Thursday
November 29, 1979
Part III
Environmental
Protection Agency
Nationaj Interim Primary Drinking Water
Regulations; Control of Trihalomethanes in
Drinking Water; Final Rule
-------
68624
Federal Register / Vol. 44, No. 231 / Thursday, November 29,1979 / Rules and Regulations
ENVIRONMENTAL PROTECTION
AGENCY
40 CFR Part 141
[FRU1312-2]
National Interim Primary Drinking
Water Regulations; Control of
Trihalomethanes in Drinking Water
AGENCY: Environmental Protection
Agency (EPA).
ACTION; Final Rule.
SUMMARY: This amendment to the
National Interim Primary Drinking
Water Regulations establishes a
Maximum Contaminant Level (MCL) of
0.10 mg/1 and associated monitoring and
reporting requirements for total
trihalomethanes (TTHMs), including
chloroform, that are introduced into
drinking water by the reaction of
naturally occurring substances with
chlorine in the course of water
treatment. The proposed requirement to
utilize granular activated carbon (GAG)
or equivalent technology in those public
water systems subject to significant
contamination by synthetic organic
chemicals has been separated from this
promulgation and will be reproposed for
additional public comment in the near
future.
EFFECTIVE DATES: For community water
systems serving 75,000 or more persons,
monitoring must begin 1 year following
promulgation and the effective date of
the MCL is 2 years following
promulgation. For community water
systems serving 10,000 to 75,000 persons,
monitoring must begin within 3 years
from the date of promulgation and the
effective date of the MCL is 4 years from
the date of promulgation. Effective
immediately, systems that plan to make
significant modifications to their
treatment processes for the purpose of
complying with the TTHM MCL are
required to seek and obtain State
approval of their treatment modification
plans.
FOR FURTHER INFORMATION CONTACT:
Joseph A. Cotruvo, Director, Criteria and
Standards Division, Office of Drinking
Water (WH-550), Environmental
Protection Agency, 401M Street, S.W.,
Washington, D.C. 20460. (202-472-5016).
SUPPLEMENTARY INFORMATION:
History of Rulemaking
On July 14,1976, EPA published an
Advance Notice of Proposed .
Rulemakin&IANPRM), entitled "Control
Options' for Organic Chemicals in
Drinking Water,* (41FR 28991 et seq.].-
The ANPRM summarized the many
facets of the issue of organic chemicals
in drinking water including the
legislative background, health effects
data, the state of available control
technology and costs. Advantages and
disadvantages of various regulatory and
non-regulatory options were examined,
and the ANPRM solicited comments and
information regarding the problem and
options presented. On February 9,1978,
the EPA published a proposed rule (43
FR 5756, et seq.) To amend the National
Interim Primary Drinking Water
Regulations to include an MCL and
associated monitoring and reporting
requirements for TTHMs. At the same
time, a requirement for the use of GAG
or equivalent technology was proposed
for application to those drinking water
sources subject to significant
contamination by synthetic organic
chemicals of industrial origin.
Subsequently, on July 6,1978, EPA
published a Supplemental Notice of
Proposed Rulemaking (43 FR 29135, et
seq.) soliciting comment on EPA's
reassessment of the economic impact
analysis for the proposal, providing
additional documentation in support of
the proposal, clarifying certain aspects
concerning the effects of organic
chemicals in drinking water, and
extending the public comment period
from July 31,1978, to September 1,1978.
The two Federal Register Notice
preambles and the supporting
documentation cited therein provided a
detailed discussion of EPA's rationale
for proposing controls on organic
chemicals in drinking water. The
subjects covered included: assessments
of the sources and occurrences of, and
human exposure to, THMs and other
organic chemicals in drinking water;
discussion of the toxicology and
epidemiology studies that relate to
possible human health risks; rationale
for the selection of the MCL for TTHMs
and associated requirements: and a
discussion of the control technology,
economic impact and air pollution and
energy impacts of the proposal. EPA's
analyses of these subjects have been
revised to incorporate information
gained during the public comment
period.
A total of 598 written comments were
received in response to the proposed
regulations of which 391 addressed the
subject of THMs. In a number of cases
the commenters confused the two
different regulations being proposed for
organic chemical control. For example,
some commenters incorrectly assumed
that GAG was proposed as the
requirement for control of THMs and
objected accordingly.
Public hearings were held between
March and July, 1978, in Miami, Florida;
New Orleans, Louisiana; Boston,
Massachusetts; Los Angeles, California
St. Louis, Missouri; Louisville, Kentucky;
Washington, D.C. and Dallas, Texas. A
total of 259 witnesses testified at the
public hearings, and of these, 157
commented on the proposed regulations
for THMs. Commenters included water
utilities, state and local officials, public
interest groups, federal health regulatory
and research agencies, engineering
consulting firms and individual citizens
and scientists. In addition, there were
496 communications from members of
Congress, and both the House and
Senate Appropriations Committees, and
the Council on Wage and Price Stability
offered comments on the proposed
regulations. The National Drinking
Water Advisory Council was also
consulted for their comments on the
regulations. A number of the comments
were duplicative, in that often the same
persons or organizations submitted both
written and oral comments and such
comments often induced inquiries from
members of Congress on the same
subject. EPA has thoroughly considered
all comments received in formulating the
final regulations. A detailed breakdown
of the comments and the Agency's
responses to them are attached as
Appendices.
Legal Authority
These final regulations are issued
under the authority of the Safe Drinking
Water Act. as amended (SOWA], 42
U.S.C. 300f et seq., specifically, sections
1401,1412,1445 and 1450. They
constitute amendments to the National
Interim Primary Drinking Water
Regulations (NIPDWR). 40 CFR Part 141,
as authorized by Section 1412(a)(l).
As noted in the preamble to the
proposed regulations (43 FR at 5759),
EPA considered establishing these
regulations as Revised Primary Drinking
Water Regulations but concluded that
they would be more appropriate as
amendments to the NIPDWR. This
means that the feasibility of control
measures under the NIPDWR must be
adjudged to have been available as of
December, '1974, when the SDWA was
enacted. As prescribed by Section
1412(a](2), these Interim Regulations
protect health to the extent feasible,
using technology, treatment techniques.
and other means which the
Administrator determines are generally
available (taking costs into
consideration) on the date of enactment,
(of the SDWA). |
Although Congress clearly
contemplated the comprehensive control
of organic chemical contaminants in the
-------
Federal Register / Vol. 44. No. 231 / Thursday. November 29, 1979 / Rules and Regulations 68625
Revised Regulations, the statute
nowhere precludes EPA from
establishing requirements as
amendments to the Interim regulations
even after the issuance of the report of
the National Academy of Sciences under
Section 1412(e). The statute does not
require that all regulations subsequent
to the NAS report be issued as Revised
Regulations. All that is required is that
the applicable statutory criteria be met.
Given Congress' early concern with the
presence of organic chemicals in
drinking water, the availability of
control measures to reduce the level of
TTHMs to 0.10 mg/1 since 1974, and
EPA's Finding that THMs "may have an
adverse effect on the health of persons,"
amending the Interim Regulations to
include these requirements as a first
step toward controlling organic chemical
contaminants in drinking water is
clearly authorized at this time.
n On February 10,1978, one day after
3 the publication of EPA's proposal in this
oo rulemaking in the Federal Register, the
''" United States Court of Appeals for the
^ District of Columbia Circuit issued its
.. opinion in Environmental Defense Fund
Cv. Castle. No. 75-2224, 578 F.2d 337. In
C.'that case, EOF sought more
^comprehensive control by EPA of
organic chemicals in the N1PDWR that
were promulgated in December 1975.
Following a review of the statutory
provisions and the legislative history
regarding the scope of the Interim
Regulations, the Court found that EPA
could exercise a degree of
administrative discretion in deciding
whether to control organic chemical
contaminants under the NIPDWR. The
Court also stated:
As we have indicated above, we believe
the legislature contemplated that the interim
regulations would, where feasible, control
every contaminant that may prove injurious
to health. The failure of the challenged
regulations to do so thus becomes suspect. In
light of the clear language of the legislative
history, the incomplete state of our
knowledge regarding the health effects of
certain contaminants and the imperfect
nature of the available measurement and
treatment techniques cannot serve as
justification for delay in controlling
contaminants that may be harmful. (578 F.2d
at 345).
The Court deferred final resolution of
the issue by remanding the record to
EPA for a report regarding "significant
changes that have occurred, since the
promulgation of the interim regulations.
in (EPA's) assessment of the problem of
controlling organic contaminants in
drinking water," and to advise the Court
"as to whether it plans to propose
amended interim regulations in light of
newly acquired data" (emphasis added)
(578 F.2d at 346). This evidenced the
Court's recognition that amendments to
the Interim Regulations were not
restricted to mere modifications to
existing requirements, as argued by one
commenter. Following EPA's submission
of its February 9,1978, proposed
regulations, the Court affirmed EPA's
earlier rulemaking action without
prejudice to the filing by EOF of a
petition to review any action or inaction
of the EPA concerning proposed
regulations dealing with organic
contaminants and without prejudice to
the Tiling by EDF of a motion to recall
the mandate should circumstances
warrant such action. (Court's order,
dated July 14,1978]. These final
regulations directly address the Court's
concerns as they were set forth in that
opinion.
Summary of the Regulations
Section 141.12 of the Interim
Regulations has been amended to add a
new maximum contaminant level of 0.10
mg/1 for TTHMs. TTHMs in § 141.2 are
defined as the arithmetic sum of the
concentrations of the THM compounds
(trichloromethane (chloroform),
dibromochlorometnane.
bromodichloromethane and
tribromomethane (bromoform)) rounded
to two significant figures. This MCL is
applicable to all community water
systems serving 10,000 or more persons
that add a disinfectant to their treatment
process. The effective dates of the MCL
are specified at § 141.6 as two years
from the date of promulgation for those
systems serving a population of 75,000
persons or more and four years from the
date of promulgation for those systems
serving a population of 10,000 to 75,000.
At this time, systems serving fewer than
10,000 persons are not covered by these
regulations unless States exercise their
discretion and expand their coverage to
these smallest systems.
Under new Section 141.30, systems
serving 75.000 or more persons are
required to begin monitoring within one
year from the date of promulgation of
this regulation and systems serving from
10,000 to 75,000 persons are required to
begin monitoring within three years
from the date of promulgation. No
monitoring is required for systems
serving fewer than 10,000 persons under
the federal regulations, but the States
may extend coverage at their discretion.
The minimum total number of samples
. required to be taken by the system is
required to be determined on a per plant
basis, with the exception that wells
drawing raw water from a single aquifer
may, with State approval, be considered
on treatment plant. Thus, if a system has
only one treatment plant, the minimum
number of samples is four samples per
quarter; if it has two treatment plants,
the minimum is eight samples per
quarter, if it has three treatment plants.
the minimum is twelve samples per
quarter. All samples taken at the
established frequency (e.g., quarterly.
annually) must be collected on the same
day.
Community water systems using
surface sources and systems using
ground water sources are, at a minimum,
required to monitor for TTHMs at
quarterly intervals, with a minimum of
four samples each quarter for each
treatment plant used by the system.
Each quarter, the system's sampling
scheme must insure that at least 25% of
the samples are taken at locations
within the distribution system reflecting
maximum residence time of the water in
the system, and that no more than 75%
of the samples are taken at other
representative locations within the
distribution system. In selecting
representative sampling locations for
TTHM monitoring, the regulations
provide that the system shall take into
account the number of persons served,
source of raw water and treatment
methods used. To the extent possible,
representative sampling for systems
with more than one treatment plant
should reflect the distributed water from
each plant separately.
Systems are further required to
average the results of all analyses
performed per quarter and to report the
results to the State, and to EPA if such
monitoring requirements have not yet
been adopted by the State with primary
enforcement responsibility. All samples
collected must be used in computing the
average, unless the analytical results are
invalidated for technical reasons by a
responsible official. Compliance will
then be determined based upon a
running annual average of the quarterly
samples.
The regulations also provide that this
sampling frequency of four samples for
TTHMs per quarter per year may be
reduced by the State to a minimum of
one sample for TTHMs per quarter per
year (for each plant used by the system)
if, after the system has monitored for at
least one full year in accordance with
the original schedule, it can demonstrate
to the State that the water it serves is
consistently below the TTHM MCL of
0.10 mg/1. This minimum single TTHM
sample must be taken at a point in the
distribution system that reflects
maximum residence time to insure
adequate protection. The system would
be required to immediately revert back
to the "four samples per quarter"
sampling frequency if the single TTHM
-------
68626 Federal Register / Vol. 44. No. 231 / Thursday. November 29, 1979 / Rules and Regulations
sample exceeds the standard and such
results have been confirmed by at least
one check sample, or in the event of any
significant change in its source of water
or treatment program. The system must
continue such program for at least one
year before it could be eligible for
reduced monitoring again. The
regulations also authorize the States
(and EPA. where the State does not
have primary enforcement
responsibility) to increase the
monitoring frequencies at their
discretion where such is deemed
necessary and appropriate to insure
consistent compliance with the MCL
throughout the distribution system.
Special consideration is given in the
regulations to community water systems
which draw their water exclusively from
groundwater sources by allowing them
to have their monitoring requirements
reduced by the State at the outset based
upon a judgment by the State that such
systems are not likely to be subject to
TTHM contamination. The regulations
require that such a system must
demonstrate to the satisfaction of the
State based on at least one sample for
each treatment plant used by the system
that it has a maximum total
trihalomethane potential (MTP) of less
than 0.10 mg/1. Thus, if the results from
at least one MTP sample are less than
0.10 mg/1 and after an examination of
local conditions, the State may reduce
the monitoring requirements of such a
ground water system to not less than
one sample for MTP per year.
"Maximum total trihalomethane
potential" is defined under new
§ 141.2(s). Any system using exclusively
groundwater sources whose MTP is
equal to or greater than 0.10 mg/1, which
results have been confirmed by a check
sample, must comply with the four
TTHM samples per quarter per year
requirement for at least one full year.
Thereafter, the monitoring may be
reduced by the State to one TTHM
sample per quarter if the TTHM levels
are consistently less than 0.10 mg/1. or
to one MTP sample per year if the MTP
is shown to be less than 0.10 mg/1.
Systems are required to report to the
State (and EPA until the State adopts
these regulations] the results of each
quarterly sampling within 30 days of
receipt of such results. Once the MCL
takes effect, public notification as well
as reporting to the State is required
whenever the running average of
quarterly samples during the previous 12
months indicates that the MCL of 0.10
mg/1 has been exceeded.
To ensure the continued
microbiological quality of the drinking
water as TTHM levels are being
reduced, water systems are required to
seek and receive State approval of their
plans to make significant modifications
to their treatment processes. State
approval shall be conditional upon
inclusion of additional monitoring and
other requirements prescribed by the
State to assure microbiological quality
in accordance with the guidance
provided by EPA. Finally, analyses must
be performed by approved laboratories
and in accordance with EPA specified
methods.
Trihalomethanes
As explained in the preamble to the
proposed regulations, the THMs found
in drinking water are members of the
family of organohalogen compounds
which are named as derivatives of
methane, where three of the four
hydrogen atoms have been replaced by
three atoms of chlorine, bromine or
iodine. Ten distinct compounds are
possible by various combinations of
three halogenated atoms, one hydrogen
and carbon atom. Current analytical
methodology applied to drinking has
thus far detected chloroform
(trichlorome thane),
bromodichloromethane,
dibromochloromethane, bromoform
(tribromomethane) and
dichloroiodomethane and monitoring
methods are currently available for the
brominated and chlorinated THMs but
not the iodinated THMs because of
chemical instability.
The principal source of chloroform
and other trihalomethanes in drinking
water is the chemical interaction of the
chlorine added for disinfection and
other purposes with the commonly
present natural humic and fulvic
, substances and other precursors. The
actual levels of TTHMs in drinking
water, however, will vary depending
upon the season, chlorine contact time,
water temperature, pH, type and
chemical composition of raw water and
treatment methodology. Since the
natural organic precursors are more
commonly found in surface waters,
water taken from a surface source is
more likely than ground water (with
notable exceptions) to produce high
THM levels.
Generally, the THM producing
reaction is as follows:
Chlorine + (Bromide ion or iodide ion) +
Precursors = Trihalomethanes and other
Halogenated Compounds
Chloroform is the most common THM
found in drinking water and it is also
usually present in the highest
concentration. In a number of cases, the
concentrations of the brominated THMs
were found to far exceed the chloroform
concentrations. The mixed THMs
appear to form by way of an initial
oxidation of bromide ion in solution by
added chlorine, followed by rapid
bromination of the organic precursors.
Bromine and chloroform may also be
introduced as contaminants of chlorine.
Chloroform and other THMs were first
reported in drinking water in late 1974.
EPA initiated the National Organics
Reconnaissance Survey (NORS) of 80
water utilities, which confirmed that
THMs were being formed during
chlorination in drinking water treatment
process. Concentrations in finished
water appeared to be roughly related to
the amounts of natural chemicals
present in the water.
In late 1975, EPA initiated the
National Organics Monitoring Survey
(NOMS) in 113 cities. The NOMS
demonstrated that considerable
amounts of THMs could form in the
water after it has entered the
distribution systems on the way to the
consumer's tap. It also showed that
THMs far exceeded the concentrations
of other synthetic organic contaminants
in finished drinking water, and that
brominated THMs could also exceed the
chloroform concentrations. Other
studies have shown that the TTHMs are
only a portion of the chlorinated
chemicals generated in water after
chlorination. Additional information is
contained in EPA's "Statement of Basis
and Purpose" accompanying this
regulation.
Review of Major Issues
During this rule-making, EPA
specifically solicited and received
comments on the following major issues:
The rationale for setting an MCL for
TTHMs and the magnitude of the MCL;
the feasibility of and timing for phased
reduction of the MCL; the concept of
phasing the application of the MCL
based upon system size; an alternative
of making the MCL applicable to all
public water systems and to phase the
implementation by a deferred
monitoring schedule linked to
population size; the method for
determining compliance, including the
number, frequency and location of
sampling sites and the averaging of
results; the availability of technology to
achieve compliance, and the need for
restrictions to assure that biologically
safe water would be maintained in the
course of achieving TTHM reduction;
and the costs incurred by public water
systems to achieve compliance with the
MCL
Magnitude and Rationale for the MCL
These final regulations adopt
unchanged EPA's proposed MCL for
-------
Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68627
TTHMs of 0.10 mg/1. The majority of
""immenters responding to this issue felt
at setting an MCL of 0.10 mg/1 for
rHMs lacked supporting justification,
jth in terms of establishment of the
need for a regulation'to protect public
health and also the numerical value that
was proposed, while others supported
the proposed MCL and some
recommended that a lower MCL be
selected. Many argued that an
unenforceable goal instead of an MCL
should be established, or that the MCL
should be higher than 0.10 mg/1.
The Coalition for Safe Drinking Water
(CSDW), a member organization of both
municipal and investor-owned water
utilities formed specifically to comment
on EPA's proposed regulations,
recommended that an MCL be
established only for chloroform and that
the MCL should be no lower than 0.3
mg/1. The CSDW presented a number of
witnesses at the various public hearings
and submitted voluminous written
comments on the THM regulation.
Among the arguments presented were:
That chloroform is not known to be a
human carcinogen; that other THMs are
not known to be animal carcinogens;
that the bioassy of chloroform
conducted by the National Cancer
stitute was flawed; that a threshold
irel could be established for
ircinogenic nsk; that the
epidemiological studies purporting to
indicate human risk were flawed or
misinterpreted; that the cancer risks
from chloroform could be considerably
lower than those computed using the
conservative linear or multi-stage
models. One (Roe) stated that
chloroform might be beneficial. EPA
evaluated the CSDW's comments but
found their arguments unpersuasive. A
detailed analysis of the CSDW's
comments is contained in EPA's
response to comments. Appendix A. A
summary of their specific comments is
presented in Appendix B.
Comments from the National Cancer
Institute (NCI), National Academy of
Sciences (NAS), the National Drinking
Water Advisory Council (NOWAC), the
National Institute of Environmental
Health Sciences (NIEHS) and federal
regulatory agencies such as the
Occupational Safety and Health
Administration (OSHA), Food and Drug
Administration (FDA) and the Consumer
Product Safely Commission (CPSC),
generally supported EPA's proposal. A
summary of their specific comments is
"resented in Appendix B. They stated
at sufficient scientific evidence had
•en accumulated to conclude that
..iloroform is an animal carcinogen as
shown from a properly conducted
bioassay and should be presumed to be
a risk to humans and that, as such,
prudent public health policy warrants
reasonable measures to reduce human
exposure. The NDWAC also specifically
concurred with the 0.10 mg/1 MCL
proposal for TTHM. The Environmental
Defense Fund (EOF) suggested that a
lower MCL would be feasible.
EPA's decision to regulate THM levels
in drinking water is based on a number
of factors which were extensively
discussed in the preambles to its
proposal notices of February 9 and July
6,1978. They include, in summary, the
potential human health risks of
chloroform and other THMs; the fact
that drinking water is the major source
of human exposure to THMs; the fact
that THMs are the most ubiquitous
synthetic organic chemicals found in
drinking water in the U.S. and are
generally found at the highest
concentrations of any such chemicals;
the fact that THMs are introduced in the
course of water treatment as by-
products of the chlorination process and
thus are readily controllable; that low
cost and feasible means have been
generally available since 1974 to reduce
their concentrations in drinking water
that monitoring is feasible; and that the
THMs are also indicative of the
presence of a host of other halogenated
and oxidized, potentially harmful by-
products of the chlorination process that
are concurrently formed in even larger
quantities but which cannot be readily
characterized chemically.
In concluding that exposure to THMs
in drinking water poses a human health
risk, EPA followed the four principles on
human risk assessment set forth in the
1077 report of the National Academy of
Sciences, "Drinking Water and Health,"
which EPA feels are representative of
the consensus of scientific opinion. As
stated in the proposal, they are as
follows:
1. Effects in animals, properly
qualified, are applicable to man.
2. Methods do not now exist to
establish a threshold for long-term
effects of toxic agents.
3. Exposure of experimental animals
to toxic agents in high doses is a
necessary and valid method of
discovering possible carcinogenic
hazards in man.
4. Material should be assessed in
terms of human risk, rather than as
"safe" or "unsafe."
In the specific case of chloroform and
other THMs, EPA has relied primarily
on animal studies demonstrating the
toxicology of chloroform. These are
described in the NAS report, "Drinking
Water and Health", and in the
"Statement of Basis and Purpose"
accompanying this regulation. The
bioassay results from studies conducted
by the NCI have demonstrated the
carcinogenicity of chloroform in both
rats and mice. Dr. Arthur Upton,
Director of NCI. concluded in his
comments that chloroform and other
chemicals have been "proven as
carcinogens in bioassays." Mechanisms
for the metabolism and toxicity of
chloroform are being investigated and
include information demonstrating
covalent binding of chloroform
metabolites to DNA and the probable
intermediate formation of phosgene as a
metabolite.
EPA has also concluded that the
available epidemiological evidence
relative to THM concentrations or other
drinking water quality factors and
cancer morbidity/mortality has not been
conclusive but is hypothesis generating
and at least suggestive of a health risk.
The NAS in its review of 13 preliminary
epidemiological studies affirmed EPA's
interpretation and concluded that the
risks were probably small but that
important confounding factors could not
be distinguished in indirect ecological
studies to allow a precise evaluation of
the contributions from THMs. They
pointed out the lack of sensitivity of
epidemiological procedures due to lack
of exposure data for individuals,
population diversity and mobility,
inability to control for all known
contributing variables such as smoking,
occupational exposures, diet, alcohol
consumption, socio-economic and
urbanization factors, and the usual 20-
40 year latency period required for most
cancers. The NAS also pointed out that
sufficient evidence was available from
animal toxicologystudies to conclude
that exposure to chloroform did pose a
risk to human health. Additional studies
are underway. Since epidemiology per
se cannot "prove" causality, and
because it may well be impossible to
epidemiologically establish a strong
causal association that THMs and
related chemicals in drinking water
contribute to higher cancer rates, EPA
has extrapolated from the results of
animal studies to assess the risk posed
by THMs to humans.
