, and J. Huber.
  1991. Pricing Environmental Health Risks:
  Survey Assessments of Risk—Risk and
  Risk-Dollar Trade-Offs for Chronic
  Bronchitis. Journal of Environmental
  Economics and Management. 21:32-51.
 Wang, L., T.J. Sorg, A.S.C. Chen, and K.
  Fields. 2000. Arsenic Removal by Full
  Scale Ion Exchange and Activated  Alumina
  Treatment Systems. AWWA Inorganic
  Contaminants Workshop, Albuquerque,
  NM, February 27-29, 2000.
Welch, A.  H., M. Lico and H. J. 1988. Arsenic
  in Ground water of the Western United
  States. Ground Water. 26(3):333-347.
Webster, R.C., H.I. Maibach, L.  Sedik, J.
  Melendres, and M. Wade. 1993. In Vivo
  and In Vitro Percutaneous Absorption and

-------
38978
Federal Register/Vol. 65, No. 121/Thursday, June  22,  2000/Proposed Rules
  Skin Decontamination of Arsenic from
  Water and Soil. Fundamental and Applied
  Toxicology. 20:336-340.
Winship, K. A. 1984. Toxicity of Inorganic
  Arsenic Salts. Adverse Drug Reactions and
  Acute Poisoning Reviews. 3:129-160.
World Health Organization. 1981.
  Environmental Health Criteria 18 Arsenic.
  United Nations Environment Programme,
  International Labour Organisation, and the
  World Health Organization.
Wu, M. M., T. L. Kuo, Y. H. Hwang and C.
  J. Chen. 1989. Dose-Response Relation
  Between  Arsenic Concentration in Well
  Water and Mortality From Cancers and
  Vascular Diseases. American Journal of
  Epidemiology. 130(6):1123-1132.
Yeh, S. 1973. Skin Cancer in Chronic
  Arsenicism. Human Pathology. 4(4):469-
  485.
Zaldivar, R. 1974. Arsenic Contamination of
  Drinking Water and Food-Stuffs Causing
  Endemic Chronic Poisoning. Beitr.
  Pathology. 151:384-400.
Zaldivar, R., L. Prunes and G. Ghai. 1981.
  Arsenic Dose in Patients with Cutaneous
  Carcinoma and Hepatic Hemangio-
  Endothelioma After Environmental and
  Occupational Exposure. Archives of
  Toxicology. 47:145-154.

List of Subjects

40 CFR Part 141
   Environmental protection, Chemicals,
Indians—lands, Intergovernmental
relations, Reporting and recordkeeping
requirements, Water supply.

40 CFR Part 142
   Environmental protection,
Administrative practice and procedure,
Chemicals, Indians—lands, Reporting
and recordkeeping requirements, Water
supply.
   Dated: May 24, 2000.
Carol M. Browner,
Administrator.
   For  reasons set out in the preamble,
the Environmental Protection Agency
proposes  to amend 40 CFR parts 141
and 142 as follows:

 PART 141—NATIONAL PRIMARY
 DRINKING WATER REGULATIONS

   1. The authority citation for part 141
 continues to read as follows:
   Authority: 42 U.S.C. 300f, 300g-l, 300g-2,
 300g-3, 300g-4, 300g-5, 300g-6, 300J-4,
 300J-9, and 300J-11.

 Subpart A—General

 §141.2 [Amended]
   2. Section 141.2 is amended by
 revising the definition heading for
 "Point-of-entry treatment  device" to
 read "Point-of-entry treatment device
 (POE)" and revising the definition
 heading for "Point-of-use  treatment
 device" to read "Point-of-use treatment
 device (POU)".
                        3. Section 141.6 is amended by:
                        a. In paragraph (a) by revising the
                     reference "(a) through (i)" to read "(a)
                     through (k)".
                        b. Revising paragraph (c).
                        c. Adding paragraphs (j) and (k).
                        The revisions and additions read as
                     follows:

