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
------- |