EPA has also concluded that it would
be Inappropriate at this time to
distinguish between an MCL for
chloroform and other THMs. As a family
of compounds, the THMs are similar in
chemical composition and nature and
are formed concurrently during the .
chlorination of drinking water.
Brominated THM levels greater than 0.6
mg/1 have been detected in some
drinking waters. Their relative
distribution in finished water is a
-------
68628 Federal Register / Vol. 44. No. 231 / Thursday. November 29, 1979 / Rules and Regulations
function of the organic and halide
precursor concentrations which can be
highly variable and unpredictable. The
other THMs are under further study in
the NCI bioassay program because of
human exposure and structural
similarity to chloroform. Mutagenicity
studies in Salmonella typhimurium
bacterial test systems have shown that
brominated and iodinated THMs are
more mutagenic than chloroform. The
gas chromatographic analytical method
concurrently analyses all four THMs.
and treatment methods that would be
employed would simultaneously reduce
all of the THMs.
Excluding brominated THMs from
these regulations would permit a
substantial number of communities with
low chloroform levels, but otherwise
high THM and other by-product
contamination, to avoid any
improvement of treatment practice and,
by implication, water quality.
Even though the toxicology of each of
the other THMs has not at this time
been as thoroughly studied by the
scientific community as chloroform, the
available lexicological information, their
structural similarities to chloroform, and
the fact that effective treatment is
generally available to reduce public
exposure to these potentially harmful
contaminants as well as for chloroform,
leads EPA to conclude that it would be
inappropriate to exclude them from
regulation.
Commenters had suggested that an
MCL of 0.30 mg/1 for chloroform could
be computed as a "safe" level for human
consumption by incorporating an
uncertainty factor of 2,000 into Roe's "no
observed effect dose." EPA has
concluded that such an approach is
totally inappropriate when dealing with
human risk from chronic exposure to a
potential carcinogen. That approach
assumes the existence of a threshold
level for carcinogens below which no
risk would exist. It is thus inconsistent
with the principles staled by the NAS in
"Drinking Water and Health." In
addition, 0.30 mg/1 is well above the
levels that are currently achievable in
the large majority of, public water
systems by generally available methods
that are technically and economically
feasible. The comment was rejected.
These comments and the Agency
responses are detailed in Appendix A.
Because of the technical inability to
determine a "safe" level for a
carcinogen and the conclusion,
therefore, that some risk must be
assumed at any dose, regulatory
agencies have attempted to minimize
human exposure to carcinogens to the
extent feasible. This approach was
endorsed in the comments received from
the National Cancer Institute, National
Institute of Environmental Health
Science, National Academy of Sciences,
Consumer Product Safety Commission,
National Institute of Environmental
Health Sciences. Food and Drug
Administration, Occupational Safety
and Health Administration, as well as
the National Drinking Water Advisory
Council. See Appendix B.
EPA's selection of an interim MCL of
0.10 mg/1 was based on a balancing of
public health considerations and the
feasibility of achieving such levels in
public water systems in the United
States. This balancing reflects the
existing and generally available
technology for water treatment which
relies heavily on the proven use of
chlorine to produce biologically safe
water. It includes the existence of
monitoring methods and trained
personnel, economic considerations, and
the limited amount of technical
assistance available from EPA and the
States, but primarily the risks that may
be introduced in some cases from
possibly inadvisable and improperly •
managed fundamental changes in
disinfection practice.
Thus, the interim MCL should not be
construed as an absolutely "safe" level,
but rather a feasible level achievable
with water treatment technology
available since 1974. The preponderance
of the current scientific thought on
human exposure to substances that have
been demonstrated lo be carcinogens in
animals in appropriate tests is that they
be considered potential carcinogenic
risks to humans. The presumptions are
that human health risk is related to the
extent of exposure and that no threshold
level without risk can be experimentally
demonstrated for a genetically diverse
population. Translated into regulatory
policy, exposure should be minimized so
as to minimize unnecessary risks.
Therefore, public water systems should
strive to reduce TTHMs and related
contaminant concentrations to levels as
low as is economically and
technologically feasible without
compromising protection against the
transmission of pathogenic
microorganisms via drinking water. •
The latest comprehensive information
on concentrations of TTHMs in the U.S.
drinking water was obtained from the
National Organics Monitoring Survey
(NOMS) of 113 communities sampled 3
times in 1975-77, This represented a
wide range of water types including
both surface and ground waters, and
waters with minimal and substantial
TTHM formation potentials. Mean levels
of TTHM for Phase II and Phase III were
0.12 mg/1 and 0.10 mg/1, respectively, in
samples allowed to react to completion
(terminal). Averages of both
dechlorinated and terminal samples
could be considered estimates of likely
concentrations to beHbund at the tap of
the average consumer. These were 0.09
mg/1 and O.OB mg/1, respectively, in
Phase II and Phase III. However,
maximum TTHM levels ranged as high
as 0.70 mg/1 and 0.78 mg/1 in terminal
samples. Therefore, an interim MCL of
0.10 mg/1 will result in substantial
reductions of TTHM concentrations in
many water systems now exceeding the
MCL.
Many commenters conceded that
TTHMs were undesirable constituents
of drinking waters, but preferred that a
goal rather than an enforceable MCL
should be established. In other words, it
was suggested that compliance with a
TTHM limit should be optional.
However, neither the SDWA nor the
facts at hand support such a course of
action at this time. The SDWA provides
for goals only in the case of the
Administrator's list of recommended
MCLs (Section 1412(b)(l)(B]). and. even •
then, the goal is to be selected as the
value that would result in no known or
anticipated adverse health effects and
would allow an adequate margin of
safety. Revised regulations must specify
MCLs that come as close to the
recommended levels as is feasible using
the best technology, treatment
techniques and other means which the
Administrator Finds are generally
available (taking costs into
consideration) (section 1412(b)(3)).
The SDWA clearly requires that EPA
take regulatory action by establishing
enforceable standards, not merely
health goals. Since the issuance of EPA a
ANPRM and proposal in this
rulemaking, only a limited number of
systems have voluntarily reduced the
levels of TTHMs in their water supplies.
Only in the presence of a mandatory
requirement can EPA expect the full
commitment in time and resources by
community water systems and the
oversight by State regulatory agencies
necessary to achieve compliance
nationally.
MCL Summary
Thus, based on the foregoing
considerations set forth in the
rulemaking record, the Administrator
believes that an MCL for TTHMs of 0.10
mg/1 in the Interim Regulations will
protect human health to the extent
feasible as prescribed by Section
1412(a](2) of the SDWA. Since the
optimum and only totally "safe" dose fo|
any carcinogen would be zero. EPA
strongly encourages all public water
systems, not only those that exceed the
-------
Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68629
interim MCL, to implement measures to
minimize the amounts of TTHMs and
slated by-products in finished water.
THM levels in finished water are a
mction of the raw water quality
(precursor content) and the sequence of
treatments applied Based upon the
performance of developing technologies,
it appears that ulitmately many public
water supplies with currently high
TTHM levels may be able to achieve
TTHM concentrations as low as 0.010 to
0.025 mg/1 and EPA suggests those
values as future goals. The MCL will be
reconsidered in the Revised National
Primary Drinking Water Regulations
based upon an updated assessment of
technological and economic feasibility,
implementation experience and
additional toxicological information.
Population Coverage and Phase-In of
the MCL and Monitoring Requirements
The proposed regulations would have
initially applied the MCL only to those
community water systems serving 75,000
or more people, and would have only
required that monitoring data be
collected for one year in communities
serving between 10,000 and 75,000
people. Systems smaller than 10,000
would not be initially covered. The
oposed effective date of the MCL was
I months after promulgation.
EPA solicited comments on
tentative approaches for coverage and
implementation, for example by
applying the MCL to all systems and
phasing-in implementation through a
deferred monitoring schedule (i.e.,
systems larger than 75,000 required to
begin monitoring within one year of
promulgation, 10,000-75,000 within three
years of promulgation, and all other
communities within five years).
The majority of commenters felt that
the regulations should not be limited to
the larger than 75,000 population
community water systems, although
some agreed that some phasing
mechanism would be appropriate. The
NOW AC suggested that utilities serving
10,000 to 75,000 should be included
beginning three years after
implementation of regulations in the
larger than 75,000 group. The NDWAC
also recommended in its initial
comments that implementation in
communities smaller than 10,000 should
be at the option of the State.
EPA has concluded thai the coverage
of these regulations should be expanded
to include community water systems
~rving 10,000 or more persons. Systems
rving 75,000 or more people are
quired to comply within two years of
omulgation, and systems serving
between 10,000 and 75,000 are required
to comply within four years of
promulgation.
This still means (hat systems serving
fewer than 10,000 people are not
required to comply with the TTHM
MCL However, EPA does not believe
that this approach will result in those
persons served by the smallest systems
being afforded reduced health
protection. This is because the great
majority (about 60%) of these smallest
systems are served by ground water
sources that are low in THM precursor
content. The proportion of small
community water systems that utilize
chlorine is less than that of large
systems and transport time within the
distribution system, which increases the
extent of TTHM formation, is generally
shorter in small systems. Therefore,
their drinking water is less likely to be
subject to TTHM contamination.
Moreover, the smallest systems incur
a greater risk of adversely affecting the
microbiological quality of their drinking
water when steps are taken to reduce
TTHMs. The majority of waterborne
disease outbreaks attributable to
inadequate treatment practice still occur
in the smallest systems. Such systems
also have limited or no access to the
resources and professional expertise
needed for TTHM control. Thus, EPA
believes that it would be premature to
divert their already sparse resources
away from improving their disinfection
practices by requiring compliance with a
TTHM MCL at this time.
It is imperative that any changes in
current treatment practice must be
carefully supervised and supported by
technical assistance from the States or
EPA. However, it is not administratively
feasible for the States and EPA to
adequately supervise the approximately
57,000 systems which each serves
communities of fewer than 10,000
people.
The approximately 60,000 community
water systems in the U.S. range in size
from 25 persons to several million and
serve a total of about 213 million people.
The 390 systems exceeding 75,000
population serve about 101 million
people, and the 2,300 systems between
10,000 and 75,000 serve an additional 66
million people. Thus, the final
regulations cover approximately 80% of
the U.S. population served by
community water systems. Most of these
larger systems have at least potential
access to the technical personnel
needed to safely and successfully carry
out any fundamental changes in
disinfection practice. The smallest
systems serve only 20% of the
population but comprise a sufficiently
large number of systems to make careful
supervision effectively impossible in the
short-term. Nevertheless, EPA does not
intend that these smallest systems be
excluded from coverage of the TTHM
regulations indefinitely.
EPA considered specifying monitoring
requirements for these smallest systems
and/or making the MCL applicable to
such systems with an extended
timeframe for compliance. However,
considerable additional time would
have been necessary to insure
availability of laboratory capability to
handle the increased number of TTHM
analyses and adequate Slate and EPA
technical assistance. Therefore, it did
not seem prudent to specify
requirements now for which compliance
would be required so far in the future.
The considerable experience that will be
gained from the efforts of the larger
systems to comply with the TTHM MCL
will serve to make compliance by the
smaller systems more feasible. For that
reason, EPA expects that small systems
will be subject to a TTHM MCL under
the Revised Primary Drinking Water
Regulations when they are established.
In those States which choose to exercise
their discretion to extend coverage to
the small systems, EPA expects that
additional phasing may be appropriate
within this size category based on
greatest likelihood of TTHM
contamination, such as by first including
those systems with surface water
supplies.
Implementation Timing
The majority of commenters on the
question of the timing of the effective
date of the MCL felt that IB months after
promulgation was inadequate to allow
for design and implementation of the
most cost-effective treatment system for
compliance. They stated that eighteen
months would only be adequate if minor
modifications were needed. EPA has
reevaluated the treatment methods most
likely to be used and has concluded that
in most cases relatively minor technical
modifications will be sufficient to
substantially reduce TTHM levels below
the MCL. Therefore, a delay in the
effective date would not have been
justified on this ground.
Other commenters pointed out that
insufficient laboratories were available
to analyze TTHM samples and that a
quality assurance program would need
to be developed; some suggested that
monitoring should be delayed for those
reasons. EPA agrees with those
commenters concerned about the
availability of sufficient numbers of
laboratories capable of providing
acceptable analytical data. At this time,
only relatively few laboratories have
demonstrated the capability of
-------
68630 Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations
consistently producing data with the
required accuracy and precision.
EPA has, therefore, decided to extend
the time frame for initiation of the
monitoring requirement for systems
serving 75,000 or more persons from the
proposed three months after
promulgation to one year after
promulgation. This will allow additional
time for State and private laboratories
to develop their capabilities and to
become certified to provide data in
support of compliance determinations.
Since the effective date for initiation of
monitoring is one year after
promulgation and one year of
monitoring results is required to
determine compliance, the effective date
of the MCL for those systems is
established as 2 years after
promulgation. To accommodate the
large incremental monitoring load,
application of the monitoring
requirements to the approximately 2,300
systems serving 10.000 to 75.000 persons
is established at 3 years following
promulgation and the effective date of
the MCL in this population range is 4
years after promulgation. Despite these
extended deadlines, EPA encourages
water systems to initiate monitoring and
corrective measures sooner than this
schedule whenever it is feasible to do
so, especially where high TTHM levels
are suspected.
EPA will immediately initiate an
interim certification program for State
laboratories (and others if appropriate)
that will be based on their ability to
analyze Performance Evaluation
samples which will be provided by
EPA's Environmental Monitoring and
Support Laboratory (EMSL). Two
analytical methods (Purge and Trap and
Liquid-Liquid Extraction) have been
approved under § 141.30(e) of the
regulations and the written procedures
are available on request from EPA's
EMSL. 26 W. St. Clair Street, Cincinnati.
Ohio 45268.
To qualify for Interim Certification,
laboratories will be required to
demonstrate their ability to analyze the
Performance Evaluation samples
provided to them to within 20% of the
"true value" for each of the THMs as
well as for the total of the THMs in the
samples, using at least one of the
approved methods. As the certification
program develops and more laboratories
gain expertise, it is likely that the
precision and accuracy requirements
will become more stringent. A quality
assurance program will be established
to insure that continued certification is
dependent upon the laboratories'
continued ability to perform quality
analyses.
State Primacy and Exemptions
The time frame of these amendments
to the NIPDWR will significantly affect
two other statutory provisions of the
SOW A: continuation of State primary
enforcement responsibility (or primacy)
under Section 1413 and the issuance of
exemptions from MCLs under Section
1413.
With respect to State primacy, the
Agency will shortly be proposing
amendments to its State implementation
regulations, 40 CFR Part 142, which will
provide primacy States adequate time to
amend their regulations without
jeopardizing primacy while more
stringent federal regulations take effect.
States are encouraged to begin the
process of amending their regulations as
quickly as possible. However, no action
to withdraw primacy will be taken
pending the establishment of new EPA
regulations under Part 142.
Under Section 14l6(b)(2](B) of the
SDWA, schedules attendant to
exemptions from the NIPDWR must
require compliance by no later than
January 1,1981 (or January 1,1983, for
systems that enter into enforceable
agreements to become part of a regional
water system). This will, in most cases,
preclude the issuance of exemptions
from the requirements promulgated
today. Since the issuance of exemptions
is discretionary with the State, or EPA
where the State does not have primary
enforcement responsibility, the
unavailability of exemptions perse is
not believed to be a fatal deficiency in
the regulations. Nevertheless, EPA
recognizes that some systems may not
achieve compliance by the effective
dates despite their best efforts. EPA is
planning to seek from Congress an
extension of the exemption deadlines as
they may apply to these regulations
when the Agency's implementation of
the Act is the subject of oversight
hearings. The States and EPA may also
exercise their enforcement discretion in
those cases where compliance with the
MCL for TTHMs is not achieved before
the applicable effective date despite the
system's good faith efforts to comply.
Summary
Therefore, EPA has accepted the
recommendation of the NDWAC and
many other commenters to broaden the
coverage of the THM regulations and to
phase-in its implementation as follows:
Water systems serving more than
75,000 are required to be in compliance
by two years from the date of
promulgation cf these regulations.
Systems serving between 10,000 and
75,000 are required to be in compliance
by four years from the dale of
promulgation.
Monitoring must be initiated no later
than one year from the promulgation
date by those water systems 75,000 or
larger, and three years from
promulgation by those systems in (he
10,000 to 75,000 population range.
However, EPA urges that compliance
and monitoring be accelerated in those
water systems where this is feasible and
where assistance is available from the
primacy authority, especially where high
TTHM levels are suspected.
Compliance with the MCL and
monitoring in communities smaller than
10,000 would only be required if the
primacy State adopts regulations that
are more expansive than these federal
regulations. EPA will consider
expanding the coverage of THM
regulations to include smaller systems
when it establishes Revised Primary
Drinking Water Regulations.
Monitoring Requirements
The proposed monitoring
requirements for systems exceeding
75,000 population included quarterly
sampling consisting of at least five
water samples collected on the same
day. The sampling locations were to be
representative of TTHM concentrations
at the consumer's tap; no more than 20%
to be collected at the entry point of the
distribution system, no less than 20% at
the extremes of the system and the
remaining 60% representative of
population density throughout the
distribution system. Compliance would
be determined by averaging the
quarterly values from the preceding 12
months. Surveillance monitoring only for
one year was proposed for systems
between 10,000 and 75,000 population.
This consisted of two samples per
quarter to be collected at the entry to
the distribution system. One sample
would be dechlorinated and the other
stored for seven days to permit
completion of the chlorine lion reaction.
These final regulations eliminate any
distinction (except for timing) between
the largest and medium size systems
and modify the requirement somewhat.
The majority of the comments on this
issue were in agreement with the
concept of determining compliance by
an annual average of quarterly samples.
Others disagreed, arguing that averaging
might mask fluctuations, and some felt
that averaging results in the distribution
system would result in higher exposures
to those populations residing in (he
extremes of the system. A few fell that
the extreme values rather than averages
should be used to compute compliance.
Some commenters suggested that
systems using deep ground water should
-------
Federal Register / Vol. 44. No. 231 / Thursday. November 29. 1979 / Rules and Regulations 60631
be exempted because of probable low
THM formation potential. Others
disagreed with a continued monitoring
requirement, even at a reduced
frequency, after it had been established
that TTHM concentrations were
unlikely to approach or exceed the MCL.
A number agreed with monitoring
requirements but objected to public
notification of results.
The intent of the monitoring
requirements is to provide a reasonable
representation of the normal
concentrations of TTHMs and related
chemicals at the tap of the typical
consumer. Data has shown that there
can be wide variation of TTHM
concentrations particularly in surface
waters and groundwaters with high
precursor levels on a day to day basis
and that levels at various points in a
distribution system can differ markedly.
The variations can be due to a number
of factors that include seasonal or other
changes in precursor concentrations in
the raw water, the amount of
precipitation and surface run-off, the
treatment method, the presence of
combined or free residual chlorine.
chlorine contact time, pH, temperature
and transit time during distribution.
EPA feels that it would be
unreasonable at this time to demand the
kind of pinpoint control that would be
necessary to maintain TTHM levels
below a particular figure at all times and
at all locations in the distribution
system of every water system. This
Interim Regulation is intended to reduce
the extremes of TTHM concentrations
that have been found in some of the
nation's public water systems, and thus,
to reduce the variability that may occur
within a given distribution system.
TTHMs in drinking water do not present
acute or short-term risks but rather
chronic or lifetime risks that increase
with long-term exposure. Therefore
some variations are tolerable and
probably do not contribute to a change
in overall risk. Thus, EPA has concluded
that an averaging approach is
appropriate and the use of a 12 month
running average for computing
compliance is retained in the
regulations.
The frequency of monitoring must be
based upon its usefulness for
determining the concentrations of
TTHMs in finished water. It should also
reflect the potential for variability of the
contaminant concentration, and this is
highly dependent upon site-specific
factors such as distance from the
treatment plant, source water quality
and treatment methods used. These
factors are particularly important in
selecting sampling locations which will
be truly representative of water served
to consumers regardless of their location
within the distribution system,
especially when a system uses more
than one treatment plant.
The consensus of the comments was
that quarterly monitoring was adequate
in most cases but many argued for more
samples. Quarterly monitoring has been
retained in the regulation because EPA
considers this to be the minimum
acceptable frequency in those places
where the water has a potential for
seasonal variability in TTHM levels.
EPA strongly urges that States review
each water system's monitoring program
to insure that the monitoring is reflective
of seasonal and other variation factors.
More frequent monitoring should be
required where this is necessary for
adequate consistent year-round control
of TTHM levels below the MCL Such
discretion to require more frequent
monitoring is provided for in these
regulations.
In further response to those comments
encouraging more frequent monitoring to
reflect variations of water quality in the
distribution system, EPA agrees that
some conditions lead to a greater
potential for wide variations of TTHM
levels. For example, if a community
water system uses more than one
treatment plant to provide water,
different water sources may be used as
well as different treatment processes,
leading to the possibility of widely
differing TTHM levels in parts of the
distribution system. For this reason, the
proposed sampling scheme has been
changed to increase the weighting of
distribution system samples. Samples
taken at the entry point to the
distribution system can no longer be
included in the quarterly or annual
averages. No less than 25% of the
samples shall be collected at locations
within the distribution system reflecting
maximum residence time of the water in
the system and no more than 75% from
representative locations within the
distribution system taking into account
number of people served, source of
water and treatment methods used.
Thus, the required number of samples is
reduced by 20% yet the results should be
more representative of tap levels
throughout the system, because the
deleted entry point sample would not
have reflected TTHM levels for a
substantial portion of the population
served. Of course, these compliance
monitoring requirements do not preclude
water systems from utilizing plant
samplings for process control.
Moreover, a minimum of four
compliance samples is required each
quarter for each treatment plant used by
the system, except that wells drawing
raw water from a single aquifer may.
with State approval, be considered one
treatment plant for the purpose of
determining the minimum number of
samples required to be taken by the
system. By determining the minimum
number of samples per system based
upon the number of separate treatment
plants used by the system, sampling
locations should be selected to reflect
water quality in identifiable portions of
the distribution systems associated with
each plant to the extent possible. Larger
systems are those most likely to have
more than one treatment plant, and
therefore more samples are both
desirable in insuring consistent water
quality throughout the distribution
system and not likely to significantly
increase the per capita cost of
monitoring. However, it would not be
reasonable to increase the number of
samples to be taken proportionate to the
number of wells drawn from a single
aquifer even though each well might
literally be considered a single
treatment plant; water quality is likely
to be consistent throughout the aquifer
and many systems have a large number
of wells. Therefore, with State approval,
wells drawing raw water from a single
aquifer may be deemed to be a single
treatment plant for purposes of
determining the minimum number of
samples required to be taken by the
system. The regulations do not provide
for similar flexibility for systems
drawing water from a single surface
source due to the likelihood of much
greater variability in raw water quality
and treatment methods at different
plants.
The sampling locations are important
because TTHM levels will likely be
higher in those parts of the distribution
system where residence time of the
water is longest, which is served by
surface water sources, and where
chlorination. as opposed to other
disinfection practices, is used. Even
though the samples will be averaged for
determining compliance with the MCL,
EPA expects that sampling will be
conducted in such a way so as to insure
that all parts of the distribution system
are serving water to consumers in
compliance with the MCL. Thus, where
a system draws its raw water from
multiple sources, or has more than one
treatment plant utilizing different
treatment methods, high THM levels in
specific parts of the distribution system
should be identified where possible, and
such levels reduced to the extent
feasible. EPA intends to address more
comprehensively the problems of
systems with multiple source waters and
-------
68632 Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations
multiple plants with differing treatment
programs, when it proposes Revised
Primary Drinking Water Regulations in
the future.
EPA also recognizes that there are a
number of public water systems, such as
those utilizing ground waters and some
surface water supplies, where, because
of the consistent quality of the source
water and the treatment method
employed, the probability that finished
water would approach or exceed the
MCL is remote. After a satisfactory
record has been established, through
one year of monitoring at a frequency of
four TTHM samples per quarter, a water
system may request that the State allow
a reduction of the monitoring frequency.