                     §141.6  Effective dates.
                     *****
                        (c) The regulations set forth in
                     §§ 141.11(d); 141.21(a), (c) and (i);
                     141.22(a) and (e); 141.23(a)(3) and (a)(4);
                     141.23(f); 141.24(e) and (f); 141.25(e);
                     141.27(a); 141.28(a) and (b); 141.31(a),
                     (d) and (e); 141.32(b)(3); and 141.32(d)
                     shall take effect immediately upon
                     promulgation.
                      *****
                        (j) The arsenic MCL listed in § 141.62
                     is effective [THREE YEARS AFTER
                     PUBLICATION DATE OF THE FINAL;
                     RULE]. Compliance with the arsenic
                     MCL listed in § 141.62 is required for
                      community water systems serving
                      10,000 people or less on [DATE 5
                     YEARS AFTER PUBLICATION DATE
                      OF THE FINAL RULE], and for all other
                      community water systems on [DATE 3
                      YEARS AFTER PUBLICATION DATE
                      OF THE FINAL RULE] for
                      §§141.23(a)(4), (a)(4)(i), (a)(5), (c), (f)(l),
                      (g), (i), (k)(l), (k)(2), and (k)(3)(ii);
                      141.62(b)(16) and (c); 141.203, and
                      revisions to arsenic in Appendices A
                      and B of Subpart Q of this part for the
                      public notification rule. However, the
                      reporting date for the arsenic MCL listed
                      in Appendix A of Subpart O of this part
                      of the consumer confidence rule
                      requirements and the arsenic reporting
                      requirements in § 141.154(b) are
                      [THIRTY DAYS AFTER PUBLICATION
                      DATE OF THE FINAL RULE]. Non-
                      transient non-community water systems
                      will be subject to the sampling,
                      monitoring, and reporting requirements
                      of §§141.23(a), 141.23(c)(lH6),
                      141.23(0, 141.23(g), 141.23(k), 141.203,
                      and 141.209 for arsenic exceeding the
                      MCL listed in § 141.62 [DATE 3 YEARS
                      AFTER PUBLICATION DATE OF THE
                      FINAL RULE].
                         (k) Compliance with §§ 141.23(c)(9),
                      141.24(0(15)(ii), 141.24(0(22) and
                      141.24(h)(20) regulations for inorganics
                      and organics other than total
                      trihalomethanes and sampling
                      frequencies for new systems and new
                      sources of water is required on [DATE
                      3 YEARS AFTER PUBLICATION DATE
                      OF THE FINAL RULE].

                      Subpart B—[Amended]

                         4. Section 141.11 is amended by
                      revising the second sentence of
                      paragraph (a) and revising paragraph (b)
                      to read as follows:
§ 141.11  Maximum contaminant levels for
inorganic chemicals.
  (a) *  * * The analyses and
determination of compliance with the
0.05 milligrams per liter maximum
contaminant level for arsenic use the
requirements of § 141.23(1).
  (b) The maximum contaminant level
for arsenic is 0.05 milligrams per liter
for community water systems serving
10,000 people or less until [DATE 5
YEARS AFTER PUBLICATION DATE
OF THE FINAL RULE], and for all other
community water systems until [DATE
3 YEARS AFTER PUBLICATION DATE
OF THE FINAL RULE]. Non-transient
non-community water systems will be
subject to sampling, monitoring and
reporting requirements for arsenic as of
[DATE 3 YEARS AFTER PUBLICATION
DATE OF THE FINAL RULE]; however,
they will not be subject to
§§ 141.23(c)(7) and (8) and
141.62(b)(16).
 Subpart C—[Amended]

   5. Section 141.23 is amended by:
   a. Adding a new entry for "Arsenic"
 in alphabetical order to the table in
 paragraph (a)(4)(i) and footnotes 6 and
 7.
   b. Adding "arsenic," before "barium,"
 in paragraph (a)(5).
   c. Adding "arsenic," before "barium,"
 in paragraph (c) introductory text.
   d. Adding paragraph (c)(9).
   e. Revising the  words "asbestos,
 antimony," to read "antimony, arsenic,
 asbestos," in paragraph (0(1)-
   f. Adding "arsenic," before
 "asbestos," in paragraph (i)(l).
   g. Adding one sentence at the end of
 paragraph (i)(l).
   h. Revising paragraph (i)(2).
   i. Add paragraph (i)(5).
   j. Revise "arsenic" entry in the table
 in paragraph (k)(l).
   k. Adding "arsenic," before
 "asbestos," in paragraph (k)(2)
 introductory text.
   1. In the table to paragraph (k)(2) by
 adding in alphabetical order a new entry
 for "Arsenic".
   m. Adding "arsenic," before
 "asbestos," in paragraph (k)(3)
 introductory text.                !
   n. Adding in alphabetical order a new
 entry for "Arsenic" to the table in  ,
 paragraph (k)(3)(ii).
   The revisions and additions read as
 follows:

 § 141.23 Inorganic chemical sampling and
 analytical requirements.
   (a) * * *
   (4) * * *
   (i) *  *  *

-------
                   Federal Register/Vol.  65, No.  121/Thursday, June 22, 2000/Proposed Rules
                                                                         38979
                                    DETECTION LIMITS FOR INORGANIC CONTAMINANTS
           Contaminant
                                   MCL (mg/l)
                                Methodology
                              Detection
                             Limit (mg/l)
Arsenic
                                        0.005  Atomic Absorption; Furnace 	i	I	
                                               Atomic Absorption; Platform-Stabilized Temperature
                                               Atomic Absorption; Gaseous Hydride 	
                                               ICP-Mass Spectrometry 	
                                                                            0.001
                                                                          6 0.0005
                                                                            0.001
                                                                          7 0.0014
  6The MDL reported for EPA Method 200.9 (Atomic Absorption; Platform—Stabilized!Temperature) was determined using a 2x concentration
step during sample digestion. The MDL determined for samples analyzed using direct analysis (i.e., no sample digestion) will be higher Usinq
multiple depositions, EPA 200.9 is capable of obtaining a MDL of 0.0001 mg/L.
  7 Using selective ion monitoring, EPA Method 200.8 (ICP-MS) is capable of obtaining a MDL of 0.0001 mg/L.
  (c) *  * *
  (9) All new systems or systems that
use a new source of water that begin
operation after [EFFECTIVE DATE OF
THE FINAL RULE] must demonstrate
compliance with the MCL within a
period of time specified by the State.
The system must also comply with the
initial sampling frequencies specified by
the State to ensure a system can
demonstrate compliance with the MCL.
Routine and increased monitoring
frequencies shall be conducted in
accordance with the requirements in
this section.
*****
  (i)* *  *
  (1)  *  * * If a system fails to collect
the required number of samples,  :
compliance (average concentration) will
be based on the total number of samples
collected.
  (2)  For systems which are monitoring
annually, or less frequently, the system
is out of compliance with the maximum
contaminant levels for antimony,
arsenic, asbestos, barium, beryllium,
cadmium, chromium, cyanide, fluoride,
mercury, nickel,  selenium or thallium if
the level of a contaminant is greater
than the MCL. If confirmation samples
are required by the State, the
determination of compliance will be
based on the annual average of the
initial MCL exceedance and any State-
required confirmation samples. If a
system fails to collect the required
number of samples., compliance (average
concentration) will be based on the total
number of samples collected.
*****

  (5) Arsenic sampling results will be
reported to the nearest 0.001 mg/L.
*****
  (k) *  * *
  (D*  * *
                      Contaminant and methodology13
                                   EPA
                                             ASTM3
                                                            SM4
                                                                        Other
Arsenic 14:                                                                 ;
    ICP — Mass Spectrometry [[[  :   2200.8
    Atomic Absorption; Platform [[[     2200.9
    Atomic Absorption; Furnace [[[  . ...............  D-2972-93C  3113B.
    Hydride Atomic Absorption [[[  '. ...............  D-2972-93B  3114B.
  2 "Methods for the Determination of Metals in Environmental  Samples — Supplement I",  EPA/600/R-94/1 1 1 ,  May 1994. Available at NTIS,
     —
  3 Annual Book of ASTM Standards,-\QQ4 and 1996, Vols. 11.01 and 11.02, American Society for Testing and Materials. The previous versions
of D1688-95A, D1688-95C (copper), D3559-95D (lead), D1293-95  (pH),  D1125-91A (conductivity) and D859-94 (silica) are also approved
These previous versions D1688-90A, C; D3559-90D, D1293-84, D1125-91A and D859-88, respectively are  located in the Annual Book of
ASTM Stanc/ards,-\994, Vols. 11.01. Copies may be obtained from the American Society for Testing and Materials  100 Barr Harbor Drive West
Conshohocken, PA 19428.
  4 18th  and 19th editions of Standard Methods for the Examination of Water and Wastewater,^92 and 1995, respectively, American Public
Health Association; either edition may be used. Copies may be obtained from the American Public Health Association, 1015 Fifteenth Street NW
Washington, DC 20005.