Upon the State's examination of at least
one year of compliance data and a
finding by the State that local conditions
are such that TTHM concentrations are
consistently below the maximum
contaminant level, the system's
monitoring frequency may be reduced to
a minimum of one TTHM sample per
quarter taken at a point in the
distribution system that reflects the
maximum residence time of the water
served. Should the system experience a
significant change in either its source of
water or its treatment program, it must
immediately reinstitute the four samples
per quarter monitoring program initially
required and continue on that program
for at least another year before its
sampling frequency could be reduced
again BO that the data baseline can be
re-established. The original sampling
requirements must also be reinstated
immediately if the results from any
analysis for TTHMs are found to exceed
0.10 mg/1 and such results are confirmed
by at least one check sample taken
promptly after the results of the first
analysis are received.
The State's decision to reduce a
system's monitoring frequency must be
made on a case-by-case basis taking
into account such factors as the
monitoring data, the quality and
stability of the source of raw water, low
total organic carbon (TOG) values, low
maximum TTHM potential (MTP) during
the time period when THM formation
would most likely be at a maximum and
the type of treatment employed. Except
in certain ground water cases,
monitoring cannot be reduced to less
than one TTHM sample per quarter.
This minimum monitoring is deemed
necessary and is sufficient to
demonstrate that conditions have not
changed to the extent that the MCL
might be exceeded. Intermittent use of
another water source may also require
additional monitoring at the discretion
of the State. This flexibility is included
in the regulations to allow States to
modify the generally applicable
monitoring requirements where
appropriate only on a case-by-case
basis to insure adequate public health
protection. Figure 1 presents the basic
steps to be followed by those systems
(other than special ground water cases
discussed below) that seek State
approval to have their monitoring
requirements reduced from four samples
to one sample of TTHMs per quarter per
year. "Maximum total trihalomethane
potential (MTP)" is defined as the
maximum concentration of TTHMs
produced in a given water containing
excess free chlorine after seven days at
a temperature of 25° C. Determination of
maximum TTHM potential should not be
confused with measurement of terminal
TTHM concentrations. The latter is
measured under the ambient conditions
of the distribution system with regard to
temperature and storage time.
BILLING CODE 6560-01-M
-------
Federal Register / Vol. 44. No. 231 / Thursday. November 29.1979 / Rules and Regulations 68633
FIGURE 1
CONSIDERATIONS FOR REDUCED MONITORING REQUIREMENTS
SURFACE WATER SYSTEMS
THE MINIMUM MONITORING REQUIREMENT IS FOUR SAMPLES PER
QUARTER PER PLANT. REDUCED MONITORING REQUIREMENTS MAY BE
APPROPRIATE IN CERTAIN CASES; UPON WRITTEN REQUEST FROM THE
PUBLIC WATER SYSTEM, STATES MAY REDUCE THE REQUIREMENTS
THROUGH CONSIDERATION OF APPROPRIATE DATA AS FOLLOWS:
SURFACE WATER SYSTEM
I
4 SAMPLES PER QUARTER
FOR TTHM
I
ONE YEAR OF DATA:
TTHM CONSISTENTLY
BELOW 0.10 MG/L
I
| NO
CONTINUE 4
SAMPLES PER QUARTER
YES
I
STATE JUDGMENT ON
REDUCED MONITORING*
MINIMUM: 1 SAMPLE PER
QUARTER FOR TTHM
TTHM > 0.10 MG/L
'FACTORS FOR CONSIDERATION:
• MONITOR ING DATA, MTP, TTHM, TOC
• QUALITY AND STABILITY OF SOURCE WATER
• TYPE OF TREATMENT
BILLING CODE 6560-01-C
-------
68684 Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations
Ground Water Sources
As several commenters suggested and
EPA agrees, many, if not most, ground
waters contain such small amounts of
precursor organic compounds (as
demonstrated by low total organic
carbon levels and low measured
maximum TTHM potential) and are so
stable, as to virtually preclude the
possibility of generating TTHM levels
approaching or exceeding 0.10 mg/1 even
when free chlorine is employed as a
disinfectant. For this reason, the
regulations provide that the monitoring
frequency applicable to systems using
exclusively ground water sources may
be reduced at the outset so that they
may be relieved from the more rigorous
monitoring program of four samples, or
even one sample, per quarter per year
which is applicable to systems using
surface water sources in whole or in
part.
Thus, a system that draws its water
exclusively from ground water sources
may have its monitoring requirements
reduced by the State if the results from a
single sample taken at a point in the
distribution system reflecting maximum
residence time of (he water in the
system and analyzed for maximum
TTHM potential (MTP) are less than 0.10
mg/1 and the State determines in writing
that, based on an examination of the
local conditions, the system is not likely
to approach or exceed the TTHM MCL
The State is expected to consider such
factors as monitoring data, the quality
and stability of the system's raw water
source, low TOC values, low maximum
TTHM potential during the time period
when THM formation would most likely
be at a maximum and the type of
treatment employed. Such sampling
frequency cannot be reduced lo less
than one sample for MTP per year. If
such a system experiences a significant
change in its source of water or
treatment program, it must immediately
take an additional sample for MTP
analysis to determine whether it should
be authorized to continue on the
reduced monitoring program following
the change. If the MTP is ever greater
than 0.10 mg/1 and such results are
confirmed by a check sample taken
promptly after the results of the original
sample are received, the system must
immediately begin taking and analyzing
four samples per quarter per year for
one full year. The year's results would
then be averaged for determining
whether (he system was in compliance
with the TTHM MCL "Maximum total
trihalomethane potential" is defined in
the regulations at new § 141.2(s).
Figure 2 presents the basic steps to be
followed by those systems using
exclusively ground water sources that
seek to have their monitoring frequency
reduced at the outset to one sample
analyzed for MTP per year, as opposed
to the four samples for TTHMs per
quarter per year otherwise applicable.
BILLING CODE «5SO-Ot-*l
-------
Federal Register / Vol. 44, No. 231 / Thursday. November 29.1979 / Rules and Regulations 68635
FIGURE 2
CONSIDERATIONS FOR REDUCED MONITORING REQUIREMENTS
GROUNDWATER SYSTEMS
THE MINIMUM MONITORING REQUIREMENT IS FOUR SAMPLES PER
QUARTER PER PLANT; SYSTEMS USING MULTIPLE WELLS DRAWING RAW
WATER FROM A SINGLE AQUIFER MAY WITH STATE APPROVAL BE
CONSIDERED AS ONE TREATMENT PLANT. REDUCED MONITORING
REQUIREMENTS MAY BE APPROPRIATE IN CERTAIN CASES; UPON
WRITTEN REQUEST FROM THE PUBLIC WATER SYSTEM. STATES MAY
REDUCE THE REQUIREMENTS THROUGH CONSIDERATION OF APPROPRIATE
DATA AS FOLLOWS:
GROUNDWATER SYSTEM
CHANC
TREAT
OR SOI
SAMPLE FOR MTP
1
MTP <0.10MG/L
1
»| MTP >0.10 MG/L
•E IN STATE JUDGMENT ON
Mr-NT - REDUCED MONITORING*
JRCE MJ
MIMUM: 1 SAMPLE
R YEAR FOR MTP
i
i
4 SAMPLES PER QUARTER
FOR TTHM
ONE YEAR OF DATA:
TTHM CONSISTENTLY
BELOW 0.10 MG/L
NO
CONTINUE 4
SAMPLES PER QUARTER
CHANGE IN
TREATMENT
OR SOURCE
STATE JUDGMENT ON
REDUCED MONITORING*
MINIMUM: 1 SAMPLE PER
QUARTER FOR TTHM •
TTHM >0.10MG/L
"FACTORS FOR CONSIDERATION:
• MONITORING DATA. MTP. TTHM, TOC
•QUALITY AND STABILITY OF SOURCE WATER
• TYPE OF TREATMENT
BILUNO CODE 6560-01-C
-------
Federal Register / Vol. 44, No. 231 / Thursday. November 29. 1979 / Rules and Regulations
Technical Feasibility of TTHM
Reduction
In establishing an MCL for TTHMs,
EPA is not required to specify any
particular method to achieve that
standard. However, in establishing
Interim Regulations. EPA must Find that
technology was generally available in
1974 to achieve the MCL. Thus, the
preamble to the proposal did discuss a
number of approaches that could be
utilized to achieve the MCL depending
on the individual circumstances. The
"Interim Treatment Guide for the
Control of Chloroform and Other
Trihalomethanes" was also published
and made available to commenters to
provide information on successful
techniques that should be considered. It
is incorporated by reference as part of
the Statement of Basis and Purpose for
these regulations.
Three general alternatives have been
presented:
(1) Use of a disinfectant (oxidant) that
does not generate (or produces less)
THMs in water:
(2) Treatment to reduce precursor
concentrations prior to chlorination; and
(3) Treatment to remove THMs after
formation. Many possible choices exist
within each category. For example,
alternate disinfectants or oxidants that
might be considered include ozone,
chlorine dioxide, and chloramines
(combined chlorine]. Precursor reduction
processes include off-line raw water
storage, aeration, improved coagulation,
ion exchange resins, granular activated
carbon (GAG), powdered activated
carbon (PAG), and ozone enhanced
biological activated carbon [BAG].
TTHM reduction has also been achieved
by merely moving the chlorine addition
point to later stages in the conventional
treatment process, and by substituting
prechlorination with some other
preoxidation process. TTHM removal
processes include GAG, aeration or
macroreticular resins. A combination of
these methods may be necessary to
comply with the TTHM MCL.
Few comments discussed the
feasibility of the available treatments,
and three suggested that additional
research should be performed on the
subject EPA has concluded that many
methods have been shown to be
effective for meeting the 0.10 mg/l MCL
for TTHMs and it remains only for the
individual water systems to select the
one or more procedures that are optimal
for their particular water characteristics.
Which treatment method (or
combination of treatment methods) is
ultimately selected by a water supplier
to achieve compliance with the MCL
must be based upon a case-by-case
assessment of the system's entire
treatment process, and an evaluation of
the precursor content of its raw water
source and TTHM formation potential
as well as the need to assure optimal
biological quality of drinking water
derived from contaminated sources.
In determining what technologies
were "generally available" in 1974 for
achieving the standard, EPA has taken
cost into consideration. The legislative
history of the SDWA clearly requires
that the reasonableness of costs must be
based on "what may reasonably be
afforded by large metropolitan or
regional public water systems" (House
Report No. 93-1185, p. IB). Moreover, the
Administrator must assume that most
intake waters are sufficiently
uncontaminated so that the MCLs can
be met with the application of those
technologies found to be "generally
available" at reasonable cost in 1974
(House Report No. 93-1185, p. 13).
EPA has estimated the costs of
various treatment methods available in
1974 to achieve compliance with the
TTHM MCL of 0.10 mg/l. They appear in
the report prepared for EPA by Gulp/
Wesner/Culp entitled, "Estimating Costs
for Water Treatment As a Function of
Size and Treatment Efficiency" and
EPA's "Interim Treatment Guide for the
Control of Chloroform and other
Trihalomethanes." The cost assumptions
in those documents in large part serve
as the basis for EPA's Economic Impact
Analysis for these regulations. These
documents are incorporated by
reference as part of the Agency's
Statement of Basis and Purpose for
these regulations.
Based on these documents, EPA has
concluded thai the use of any of the
alternative disinfectants discussed
above has been clearly available at
reasonable costs since 1974 to any large
public water system to achieve the MCL
of 0.10 mg/l. Alternatives, such as
changing the point of disinfection, off-
line raw water storage and improved
coagulation are also relatively
inexpensive and are also found to be
"generally available" at reasonable
costs.
With respect to the use of adsorbants,
the reasonableness of costs will be
dependent upon the particular
operational parameters that are
employed. For purposes of establishing
these regulations, EPA assessed the
costs that would be incurred by systems
utilizing GAG as a replacement for their
existing filter media, with a regeneration
frequency of one year. Although most
systems are expected to select the less
expensive treatment methods where
they are effective in achieving
compliance with (he MCL, the use of
GAG under these operating conditions
has also been found to be "generally
available" at reaapnable cost since 19741
for achieving the standard. Systems witr
very high raw water TOG may need to
use GAG with more stringent operating
parameters or additional treatment
methods to achieve the MCL. For this
reason, EPA has also assessed the cost
of using biological activated carbon
(ozone plus GAG) with a regeneration
frequency for the carbon of two years;
this cost has also been found to be
reasonable.
Disinfectant Restrictions and the
Standard Plate Count
Restrictions were proposed on the
excessive use of chlorine dioxide
because of possible by-product chlorite
toxicity, and also on misuse of
chloramines because of their low-
potency as disinfectants compared to
free chlorine. The proposal also
admonished those considering
modifications to their treatment process
to reduce TTHMs that any such
modification must not in any way affect
the microbiological quality of drinking
water so as to increase the possibility of
transmission of infectious disease. Also,
EPA espoused the fundamental principle
that water treatment should aim at
producing water of high quality and low
chemical content prior to application of
the oxidant, so as to maintain pathogen
control while minimizing oxidant use
and by-product demand.
Because of possible adverse effects on
finished water quality from ill-advised
treatment modification, the following
three conditions were specifically
proposed to apply in cases where
changes to current treatment practice
would be utilized to reduce TTHMs:
1. The total quantity of chlorine
dioxide added during the treatment
process should not exceed 1 milligram
per liter of water.
2. Chloramines should not be utilized
as the primary disinfectant. Chloramines
may be added for the purpose of
maintenance of an active chlorine
residual in the distribution system only
to water that already meets primary
drinking water regulations.
3. Monitoring for general bacteria
populations (Standard Plate Count)
should be performed as determined by
the State but at least daily for at least •
one month prior to and six months
subsequent to the modifications.
These restrictions have been deleted
from the final regulations to provide the
States with greater discretion to
prescribe requirements as necessary on
a case-by-case basis. This should not be
construed to reflect EPA's lack of
concern regarding microbiological
-------
Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68637
quality. As described below, EPA is
requiring that water systems obtain
State approval of any proposed
'significant modifications to their
treatment process. Once a system's plan
is approved, the system must follow the
plan. Moreover, these regulations
prescribe those minimum conditions
which must be satisfied by the plan
before State approval can be granted.
EPA will also publish guidance for the
States that will serve as a useful
reference in approval of system plans.
This approach is believed to be more
reasonable than the inclusion of specific
nationally applicable restrictions which
may or may not be applicable in every
case. Because systems will begin making
modifications to their disinfection
processes immediately upon
promulgation of this regulation (and in
fact, some systems have already begun
to make such changes), EPA has
determined that good cause exists to
make the requirements of § 141.30(f)
(approval of system treatment
modification plans) effective
immediately upon promulgation. This is
necessary to ensure that all system
treatment modifications are made
subject to close State and EPA
supervision at the earliest possible time.
Chlorine Dioxide
Oxidation/reduction reactions of
chlorine dioxide in water produce
chlorite and some chlorate and
ultimately chloride ions. Preliminary
studies with cats and rats had indicated
that excessive exposures (above 10 mg/
1) to chlorite had resulted in deleterious
effects on red blood cells in some
animals. A limit on applied chlorine
dioxide of 1 mg/1 was proposed to
provide a margin of safety from the
possible effects of ingested chlorine
dioxide and chlorite and chlorate, and
assumed that a portion of the chlorine
dioxide would be spontaneously
reduced to chloride which is not toxic.
In a more recent study in a human
population using drinking water treated
seasonally with chlorine dioxide,
statistically significant blood effects
were not found at concentrations of
approximately 5 mg/l of oxidant in
water; however, this was a short
duration test that terminated earlier
than expected. One individual shown to
be deficient in glucose 6-phosphate
dehydrogenase, a genetic defect that is
present in a small percent of the U.S.
population that would possibly be
sensitive to oxidants, showed an effect,
but, it was within the range of effects of
some of the normal population.
Only ten comments were received on
the proposed chlorine dioxide restriction
and nine were opposed claiming
insufficient evidence of adverse health
risk. Several suggested acceptable levels
as high as 2 or 3 mg/1, but did not submit
supporting data.
EPA has concluded that while there is
evidence that exposure to chlorine
dioxide by-products can result in
detectable if not clinically significant
blood effects, restrictions should be
more appropriately placed on the
residual oxidants (CIO,. C1O-, and
ClO-jJ in the water rather than on the
amount of CIO, added. The extent of the
oxidation/reduction of the added ClOi
and the formation of the intermediate
chlorite and chlorate would be a
function of the reducing agents present
in the water, and the chlorine dioxide
that would be completely reduced to
chloride is of no lexicological
significance.
In the 1979 update of "Drinking Water
and Health", the NAS reviewed the data
as of 1978 and estimated acceptable
exposure values of 0.3B mg/1 and 0.21
mg/1 for chlorine dioxide and chlorite
respectively. These were computed from
data in rats and cats and incorporated
an uncertainty factor of 100. The NAS
also noted that the computed value for
chlorine dioxide was consistent with
EPA's proposal limiting the amount
added to 1 mg/1 assuming 50%
conversion to chloride. Very recent
incomplete data obtained from
controlled studies with normal male
volunteers detected slight but not
clinically significant effects at higher
than normal doses. These experiments
are continuing and will produce more •
definitive results within the next year.
Therefore, although the restriction on
chlorine dioxide addition has been
deleted from the regulation, EPA feels
that whenever chlorine dioxide is used
residual oxidants should be monitored
and kept below 0.5 mg/1. EPA will
consider establishing an MCL for
chlorine dioxide, chlorite and chlorate
or the aggregate as total oxidant for
inclusion in the Revised Regulations
after further studies have been fully
evaluated.
Chloramines
Chloramine (combined chlorine) has
been shown to be a simple and readily
available means of reducing the
formation of THMs in many water
supplies in those cases where raw water
quality and treatment methods permit.
The proposal to restrict the use of
chloramines in THM control in
inappropriate circumstances was based
upon the well known fact that
chloramines, in themselves, are very
weak disinfectants for bacteria, virus
and protozoa compared to free chlorine
as HOC1, ozone and chlorine dioxide.
Thus, the use of chloramines as a
primary disinfectant, (i.e., to kill or
inactivate pathogens in raw water), may
increase the risk of pathogens reaching
the consumer. The proposed restriction
would not have affected the use of
chloramines for disinfection
maintenance in distribution systems.
Opponents of the restriction argued
that chloramines had been effectively
used in many systems. Other
commenters agreed with the proposal
that chloramines should be restricted
from use as a primary disinfectant.
Those opposed to the restriction did not
distinguish between the common use of
chloramines to maintain an active
combined chlorine residual (as a
secondary disinfectant by EPA's
definition) and total reliance on
chloramines (as a primary disinfectant].
None of the commenters contradicted
the experimental fact that chloramines
are much less efficient bacteriocides
and virocides than chlorine (HOC1),
ozone, and chlorine dioxide. The
NOW AC felt that the proposed
limitation was unduly restrictive.
Providing the necessary barrier
against waterborne disease
transmissions is the function of the total
process of providing water to the
consumer. This process begins with
selection of the best available source,
and its protection from contamination
and is followed by the treatment train,
that may consist of off-line storage,
coagulation, sedimentation and filtration
and/or lime treatment and pH
adjustment, along with several
increments of oxidant (disinfectant). It
concludes with protecting the finished
water in transit by maintenance of the
integrity of the distribution system. EPA
recognizes the use history as well as the
risks inherent in misuse of chloramines
and has concluded that the decision is
best made on a case-by-case basis by
the State or primacy authority in its
review and approval of a water system's
plan under § 141.30(0 to provide the
necessary supervision. This subject is
also included in EPA's guidance to the
States for approval of system treatment
modification plans.
Standard Plate Count
The presence of coliform bacteria is
considered to be the most reliable
indicator of possible fecal
contamination and associated enteric
microorganism. Current National Interim
Primary Drinking Water Regulations (40
CFR 141.21, 40 FR 59556) require
monitoring for coliforms on a frequency
based upon population served in the
community water system and include an
MCL of 1 coliform per liter as
determined by the membrane filter
-------
68638 Federal Register / Vol. 44, No. 231 / Thursday, November 29. 1979 / Rules and Regulations
technique. Nevertheless, certain
bacteria, viruses and cysts are more
resistant to disinfectants and are
capable of surviving in water longer
than the coliform indicator organisms.
Because of the possibility that, in the
course of applying treatment
modifications to reduce TTHMs, some
water systems might.be tempted to
utilize less efficient disinfectants such
as chloramines or shorter contact times
with free chlorine, the proposal
contained a requirement to utilize the
Standard Plate County (SPC) analysis
during transition periods when current
treatment practice was being modified.
This was intended to be applied as a
more sensitive Indicator of general
biological quality to signal the
possibility of a deterioration of
treatment effectiveness and therefore
increased potential of undetected
pathogens.
Of the comments on this issue, more
than half opposed or questioned the
significance of the SPC as an indicator
of water quality. However, somewhat
less than half of the commenters agreed
with the proposal that SPC should be
required during treatment modification.
A few suggested that SPC should be
required only for those water sources
receiving discharges of municipal waste.
Others felt that SPC should be used at
the discretion of the State. The NEWAC
recommended that the SPC should not
be a regulatory requirement but rather a
matter of Stale discretion.
In "Drinking Water and Health," the
Safe Drinking Water Committee of the
NAS underscored the usefulness of SPC
applied in conjuction with total coliform
tests to measure the sanitary quality of
drinking water. The Committee
recommended use of SPC to:
1. Provide a method for monitoring for
changes in the microbiological quality of
finished water;
2. Determine whether the normal flora
with coliform detection; and
3. Monitor the effectiveness of a
disinfectant or treatment practice within
the plant and distribution system and
provide an indication of filter-effluent •
quality deterioration and the occurrence
of the breakthrough of microorganisms.
EPA remains convinced that the SPC
is an appropriate adjunct to coliform
monitoring and a sensitive indicator of
process performance and distribution
system integrity, and that it should be
employed particularly during periods
when treatment modifications are being
introduced. Many public water systems
have extensively used the test as a
routine quality monitor. Its application
is particularly essential in drinking
water drawn from raw water sources
contaminated by sewage effluent. SPC
has been deleted as a requirement from
these regulations, but should be a
condition for State approval of system
plans where disinfection process
modifications are contemplated. SPCs
are therefore included in the guidance to
States for approval of system treatment
modification plans.
Microbiological Considerations—State
Approval of System Treatment
Modification Plans to Reduce TTHMs
Historically, the States have had the
responsibility of ensuring that drinking
water in public water systems has
received adequate treatment before it is
distributed. When systems alter
traditional treatment practices to reduce
TTHMs, States must continue to
exercise control to assure that water is
provided to the consumer by public
water systems that is microbiologically
and chemically safe and of optimal
quality. Where States lack primacy
enforcement responsibility, that
responsibility falls to the EPA Regional
Office.
The goal of disinfection has been and
still is to produce water that is
biologically safe to drink; this goal is
attained by killing pathogens in the
water. However, potentially harmful
chemicals are now known to be
produced during disinfection. Quality
control thus necessitates careful
consideration of all appropriate factors .
for each public water system modifying
disinfection processes to control
production of those chemicals, and
States should exercise their full
authority to see that the public is
protected.
The National Academy of Sciences'
reports, "Drinking Water and Health"
and "The Disinfection of Drinking
Water" and the Office of Drinking
Water (EPA) Report, EPA-570/9-7ft-002.
"Evaluation of the Microbiology
Standards for Drinking Water" address
the principles of drinking water
disinfection and their effect on microbial
problems. These documents, along with
the guidance accompanying this
regulation, should be consulted early in
the development of the public water
supply's program to reduce TTHM
formation.
The basic principle in achieving
compliance with the TTHM MCL is that
as TTHM control practices are
conceived and put into practice, the
water supplied to the consumer must be
of optimal quality. Systems must be
carefully supervised to ensure that
water quality is not allowed to
deteriorate as a result of changes in
treatment practice, thereby creating
risks to the public health from particular
chemicals or infectious agents. The
integrity of the bacteriological quality of
the drinking water must not be
compromised.
EPA is therefore requiring that public
water systems contemplating significant
changes in treatment practice lo control
TTHMs submit an action plan to the
State for approval and after approval
has been received, to follow the
conditions set forth in the approved
plan, that will be based upon the
guidance provided by EPA.
The following summarizes the major
principles set forth in the EPA guidance
to the States:
1. Prior to any significant
modification, the entire system should
be evaluated to detect the presence of
sanitary defects and to determine the
risks from breakthrough of
microbiological contaminants in the
source water, through treatment and in
the distribution system. Virus studies
are essential where source waters are
heavily contaminated with sewage
effluents.