  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 (/:a,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 31 13 B; and lead by Method D3559-90D unless multiple in-furnace depositions are made.
  14 If ultrasonic nebulization is used in the determination of arsenic by Methods 200.7, 200.8, or SM 3120 B, the arsenic must be in the penta-

-------
38980
Federal  Register/Vol. 65, No.  121/Thursday, June 22, 2000/Proposed  Rules
     Contaminant
                         Preservative1
                                                                                 Container2
                                                                                                      Time3
Arsenic
    Cone HNO3 to pH <2	  P or G
                                                                                                6 months.
  1 When indicated, samples must be acidified at the time of collection to pH <2 with concentrated acid or adjusted with sodium hydroxide to pH
> 12. When chilling is indicated the sample must be shipped and stored at 4°C or less.
  2 P = plastic, hard or soft; G=glass, hard or soft.
  3 In all cases samples should be analyzed as soon after collection as possible. Follow additional (if any) information on preservation, con-
tainers or holding times that is specified in method.
    (3) * * *
    (ii) * * *
                  Contaminant
                                             Acceptance limit
Arsenic	  ±30 at >0.005 mg/l
  6. Section 141.24 is amended by:
  a. Adding one sentence to the end of
paragraph (f)(15)(i).
  b. Removing the last sentence of
paragraph (f)(15)(ii) and adding in its
place two new sentences.
  c. Adding paragraph (f)(22).
  d. Adding a sentence to the end of
paragraph (h)(ll)(i).
  e. Removing the last sentence of
paragraph (h)(ll)(ii) and adding in its
place two new sentences.
  F. Adding paragraph (h)(20).
  The revisions and additions read as
follows:

§ 141.24  Organic chemicals other than
total trihalomethanes, sampling and
analytical methods.
*****
  (f)* * *
  (15) *  *  *
  (i) * * * If a system fails to collect the
required number of samples,
compliance (average concentration) will
be based on the total number of samples
collected.
  (ii)  *  * * If confirmation samples are
required by the State, the determination
of compliance will be based on the
annual average of the initial MCL
exceedance and any State-required
confirmation samples. If a system fails
to collect the required number of
samples, compliance (average
concentration) will be based on the total
number of samples collected.
*****
  (22) All new systems or systems that
use a  new source of water that begin
operation after [DATE THREE YEARS
AFTER PUBLICATION DATE OF
FINAL RULE] must demonstrate
compliance with the MCL within a
                     period of time specified by the State.
                     The system must also comply with the
                     initial sampling frequencies specified by
                     the State to ensure a system can
                     demonstrate compliance with the MCL.
                     Routine and increased monitoring
                     frequencies shall be conducted in
                     accordance with the requirements in
                     this section.
                       (h) *  *  *
                       (11)* *  *
                       (i) * * * If a system fails to collect the
                     required number of samples,
                     compliance (average concentration) will
                     be based on the total number of samples
                     collected.
                       (ii) *  *  * If confirmation samples are
                     required by the State, the determination
                     of compliance will be based on the
                     annual average of the initial MCL
                     exceedance and any State-required
                     confirmation samples. If a system fails
                     to collect the required number of
                     samples, compliance (average
                     concentration) will be based on the total
                     number of samples collected.
                     *****

                       (20) All new systems or systems that
                     use a new source of water that begin
                     operation after [DATE THREE YEARS
                     AFTER PUBLICATION OF THE FINAL
                     RULE] must demonstrate compliance
                     with the MCL within a period of time
                     specified by the State.  The system must
                     also comply with the initial sampling
                     frequencies specified by the State to
                     ensure a system can demonstrate
                     compliance with the MCL. Routine and
                     increased monitoring frequencies shall
                     be conducted in accordance with the
                     requirements in this section.
Subpart F—[Amended]

  7. In § 141.51(b) , the table is amended
by adding in alphabetical order an entry
for Arsenic to read as follows:

§ 141.51 Maximum contaminant level goals
for inorganic contaminants.
*    *    *    *    *
  (b) * * *
                                                                Contaminant
                       MCLG (mg/l)
Arsenic 	  zero
Subpart G—[Amended]

  8. Section 141.60 is amended by
adding paragraph (b)(4) to read as
follows:

§ 141.60 Effective dates.
*****
  (b) * * *
  (4) The compliance date for
§ 141.62(b)(16) is [DATE 5 YEARS
AFTER PUBLICATION DATE OF THE
FINAL RULE] for community water
systems serving 10,000 people or less,
and [DATE 3 YEARS AFTER
PUBLICATION DATE OF THE FINAL
RULE] for all other community water
systems.
  9. Section 141.62 is amended by:
  a. Revising the second sentence of
paragraph (b).
  b. Adding entry "(16)" to the table in
paragraph (b).
  c. Adding an entry and footnote for
"Arsenic" in alphabetical order to the

-------
                  Federal Register/Vol. 65,  No. 121/Thursday,  June 22, 2000/Proposed  Rules
                                                                                                          38981
 table in paragraph (c) and revising the     §141.62  Maximum contaminant levels for
 table heading.
   d. Adding paragraph (d).
   The revisions and additions read as
 follows:
                                      inorganic contaminants.
                                      *****
                                       (b) *  *  * The maximum contaminant
                                      level specified in paragraphs (b)(l) and
                                      (b)(16) of this section  only apply to
                                      community water sy.stems. *  *  *
                                                                                  Contaminant
                                                                                                      MCL (mg/l)
                                                                               (16) Arsenic	  0.005
    (c)
                            BAT FOR INORGANIC COMPOUNDS LISTED IN SECTION 141.62(B)
                       Chemical name
                                                                                    BAT(s)
 Arsenic4
                                                           1, 2, 5, 6, 7, 91
  4BATs for Arsenic V. Pre-oxidation may be required to convert Arsenic III to Arsenic V.
    (d)  The Administrator,  pursuant  to  section 1412  of 'the Act, hereby  identifies in the following table the affordable
 technology, treatment technique, or  other means available to systems serving  10,000  persons or fewer for achieving
 compliance with the maximum contaminant level for arsenic:

                        SMALL SYSTEM COMPLIANCE TECHNOLOGIES (SSCTs)1  FOR ARSENIC 2
Small
Activated Alumina (centralized) 	
Activated Alumina (Point-of-Entry)4 	
Activated Alumina (Point-of-Use) 4 	
Coagulation/Filtration 	
Coagulation-assisted Microfiltration 	
Ion Exchange 	
Lime Softening 	
Oxidation/Filtration 5 	
Reverse Osmosis (centralized) 	
Reverse Osmosis (Point-of-Use)4 	

System Compliance Techho.logy '.










Affordable for listed small system
categories 3
All size categories
All size 'categories
All size categories
501-3,300, 3,301-10,000
501-3,300, 3,301-10,000
All size categories
501-3,300, 3,301-10,000
All size categories
501-3,300, 3,301-10,000
All size cateaories

'Section 141 2(b)(4)(E)(ii) of the SDWA specifies that SSCTs must be affordable and technically feasible for small systems
2 SSCTs for Arsenic V. Pre-oxidation may be required to convert: Arsenic III to Arsenic V

                                                                  ™te' M fewer than 501' ® 
-------
38982
Federal Register/Vol.  65, No.  121 / Thursday, June  22, 2000/Proposed Rules
Contaminant (units)
  Traditional
    MCL
   in mg/L
To convert
 for CCR,
multiply by
 MCL in
CCR units
                                                          MCLG
Major sources in drinking
        water
Health effects language
Inorganic contami-
  nants:
Arsenic (ppb) ..
       0.005
                                     1000
                                 0  Erosion of natural deposits;
                                      Runoff from orchards; Run-
                                      off from glass and elec-
                                      tronics production wastes.
                                                   Some people who drink water
                                                     containing arsenic in excess
                                                     of the MCL over many
                                                     years could experience skin
                                                     damage or problems with
                                                     their circulatory system, and
                                                     may have an increased risk
                                                     of getting cancer.
  Key:
  • '              *                *
  ppb « parts per billion, or micrograms perjiter (ug/l)
Subpart Q—[Amended]
  12. Section 141.203(a), published at
65 FR 26036 on May 4, 2000, and
                       effective June 5, 2000. is; amended by
                       adding entry (4) in numerical order to
                       Table 1 to read as follows:
                                                  § 141.203  Tier 2 Public Notice—Form,
                                                  manner, and frequency of notice.