2. A comprehensive evaluation of
existing treatment practices and
available options should be conducted
to determine the most effective
treatment modifications that would
result in optimum finished water
biological quality and TTHM control.
Any system deficiencies that are found
during the examination should be
promptly corrected.
3. A baseline water quality survey of
source water, water undergoing
treatment prior to disinfection and water
within the distribution system
particularly in the extremes of the
system and in deadends should be
conducted prior to the initiation of the
TTHM control practices at a sufficient
frequency and time span to establish an
understanding of the water quality.
Measured parameters should include
coliform and fecal coliform bacteria,
fecal streptococci, standard plate count
incubated at 35* C and 20° C, phosphate,
ammonia nitrogen, TOC and others
directed by the State based on the
particular characteristics of local water
quality. In systems using poor quality
source water, for example, a weekly or
more frequent sampling frequency may
be necessary.
4. Following modification, the water
quality survey (in item 3 above) should
be continued for one year to determine
the performance of the treatment system
for all seasons. The parameters in the
baseline study should continue to be
examined using samples from the same
locations.
5. Treatment practices for THM
control should also provide effective
post disinfection to control microbial
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations 68639
populations, and an active disinfectant
.residual should be maintained in all
parts of the distribution system.
6. If the present point of chlorination
is altered, the supply should maintain
proper pH control and allow sufficient
contact time for optimal disinfection.
7. Monitoring for chlorate, chlorite and
chlorine dioxide should be performed
when chlorine dioxide is used as a
disinfectant. Residual concentrations of
total residual oxidants (except for HOC1
derivatives) in the water should not
exceed 0.5 mg/1 in the interim until
further EPA studies are completed.
8. Chloramines are less efficient as
disinfectants particularly for virus and
protozoans as compared to chlorine,
chlorine dioxide and ozone. If
chloramines are used with contaminated
source water, the total treatment
process should be capable of
compensating for any potential
reduction in disinfection efficiency.
9. Ozone is not an appropriate
disinfectant for high TOG containing
waters unless the potential for post
treatment biological growth can be
controlled such as by the use of
processes that control biodegradable
chemicals in the source water and the
finished water.
10. Systems presently utilizing pre-
chlorination for disinfection purposes
must be certain that alternative
pretreatment practices are sufficient to
protect the public if changes are
introduced.
11. Any oxidant (disinfectant) used to
treat drinking water will interact with
chemicals already in the water to form
undesirable by-products in the finished
water. Therefore the basic principle
should be to maximize precursor
removal prior to the addition of the
oxidant so as to minimize a disinfectant
demand and by-product formation.
Otherwise, an excessive disinfectant
demand could reduce the efficiency of
any disinfectant practice and add, in the
process, substantial amounts of
undesirable and perhaps toxic
compounds.
12. Varied and extensive modification
of existing treatment processes often
result in changes in the chemical and
microbial quality of treated water.
Increased monitoring of coliform
bacteria and the use of other indicators
of the sanitary quality of water (e.g.,
SPC) are advisable.
Individual system plans for TTHM
control should include the design of the
vulnerability and baseline data surveys
and the additional surveillance
monitoring to assure maintenance of
biological quality with the altered
treatment system and must be approved
by the State prior to their
implementation. The plan should also
include information on current treatment
practices and their performance and
other information as directed by the
State. EPA believes that if States and
public water systems follow the
guidance and technical assistance is
provided as needed, TTHM control will
be safely achieved.
Economic Impact Assessment
The economic impact of these
regulations was projected based on the
three principal control options available
to the approximately 2,700 community
water systems serving more than 10,000
people required to comply with the
regulatory requirements—modifying
chlorination or associated treatment
procedures, changing disinfectants,
using an adsorbent or some
combination of the above. The
calculation of total national cost
projections for the TTHM regulation
required an estimate of the number of
systems choosing each control option
and the incremental costs associated
with each option considered. An
incremental expense will accrue to all
systems covered, whether or not
treatment is required, to cover
monitoring expenses. These expenses
for all systems covered are included in
the following estimates of total costs for
the TTHM regulation.
This analysis employed a
probabilistic and structured approach
for determining the choice of control
options that each public water system
would make since no empirical method
exists for predetermining that choice. A
logical sequence of decision points was
designed to distribute the systems
anticipated to be covered by the
regulation according to the most likely
path they would follow. The decision
made at each point is consistent with
the following criteria:
1. The treatments currently used: If a
system does not add chlorine it will not
be affected by a THM regulation, and
therefore will require no new treatment
2. Water source used: If a system uses
surface water (except the Great Lakes
and some high quality mountain water)
as its primary source, it is more likely to
exceed a given level of THM
contamination. Hence the number of
water systems using water from ground
or surface sources affects the number of
systems which will exceed the MCL and
will therefore require treatment.
3. Degree to which water quality
exceeds MCL If the presence of TTHMs
is only slightly in excess of the initial
MCL, then minimal modifications to
current treatment procedures may be
adequate for compliance. As the level of
contamination increases, a system must
consider more significant (and costly)
treatment techniques.
4. Economic considerations: The
presumption was that systems would
adopt the least costly treatment strategy
that satisfies the regulations.
5. Treatment effectiveness: Many
systems with TTHM concentrations only
slightly above the MCL can comply by
modifying treatment procedures. Others
may need to change disinfectants.
Finally, precursor concentrations
resulting in very high THM formation
potentials can probably be best
controlled by the use of adsorbents. This
is because of the likelihood that high
disinfectant demand waters cannot be
disinfected adequately without
generating considerable amounts of by-
products of unknown hazard or without
exceeding the MCL. Consequently, some
of those systems with very high levels of
TTHMs are projected to use adsorbents.
Based on all information available to
EPA of the 390 public water systems
that serve more than 75,000 people, 61
purchase the majority of their water
from other systems that are presumed to
provide treatment. Thus, a total of 329
systems would be initially affected
although 7 of these were excluded
because they do not presently add a
disinfectant. Of the remaining 322, some
95 systems were estimated to have
TTHM levels above 0.10 mg/1 and hence
would require changes in their treatment
processes.
Since the final regulation phases in
coverage to include systems serving
between 10,000 and 75,000 people, the
economic analysis has also included the
costs these systems will bear in
achieving compliance. Of the 2,295
public water systems that serve
between 10,000 and 75,000 people, 355
are known to purchase the majority of
their water from other systems that are
presumed to provide treatment. Thus a
total of 1,940 systems between 10,000
and 75,000 population would be initially
affected, although 281 of these are
excluded because they do not presently
add a disinfectant Of the remaining
1,659, some 420 systems were estimated
to have TTHM levels above 0.10 mg/1
and hence would require changes in
their treatment processes to comply by
the applicable effective date in the
regulation.
The following projections were made
based upon information presented
during the comment period primarily
from the water utilities and consultants.
Of the systems estimated to be in the
range of 1 to 1.5 times the MCL, 60
percent were expected to modify their
chlorination procedures and 40 percent
were expected to change disinfectants.
Of the systems with TTHM levels in the
-------
68640 Federal Register / Vol. 44, No. 231 / Thursday. November 29. 1979 / Rules and Regulations
range of 1.5 to 2.5 times the MCL, 25
percent were expected to change their
chlorination procedures with 75 percent
changing disinfectants. Finally, of the
systems exceeding 2.5 times the MCL, 80
percent were anticipated to change
disinfectants and the remaining 20
percent would likely use an adsorbent
On the basis of the above assumptions,
national cost estimates for compliance
with these final regulations are as
follows:
Summary of Estimated Total Coats lor an MCL
Regulation With the Trlhalomethane
Concentration of 0.10 mg/l
|ln millions of 1980 dollars)
Catego
MroTnglo
poput&tion served by
average system
10.000-
75.000
Over
75.000
Total
Capital Expenditures
Operation and Maintenance
Revenue Requirements
Annual per Caita Costs of
Treatment < (doDare)
Increase in Annual
Residential Bill ' (dollars)
'Includes ontv systems oroh
$40
5
e
060
120
acted to Incur to
S45
5
10
090
180
'B&lfnBflt C
$85
to
19
070
140
sosta
Per capita costs will vary depending
upon the type of treatment selected, the
system size, and many other factors.
Given an MCL of 0.10 mg/l, the range of
annual residential bill increases for a
typical family of 3 would be from $0.32
to $1.89 for systems using an alternative
disinfectant and $4.44 to $11.18 for
systems using an adsorbent in
combination with ozonation assuming a
720 day regeneration cycle.
The costs presented in this final
analysis are considerably lower than
EPA's previous national cost estimates
for the TTHM regulations as set forth in
the February 9.1978, notice and later
revised in the July 6,1978. supplemental
notice, even though they are now stated
in 1980 dollars while the August 1977
report accompanying the proposed
regulations used 1976 dollars. The
differences causing this reduction result
from numerous changes in the
underlying data, based on information
received during the comment period,
including: (a] Revised estimates of the
number of systems using disinfectants;
(b) revised estimates of the level of
TTHMs in a given ground or surface
system; (c) changes in the probabilities
assigned to branches of the decision tree
used to select among control options
with more systems using chloramines
and many fewer using CAC; (d)
revisions of unit cost data to reflect
inflation to 1980 dollars and increases in
assumed levels of professional fees
(resulting in an approximate 28 percent
increase in costs}; (e) changes in the
GAC costs to reflect longer projected
regeneration cycles (from 60 days to 360
days for GAC alone and 720 days for
GAC and ozone], more off-site
regeneration at regional facilities and
use of GAC in existing filter beds.
Detailed analysis of the costs of various
options and the underlying data are
contained in the "Economic Impact
Analysis of a Promulgated
Trihalomethane Regulation for Drinking
Water," available on request, and
incorporated by reference as part of the
Statement of Basis and Purpose for this
regulation.
Although the typical economic
impacts appear to be reasonable, it is
possible that some utilities will have
unique problems which lead to financial
hardships. This would take the form of
an inability to raise capital needs for
improvements in treatment necessary to
comply with the TTHM regulation.
Should a situation arise, opportunities
exist which can ease these financing
difficulties. The Office of Drinking
Water provides technical assistance in
this area, and interested parties should
contact: Victor ]. Kinun, Deputy
Assistant Administrator for Drinking
Water (WH-550), Environmental
Protection Agency, 401M Street, SW.,
Washington, D.C. 20460 for additional
information.
Energy Impact Assessments
The TTHM regulation will have a
negligible impact on annual domestic
energy consumption. The total energy
requirements associated with the
regulation are SOBXlO'BTU's, or 0.0007
percent of 1977 U.S. energy
consumption. The annual energy
requirements of the various treatment
alternatives selected by utilities to meet
the MCL for TTHMs are as follows:
Electric power, 39.9 million kilowatt-
hours; diesel fuel, 64,000 gallons; and
natural gas, 76.4 million cubic feet. In
1980 dollars these total annual energy
requirements are estimated to cost $2.3
million per year. The annual electric
power demand of 39.9 kwhr is
approximately 0.002 percent of 1977
total domestic electric power sales. The
annual diesel fuel demand represents
only 0.00002 percent of the 1977 total
domestic demand for refined oil
products. At 76.4 million cubic feet, the
annual natural gas demand represents
less than 0.004 percent of the 1977
domestic natural gas demand.
Approximately 87 percent of the
electric power demand is due to ozone
disinfection processes. GAC treatment
and ozonation together represent 98
percent of the total electric power
demand.
The diesel fuel and natural gas
requirements are created by the GAC
regeneration process. For those water
utilities without on-site GAC
regeneration, transport of GAC to
remote processing sites will require
diesel fuel. The regeneration process
itself requires either oil or natural gas as
an energy source. In preparing these
energy demand estimates, EPA assumed
that only natural gas would be used in
GAC regeneration furnaces. The energy
impacts of this regulation are reduced
from those associated with the proposal
because fewer systems are expected to
resort to the more energy intensive
treatment methods to achieve
compliance with the MCL.
Evaluation Plan
As noted previously, these regulations
are considered to be an initial step in
controlling disinfection by-products,
with TTHMs being a surrogate. As the
regulations are implemented, an
extensive data collection effort will
begin through the self-monitoring
programs at the applicable public water
systems. These data will include levels
of TTHMs associated with disinfection
of various types of raw water sources
and the specific technologies utilized for
control of TTHMs.
Compliance with the regulations will
be determined by State program staffs
and the compliance data will be
included in the Model State Information
System and Federal Data Reporting
Systems (computer systems}. This will
allow easy access to evaluation of
national compliance with the
regulations.
The compliance data will be
evaluated along with results of ongoing
research and development efforts which
are examining the toxicology of
disinfection by-products and available
treatment alternatives for control. The
evaluation will be used to determine the
appropriateness of the level of the MCL
and will be the basis of further
regulatory actions controlling
disinfection by-products. These
evaluations will be conducted no later
than three years after the promulgation
of the regulations. The Director, Criteria
and Standards Division, Office of
Drinking water, should be contacted if
further information is desired.
Under Executive Order 12044. EPA is
required to judge whether a regulation is
"significant" and therefore subject to the
procedural requirements of the Order or
whether it may follow other specialized
development procedures. EPA labels
these other regulations "specialized." I
have reviewed this regulation and
determined that it is a specialized
-------
Federal Register / Vol. 44. No. 231 / Thursday. November 29. 1979 / Rules and Regulations 68641
regulation not subject to the procedural
requirements of Executive Order 12044.
Dated: November 5,1970.
Douglas M. Coslle,
Administrator.
Accordingly. Part 141, Title 40 of the
Code of Federal Regulations is hereby
amended as follows:
1. By amending § 141.2 to include the
following new paragraphs (p) through
(»):
§141.2 Definitions.
*****
(p) "Halogen" means one of the
chemical elements chlorine, bromine or
iodine.
(q) "Trihalomethane" (THM) means
one of the family of organic compounds,
named as derivatives of methane,
wherein three of the four hydrogen
atoms in methane are each substituted
by a halogen atom in the molecular
structure.
(r) "Total trihalomethanes" (TTHM)
means the sum of the concentration in
milligrams per liter of the
trihalomethane compounds
(trichloromethane [chloroform],
dibromochlorome thane,
bromodichlorome thane and
tribromomethane [bromofonn]), rounded
lo two significant figures.
(s) "Maximum Total Trihalomethane
Potential (MTP)" means the maximum
concentration of total trihalomethanes
produced in a given water containing a
disinfectant residual after 7 days at a
temperature of 25' C or above.
(t) "Disinfectant" means any oxidant,
including but not limited to chlorine,
chlorine dioxide, chloramines, and
ozone added to water in any part of the
treatment or distribution process, that is
intended to kill or inactivate pathogenic
microorganisms.
2. By revising § 141.6 to read as
follows:
1141.6 Effective dates.
(a] Except as provided in paragraph
(b) of this section, the regulations set
forth in this part shall take effect on
June 24,1977.
(b] The regulations for total
trihalomethanes set forth in § 141.12(c)
shall take effect 2 years after the date of
promulgation of these regulations for
community water systems serving 75,000
or more individuals, and 4 years after
the date of promulgation for
communities serving 10,000 to 74,999
individuals.
3. By revising the introductory
paragraph and adding a new paragraph
(c) in § 141.12 to read as follows:
§ 141.12 Maximum contaminant levels for
organic chemicals.
The following are the maximum
contaminant levels for organic
chemicals. The maximum contaminant
levels for organic chemicals in
paragraphs (a) and (b] of this section
apply to all community water systems.
Compliance with the maximum
contaminant levels in paragraphs (a)
and (b) is calculated pursuant to
§ 141.24. The maximum comtaminant
level for total trihalomethanes in
paragraph (c] of this section applies only
to community water systems which
serve a population of 10,000 or more
individuals and which add a
disinfectant (oxidant] to the water in
any part of the drinking water treatment
process. Compliance with the maximum
contaminant level for total
trihalomethanes is calculated pursuant
to § 141.30.
*****
(c] Total trihalomethanes (the sum of
the concentrations of
bromodichloromethane,
dibromochloromethane,
tribromomethane (bromofonn) and
trichloromethane (chloroform)]
0.10 mg/1.
4. By revising the title, the
introductory text of paragraph (a] and
paragraph (b) of § 141.24 to read as
follows:
§141.24 Organic chemicals other than
total trihalomethanes, sampling and
analytical requirements.
[a] An analysis of substances for the -
purpose of determining compliance with
§ 141.12(a) and 8 141.12(b) shall be made
as follows:
(b) If the result of an analysis made
pursuant to paragraph (a) of this section
indicates that the level of any
contaminant listed in § 141.24 (a] and (b)
exceeds the maximum contaminant ^
level, the supplier of water shall report
to the State within 7 days and initiate
three additional analyses within one
month.
5. By adding a new § 141.30 tc read as
follows:
g 141.30 Total trihalomethanes sampling,
analytical and other requirements.
(a] Community water system which
serve a population of 10,000 or more
individuals and which add a
disinfectant (oxidant] to the water in
any part of the drinking water treatment
process shall analyze for total
trihalomethanes in accordance with this
section. For systems serving 75,000 or
more individuals, sampling and analyses
shall begin not later than 1 year after the
date of promulgation of this regulation.
For systems serving 10,000 to 74.999
individuals, sampling and analyses shall
begin not later than 3 years after the
date of promulgation of this regulation.
For the purpose of this section, the
minimum number of samples required to
be taken by the system shall be based
on the number of treatment plants used
by the system, except that multiple
wells drawing raw water from a single
aquifer may, with the State approval, be
considered one treatment plant for
determining the minimum number of
samples. All samples taken within an
established frequency shall be collected
within a 24-hour period.
(b)(l) For all community water
systems utilizing surface water Sources
in whole or in part, and for all
community water systems utilizing only
ground water sources that have not been
determined by the State to qualify for
the monitoring requirements of
paragraph (c) of this section, analyses
for total trihalomethanes shall be
performed at quarterly intervals on at
least four water samples for each
treatment plant used by the system. At
least 25 percent of the samples shall be
taken at locations within the
distribution system reflecting the
maximum residence time of the water in
the system. The remaining 75 percent
shall be taken at representative
locations in the distribution system,
taking into account number of persons
served, different sources of water and
different treatment methods employed.
The results of all analyses per quarter
shall be arithmetically averaged and
reported to the State within 30 days of
the system's receipt of such results.
Results shall also be reported to EPA
until such monitoring requirements have
been adopted by the State. All samples
collected shall be used in the
computation of the average, unless the
analytical results are invalidated for
technical reasons. Sampling and
analyses shall be conducted in
accordance with the methods listed in
paragraph (e) of this section.
(2) Upon the written request of a
community water system, the monitoring
frequency required by paragraph (b)(l)
of this section may be reduced by the
State to a minimum of one sample
analyzed for TTHMs per quarter taken
at a point in the distribution system
reflecting the maximum residence time
of the water in the system, upon a
written determination by the State that
the data from at least 1 year of
monitoring in accordance with
paragraph (b](l] of this section and local
conditions demonstrate that total
trihalomethane concentrations will be '
consistently below the maximu-
contaminant level.
-------
68642 Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations
(3) If at any time during which the
reduced monitoring frequency
prescribed under this paragraph applies,
the results from any analysis exceed
0.10 mg/1 of TTHMs and such results are
confirmed by at least one check sample
taken promptly after such results are
received, or if the system makes any
significant change to its source of water
or treatment program, the system shall
immediately begin monitoring in
accordance with the requirements of
paragraph (b](l) of this section, which
monitoring shall continue for at least 1
year before the frequency may be
reduced again. At the option of the
State, a system's monitoring frequency
may and should be increased above the
minimum in those cases where it is
necessary to detect variations of TTHM
levels within the distribution system.
(c)(l) Upon written request to the
State, a community water system
utilizing only ground water sources may
seek to have the monitoring frequency
required by subparagraph (1) of
paragraph (b) of this section reduced to
a minimum of one sample for maximum
TTHM potential per year for each
treatment plant used by the system
taken at a point in the distribution
system reflecting maximum residence
time of the water in the system. The
system shall submit to the State the
results of at least one sample analyzed
for maximum TTHM potential for each
treatment plant used by the system
taken at a point in the distribution
system reflecting the maximum
residence time of the water in the
system. The system's monitoring
frequency may only be reduced upon a
written determination by the State that,
based upon the data submitted by the
system, the system has a maximum
TTHM potential of less than 0.10 mg/1
and that, based upon an assessment of
the local conditions of the system, the
system is not likely to approach or
exceed the maximum contaminant level
for total TTHMs. The results of all
analyses shall be reported to the State
within 30 days of the system's receipt of
such results. Results shall also be
reported to EPA until such monitoring
requirements have been adopted by the
State. All samples collected shall be
used for determining whether the system
must comply with the monitoring
requirements of paragraph (b) of this
section, unless the analytical results are
invalidated for technical reasons.
Sampling and analyses shall be
conducted in accordance with the
methods listed in paragraph (e) of this
section.
(2) If at any time during which the
reduced monitoring frequency
prescribed under paragraph (c)(l) of this
section applies, the results from any
analysis taken by the system for
maximum TTHM potential are equal to
or greater than 0.10 mg/1, and such
results are confirmed by at least one
check sample taken promptly after such
results are received, the system shall
immediately begin monitoring in
accordance with the requirements of
paragraph (b) of this section and such
monitoring shall continue for at least
one year before the frequency may be
reduced again. In the event of any
significant change to the system's raw
water or treatment program, the system
shall immediately analyze an additional
sample for maximum TTHM potential
taken at a point in the distribution
system reflecting maximum residence
time of the water in the system for the
purpose of determining whether the
system must comply with the monitoring
requirements of paragraph (b) of this
section. At the option of the State,
monitoring frequencies may and should
be increased above the minimum in
those cases where this is necessary to
detect variation of TTHM levels within
the distribution system.
(d] Compliance with § 141.12(c) shall
be determined based on a running
annual average of quarterly samples
collected by the system as prescribed in
subparagraphs (1) or [2] of paragraph (b)
of this section, If the average of samples
covering any 12 month period exceeds
the Maximum Contaminant Level, the
supplier of water shall report to the
State pursuant to § 141.31 and notify the
public pursuant to § 141.32. Monitoring
after public notification shall be at a
frequency designated by the State and
shall continue until a monitoring
schedule as a condition to a variance,
exemption or enforcement action shall
become effective.
(e) Sampling and analyses made
pursuant to this section shall be
conducted by one of the following EPA
approved methods:
(1) "The Analysis of Trihalomethanes
in Finished Waters by the Purge and
Trap Method." Method 501.1, EMSL,
EPA Cincinnati, Ohio.
(2) "The Analysis of Trihalomethanes
in Drinking Water by Liquid/Liquid
Extraction,"Method 501.2, EMSL, EPA
Cincinnati, Ohio.
Samples for TTHM shall be
dechlorinated upon collection to prevent
further production of Trihalomethanes,
according to the procedures described in
the above two methods. Samples for
maximum TTHM potential should not be
dechlorinated, and should be held for
seven days at 25° C prior to analysis,
according to the procedures described in
the above two methods.
({] Before a community water system
makes any significant modifications to
its existing treatment process for the
purposes of achieving compliance with
§ 141.12(c), such system must submit
and obtain State approval of a detailed
plan setting forth its proposed
modification and those safeguards that
it will implement to ensure that the
bacteriological quality of the drinking
water served by such system will not be
adversely affected by such modification.
Each system shall comply with the
provisions set forth in the State-
approved plan. At a minimum, A Slate
approved plan shall require the system
modifying its disinfection practice to:
(1) Evaluate the water system for
sanitary defects and evaluate the source
water for biological quality;
(2) Evaluate its existing treatment
practices and consider improvements
that will minimize disinfectant demand
and optimize finished water quality
throughout the distribution system;
(3) Provide baseline water quality
survey data of the distribution system.
Such data should include the results
from monitoring for coliform and fecal
coliform bacteria, fecal streptococci,
standard plate counts at 35* C and 20* C,
phosphate, ammonia nitrogen and total
organic carbon. Virus studies should be
required where source waters are -
heavily contaminated with sewage
effluent;
(4) Conduct additional monitoring to
assure continued maintenance of
optimal biological quality in finished
water, for example, when chloramines
are introduced as disinfectants or when
pre-chlorination is being discontinued.
Additional monitoring should also be
required by the Stale for chlorate,
chlorite and chlorine dioxide when
chlorine dioxide is used as a
disinfectant. Standard plate count
analyses should also be required by the
State as appropriate before and after
any modifications;
(5) Demonstrate an active disinfectant
residual throughout the distribution
system at all times during and after the
modification.
This paragraph (f) shall become
effective on the date of its promulgation.
Appendix A—Summary of Public
Comments and EPA Responses on
Proposed Amendments to the National
Interim Primary Drinking Water
> for Control of
alomethanes in Drinking Water
The following is a summary and
discussion of the principal public
comments to EPA's proposed
regulations for the control of
-------
Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68643
trihalomethanes (THMs) in drinking
rater and EPA's responses to them.
rtany comments have already been
ddressed in the preamble which should
je referred to for additional explanation
of the agency's responses. In its
February 9.1978, notice of proposed
rulemakmg, EPA specifically solicited
comments on the following six
questions:
1. The reasonableness of the concept
of phasing the application of the
regulation by making the MCL
mandatory initially only for large water
systems and for the time being requiring
monitoring only in others, and no
requirements in the smallest systems.
Should the regulations differentiate in
their application between ground and
surface water supplies? Are monitoring
frequencies sufficient to identify
locations with high TTHM levels?
An alternative approach on which
public comments are solicited would be
to make the MCL applicable to all public
water systems and affect phasing of
implementation by establishing a
deferred monitoring schedule. Systems
serving more than 75,000 people would
be required to begin monitoring within
one year of promulgation, systems
serving between 10,000 and 75.000
would be required to begin monitoring
l/vithin three years and all other
communities within five years.
2. The magnitude of the MCL at 0.10
mg/1. Does the current information
warrant more restrictive regulations at
this time, for example, 0.050 mg/1 or
less? How rapidly can the MCL be
reduced to lower feasible levels?
3. The feasibility and timing of the
treatment modifications that will be
necessary to achieve compliance. Will
18 months provide adequate time for
most impacted systems to take steps to
come into compliance?
4. The economic impact on large,
medium, and small water systems either
for the proposed regulation or for more
restrictive regulations. Are EPA's
estimates of the cost of compliance
reasonable?
5. The concept of averaging the
concentrations of the TTHMs for
compliance—both the annual averaging
of quarterly samples, and the averaging
of representative samples within the
distribution system.
6. The use of the Standard Plate Count-
as a more sensitive indicator of
microbiological quality while treatment
modifications are being introduced and
the limitations on chlorine dioxide and
ichloramines.
In addition, the proposed regulations
'generated comments on. other issues,
including such issues as whether the
States with primary enforcement
responsibility had been provided
sufficient time to make State regulations
consistent with the federal regulations
by the effective date. The majority of
commenters did not address all of the
issues that were posed by EPA; many
commented on just a few issues or only
on a single issue.
In all, EPA received 598 written
comments and 259 oral statements were
presented in the eight public hearings.
The total of 857 comments came from
various interested parties, including 390
from water utilities, 32 from private
industries, 28 from consulting engineers,
95 from special interest groups, 80 from
private individuals, 33 from educational
institutions, 13 from Federal government
agencies, 98 from local governments. 75
from local and State health and
environmental departments, and 13 from
other groups including some members of
Congress. An additional 498
communications from members of
Congress were received and responded
to directly. Many of the comments were
duplicative; some commenters presented
both written and oral comments, or the
comments were repeated in substance
by many commenters, including
members of Congress. In a number of
cases, commenters simply endorsed the
official position taken by a particular
organization. For example, 124 water
utilities and local governments
responded by endorsing the position of
the American Water Works Association
(AWWA) which recommended an
alternative program for the control of
organic chemical contamination in
drinking water. Comprehensive
comments were also received from the *
Coalition for Safe Drinking Water
(CSDW), a member organization of both
municipal and investor-owned water
utilities formed specifically to comment
on EPA's proposed regulations, Calgon
Corporation, a large manufacturer of
carbon, and the National Drinking
Water Advisory Council. These and
other major comments are summarized
in Appendix B. The following discussion
summarizes comments received on the
proposed regulations and the Agency's
responses to those comments.
1. A majority of public comments
disagreed with EPA's proposal to limit
the applicability of the TTHM MCL to
systems serving greater than 75,000
people. Most commenters preferred to
have all water systems included under
the regulation if control of chloroform
was indeed deemed necessary (many of
them did not feel any regulation was
necessary). Phasing-in the applicability
of the regulation to smeller systems in
time was also opposed by some
commenters, but a large number thought
such a phasing approach to be logical.
The population cut-off of 75,000
received a total of 158 comments.
Among the commenters, 132 felt that the
regulations should be applied to all
systems regardless of size; 22
commenters thought the population cut-
off and phasing approach were
reasonable. The main reason given by
those who opposed the population cut-
off was that they felt such an approach
was contradictory to the intent of the
SDWA which was to protect all persons
served by community water systems.
Therefore, these commenters said that if
there was a health concern, all systems
should be required to comply with the
TTHM MCL, not just those who are
served by a large water system. The
commenters who thought that the
population cut-off and phasing approach
were reasonable cited as their reasons
economic and technical feasibility,
realizing that the larger water utilities
would be better financed and staffed.
In response to the comments, EPA has
accepted the recommendation of the
National Drinking Water Advisory
Council and many other commenters to
broaden the coverage of the TTHM
regulations to include those systems
serving as few as 10,000 people and to
phase-in the effective dates of the MCL
by system size as follows:
• Water systems serving 75,000 or
more people are required to be in
compliance with the TTHM MCL within
two years from the date of promulgation
of the regulations.
• Systems serving between 10,000 and
75,000 people are required to be in
compliance by four years from the date
of promulgation.
This still means systems serving fewer
than 10,000 persons are not covered by
these regulations. EPA does not believe
that this approach violates the intent of
the SDWA to protect all persons served
by community water systems. The great
majority of smallest systems are served
by ground water sources that are low in
THM precursor content. Therefore, their
drinking water is less likely to be
subject to significant THM
contamination. EPA is also concerned
that measures taken by the smallest
systems to reduce THM levels are more
likely to result in drinking water of poor
microbiological quality since they
generally lack the expertise and access
to technical assistance necessary for
careful supervision of alterations in
disinfection practice. Commenters are
referred to the preamble to these
regulations for a more complete
discussion of EPA's rationale for
excluding these smallest systems from
-------
68644 Federal Register / Vol.' 44. No. 231 / Thursday.' November 29.' 1J979 / Rules and Regulations
the coverage of these amendments to
the Interim Regulations.
As discussed in the preamble, EPA's
decision to phase-in the effective date of
the MCL by system size has been based
in part on die present limited laboratory
capability available for TTHM analyses
and the need for careful supervision of
any alterations to the disinfection
process. The systems in the 10,000 to
75,000 population range will be able to
draw upon the experience gained by the
first group of largest systems who must
achieve compliance in the shortest
feasible time-frame. By that time,
laboratory resources and technical
assistance from the States and EPA will
be available to handle the increased
number of systems. It was believed to
be unreasonable to make the regulations
effective for all systems at once for
these reasons.
2. Thirty-seven comments were
received on whether the regulations
should differentiate between surface
and ground water sources. Twenty-five
opposed the idea of differentiation and
said that the regulations should be
based on water quality rather than.
water sources. Nine believed
differentiation between sources was a
good approach because in general
ground water contains relatively less
precursor material than surface water
and therefore has less chance to •
produce TTHMs during chlorination
practice. Three thought that the States
should make the decision whether to
distinguish between surface and ground
water.
In response to these comments, the
TTHM MCL applies equally to ground
and surface water supplies within the
population range covered. Water quality
serves as the basic distinguishing factor
to the extent that only those systems
that exceed the MCL will be required to
take steps to reduce TTHM levels in the
finished drinking water. However, the
monitoring requirements have been
modified from the proposal to
accommodate the valid concerns of
some commenters that systems with
relatively stable ground water sources
should not be required to incur the
expense of regular monitoring where it
is demonstrated that TTHM levels are
not likely to approach or exceed the
MCL. As discussed more fully in the
preamble, the States have been
accorded some flexibility to modify the
monitoring requirements on a case-by-
case basis under such circumstances.
3. Four comments were received on
the monitoring and compliance
timeframes established in the proposal.
One of these commenters asked what
would happen at the end of one year of
monitoring for systems serving 10.000 to
75,000 people. He questioned why no
action would be required if the TTHM
levels exceeded the MCL. One
commenter suggested that monitoring
requirements be extended to systems
which serve less than 10,000 population
and report the results to customers as
well as authorities. One commenter
suggested that water systems serving
more than 75,000 should start monitoring
within 6 months, systems serving 10,000-
75,000 should start monitoring within 1
year while the rest of the communities
should begin monitoring within 3 years.
One commenter felt that more discretion
should be left to the States to determine
which systems should be brought into
compliance first.
EPA has responded to the comment
concerning compliance by those systems
serving between 10,000 and 75,000
persons by applying the TTHM MCL to
those systems within 4 years of the
promulgation of these regulations. Thus,
systems in that size category that
exceed the MCL would be required to
take measures to reduce TTHM levels in
their drinking water.
The monitoring requirements have not
been extended to systems serving fewer
than 10,000 people in the final
regulations. Monitoring and public
notification of the results were not
believed to be warranted unless and
until those smallest systems were also
going to be required to reduce TTHM
levels when the monitoring results
showed that the MCL was exceeded.
EPA was also concerned about the
availability of laboratories for
conducting TTHM analyses for the
approximately 57,000 systems that fall
within this size category. EPA's
rationale for excluding these systems
from the coverage of the MCL has
already been addressed in response to
other comments and in the preamble to
these regulations.
The alternative monitoring timeframe
suggested by one commenter was
presumably intended to lengthen the
timeframe that EPA had originally
proposed as well as to require
monitoring by the smallest size systems
within a definite timeframe. In these
final regulations, EPA has expanded the
timeframe it originally proposed by
requiring the largest systems to begin
monitoring within one year from the
promulgation of these regulations and
the next size category within 3 years.
EPA found that requiring the largest
systems to begin monitoring within three
to six months would not have provided
adequate time for sufficient numbers of
laboratories to become properly
certified to perform quality TTHM
analyses. An additional two years was
believed to be necessary to insure the
existence of quality laboratory
capability to accommodate the
approximately 2,300 more systems in the
next size category. EPA's reasons for not
requiring monitoring by the smallest size
systems have already been discussed.
With respect to the comment ,
suggesting that the States should have
more discretion to determine which
systems should be brought into
compliance first, this regulation does not
impair the State's prerogative to give
highest enforcement priority to those
systems with, for example, the highest
TTHM levels. However, applying a
uniform effective date for the MCL to
the largest size systems first insures a
fair application of the regulation among
systems and achieves public health
protection for the most people in the
shortest timeframe. While it is the
State's responsibility to enforce
compliance with the MCL, it is each
system's responsibility to achieve
compliance by the applicable date.
4. Other monitoring-related issues
submitted by commenters included:
Seven commenters said that the
proposed timing for monitoring was
inadequate; several commenters said
that it was premature at this time to
require the water utilities to monitor for
TTHMs while other commenters urged
EPA to establish a deferred monitoring
schedule; and two commenters felt that
the monitoring requirement and the
setting of a MCL should be a two-step
action including initial monitoring
followed by setting the MCL. One
commenter believed that it was
necessary to establish an occurrence
data base prior to setting a MCL and
recommended that monitoring must
span at least a 2 to 3 year period in
order to determine the varying
concentrations of these contaminants.
As noted previously, the effective date
of the monitoring requirements has been
extended to one year and three years for
the two size categories, respectively.
This extension will allow adequate time
for development of laboratory
capabilities. In regard to the two step
approach suggested by two commenters
and the establishment of an occurrence
data base prior to setting an MCL, the
EPA agrees with the commenter's
concept and has included both steps in
the regulations: monitoring followed by
compliance with the MCL. A sufficient
data base has been established for
setting the MCL and monitoring for one
year prior to the effective date of the
MCL will provide-more precise
information on variations in TTHM
levels. Of course, systems may, at their
-------
Federal Register / Vol. 44, No. 231 / Thursday. November 29. 1979 / Rules and Regulations 68645
jption, begin monitoring prior to the
effective date.
5. With regard to EPA's proposed
monitoring frequencies for TTHMs of
Five analyses per quarter, 37 comments
were received. Eleven comments said
that the proposed monitoring
frequencies were reasonable. Twenty-
two felt that quarterly sampling was
insufficient, and some suggested more
frequent sampling, such as one sample
every month. Two commenters thought
the proposed frequencies were too
frequent and suggested that monitoring
be conducted twice a year. Two
commenters suggested that the
frequency should be proportionate to the
population served and at regular
intervals.
EPA has retained the quarterly
sampling requirements of the proposal
as the minimum acceptable frequency
for determining the effect of differing
treatment practices and seasonal
variations in raw water quality on
TTHM concentrations in the finished
drinking water. Four instead of five
samples per quarter are required based
on the number of treatment plants used
by the system. Thus, more samples must
be taken by those larger systems most
likely to utilize more than one plant.
This also allows for more representative
sampling since TTHM levels may vary
depending upon the system's raw water
source or treatment program at a
particular plant. Systems may seek State
approval to have multiple wells drawing
raw water from a single aquifer
considered as a single treatment plant
for the purpose of determining the
minimum number of samples.
In response to those comments
seeking more frequent sampling,
generally, the final regulations provide
that the States may require more
frequent sampling where it is necessary
to insure adequate and consistent
control of TTHM levels below the MCL
in the water served to all consumers of
the system. EPA also recognizes that, in
some situations, quarterly sampling
should not reasonably be required
because the maximum TTHM potential
in some ground waters is consistently
well below the TTHM MCL. Thus, the
final regulations also allow the States to
exercise their discretion to reduce the
monitoring frequency in those situations.
The requirements of these regulations
have thus been fashioned to establish a
minimum regular monitoring frequency
while providing for case-by-case
flexibility, recognizing that the optimal
monitoring frequency for TTHM control
will depend largely on site-specific
circumstances.
6. Many comments were received
charging that EPA's action of setting a
TTHM MCL of 0.10 mg/1 was arbitrary,
premature and lacking in supporting
data. 243 comments suggested that EPA
adopt 0.10 mg/1 TTHM as a goal rather
than a regulation while additional data
were being collected and more research
on the health effects of the TTHMs was
being conducted.
EPA believes that a TTHM MCL of
0.10 mg/1 is adequately supported by the
evidence in the rulemaking record
demonstrating that THMs "may cause
any adverse effect on the health of
persons" (Section 1401) and that such a
standard "shall protect health to the
extent feasible, using technology,
treatment techniques, and other means,
which the Administrator determines are
generally available (taking costs into
consideration] on the date of
enactment" of the SOW A, as required
by Section 1412. Although new
information will always be forthcoming
on any regulatory subject, EPA must
make the critical decision of when a
sufficient basis is established to support
regulatory action in order to comply
with the protective intent of the SOW A.
Citing the House Report accompanying
the Act, the United States Court of
Appeals for the District of Columbia
Circuit has noted that "controls were
not to be delayed pending the
development of more refined data on
health effects and more efficient
detection and treatment technology"
(EOF v. Costle, 578 F.2d 337,344 (D.C.
Cir. 1978). As discussed in the preamble
to these regulations, EPA's mandate to
protect the public health to the extent
feasible does not contemplate the mere
establishment of "goals" which utilities
may choose to ignore when the evidence
demonstrates that protective action is
warranted.
7. Ten comments suggested that if a
MCL were to be set for TTHMs, the
MCL should be 0.30 mg/1. Other
comments suggested higher TTHM
MCLs than EPA's 0.10 mg/1 ranging from
0.25 nfg/1 to 15 mg/1. Although most of
these suggested MCLs were offered
without supporting data, two
commenters submitted suggested MCLs
based upon their own studies or
formulas. One commenter suggested a
MCL of 0.3 mg/1 for chloroform based'
upon his studies on dogs, rats and mice
in the laboratory while another
commenter calculated an MCL for
chloroform in drinking water of 0.429
mg/1. Thirty-four comments supported
the proposed MCL of 0.10 mg/1 for
TTHM while 11 comments said that a
MCL of 0.10 mg/1 should be lower but
did not provide supporting data.
In establishing a TTHM MCL of 0.10
mg/1 as an Interim Regulation. EPA has
struck a reasonable balance between
requiring the reduction of TTHM levels
in drinking water to protect the public
health and what public water systems
could reasonably have been expected to
achieve in 1974, taking into account
technological and economic feasibility.
EPA has also been mindful of the fact
that corrective measures taken to
comply with a TTHM MCL have the
potential for adversely impacting the
microbiological quality of a system's
drinking water. Although technologies
are availabje to reduce TTHM levels
below 0.10 mg/1, EPA believes that a
more stringent standard at this time
would unnecessarily jeopardize the
overriding need for quality disinfection.
Moreover, EPA expects that many
systems striving to comply with the
standard of 0.10 mg/1 will, in fact,
achieve lower TTHM levels as well as a
reduction in other potentially harmful
disinfection by-products. Thus, EPA's
approach to the regulation of THMs, as
discussed more fully in the preamble to
the regulations, has been both deliberate
and cautious.
EPA does not believe that a less
stringent MCL is warranted. Based upon
EPA's occurrence data, if a less stringent
standard were established, very few
systems would be required to reduce the
TTHM levels in their drinking water,
resulting in no improvement of water
quality served to their consumers. While
this would relieve many systems from
any costs, it would clearly not further
the protective intent of the SOW A. EPA
has determined that treatment methods
have been generally available since 1974
at reasonable cost to reduce TTHM
levels to 0.10 mg/1, and therefore, a
higher standard would not be justified.
As to those commenters who
suggested that an MCL of 0.3 mg/1 for
chloroform could be computed as a
"safe" level for human consumption by
incorporating an uncertainty factor of
2,000 into Roe's "no observed effect
dose." EPA has concluded that such an
approach is inappropriate when dealing
with human risk from chronic exposure
to a potential carcinogen. That approach
assumes the existence of a threshold
level below which no risk would exist. It
is thus inconsistent with the principles
stated by the NAS in its report.
"Drinking Water and Health". In
addition, 0.3 mg/1 is well above the
levels that are currently achievable in
the large majority of public water
systems by generally available methods
that are technically and economically
feasibje. Roe's study has been
specifically addressed elsewhere in this
Appendix.
-------
68646 Federal Register / Vol. 44, No. 231 / Thursday. November 29, 1979 / Rules and Regulations
8. Sixteen comments responded
specifically to the question of whether
the current information warrants more
restrictive regulations at this time and
how rapidly the MCL could be reduced
to lower feasible levels. Except for one
commenter who said that a TTHM MCL
of 0.05 mg/1 would be technically
feasible today at reasonable cost, the
other 15 commenters all said that a more
restrictive regulation was unnecessary
due to questions regarding the health
basis of 0.10 mg/1. Further, they
expressed serious doubts that a much
lower MCL could be met without
extensive modification in treatment
processes. Several comments
disapproved of the agency's intention to
make the MCL more stringent in the
future, noting that it might be difficult
for water utilities to cope with a moving
target since the economics of system
improvements frequently depend upon
the level of control sought. State
activities would be seriously disrupted
because utilities would have to re-
modify their treatment processes
whenever new standards were set
(modifications would require State
approval), and the States would have to
change their regulations to retain
primary enforcement responsibility.
EPA has already explained its
rationale for not imposing a more
restrictive standard for TTHMs at this
time in its response to other comments
and in the preamble to these regulations.
EPA's health basis for these regulations
is also discussed elsewhere in the
preamble and in this Appendix. EPA
agrees that reducing TTHM levels to
0.05 mg/1 would necessarly result in
increased costs greater than those
estimated to achieve EPA's MCL of 0.10
mg/1; it is, however, EPA's concern for
the potential adverse impact on
disinfection practices and
microbiological quality rather than the
increased cost that has let EPA to
conclude that a more stringent standard
is not justified at this time.
When EPA establishes Revised
Primary Drinking Water Regulations, the
Act clearly authorizes and indeed
requires, more stringent and more
comprehensive regulations of those
contaminants which may have an
adverse effect on human health,
including TTHMs. Congress
contemplated that, as new technologies
were developed to reduce the level of
contaminants in drinking water, EPA's
regulations would be reevaluated
accordingly. Since new information
regarding health effects and treatment
technology will continue to be
generated, it would be unrealistic to
expect that EPA's requirements would
remain static. However, EPA recognizes
the increased burden placed on water
utilities and the States when more
stringent regulations are promulgated;
when this occurs, adequate opportunity
for public comment and time for
compliance with any more stringent
regulation will be provided.
9. On the question of feasibility of
compliance with EPA's proposed TTHM
MCL, three commenters said that more
research is needed to study the
feasibilities of different treatment
processes for the removal of TTHMs.
One expressed the need for EPA's
assistance in evaluating the appropriate
treatment for his system. One suggested
that ozone in combination with a
chlorine residual, when the two are
properly used together as part of a total
treatment scheme, often results in a
significant reduction in the ultimate
TTHM levels. One said that granular
activated carbon (CAC) is good for
TTHM removal as well as taste and
odor.control. One stated that the type of
treatment modification used for
compliance with the MCL should be
determined by the water utility.
EPA believes that despite the ongoing
research being conducted on control of
THMs in drinking water, sufficient
evidence exists to demonstrate that
technology and treatment methods were
generally available in 1974 at
reasonable cost for water systems to
achieve TTHM levels of 0.10 mg/1. Such
methods include both relatively
inexpensive alterations of a system's
disinfecton practices, which will be
sufficient in most cases to reduce TTHM
levels to below the standard, as well as
more complex treatment modifications,
such as those suggested by two
commenters. EPA's findings regarding
the feasibility of TTHM control are fully
set forth in the report "Interim
Treatment Guide for the Control of
Chloroform and Other
Trihalomethanes," which has been
incorporated by reference as part of the
Agency's Statement of Basis and
Purpose for these regulations.
A1978 report prepared by J. S.
Zagorski, G. D. Allgeier and R. L.
Mullins, jr., "Removal of Chloroform
from Drinking Water," studying the
reduction of chloroform formation upon
subsequent chlorination, reported that
various common treatment processes
including sedimentation; sedimentation >
followed by chemical coagulation and
precipitative softening; sedimentation,
chemical coagulation, precipitative
softening and rapid sand filtration; and
sedimentation followed by chemical
coagulation, precipitative softening,
rapid sand filtration and GAC
adsorption resulted in substantial
reductions of the chloroform formation
potential. They also reported that both
alum and polymers at moderately large
dosages were capable of reducing the
potential of Ohio River water to form
chloroform and other THMs. Both
ozonation and powdered carbon at high
doses also reduced THM formation
potential. In the plant-scale studies, the
same investigators also reported that
moving the point of chlorination from
the head of pre-sedimentation reservoirs
to the head of the coagulation process
significantly reduced the concentration
of CHCI3 in finished water, and that
ammoniation at the head of precipitative
softening ceased the THM formation
reaction and markedly reduced the level
of THMs in softened water. Aeration
also was able to reduce chloroform in
finished water.
As explained in EPA's response to
other comments and in the preamble, in
light of currently available information,
EPA need not wait for the results of
additional research before establishing
regulations to control TTHMs. Rather,
any new information will be considered
by EPA when it develops Revised
Primary Drinking Water Regulations.
EPA agrees with the comment that the
type of treatment modification used to
comply with the TTHM MCL must be
determined by the water utility that has
the ultimate responsibility to select a
method for achieving compliance. Many
commenters appeared to erroneously
confuse the TTHM regulation with
EPA's proposal of a specific treatment
technique for control of pollution-related
synthetic organic chemicals in drinking
water. Nevertheless, technical
assistance will be provided by EPA and
the States on a case-by-case basis.
Systems that modify their treatment
processes to comply with the TTHM
MCL are also required to obtain State
approval of their plans prior to
implementation to insure proper
supervision of alterations in disinfection
practice.