                                                    (a) *  *  *
      TABLE 1 TO §141.203.—VIOLATION CATEGORIES AND OTHER SITUATIONS REQUIRING A TIER 2 PUBLIC NOTICE
(4) Non-transient non-community water systems exceeding the arsenic MCL.
  13. Appendix A to Subpart Q,
published at 65 FR 26040 on May 4,
2000, effective June 5, 2000, is amended
                       in the table by revising t he entry for "2.
                       Arsenic" under B. Inorganic Chemicals
                       (lOCs), revising endnote 1 and adding
                       endnotes 18 and 19 to read as follows:  ;
                                                  Appendix A to Subpart Q of Part 141.-
                                                  NPDWR Violations and Other
                                                  Situations Requiring Public Notice1
                                                                MCL/MRDL/TT violations2
                          Contaminant
                                             Tit?r of pub-
                                             lic notice re-
                                                quired
                                                                             Citation
                                                         Monitoring & testing procedure viola-
                                                                      tions

                                                        Tier of pub-
                                                        lic notice re-         Citation
                                                          quired
B. Inorganic Chemicals (lOCs)

        *                *

2. Arsenic 	'..
                                                                                         19l4l.23(a),:(c)
Appendix A—Endnotes

  1. Violations and other situations not listed
in this table (e.g., reporting violations and
failure to prepare Consumer Confidence
Reports), do not require notice, unless
otherwise determined by the primacy agency.
Primacy agencies may, at their option, also
require a more stringent public notice tier
(0,g,, Tier 1 instead of Tier 2 or Tier 2 instead
of Tier 3) for specific violations and
situations listed in this Appendix, as
authorized under § 141.202{a) and
§141.203(a).
                         2. MCL—Maximum contaminant level,
                       MRDL—Maximum residual disinfectant
                       level, TT—Treatment technique.
                       *****
                         18. The arsenic MCL citations apply [DATE
                       5 YEARS AFTER PUBLICATION DATE OF
                       THE FINAL RULE] for community water
                       systems serving 10,000 people or less and
                       [DATE 3 YEARS AFTER PUBLICATION
                       DATE OF THE FINAL RULE] for all other
                       community water systems and non-transient
                       non-community water systems. Until then,
                       the citations are § I41.11(b) and § I41.23(n).
                         19. The arsenic Tier 3 violation MCL
                       citations apply [DATE 5 YEARS AFTER
                       PUBLICATION DATE OF THE FINAL RULE]
                       for community water systems serving7,10,000
                                                  people or less and [DATE 3 YEARS AFTER
                                                  PUBLICATION DATE OF THE FINAL RULE]
                                                  for all other community water systems. Until
                                                  then, the citations are § 141.23(a,l).

                                                    14. Appendix B to Subpart Q
                                                  published at 65 FR 26043 on May 4,
                                                  2000, effective June 5, 2000, is amended
                                                  in the table by revising entry "9.
                                                  Arsenic" and adding footnote 23  to read
                                                  as follows:

                                                  Appendix B to Subpart Q of Part 141.—
                                                  Standard Health Effects Language  for Public
                                                  Notification

-------
                  Federal Register/Vol.  65, No.  121/Thursday, June 22,  2000/Proposed Rules
                                                                           38983
           Contaminant
MCLG1
 mg/L
                                             MCL2 mg/L
Standard health effects language for public notification
 9. Arsenic23
                                                  0.005  Some people who drink water containing arsenic in excess of the MCL
                                                          over many years could experience skin damage or problems with thier
                                                          circulatory system, and may have an increased risk of getting cancer.
 Appendix B—Endnotes
  1. MCLG—Maximum contaminant level
 goal.
  2. MCL—Maximum contaminant level.
 *****
  23. These arsenic values apply [DATE 5
 YEARS AFTER PUBLICATION DATE OF
 THE FINAL RULE] for community water
 systems serving 10,000 people or less and
 [DATE 3 YEARS AFTER PUBLICATION
 DATE OF THE FINAL RULE] for all other
 community water systems and non-transient
 non-community water systems. Until then,
 the MCL is 0.050 mg/L and there is no
 MCLG.