Significant reductions in THMs can
normally be achieved by making
relatively minor modifications to
existing water treatment systems, such
as maximizing the efficiency of
precursor removal during coagulation/
filtration or changing the point of
chlorination. Where minor modifications
to existing treatment methods prove
insufficient to bring the system into
compliance with the MCL, the system
may need to use an adsorbent
technology, such as GAC, to reduce
precursors and thereby achieve
compliance with the MCL. Thus, each
system will probably be using a
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations 68647
combination of the available treatment
options that will be most effective for its
situation. Because of these treatment
alternatives, total reliance upon an
adsorbent for reduction of the THMs to
below the MCL will not likely occur. The
EPA has estimated that of the
approximately 2,700 systems serving
more than 10,000 people required to
comply with the MCL, approximately 25
systems may ultimately need to install
adsorbent technology to control THMs.
10. One commenter stated that GAC
has never been tested or proven in full-
scale operation in the United States and
therefore constitutes a nationwide
experiment in water treatment.
The availability and efficacy of GAC
technology has been clearly
demonstrated by the large extent of use
by numerous facilities in the United
States as well as overseas. GAC
technology has been used for many
years in the water treatment industry,
and today over 60 drinking water plants
presently use GAC in their treatment
facilities. Extensive use of GAC is
practiced in the food and beverage
industry for removal of organic
contaminants from process waters and
in the treatment of industrial and
municipal waste waters prior to
discharge to receiving waters. GAC for
removal of organic chemical
contaminants has been in use by
numerous European municipal drinking
water plants since the 1960's and as
industrial activity continues to increase,
more facilities using CAC are being
installed.
Most drinking water plants in the U.S.
have been using GAC as a replacement
for the media in their existing filters for
the stated purpose of removal of taste
and odors. However, with the
development of more sophisticated
analytical procedures which are capable
of detecting and measuring levels of
organic chemicals (including THMs and
THM precursors] in drinking water, EPA
now knows that such chemicals are
actually being removed by GAC and
that their presence, previously
undetectable by analytical
measurement, was being manifested
through taste and odor problems.
Commenters nevertheless question
the availability of means for the
regeneration of GAC and use of GAC in
post contactors for removal of organic
compounds. Regeneration of GAC has
been demonstrated in numerous
locations including a full scale operation
at a drinking water facility in the late
1960's in the U.S. Some European
drinking water plants have also been
regenerating GAC for several years. The
frequency at which drinking water
plants in the U.S. replace the GAC
ranges from less than six months to two
to three years. The GAC is usually
removed from the facility and replaced
by virgin carbon.
In addition to its use by numerous and
varied types of drinking water systems
in the U.S. and overseas, GAC has been
widely and successfully used for the
treatment of municipal waste waters for
removal of organic chemical pollutants.
For example, since the mid-1980's, the
municipality of Lake Tahoe has used
GAC in contactors with on-site
regeneration. Thus, regeneration
technology has been applied both on
site and at central furnace facilities.
Frequency of regeneration will
necessarily be dependent upon TTHM
reduction needed on a case-by-case
basis. Numerous drinking water
treatment plants are presently operating
modules of full scale GAC systems or
pilot plants to more fully correlate GAC
performance with various regeneration
frequencies.
11. One commenter stated that the
GAC treatment process may result in
serious problems and these may
outweigh the alleged environmental
benefits associated with GAC treatment.
These problems include potential air
pollution from regeneration and the
waste water associated with air
pollution scrubbers as well as waste
water from backwash and drainage from
carbon slurries.
GAC is normally regenerated at
furnace temperatures of 750° C to 900* C
and at these temperatures, data do not
show that most pollutants are oxidized
to other than harmless compounds. EPA
has considered potential waste disposal
problems including air and water
pollution relating to GAC reactivation
and has found that techniques are
available to control wastes from these
facilities.
In regard to discharge of backwash
water or drainage from carbon slurries
(if at the water treatment plant), no
additional water is expected to be
necessary. In fact, less water is normally
used in backwashing with GAC than
with conventional media in the filter.
Any drainage from carbon slurries at the
off-site GAC regeneration facility is not
large in volume and normally is
discharged to municipal treatment
plants.
12. Several commenters were
concerned that the use of GAC may
constitute a larger health hazard than
means for improvement of water quality.
The alleged health hazards associated
with GAC included desorption.
chromatographic effect (competitive
displacement], resorption (leaching) of
heavy metals and polycyclic aromatic
hydrocarbons contained in the virgin or
regenerated carbon, release of carbon
fines, promotion (catalytic reactions) on
the carbon itself of hazardous
compounds due to chemical reactions
between chlorine and organic
compounds, bacterial growth on the
carbon and air pollution from
regeneration facilities. Commenters also
noted that indirect hazards were
associated with GAC usage through the
manufacture of GAC and the production
of energy necessary to operate GAC
facilities. They said these industries,
such as the coal industry, pose a high
risk of morbidity and mortality to the
workers. Because of these concerns,
they urged that additional research and
testing should be conducted prior to
implementation of GAC in this country's
major waterworks. It was suggested that
toxicological evaluations be conducted
using concentrated effluents from GAC
to assess these potential hazards.
EPA has evaluated the potential
hazards associated with the use of GAC.
The items listed can be shown to occur
under specific laboratory conditions
directed at obtaining a specific'reaction,
such as the promotion reaction or the
chromatographic effect, but no
significant hazard is expected under
actual use conditions so long as proper
operating procedures are followed. For
example, use of GAC for THM control
will not result in desorption of TTHM to
levels above the MCL since the GAC
would be regenerated at the point where
THM levels in the effluent approached
those in the influent. Also, bacterial
growth on GAC is common, is frequently
encouraged by adding oxygen to the
influent waters, and assists in reduction
of precursor compounds. Control of
bacteria in the finished drinking water is
effectively accomplished by disinfection
and the alleged slugs of bacteria
breaking through the GAC do not occur
with proper operation; in any event.
proper disinfection with a residual
throughout the distribution system
would eliminate this potential hazard.
In addition, present data have not
shown a health hazard associated with
the use of GAC in its many applications
in drinking water treatment.
Nevertheless, EPA is continuing to
conduct research on these questions. For
example, short term bioassay studies
are being conducted with animals using
concentrated raw and finished waters to
assess the toxicological significance of
various disinfectants, such as chlorine
and ozone, and the use of various
treatment technologies, including GAC.
However, the methodologies used in
these studies are only now being
developed and must be verified by more
established methods.
-------
68648
Federal Register / Vol. 44. No. 231 / Thursday. November 29. 1979 / Rules and Regulations
13. Twenty-seven comments were
received discussing the proposed
effective date of the TTHM regulations.
In general, the commenters thought that
the compliance dates for either the
monitoring requirement or the MCL
were unreasonable. A number of these
commenters had apparently confused
the effective date for the TTHM
regulations with that for the treatment
technique requirement and commented
accordingly.
Specifically, 11 commenters said the
allowed time for compliance with the
proposed regulations was unreasonable
without specifically referring to whether
the comment was addressed to the
monitoring schedule or the MCL. Nine
commenters, however, submitted
specific time-tables that they felt would
be required for compliance with the
proposed TTHM regulations ranging
from monitoring beginning 3 months
after promulgation of the regulations to
as long as 8 years for the completion of
plant modifications.
One commenter submitted his
suggestion of a specific time-table
including the following: (1) Request for •
variance or exemption should be
submitted no later than the effective
date, (2) design specifications should be
submitted to States for approval no later
than 18 months after the effective date,
(3) by no later than 24 months after the
effective date, final design plans and
specifications should be submitted to
States for approval. (4) construction
should be completed and operation
should begin no later than 4 years after
effective date, and (5) operational data
should then be submitted to States for
evaluation. One commenter suggested
postponement of the regulations and
instead conducting a two-year
comprehensive monitoring program. One
commenter felt that the proposed time-
table of the TTHM regulation was
adequate.
Thirty-four commenters said that
EPA's proposed effective date, allowing
18 months for compliance, was
unreasonable and that it was technically
impossible for systems to design the
most cost-effective treatment system
within that timeframe. These comments
suggested allowing additional time for
compliance, ranging from 3 to 7 years.
Four thought the allowed time of 18
months was adequate. Three said the
regulation should be more flexible with
regard to the time for compliance and
the type of treatment modification used
and suggested that the States make
these decisions. One commenter said
that the allowed 18 months was
adequate if only minor modifications
were needed but that additional time
would be required if major changes to
the treatment plant were needed.
Another commenter said that whether
the allowed timing was adequate would
depend upon whether the particular
water system would need to use GAC to
remove TTHMs. One stated that the
primacy States should have a minimum
of two years to revise their regulations
to be consistent with the regulations
finally adopted by EPA before they
became effective requirements for the
water supplies. One commenter said
that although the proposed timing was
feasible, in most cases, the final
regulations should provide for a delay in
the effective date for systems that could
show the need for additional time. One
commenter said that the proposed
compliance schedule was appropriate if
the MCL were established at 0.30 mg/1.
EPA has responded to the comment
seeking more time to achieve
compliance by extending the effective
date of the TTHM MCL for systems
serving more than 75,000 people to two
years after the promulgation of these
regulations. Systems serving between
10,000 and 75.000 people have been
given four years to achieve compliance
with the MCL. Both dates take into
account the need for one year of
monitoring data to be established and
the need for adequate time to develop
quality laboratory capability for TTHM
analyses. The two-year effective date of
the MCL for the first size category also
serves to provide primacy States with
sufficient time to amend their
regulations before the MCL takes effect.
In the meantime, EPA will not allow
State primacy to be needlessly
jeopardized. The Agency will be
proposing regulations shortly as
amendments to 40 CFR Part 142 which
will allow for a reasonable amount of
time for States to conform their
regulations to the federal requirements.
The extended timeframes suggested
by some commenters do not appear to
be warranted for applicability to all
systems. It appears that these
commenters may have been erroneously
assuming that CAC was being required
for control of TTHMs in all cases. On
the contrary, EPA believes that most
systems will be able to achieve
compliance with the TTHM MCL of 0.10
mg/1 with relatively minor changes to
their existing treatment processes.
Therefore, the timeframe provided in the
final regulations should provide ample
time for compliance measures to be
implemented. However, EPA recognizes
that additional time may be needed by
those few systems that will need to
institute more complex treatment
modifications to comply with the TTHM
MCL In such cases, Section 1416
normally provides for the issuance of
exemptions. Due to the belated issuance
of these amendments to the Interim
Regulations, an extension of the
compliance deadlines presently
established in Section 1418 will be
needed to authorize exemptions from
the TTHM MCL EPA will seek a
legislative extension of the exemption
deadline. So long as good faith efforts
are being taken by systems to comply
with the TTHM MCL EPA and the
States may exercise their enforcement
discretion to insure compliance as
expeditiously as practicable.
14. Seventy comments addressed the
specific cost estimates for installation of
the technologies as well as the projected
national cost impacts of the regulations.
The majority said that EPA's estimates
were not reasonable and that the actual
costs would be considerably higher. A
few comments felt that the costs were
reasonable or "in the ball park."
Of these comments, 32 stated that the
costs for installation of the technologies
were low while five thought that the
estimates were reasonable. Some of
these felt that the EPA estimates in most
cases did not conform to local economic
conditions. Other commenters said the
EPA's costs were underestimated and
submitted cost estimates for their
particular utilities in support of their
argument. They indicated that
compliance with the MCL would require
far larger investments by the utility than
those estimated by EPA. In addition, one
commenter provided data showing that
the cost impacts would be higher
because his public water system used
225 gallons per capita per day (gpcd) as
compared to the 179 gallons per capita
per day used by EPA in the estimates.
The commenter also used maximum
'daily and hourly flows of 240 percent
and 390 percent of average daily flows,
respectively, and 65 percent of the total
year's flow occurred during the four
summer months.
EPA's analysis of the cost and
economic impact of the final regulation
is discussed in the preamble and
described in detail in the "Economic
Impact Analysis for a Promulgated
Regulation for Trihalomethanes in
Drinking Water". The costs of treatment
are based upon average national costs
and were determined from an analysis
of the costs of materials and labor rates
in various parts of the United States.
The costs of treatment represent those
of an average size utility in each of
several size categories, and serve as the
basis for assessment of the national cost
impacts. It is expected that some
utilities would experience costs that are
-------
Federal Register / Vol. 44. No. 231 / Thursday. November 29, 1979 / Rules and Regulations 68649
higher than the average system in its
size category, while others would be
lower. In order to reflect site-specific
factors for a utility, contingency factors
are incorporated into the treatment cost
estimates.
The base flows used in the cost
analysis are values representing the
average flow conditions for a certain
size range of systems. The values are
based upon a recent survey of 1,000
water systems in the United States
during which it was determined that
larger systems have higher water usage
per capita than do smaller systems. This
is a result of commercial and industrial
customers. Thus, a different flow base
was used for each size category ranging
from 155 to 210 gpcd for systems serving
one million persons or more. Capital
costs were based upon capacity flows
and O&M costs were based upon
average daily flows. The exception was
that capital costs of GAG were based
upon the average day in the peak month
which was less than the capacity flow.
Commenters are referred to EPA's
document "Economic Analysis" for
further details.
15. One comment noted that it was
difficult to determine whether EPA's
estimates of the cost for compliance
were reasonable. He felt that debt
service, the additional water treatment
plant personnel laboratory assistance
and control, and more sophisticated
monitoring equipment, were not
adequately considered. One commenter
stated that it would cost $20,000 to
$30,000 per year to conduct monitoring
for his utility. Four said that the
compliance cost for TTHM analyses
estimated by EPA at $25 per sample was
low and that the current rate for
commercial TTHM analyses was
approximately $100 per sample
exclusive of sampling and delivery
costs. Two other commenters suggested
that prices of $75 and $120 per sample,
respectively, were appropriate. Three
commenters agreed with the EPA's
estimation of monitoring costs.
EPA's analysis of the costs of
treatment specifically considered each
of the items of concern to the
commenter. Debt service is included in
the annual costs (revenue requirements)
and includes interest rates on capital of
8% and 10% for public and privately-
owned utilities, respectively. The rate
for privately-owned utilities was revised
from the 9% rate used in the cost
estimates supporting the proposed
regulations to take into account the
current and projected cost of capital.
Additional plant personnel were
included in the O&M costs and thereby
In the annual costs.
In regard to monitoring costs, the total
required monitoring costs were
estimated to be $800 per year per system
based upon four samples per quarter. As
noted in the preamble, monitoring costs
for some systems will be higher than
$800 per year because these systems
have more than one plant, thereby
necessitating (in some cases) additional
sampling. This cost estimate included
costs of analysis at $50 per sample. The
cost of sampling and mailing samples to
an outside laboratory was not
considered to be significant. No
additional sophisticated monitoring
equipment was included in the estimate;
however, it was anticipated that many
systems would purchase analytical
equipment to perform their own
analyses. While commercial rates for
TTHM analyses varied from $25 per
sample to more than $100 per sample,
$50 was used as a reasonable estimate
and this was increased from the value of
$25 per sample used in the proposed
regulations. However, it is expected that
the cost per sample will likely be lower,
since increased availability of analytical
services, competition between
laboratories and the increased number
of samples for analyses will provide •
opportunities for cost-savings.
In addition to the costs associated
with the required monitoring, additional
costs will be incurred by some systems
in the monitoring conducted to assure
that the bacteriological quality of the
drinking water will be maintained
during and after treatment modifications
for the purpose of reducing THM levels.
Costs of this monitoring will vary
between systems but will not likely
exceed approximately $5,000 at systems
with the most extensive monitoring
program. This estimate was based upon
use of outside contract laboratories, and
it is expected that most water systems
will conduct some of the analyses in
their own laboratories, thereby reducing
the costs. Nevertheless, this cost is
considered reasonable for those systems
which will need the most extensive
monitoring (e.g., for systems serving
10,000 people, this cost would be $0.50
per person), and is a one-time expense
(as opposed to continued requirements
for quarterly TTHM monitoring).
16. One commenter said that the use
of a forty-year amortization period to
determine the yearly cost for capital
improvements was unreasonable in that
the life of the water treatment facility
would be considerably less than 40
years.
Forty years was used as
representative of the average expected
life of equipment in public water
systems. While some equipment may
require replacement sooner than 40
years, other equipment has a life greater
than 40 years. While privately-owned
utilities often depreciate equipment at a
20-year rate, this is primarily for tax
advantages and does not represent the
true life of the equipment. Publicly
owned utilities most often use rates of
approximately 40 years since no tax
advantages are available. Since over
80% of water systems are covered by
this regulation and are publicly-owned,
it is reasonable to use the 40-year
amortization period as the basis of
annual costs.
17. One commenter said that EPA's
use of $5.58 per hour for labor in its EPA
cost estimates was too low, stating $7.00
per hour for labor cost would be more
appropriate. The cost estimates have
been revised and now include labor
costs at $11.75 per hour including fringe
benefits. In addition, it should be noted
that contrary to the commenter's
statement, the proposed regulations
were based upon an average labor cost
of $7.50 per hour.
18. A number of commenters argued
that the costs were underestimated
because of specific factors in the
analysis. For example, one commenter
stated, based upon the use of GAG. that
the difference between his potential
national cost estimates and EPA's
estimates could be explained primarily
by four factors. It was not clear to what
extent these, comments differentiated
between costs for GAG for TTHM
control and costs for GAG to control
other synthetic organic chemicals in the
separate treatment technique
requirement. The four specific areas of
difference noted by this commenter and
EPA's responses are as follows:
(a) EPA determined its estimated
capital costs for a system based upon
the capacity of the entire system:
whereas, the commenter estimated the
system capital costs as equal to the sum
of the capital costs for each treatment
plant based on the capacity of each
plant.
The EPA recognizes that several large
public water systems use more than one
treatment plant and thereby might be
required to install necessary treatment
at each plant if they utilize the same or
similar source waters. Due to the
limitations of available data, the cost
estimates were based upon installation
of treatment for the total flow capacity
of each water system, rather than
separate flows from each plant. EPA
does not believe that per plant costs
would significantly affect the national
cost estimates. Treatment costs depend
upon flow capacity whether apportioned
per plant or taking the system as a
whole. In some cases, costs could be
-------
68650
Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations
reduced if only the flow from a single
plant required treatment to reduce
TTHMs. These effects have been taken
into account by including contingencies
in the cost estimates. Moreover, since it
is generally the larger systems that have
multiple plants, additional costs of
treatment will be borne by a greater
number of customers, reducing the per
capita impact.
(b) EPA's estimates were based upon
the system capacity on the average day
of the peak month; whereas, the
commenter's estimates were based upon
the actual capacity of each treatment
plant.
As presented in the cost analysis and
discussed above, costs were determined
based upon system capacity except for
the use of CAC which was based upon
the average day in the peak month. This
was determined to be an appropriate
cost base rather than total plant
capacity because compliance, with the
MCL will be based upon a running
annual average of average quarterly
monitoring results and not a peak value.
(c) EPA assumed that some of the
affected systems would design facilities
for a 9-minute empty bed contact time
(EBCT); whereas, the commenter
assumed that all GAC facilities would
be designed for an 18-minule EBCT.
It is anticipated that for most systems,
9 minutes EBCT will be adequate to
achieve the MCL It is possible that
certain systems may require additional
contact time but use of an average
condition is entirely appropriate in the
development of a national cost estimate.
Use of 18 minutes EBCT as the base of
the national cost estimate would have
inflated the costs unrealistically.
(d) The commenter's estimates for
specific systems, based on the costing
out of the individual components, were
30-80% higher than EPA's proposed
estimates.
As stated previously. EPA's cost
estimates have been substantially
revised to take into account many of the
commenter's concerns. The cost
estimates have been based upon the
most accurate and recent sources of
information and cost data available and
that have been reviewed within the
industry. Differences between the
commenter's costs and EPA's proposed
cost estimates were primarily due to
differences in the base year for the
estimates (EPA was 1978 dollars and the
commenter was 1978) and differences in
EBCT (9 minutes vs. 18 minutes). In any
event, the commenter's detailed
estimates have been evaluated and the
EPA estimates have been revised
appropriately.
The commenter's O&M cost estimates
were higher than EPA's primarily
because they were based on expenses at
multiple treatment plants. Certain
specific costs, such as the price of GAC
and fuel costs, also account for portions
of the differences and have been revised
in the final cost analysis. EPA's GAC
costs were based upon current and
projected costs and ranged from $0.65 to
$0.84 per pound of GAC depending upon
the size of the public water system. Fuel
costs were also projected and included
estimates for 1980 of $0.84 per gallon for
diesel fuel, $0.0038 per cubic feet for
natural gas, and $0.038 per kilowatt-hour
for electricity. The commenter's revenue
•requirement estimates were higher than
EPA's primarily because of the higher
estimates of capital and O&M costs.
19. Two comments stated that the
costs were understated because the
increased demand for materials required
to comply with the regulations would
cause costs to rise beyond normal
inflation rates. This concern has been
evaluated and, as shown in the
economic analysis, no single chemical or
component of any of the available
treatment technologies is expected to
experience a sufficiently large demand
so as to affect its price. For example, the
initial demand for GAC (to meet these
regulations] is estimated to be four
million pounds whereas the industry has
excess GAC capacity of more than 100
million pounds per year.
20. One commenter stated that the
EPA estimates did not include costs for
land that would be necessary for
installation of the GAC facilities. As
shown in the economic analysis, costs of
land acquisition were included in the
capital cost estimates.
21. One commenter indicated that the
EPA's estimates were based upon 1976
costs. He felt that approximately 20
percent increase was needed just due to
elapsed time to date (1978) and that at
the time of construction of needed
facilities, another 50 percent inflationary
increase would be applicable.
EPA's costsjiave been revised to
reflect anticipated use of 1980 dollars to
meet the regulations. The estimates
were increased to 1980 dollars through
the use of the available cost indices
which included separate indices for
labor, steel, excavation, concrete,
manufactured equipment, pipes and
valves, electrical and instrumentation,
housing, and producer prices. These
indices took into account anticipated
inflation to 1980 and the precise index
values are presented in the economic
analysis and supporting documents.
Overall, unit costs have been increased
by approximately 36% as a result of this
change from 1976 to 1980 dollars.
22. A number of commenters stated
that the use of GAC will have
substantial financial impact upon water
supplies and that actual costs are very
difficult to predict and are understated.
For example, the average capital cost for
a system serving over one million people
was alleged to exceed $106 million with
annual costs of more than $23 million.
These commenters estimated that rate
increases for residential customers
would be in the range of 40-70% and
that these rates could double where
there were site-specific problems, such
as land acquisition. The commenters
claimed that these costs may result in
insurmountable problems at some
utilities in obtaining financing for GAC
treatment facilities. They charged that
EPA's assessment of the feasibility of
financing the GAC treatment facilities
was totally out of step with the realities
of both the financing markets and
operating needs of the public utilities.
Costs for GAC treatment are highly
dependent on the substances being
removed and the target level in finished
water. The use of GAC to control THM
precursors would not require the most
stringent design and operating
characteristics in most cases. Thus, the
cost for this application would likely be
very much less than the cost for using
GAC to control synthetic organic
chemicals. As noted in the preamble,
EPA's cost estimates for using GAC for
TTHM control were revised from those
costs supporting the proposed
regulations. For purposes of the
economic analysis supporting this final
THM regulation, EPA estimated the
costs for a system using GAC by
replacing its existing filter media with
GAC and regenerating its carbon no
more frequently than once every 12
months. Only systems with severely
contaminated raw water sources will
require the extent of GAC usage that the
estimates accompanying the original
proposal were based upon (post-
filtration contractors with two month
regeneration cycles). The data indicate
that in most cases the raw waters were
relatively uncontaminated and this was
used in determining feasibility of
treatment and reasonableness of costs
for purposes of establishing the MCL
Thus, the revised costs are significantly
lower than those in the economic
analysis of the proposed regulations. Of
course, the economic impact analysis is
based upon a specific model system and
costs will vary depending upon specific
details at each site. To a reasonable
extent, site-specific factors were
included in the revised analysis and
EPA's supporting economic document
should be consulted for details. The
document also examined the feasibility
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68651
of financing and found that financing is
available.
23. Nine commenters said that the cost
estimation should be more realistically
based upon results from controlled
experiments such as field studies. As the
commenters suggested, one of the
primary factors considered by EPA in
developing the cost estimates has been
the engineering application of the
available treatment technologies. EPA
has revised its cost estimates to reflect
the engineering costs developed by
Gulp, Wesner, and Gulp, consulting
engineers with extensive experience in
water treatment technology.
24. One commenter stated that the
costs for GAG did not include the
investment necessary for disposal of the
concentrated organics removed from the
off-gases by either landfill or
underground injection. The cost
estimates for use of CAC are based
upon off-site regeneration, and all
aspects of regeneration of GAG,
including disposal of scrubber waters
and other waste products were taken
into account.
25. Two commenters stated that the
cost estimates were low because EPA
did not include the costs of installation
of conventional treatment (coagulation.
sedimentation, filtration] followed by
THM control. The commenters indicated
that some water supplies use sources
from such places as the Adirondacks
which do not necessitate conventional
filtration but have TTHM levels at 150 to
250 mg/1. One of the commenters stated
that for his system, which serves 140,000
people, to meet the MCL filtration would
have to be installed at a capital cost of
$12 million, an annual cost of $1.3
million, and a rate increase of 60
percent.
Most public water systems use
conventional treatment technology and
thus EPA's cost estimates included only
those treatment technologies that are
additions or adjustments to such
conventional treatment. It would not be
appropriate to include the costs of
conventional technology in these
regulations since, in most cases,
compliance with other requirements of
the NIPDWR (e.g., turbidity) necessitate
use of conventional treatment.
Therefore, the cost of conventional
treatment should not be directly
attributable to this regulation.
Nevertheless, many of these systems are
expected to be able to comply with the
regulations through adjustment of
chlorination procedures or use of an
alternate disinfectant.
26. One commenter stated that the
economic impact assessment did not
take into account the costs of treating
waste water from GAG operations, such
as backwash waters, wet scrubbers and
drainage from carbon slurries. It was
estimated that 50,000 gallons of waste
water will be generated for every one
million gallons of drinking water treated
and half of that amount would need to
be discharged. This commenter
concluded that this would result in
increased flows and an approximate 4%
increase in operation and maintenance
costs at municipal waste water
treatment facilities.
EPA's estimates did take into account
disposal of any additional waste waters
from the use of GAG. For example, the
cost estimates were based upon
regeneration of carbon al an off-site,
privately owned, regeneration facility.
The costs of regenerated carbon utilized
in the estimates were based upon actual
manufacturer's estimates and operating
rates. Overall rates included costs of
GAG regeneration and all ancillary
activities such as air pollution control
and disposal of waste waters.
27. Several commenters stated that
the estimates were low because EPA did
not include the administrative,
environmental, overhead, and political
costs of implementing the regulations.
Two of these commenters felt that
additional dollars would be required for
such items as cost of processing
variances, public hearings, research
costs into health and treatment aspects
of the regulations, monitoring
compliance, laboratory instrumentation
and facilities, and laboratory
certification programs.
The Agency agrees that each of the
above items has some degree of costs
associated with it and has taken
appropriate costs into account in the
revised cost estimates. Systems would
not be expected to conduct research into
the health aspects of the regulations,
and only research into treatment aspects
to the extent necessary to determine
which treatment would be most
effective in meeting the MCL. Costs
attributable to administrative or legal
(or political) factors, processing.
variances, and public hearings are
difficult to precisely estimate. They have
been included in appropriate parts of the
estimate. Thus, administrative and legal
costs have been included in the
engineering costs at a rate of 12% of the
total treatment cost. Some of the
overhead costs have been included in
the O&M costs which include labor rates
with fringe benefits. Further, costs
associated with monitoring have been
included in the monitoring costs;
environmental costs have been
considered in GAG regeneration costs
which would take into account such
items as air pollution control equipment
and disposal of by-products; finally, any
other costs not included in those
components of the total cost have been
included in the contingency added to the
costs.
28. One commenter said that the costs
associated with the treatment cost
analysis were inflated. He slated that
the cost analysis was based upon
NOMS data which averaged values of
THM concentrations measured in over
100 finished water supplies across the
United States. The commenter believed
that the cost analysis should have been
done in two phases: one for summer
conditions and one for whiter, using
quenched values for all 117 cities, and
measured at the point in the distribution
system most distant from the source to
accurately measure the THM
concentration reaching the consumer.
EPA's national cost estimate has been
based upon NOMS which is the most
recent available data base with regard
to the levels of THMs in finished
drinking water supplies. Certainly a
more refined and extensive survey
would provide a higher degree of
confidence for its estimates; however.
for the purposes of assessing the
national cost impact of these
regulations, the NOMS data base waa
felt to be a reasonable representation of
THM occurrence.
29. One commenter estimated that the
cost to the consumers in his system
could increase 50 to 75 cents per 1,000 '
gallons and the needed treatment
modifications would also result in
reducing his filter capacity up to 70
percent. Other estimated rate increases
reported by several commenters reached
as high as 120 percent, while it was
slated by one commenter that a 5.4%
increase would be necessary for his
utility.
EPA's projected national capital
expenditures total $85 million in 1980
dollars resulting in a overall rate
increase of 2% which is a considerable
reduction from EPA's original estimates.
EPA's original estimates were $154
million (1976 dollars), equivalent to $210
million in 1980 dollars, and included
only those impacted communities larger
than 75,000 population. EPA's revised
cost estimates now include those
communities between 10,000 and 75,000
population, and assume that a total of
515 water systems would be required to
institute some type of change in current
processes. Fewer systems are expected
to use the more expensive treatment
technologies. Available technologies
range from no-cost or very low cost
changes such as improving coagulation
or moving the point of chlorination (172
systems estimated], to low to moderate
cost changes, such as modification of
-------
68652 Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations
the disinfectant (319 systems estimated),
to high cost changes, such as use of an
adsorbent like GAG [24 systems
estimated).
EPA restructured its decision tree
based upon several factors regarding the
treatment technology alternatives that
are available to meet the MCL and the
number of systems by size that would be
likely to modify or install treatment
because they exceeded (he MCL It is
not anticipated that the existing Filter
capacity, as suggested by the
commenter, would be reduced by
application of these technologies. These
projections have been derived based to
a large degree upon information
received during the comment period. For
example, considerably wider use of
chloramines and less usage'of GAG is
expected to be selected to reduce THMs.
Primarily, for those reasons, the cost
estimates have significantly changed,
and the typical costs per family (i.e.
residential bill increase) are expected in
the range of $1.40 per year. In those few
cases (24) where GAG is necessary,
costs per family have been estimated to
be up to $11.20 per year, less than $1.00
per month. After review of existing
rates, rates for other utilities, and the
specific costs involved, EPA does not
believe that such increases will have an
unreasonable impact on a family budget.
30. Twenty commentere thought that
the monitoring costs were excessive for
the water utilities to pay and they felt
that the federal government or EPA
should conduct or fund the monitoring
program. One questioned whether
Federal funds would be available to
assist in the additional financial burden
of the regulations. However, another
stated that no federal grants should be
issued to public water systems because
of their prior record of providing
services and supporting themselves from
their own resources.
Monitoring costs required by these
regulations amount to approximately
$800 per system per year. These costs
are not considered to be excessive; for
example, minimum cost per capita for
monitoring for systems serving 10,000
people will be $0.08 per year and for
systems serving one million people,
$0.0008 per year. As noted above, the
costs associated with this regulation
generally are not significant and federal
financial assistance should not be
needed in the size range covered by this
regulation. If it is needed, federal
financial assistance programs are
available for public water system
improvements. It is also probable that in
many cases the States may provide
analytical services for their
communities.
31. One commenler was concerned
that compliance with the regulations by
systems that will require major
modifications would be difficult because
of the economic and social burden; the
commenters also questioned how the
regulations relate to the President's
urban policy. Several commenters were
concerned that the burdens of increased
water rates would be difficult for those
least able to afford it; that is, low
income and high unemployment groups,
minorities, and retirees. One felt that the
required rate increases for both normal
system maintenance and to meet the
regulations might not be supported by
the customers, concluding that this could
eventually result in deterioration of the
water supply facilities because the cost
of meeting the regulations would take
needed capital away from maintenance
type programs. One felt that the cost of
the regulations would take money away
from the needy and could result in
poorer and less nutritious diets.
Because of the relatively low costs
associated with these regulations, the
impact on consumers' other needs are
not considered to be significant. EPA
believes that providing healthful
drinking water must be a high national
priority and that these regulations do
not conflict with the President's urban
policy.
32. A commenter said that it was not
clear that GAG would effectively reduce
TTHM concentrations more than
movement of the chlorination point or
changing disinfectants; the choice of
installing GAG filtration by water
treatment plant managers might produce
only slight reduction in TTHM
concentrations at a very high cost and
therefore might not be a feasible
alternative.
EPA estimates that GAG will only be
used by about 25 systems to comply
with the MCL because less expensive
technology alternatives are available,
such as changing the point of
chlorination or using an alternate
disinfectant. For these 25 systems, it is
expected that a comprehensive
evaluation of the existing treatment will
be made to determine the most cost-
effective technique for compliance with
the MCL. These systems will most likely
use a combination of the alternative
treatments, such as changing the point
of chlorination or maximizing
coagulation/filtration efficiencies. Use
of GAC for TTHM control has been
found to be effective for not only
reducing precursor compounds which
contribute to TTHM formation, but also
to some degree for removing THMs once
they are formed.
33. One commenter felt that increases
in State program grants would be
necessary for States to implement these
regulations.
These requirements are not expected
to be an undue burden upon State
programs. Implementation of these
regulations will require State review ana
approval of proposed plans for
treatment modifications for
approximately 515 systems. Because of
the relatively small number of systems
within each State, the phasing-in of the
two population segments, and the fact
that, for the most part, minor
modifications will be necessary, this is
expected to be accomplished with
minimal disruption to existing State
programs. Further, many States already
review system plans for any
modifications to existing treatment.
Compliance monitoring will also be
required but this will only be a minor
addition to the system already in use by
State programs for checking compliance
with the N1PDWR in effect.
34. One commenter stated that EPA
underestimated the costs of
implementing the regulation by
underestimating the number of impacted
systems. This commenter disagreed with
EPA's use of a specific model for the
water supply industry, assumptions
regarding the number of systems that
purchase water and use alternate
disinfectants, and assumptions and
predictions based upon NOMS for
determining the level of THMs and if
systems would be impacted. Instead,
they said EPA should have conducted
sampling at all systems and based its
estimates upon those results. They
further commented that EPA's estimate
of 390 systems serving greater than
75,000 persons was not derived from
EPA's Inventory of Systems but was
based upon a policy testing model which
left out numerous systems including all
Federal Systems (e.g. District of
Columbia) and the States of Hawaii and
Alaska. They criticized EPA for not
confirming the hypothetical results of
the model with empirical data. Finally,
they said EPA's assumptions regarding
the number of systems using specific
treatment systems such as GAC or no-
cost modifications were arbitrary.
EPA has based its assumptions
regarding the number of public water
systems upon the actual inventory of
water supply systems in the U.S. as
ascertained in the Federal Reporting
Data System (FRDS), and thus the
number of systems is as accurate as
possible. Certainly, surveys at every
plant in the U.S. as suggested by the
commenter would provide actual results
rather than an estimate of TTHM levels.
but NOMS is considered to be a valid
representation of national exposure
-------
Federal Register / Vol. 44. No. 231 / Thursday. November 29. 1979 / Rules and Regulations 68653
levels. NOMS is the most recent and
extensive data base and is adequate for
estimating national cost impacts. In
regard to disinfectant use. EPA based its
estimates upon an EPA national survey
in 1976 of drinking water plant
operations. The determination of the
number of systems that are expected to
use specific types of treatment was
discussed in the preamble and are
reasonable estimates based upon the
TTHM levels and available
technologies. Finally, the commenter
was unfamiliar with the policy testing
model which the Agency uses to support
economic and financial analysis. A
description of this model is presented in
Appendix A of the economic analysis
document. It is used only to generate the
aggregated costs and financial impacts,
based upon inputs from treatment cost
data, water supply inventory data, and
water supply operating characteristics
data.
35. One commenter stated that the
EPA should provide a cost estimate of
the stated goal of lowering the MCL at a
later time to 50 ppb or 10 ppb.
Prior to lowering the MCL to any
level, a full economic impact analysis
would have to be conducted and
available for public comment as part of
an entire rulemaking proceeding. The
0.010 to 0.025 mg/1 was merely stated as
an indication of future technological
performance potential.
36. One commenter was concerned
that EPA had underestimated the
financial implications of the TTHM
regulations on water utilities, for
example, by assuming that the rate
increase required to finance the
necessary revenue requirements would
be easily obtained. This commenter
noted that projections of future capital
requirements in addition to the cost of
the GAG process for various water
systems had not been factored into the
analysis. Another commenter stated that
in order to install GAG, water utilities
would need to raise capital through
large rate increases. They noted that
there were substantial regulatory
barriers which could preclude water
utilities from obtaining the necessary
rate increases. Even if utilities were able
to raise the capital funds, the quality of
their credit and the attractiveness of
their common stock would be severely
reduced; this would reduce their ability
to obtain external financing for normal
water supply activities.
EPA believes that the estimated costs
will not result in an undue burden upon
water utilities and therefore, revenue
requirements will be reasonably
obtained in most cases. Further, EPA did
not factor in capital requirements for
such items as system maintenance or
expansion into the analysis since these
are not directly related to the
regulations. Since implementation of
these regulations will improve drinking
water quality, utilities should be in a
favorable position to obtain rate
increases. Further, it is not expected that
bond rating of the utilities will be
significantly affected or that regulatory
barriers will seriously prevent systems
to obtain financing for complying with
these regulations.
37. Two commenters stated that EPA
was required to prepare an
Environmental Impact Statement (EIS)
in conjunction with these regulations.
They noted that EPA had not addressed
the significant primary and secondary
environmental problems associated with
the use of GAG treatment facilities and
that EPA's assessment had not
evaluated the full environmental impact
potential of the regulations so as to be
functionally equivalent to an EIS.
EPA is not required to prepare a
formal EIS for these regulations. Section
102(2)(C) of the National Environmental
Policy Act (NEPA) requires the
preparation of an EIS for "major Federal
actions significantly affecting the quality
of the human environment." However.
the courts have exempted EPA
rulemaking from this requirement where
the Agency's action in carrying out its
statutory obligations is designed to
protect the environment and amounts to
the "functional equivalent" of the
requirements of NEPA. Although the
courts have not specifically addressed
the applicability of NEPA under the
SOW A. the "functional equivalent"
standard is equally appropriate and
clearly satisfied here. This rulemaking
has involved extensive efforts by EPA.
including public participation, for
evaluating the primary environmental
impacts related to the control of TTHMs
in drinking water. The potential negative
impacts included air and water pollution
impacts of GAG and its attendant
regeneration furnaces, waste disposal
issues related to such furnaces, adverse
effects on the microbiological quality of
drinking water, as well as risks
associated with the use of GAG. Many
other environmental impacts will be
positive since human exposure to
harmful chemicals will clearly be
reduced. Moreover, the legislative
history of the SDWA indicates that
proposed provisions that would have
required literal compliance with NEPA
for actions taken under the SDWA were
rejected by Congress. The secondary
impacts were found to be too remote for
consideration in EPA's analysis but are
also believed to be negligible.
38. Two commenters stated that EPA
was required to prepare an Inflationary
Impact Statement (IIS) in conjunction
with these regulations.
EPA does not believe that it was
required to prepare an IIS for these
regulations. Under Executive Orders
Nos. 11821 and 12044. only major
regulatory actions which may have a
significant impact on inflation require
the preparation of such statements. A
major or significant regulation is one
which has associated annual costs of
greater than $100 million, causes an
increase in price of greater than five
percent, or is so designated by the
Agency's Administrator. For the TTHM
regulation, annual costs are estimated at
$19 million, and average increases in the
price of water are less than one percent.
The Administrator has not designated
this regulation as significant.
Nevertheless, EPA has conducted a full
economic and financial impact analysis
of these regulations which is reported in
the economic analysis document.
39. Comments were received
concerning the air pollution and energy
impacts associated with the use of
regeneration furnaces for GAC. These
commenters were concerned that the
regulations would promote substantial
new consumption of energy through use
of GAC as well as in secondary energy
consumption such as in the production
of the energy that will be used in GAC
regeneration, or the energy usage
associated with the manufacture and
transportation of GAC. One commenter
stated that the Agency did not address
the cost and environmental impact of
such furnaces. One commenter was
concerned about the availability and
costs of energy for on-site regeneration
of GAC as well as increased energy
consumption.
EPA issued a supplemental notice of
proposed rulemaking on July 6,1979 (43
FR 29135 at 29147) which addressed
precisely these concerns. EPA has
concluded that the air pollution and
energy impact of these regulations will
be negligible. Air pollution associated
with GAC furnaces will be minimized by
the use of scrubbers whose cost have
been included in EPA's estimated cost of
compliance for those systems that will
be required to use GAC for meeting the
TTHM MCL. Since fewer systems are
expected to have to install GAC than
EPA originally proposed, these impacts
have further been reduced. Secondary
energy impacts, such as transportation
costs, are too tangential to be estimated
with any degree of accuracy, but are
also considered to be insignificant.
Energy consumption will increase
consumption by an estimated 508 x 10"
-------
68654 Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations
BTU's per year or 0.0007 percent of
present U.S. energy consumption. These
figures do include a number of
secondary energy impacts. Commenlers
are referred to the preamble and EPA's
economic impact analysis
accompanying these regulations for
further details on these issues.
40. One commenter noted that EPA
was required lo analyze the costs of its
actions in terms of the benefits hoped to
be obtained and had failed to do so.
EPA has conducted a thorough
analysis of the costs of this regulation
and has examined in a qualitative
source the perceived benefits from
reducing levels of human exposure to
THMs. It has been determined that the
costs of this regulation are reasonable
and therefore risks associated with
exposure to THMs should be reduced
accordingly. However, EPA is not
required under the SDWA to perform a
quantitative cost/benefit analysis nor to
base regulatory decisions solely on the
basis of such an analysis. Rather EPA is
directed to establish an MCL which
requires contaminants which may have
any adverse effect on human health,
including carcinogens, to be reduced to
the extent feasible and that is the basis
of EPA's establishment of the TTHM
MCL at 0,10 tng/1 TTHMs. Further
reduction was not considered to be
feasible at this time because of the
potential trade-off of compromising the
bacteriological quality of the drinking
water due to less effective disinfection
practices. Commenlers are also referred
to the discussion that follows below.
41. Information on the relative
benefits related to the costs of the
TTHM regulation was provided in an
NAS Report, "Non-Fluorinated
Halomethanes in the Environment"
(1978). Dr. Andelman, using GAG and
aeration as the tool to demonstrate a
methodology of evaluating cost and
benefits, concluded that in the absence
of any other perspective, it was not cost-
beneficial to use GAC or aeration
simply to reduce chloroform
concentrations in drinking water.
The report stated:
From the viewpoint of economics, ihe
central policy issue in controlling human
exposure to any toxic substance is whether
Ihe benefits of reducing deaths, suffering,
illness, and other losses outweigh the costs of
controls. This involves identificalion of
population exposure levels and a
determination of when Ihe costs of additional
controls exceed the benefits of a further
reduction in exposures.
The report applied four concepts and
principles including: (1) The discounted
value of an individual's production. [2]
extrapolations from risk premiums, (3)
.osts of illness and human suffering, and
(4) the Pareto Improvement principle,
and applied the empirical estimates of
values of reducing the probability of
death to develop his benefit-cost
evaluation. The report concluded that:
Depending on the methodology thai is used
to compute costs, from these examples, the
most reasonableness estimates of Ihe per
capita value associated with reducing the
probability of death by 100 percent range
from $100,000 to $1.000.000.
EPA has reviewed this NAS report
and believes that the cost side of a
benefit-cost equation that is used in the
control of toxic substances should have
been calculated for each specific control
technique because the costs per person
benefited may vary greatly among the
available control options. The NAS
report selected only aeration and GAC
adsorption process for the control of
THM concentration in drinking water
and failed to consider other less
expensive treatment methods which
will, in fact, be used by most systems to
comply with the TTHM MCL.
The report assumed that the most
significant effect of human exposure to
chloroform in drinking water was cancer
and that all of these cancers result in
death; effects other than cancer
mortality were presumed to be
negligible. Therefore, the report said Ihe
benefits of reducing human exposure to
chloroform in drinking water could be
estimated by multiplying data on
lifetime risk of cancer by the economic
value of reducing the risk of death from
cancer in a population. The benefits also
could be calculated by multiplying the
daily per capita uptakes of chloroform
by the risk of a cancer death over an
average lifetime from a given daily dose
of the carcinogen by the economic value
of reducing the risk of a cancer death.
Based upon the above principles and
other assumptions, the report found that:
Very high concentrations of chloroform in
drinking water are associated with enough
risk of cancer lo justify the costs, on
economic grounds albne, of treatment
processes for removal of this compound. The
potential magnitude of the problem is even
greater if allowance is made for the upper
limit of risk. Furthermore, justification for
treatment rises with the value imputed to .
avoiding a death. However, the current cost
of treatment to remove chloroform from
drinking water is sufficiently high that the
economic justification for removing
chloroform from drinking water in the United
States, assuming the most probable risk.
exists only in those cases where maximum
initial concentrations of chloroform are found
in drinking water, there is maximum fluid
intake, and the risk of death is valued at
$1.000,000 or more. Using a more typical and
more statistically justifiable value of reducing
the risk of death, i e. $300.000, Ihe high cost
of removing chloroform alone cannot be
justified on economic grounds for Ihe most
probable risk conditions, even when there arc-
maximum concentrations and intake.
EPA believes Ihe analysis in this NAS'
report has everal serious shortcomings
which obviates its conclusions. As is
stated in the report itself, the analysis
was designed primarily "to demonstrate
a methodology," rather than to draw
strong conclusions about the particular
example used. In EPA's view, Ihe
following assumptions made in the
analysis bias it against regulation of
THM: (l)The risk extrapolation used for
chloroform is lower by a factor of 8.5
from that derived by the NAS in
Drinking Water and Health (the
existence of so large a discrepancy in an
estimate by the same organization using
the same model illustrates the
difficulties in making a fine-grained
comparison of risks and costs], (2) no
account is taken of the benefits of GAC
other than removal of THM, such as
removal of other disinfection by-
products, synthetic organic chemicals
present in the raw water, and
substances with objectionable taste and'
odor, and (3) it does not take into
account much cheaper technologies for
THM control. In spite of these biasing
assumptions, the analysis still concludes
that, for an assumed value per cancer
case avoided of $500,000, a community
would be justified in installing GAC for
TTHM control if its TTHM level
exceeded 164 ug/1, a conclusion which is
not at all inconsistent with an MCL of
100 ug/1.
EPA agrees that the costs and benefits
of alternative regulations should be
examined in deciding whether and how
stringently to regulate, where the
statutory framework does not prohibit
such examination. While no such
prohibitions are contained in the
SDWA, EPA believes that the
uncertainties in quantifying the health
benefits of regulatory actions,
particularly given the great scientific
uncertainties about the effects of low
levels of carcinogens, make formal cost-
benefit analysis of limited usefulness in
regulatory decision making.
The quantification of risk is sorely
limited by the lack of demonstrable
accuracy and precision of any statistical
model, the inability to identify more
than a portion of the substances that
would be generated by chlorination in
water, the inability to predict the toxic
potency of those chemicals individually
let alone as a variable complex mixture,
the inability to quantify the
contributions of these chemicals to and
their interactions with (he mass of toxic
chemicals that are part of human body
burdens, and the inability to identify
-------
Federal Register / Vol. 44, No. 231 / Thursday, November 29. 1979 / Rules and Regulations 68655
particularly susceptible high risk
segments of the population.
The costs that were used in the NAS
nalysis dwelled on GAC and aeration
which are among the most expensive
options and which only a smalt number
of water systems would need to use.
Prevention or reduction of THM
formation potential prior to introduction
of chlorine is much less costly than
removal after formation. The NAS
estimate of benefits associated with the
THM regulation considered removal
only of chloroform and none of the other
by-products, and also did not consider
any other water quality improvements.