 PART 142—NATIONAL PRIMARY
 DRINKING WATER REGULATIONS
 IMPLEMENTATION

  1. The authority citation for part 142
 continues to read as follows:
  Authority: 42 U.S.C. 300f, 300g-l, 300g-2,
 300g-3, 300g-4, 300g-5, 300g-6, 300J-4,
 300J-9, and 300J-11.

 Subpart B—Primary Enforcement
 Responsibility

  2. In § 142.16, revise paragraph (e)
 introductory text and add paragraphs (j)
 and (k) to read as follows:

 § 142.16  Special primacy requirements.
 *****
  (e) An application for approval of a
 State program revision which adopts the
 requirements specified in §§ 141.11,
 141.23, 141.24, 141.40, 141.61 and
 141.62 for a  newly regulated
 contaminant must contain the following
     (in addition to the general primacy
     requirements enumerated elsewhere in
     this part, including the requirement that
     State regulations be at least as stringent
     as the federal requirements):
     *****
       (j) An application for approval of a
     State program revision which adopts the
     requirements specified in §§ 141.11,
     141.23, 141.24, 141.32, 141.40, 141.61
     and 141.62 for an existing regulated
     contaminant must contain the following
     (in addition to the general primacy
     requirements enumerated elsewhere  in
     this part, including the requirement that
     State regulations be at least as stringent
     as the federal requirements):
      (1) If a State chooses to issue waivers
     from the monitoring requirements in
     §§ 141.23,  141.24, and 141.40, the State
     shall describe the procedures and
     criteria which it will use to review
     waiver applications and issue wavier
     determinations. The State shall provide
     the same information required in
     paragraphs (e)(l)(i) and (ii) of this
     section.  States may update their existing
     waiver criteria or use  the requirements
     submitted under the National Primary
     Drinking Water Regulations for the
     inorganic and organic contaminants
     (i.e., Phase II/V rule) in paragraph (e) of
     this section. States may simply note in
     their application any revisions to
     existing waiver criteria or note that the
     same procedures to issue waivers will
     be used.
      (2) A monitoring plan by which the
     State will assure all systems complete
              the required monitoring with the
              regulatory deadlines. States may update
              their existing monitoring plan or use the
              same monitoring plan submitted under
              the National Primary Drinking Water
              Regulations for the inorganic and
              organic contaminants (i.e. Phase II/V
              rule) in paragraph (e) of this section.
              States may simply note in their
              application any revisions to an existing
              monitoring plan or note that the same
              monitoring plan will be used. The State
              must demonstrate that the monitoring
              plan is enforceable under State law.
                (k) States establish the initial
              monitoring requirements for new
              systems and new sources. States must
              explain their initial monitoring
              schedules and how these monitoring
              schedules ensure that public water
              systems and sources that begin
              operation after [DATE THIRTY DAYS
              AFTER PUBLICATION OF THE  FINAL
              RULE]  comply with MCL's and
              monitoring requirements. States  must
              also specify the time frame in which
              new systems will demonstrate
              compliance with the MCLs.
                4. In  § 142.62(b), the table is amended
              by revising the table heading and adding
              arsenic in alphabetical order to the list
              of contaminants to read as follows:

              § 142.62  Variances and exemptions from
              the maximum contaminant levels for
              organic and inorganic chemicals
                (b)
                              BAT FOR INORGANIC COMPOUNDS LISTED IN §141.62(6)'
                      Chemical name
                                                                                    BAT(s)
Arsenic
                                                           1,2, 5, 6, 7,9
Key to BATs in Table
1 = Activated Alumina
2 = Coagulation/Filtration (not BAT for.
    systems < 500 service connections)
    5 = Ion Exchange
    6 = Lime Softening (not BAT for systems <
        500 service connections)
    7 = Reverse Osmosis
             9 = Electrodialysls
             *    *     *    . *     *

             [FR Doc. 00-13546 Filed 6-21-00; 8:45 am]
             BILLING CODE 6560-50-P

-------

-------

-------
 ~° o    ^-s
 2  E*    5r 05
 g  Q:    s... Q s.
•<  —    3
. -K C3D    
-------