The study's cost of not controlling THMs
in public water systems did not include
the considerable offset of increased cost
to consumers and society by increased
reliance on bottled water or home
devices that ostensibly reduce organic
chemicals at the tap. Morbidity costs,
lost wages and health treatment costs
were also not considered. Thus, risks
and benefits can easily be
underestimated, and costs
overestimated. Considering costs, risks
and benefits is of course an essential
part of any regulatory process, but the
judgment of an acceptable societal cost
for a human life is a matter of policy
hat requires many more complex and
ubtle factors that are not within the
Went state-of-the-art for these types of
quantitative analyses. Additional
discussion of cost-benefit analyses is
provided below in the reponse to the
comments submitted by the Council on
Wage and Price Stability.
42. The Council on Wage and Price
Stability (CWPS) said that the EPA
studies contain:
(a) No analysis of the benefits of
alternative performance standards or of
alternative population-size cut-offs, and
(b) No analysis of either the costs or
the benefits of alternative design
standards.
Consequently, CWPS believed that
the EPA analyses shed no light on the
reasonableness (i.e., the cost-
effectiveness) of these decisions. They
said that EPA provided no information
about the consistency of these
regulatory decisions with each other or
with other EPA regulations. CWPC
believed that because the resources
available for health-related programs
are limited, it is important that those
resources be allocated in a way that
maximizes the benefits (in terms of lives
saved or cases of illness or injury
avoided). This in turn would require that
"le incremental cost per cancer case
voided be at least approximately
quated for different regulations or
Different adopted standards. CWPS felt
that it was incumbent upon EPA to
support its proposed regulations with
careful risk-assessment and cost-benefit
analyses, employing the best estimates
available regarding uncertain variables,
parameters, and relationships. CWPS
made some preliminary calculations and
suggested that more lives could be
saved with no increase in costs by
tightening up on the performance
standard for THM (i.e.. lowering the
allowable concentration below 100 ug/I
and concomitantly relaxing the
population cut-off (higher than 75,000]).
CWPS said that:
(a) The incremental cost of lowering
the population cut-off from 100,000 to
75,000 (given a 100 ug/1 standard) is
$12.2 million per additional cancer case
avoided.
(b) The incremental cost of
strengthening the performance standard
from 100 ug/I to 50 ug/1 [given a
population cut-off of 75,000) is $6.3
million per additional cancer case
avoided.
(c) Thus, the cost of avoiding cancer
cases by applying the MCL to
communities with populations of 75,000
and above, which EPA had done, is
double the cost of avoiding cancer cases
by strengthening the standards to 50 ug/
1, which EPA did not propose.
(d) CWPS also said, "These
calculations do not necessarily mean
that the performance standard should be
tightened to 50 ug/I, but they do suggest
that the (two) proposed regulations are
internally inconsistent."
The CWPS comments raise two
separate types of issues with respect to
the THM regulation. The first concerns
the use of cost-benefit analysis to
determine whether a regulation is
justified and what its overall level of
stringency should be. The second
concerns whether, given that a
regulation limiting THM levels is to ba
implemented, the proposal would be the
most cost-effective way of using a given
level of social resources to reduce the
population's exposure to THMs.
On the first issue, CWPS did not draw
any conclusions as to whether the
regulation was justified, but
recommended that cost-benefit analyses
be an integral part of the Agency's
decision process.
EPA has reviewed the subject of using
cost-benefit analysis in regulatory
decision-making under the SOWA and
reached the following conclusions. First,
benefit-cost analysis is most useful to
decision-makers when benefits can be
specified with the same degree of
certainty as the costs. However, when
dealing with long-term health risks, such
as cancer-causing contaminants like
THMs, while it is possible to establish
the existence of a risk, it is beyond the
state-of-the-art of current scientific
knowledge to establish the exact degree
of risk. Crude indications of risk can be
made, and these can be used to develop
a range of health benefits associated
with a regulation, however, the range is
so broad that its use in benefit cost
analysis overwhelms these elegant and
sensitive analytical procedures. In
addition, there is little agreement on the
dollar value which should be ascribed to
the avoidance of a case of cancer. Past
estimates have ranged from $10,000 to
$158 million. Therefore, due to these two
fatal deficiencies, it is not possible to
place excessive significance on cost-
benefit analysis for the long-term health
risks related to this regulation.
Despite these inherent difficulties,
EPA conducted an analysis of regulation
alternatives. Constraints to
decisionmaking involving technical and
administrative issues tended to limit the
range of alternatives. Within this
framework, however, it was possible to
establish that for the regulation the
marginal cost of a case of cancer
avoided is approximately $200,000
(counting only the benefits of THM
reduction]. This is similar to that
suggested in the NAS report cited by Dr.
Andelman. Further discussion is
included in the Statement of Basis and
Purpose.
On the second issue, EPA agrees that
any regulation should make the most
efficient possible use of the social
resources devoted to compliance, to the
extent that it is possible to predict.
CWPS presented an analysis which
purported to show that, for the same
total cost, a greater reduction in THM
exposure might be obtained by reducing
the MCL and increasing the population
cut-off figure. However, the assumption
had been made that systems exceeding
the MCL would reduce their THM levels
precisely to the MCL; in fact, many of
the control technologies would actually
reduce THM levels to much lower levels
in practice. When account is taken of
this fact, the analysis shows that EPA's
proposed regulation is more cost-
effective than the CWPS' suggested
alternative. After staff-level discussion,
CWPS recognized this and other
technical deficiencies in its analysis in a
letter to EPA dated January 31,1979.
43. Sixty-nine comments were
received on the proposed concept of
averaging concentrations of TTHMs for
compliance. A majority of (he
commenlers approved of both the
annual averaging ofTTHM values from
quarterly samples, and the averaging of
TTHM values of representative samples
within the distribution system. However,
fourteen commenters thought that
-------
68656
Federal Register / Vol. 44. No. 231 / Thursday. November 29. 1979 / Rules and Regulations
averaging the quarterly results would
mask fluctuations in TTHM levels as
affected by seasonal and other site-
specific factors. One said that quarterly
averaging would be justified if EPA
were concerned about the chronic but
not acute effects of THMs. One said that
flexibility should be retained in the
regulation for later reconsideration of
this averaging concept. Two
conunenters said that compliance should
be determined by averaging all of the
results of samples taken in the preceding
12 months. One suggested that a
geometric mean should be used in
compiling and averaging the sampling
results. One felt that there was not
enough information to determine
whether the concept of averaging was
reasonable.
On the question of averaging results
of samples in the distribution system,
several conunenters felt that averaging
values could mask high TTHM
concentrations and fail to protect those
individuals receiving maximum doses.
Because flow patterns in the distribution
system are likely to be relatively
constant, these conunenters believed
that some residents could be unduly
exposed to consistently high levels of
TTHMs over a long period of time. One
commentcr opposed averaging the high
values of TTHM analyses from samples
taken at the extremes of a distribution
system, with the lower results from
other areas of the distribution system
because it would result in uneven
population exposure. Three others
suggested that all samples should be
taken at the extremes of the distribution
system instead of averaging all sample
results. One suggested that all samples
should be incubated to obtain terminal
TTHM and hence uniform results. One
commenter said that all samples should
be taken from the same point every time
to avoid misrepresentation. One
commenter thought that selection of
sampling locations should be based
upon results of a sanitary survey for
each system.
EPA's proposal to determine
compliance with the TTHM MCL based
upon an annual average of the sampling
results per quarter has been retained in
the final regulations. EPA recognizes
that TTHM levels may fluctuate
depending upon seasonal and other site-
specific factors. However, the MCL for
TTHMs has been established primarily
to protect the public from the adverse
effects attributable to chronic exposure
to these contaminants, rather than from
any acute effects. EPA nevertheless
retains the flexibility to amend these
regulations should new information
indicate that annual averaging of
quarterly results is not adequately
protective. On the other hand, EPA
believes that it would not be reasonable
to determine compliance by an annual
average of all samples taken since this
could clearly allow systems to mask
fluctuations in TTHM levels over the
year. In regards to use of a geometric
mean as the basis of the MCL, the
arithmetic mean is considered to be
more appropriate because it is a more
accurate representation of typical
human exposure.
With regard to those commenters who
expressed concern about EPA's
proposed sampling program, it is noted
that it would have required systems to
average a minimum of five samples per
quarter, no more than 20% of the
samples to be taken at the entry point to
the distribution system, no less than 20%
at the extremes of the distribution
system and the remaining 60% at
representative points in the system
relative to population density. In
response, these final regulations have
reduced to four the minimum number of
samples to be taken per quarter, but no
longer allow any samples to be taken at
the entry point to the distribution
system, where TTHM levels would have
likely been lowest, and where few
consumers would have actually been
exposed to such levels. EPA believes
that this sampling program will better
reflect the average TTHM levels in the
drinking water served to most
consumers.
However, EPA rejected the
suggestions to require all samples to be
taken either at the extremes of the
distribution system, or at the same point
in the distribution system each time.
Such sampling schemes would not fairly
represent the water system as a whole.
However, EPA is concerned that very
high levels of TTHMs at the extreme
ends of a distribution system be
reduced. EPA believes that by requiring
extreme sampling results to form a
larger percentage of the quarterly
average (25% as opposed to the
proposed 20%), any great differences in
TTHM concentrations in such locations
may be detected and corrected.
In response to the remaining
comments on EPA's proposed sampling
program, EPA has not required all
samples to be incubated to obtain
terminal results because this would
probably overestimate actual
concentrations at the taps of most
consumers. EPA agrees with the
comment that sampling locations must
be selected by the system on a case-by-
case basis, preferably after a sanitary
survey, depending upon the particular
configuration of its distribution system.
Systems are encouraged to work with
the States and EPA in the selection of
truly representative sampling points.
EPA has required that the number of
samples taken be commensurate with
the number of treatment plants used by
each system to allow sampling to detect
differences in TTHM levels within each
system attributable to different source
waters and different treatment methods.
Once problems are detected, systems
should reduce extreme differences of
TTHM levels within their distribution
system.
44. Twenty-three commenters
supported EPA's proposal to require use
of the Standard Plate Count (SPC) as a
more sensitive indicator (than the
coliform test) of microbiological quality
during treatment modifications. Thirty-
seven commenters felt that the SPC was
of questionable value or unreliable, and
that the SPC requirement would impose
an unnecessary administrative burden
on water utilities. Five commenters
suggested that the SPC should only be
required for those systems whose water
sources receive municipal point source
discharges, and should not be required
for all treatment modifications. Four
commenters also felt that the SPC
should only be used to confirm a
questionable microbiological count and
that the decision to use the SPC should
be left to the discretion of the State
regulatory agency.
In response to these comments, EPA
has decided to delete from these
regulations the SPC as a mandatory
requirement for all systems that make
treatment modifications to comply with
the TTHM MCL. However. EPA still
believes that compliance with the
TTHM MCL should not be achieved at
the expense of the microbiological
integrity of the water and that the SPC
can be a reliable and useful tool as an
overall indicator of water quality.
Therefore, in order to insure that
disinfection is not compromised, while
affording maximum flexibility to the
States to address case-by-case
situations, these final regulations have
included a requirement whereby
systems must seek and obtain State
approval of any planned significant
modifications to their treatment process
made to comply with the TTHM MCL
that could affect biological quality. The
States (or EPA in non-primacy States)
must therefore exercise careful
supervision over system treatment
changes by prescribing specific
measures (which would include the SPC
in appropriate cases) to insure the
continued microbiological quality of the
drinking water. The usefulness of the
SPC and other biological tests are
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations 68657
discussed in greater detail in the
ireamble to these regulations and will
je discussed in EPA's guidance to the
States concerning approval of system
treatment modification plans.
45. Ten comments were received on
EPA's proposed restriction on the use of
chlorine dioxide as an alternative
disinfectant to free chlorine. Nine
opposed the restriction of using chlorine
dioxide at a maximum dose of 1 mg/1
but provided no supporting data. One
felt that EPA should encourage the
testing and use of alternative
disinfectants while others felt that the
limit of 1 mg/1 for chlorine dioxide was
arbitrarily set and that up to 2 to 3 mg/1
chlorine dioxide should be allowed. One
commenter reported that chlorine
dioxide was effective in reducing the
TTHM concentration in his system from
284 mg/1 to 16 mg/1.
In response to these comments. EPA
has deleted from the final regulations its
proposed restriction on the amount of
added chlorine dioxide. EPA is
nevertheless concerned about the
uncertain state of knowledge concerning
the potential for adverse effects
associated with chlorite, chlorate and
chlorite ion, which are produced from
oxidation/reduction reactions of
chlorine dioxide in water. EPA will be
considering proposing limitations on the
residual oxidants (CIO,, CIO',, and
CIO",) in the finished drinking water
rather than on the amount of chlorine
dioxide added. In the meantime,
additional research on the health effects
of alternative disinfectants will
continue. MCLs may be developed for
inclusion in the Revised Regulations
after further studies have been fully
evaluated.
By requiring all systems significantly
modifying their treatment process to
comply with the TTHM MCL to obtain
State approval of their modification
plan, EPA expects that where
restrictions on chlorine dioxide are
necessary, the States will impose such
restrictions as appropriate in
accordance with EPA guidance
including monitoring for residual
oxidants and maintaining their
concentration at a low level. Where
chlorine dioxide is completely reduced
to chloride, no restrictions would be
necessary since by-products are
believed to be of no toxicological
significance. Case-by-case judgments
can also be made to impose restrictions
when the presence of reducing agents in
the raw water of a particular system
would result in excess formation of
chlorite and chlorate. Additional
discussion on the use of chlorine dioxide
as an alternative disinfectant is
contained in the preamble to these
regulations and will be contained in
additional EPA guidance to the States
for approval of system treatment
modification plans.
40. Fifty-five commenters opposed
EPA's proposed limitation on the use of
chloramines as a primary disinfectant.
They argued that chloramines would
solve some of the problems of using
chlorine for drinking water treatment
because chloramines do not react with
precursors to produce TTHMs, and
chloramines have been in use in many
water systems for many years without
any problems. Eleven commenters
agreed that chloramines should be
restricted from use as a primary
disinfectant. One of these commenters
reported that preliminary data had
indicated that chloramines may not be
effective in neutralizing viruses and
amoebic or Giardia cysts. One
commenter suggested that chloramines
may be used after the primary
disinfection step for the purpose of
maintaining an active disinfectant
residual. The NOW AC felt that the
proposed limitation was unduly
restrictive.
EPA found that most of the
commenters opposed to the imposition
of restrictions on the use of chloramines
failed to recognize that EPA's proposed
restriction was limited to prohibiting its
use as a primary disinfectant. EPA does
not disagree with those commenters
who endorsed the use of chloramines as
an effective secondary disinfectant (to
maintain an active combined chlorine
residual). Nevertheless, in response to
these comments, EPA has decided to
delete the chloramine restriction from
the final regulations, allowing
appropriate restrictions to be imposed in
necessary situations by the States in
approving system treatment
modification plans. Use of chloramines
instead of free chlorine has been shown
to be a simple and readily available
means for reducing the formation of
TTHMs in many examples. However,
they are also known to be weak
disinfectants for certain bacteria,
viruses and protozoa, compared to free
chlorine as HOC1, ozone and chlorine
dioxide. Therefore, where such
contamination is suspected, appropriate
restrictions should be imposed.
Additional information on the use of
chloramines as an alternative
disinfectant is contained in the
preamble to these regulations and in
additional EPA guidance to the States
on approval of system treatment
modification plans.
47. Eleven comments were received
opposing the concept of setting an MCL
to control TTHMs in drinking water.
Two commenters said that EPA lacked
legal authority to regulate the TTHMs
under the Amendments to the National
Interim Primary Drinking Water
Regulations (N1PDWR). One commenter
noted that the feasibility of control
measures under the NIPDWR must be
adjudged to have been available as of
December 1974, when the SDWA was
enacted. Three commenters said the
NAS report, "Drinking Water and
Health" fell far short of providing the
needed scientific definition and did not
recommend EPA to set MCLs for
TTHMs. One commenter said that the
EPA should not yield to the pressure
from some public interest groups to set a
MCL for TTHMs before the health risks
have been established. Four commenters
believed that the main reason EPA
proposed an MCL for TTHMs was
because EPA was anti-chlorination and
was trying to abolish chlorination
practice in water treatment. One of the
four suggested that instead of an MCL,
EPA should tighten chlorine
specifications so that no contamination
of the water will result during
chlorination practice. Three others
recommended that the regulations
provide guidance on the proper use of
chlorine as a disinfectant, either free or
combined, for case-by-case applications.
EPA's response to those comments
addressing the Agency's authority to
establish these regulations as Interim
Primary Drinking Water Regulations is
contained in the preamble, and
commenters are referred thereto. EPA
agrees that the feasibility of control
measures under the NIPDWR must be
based on technology generally available
as of 1974 and has found that these
regulations satisfy the statutory test.
With respect to those commenters
that cited the NAS Report "Drinking
Water and Health" to support their
position that regulation of TTHMs is
premature, EPA disagrees with their
interpretation that the NAS only
recommended further research. In fact,
the NAS concluded that: "strict criteria
be applied when limits for chloroform in
drinking water are established to protect
the public health." Moreover, Dr. Riley
Housewright of the NAS Safe Drinking
Water Committee, stated that:
"chloroform and other THMs present a
health hazard and that steps should be
taken to prevent their formation or to
remove them from drinking water." As
noted in the preamble and EPA's
responses to other comments in this
Appendix, EPA believes that sufficient
information is known about the
potential for adverse health effects from
the presence of THMs in drinking water
-------
68658 Federal Register / Vol. 44. No. 231 / Thursday. November 29, 1979 / Rules and Regulations
to warrant regulation at this time.
Although further research will continue
to be forthcoming, EPA need not wait
for definitive proof of harm before it
takes regulatory action under the
SDWA.
With respect to those commenters
who charged that EPA's establishment
of a TTHM MCL evidenced EPA's intent
to abolish chlorination as a drinking
water treatment practice, EPA disavows
such an intent. However, EPA does
believe that improper or careless use of
chlorine, as well as any other
disinfectant, can result in the
unnecessary formation of potentially
harmful by-product chemicals in the
finished drinking water. EPA
acknowledges that chlorine is currently
the most widely used, highly effective
drinking water disinfectant and expects
that use to continue. However, control of
TTHMs should lead to a more judicious
use of chlorine and will serve to
minimize human health risks from
exposure to other disinfection by-
products. EPA also agrees that better
quality control in the manufacture of
chlorine for drinking water treatment is
necessary to avoid harmful
contaminants contained therein and will
address such concerns in conjunction
with its overall review of water
treatment additives. EPA's guidance to
the States for approval of system
treatment modification plans will
contain additional information on
proper chlorine use.
48. A total of 306 comments were
received expressing a concern for the
basis of health effects data that support
the proposed TTHM regulations. The
majority of the commenters felt that the
proposed MCL was not based upon
incontrovertible health effects
information and urged that additional
health effects research and
epidemiological studies should be
conducted. Only a* few commenters said
the supporting health effects data for the
proposed THM regulations were
adequate and that the regulatory action
was justified now.
Specifically, 292 comments said that
the available health effects data, both
epidemiological studies and laboratory
animal tests, were not conclusive and
were disputed by many scientists. These
commenters, therefore, believed that the
setting of an MCL for TTHMs was not
warranted at this time. They suggested
that more research should be conducted
specifically on the toxicological
assessment procedures and the health
effects of long term exposure to low
dosage of THMs.
EPA has reviewed these comments in
light of all available health effects
information and has concluded that long
term low level exposure to TTHMs may
be harmful to human health. EPA's
conclusions are supported by comments
and statements of policy by
representatives of the National Cancer
Institute, National Academy of Sciences,
National Drinking Water Advisory
Council, National Institute of
Environmental Health Sciences, Food
and Drug Administration, Occupational
Safety and Health Administration, and
the Consumer Product Safety
Commission. These commenters
emphatically stated that EPA should not
wait for additional evidence to proceed
with regulatory action to control
chloroform and trihalomethanes in
drinking water which was warranted
now. These comments are summarized
in Appendix B.
The following discussion summarizes
the specific concerns expressed by
commenters regarding the health basis
of the regulations and presents the
Agency's responses.
49. Comments were received that
argued that chloroform poses no
potential cancer risk and there are no
available data that support the premise
of a causal relationship between the
concentrations of THMs normally found
in drinking water and cancer in humans.
They noted that the epidemiological
studies that have been conducted
concerning drinking water and a
possible connection with cancer risk in
humans were inconclusive.
EPA reviewed the available 18
epidemiological studies concerning the
relationship between cancer morbidity/
mortality and constituent concentration
in drinking water supplies. In summary,
many but not all of the preliminary
studies have found positive correlations
between some drinking water quality
factors and some cancer mortality and
morbidity statistics such that the general
hypothesis is supported. Further
evaluations are necessary due to the
confounding factors inherent in
epidemiological studies of this nature.
Therefore, EPA has relied primarily on
the results of animals studies in
concluding that TTHMs in drinking
water pose a risk to humans. Thus. EPA
does not disagree with the comment that
data do not exist to demonstrate a
causal relationship between the
concentrations of synthetic organic
chemicals including THMs in drinking
water and cancer in humans. However,
the positive correlation of cancer
morbidity/mortality and contaminants
in drinking water are suggestive and are
not inconsistent with the carcinogenic
potential of chloroform as demonstrated
by well conducted animal experiments
at high doses.
50. Some commenters opposed EPA's
reliance on animal studies for its finding
that TTHMs in drinking water pose a
health risk on the grounds that
extrapolation of results in animal cancer
studies to humans is fraught with
problems and uncertainties.
EPA recognizes the problems of
extrapolating animal data to man. The
state-of-the-art in toxicology as
illustrated by the NAS in the report
"Drinking Water and Health" is that the
effects in animals, properly qualified,
are applicable to man. Chloroform has
been shown to be carcinogenic in
experimental animals; its metabolic
pattern in animals is similar to that in
humans; EPA therefore believes that the
carcinogenic effect of chloroform as
observed in animals do indicate risks
from human exposure to TTHMs in
drinking water.
51. Some commenters argued that the
study (by National Cancer Institute
(NCI)) cited by EPA to support the
carcinogenic!ty of chloroform was "a
preliminary screening test and not a
definitive study." They said that the
study was not intended to be used to
extrapolate health effects of chloroform
to drinking water levels and that the
NCI study was inadequately controlled
and did not follow proper scientific
protocols. Since a new EPA/NCI study
is underway it was recommended that
the implementation of any regulations
be delayed until this study was
completed. They claimed that the NCI
study was not intended to be used to
extrapolate the adverse health effects of
the tested animals to the potential
human health risk posed by the low
levels of chemicals that are found in
drinking water, since many researchers
believe that the high morbidity rates in
the animal experiments suggested acute
toxicity rather than chronic toxicity.
Based on the NAS review and the NCI
report. EPA has concluded that the NCI
chloroform-carcinogen bioassay with all
its short-comings is a valid test. It has
been accepted by the other federal
agencies for regulatory purposes. The
morbidity noted took months or years to
develop and would not be an acute
effect by definition which would occur
in 3-7 days. In addition, the studies
performed as early as 1945 by
Eschenbrenner and Miller pointed out
the carcinogenic potential of chloroform
and the metabolic similarity of
chloroform in humans and animals. The
NCI study on the carcinogenic potential
of chloroform has been used by the NAS
as well as by the EPA's Cancer
Assessment Group (CAC) for risk
estimation. Additional refining studies
are continuing, but sufficient evidence
-------
Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68659
exists to indicate potential human risk
id, therefore, to reduce human
cposure.
52. Several commenters stated that Dr.
noe's studies with chloroform on dogs,
rats and four strains of mice at low dose
levels did not produce tumors in
animals. Dr. Roe recommended a level
of 300 ppb THM in drinking water based
upon his results. It was claimed that
Roe's studies showed a no observed
effect at 595,000 (drinking water
equivalent) ppb of chloroform in
drinking water. Therefore, he argued
that 300 ppb would provide margin of
safety of 2.000. It was argued that EPA
had used 500 as a margin of safety in
other regulations. Based upon his chosen
statistical extrapolation model, he found
that a THM MCL of no lower than 0.30
mg/1 (300 ppb] would provide a more
than adequate margin of safety.
however, it was also stated that this
level is still too low to be justified on a
cost-benefit basis if GAG were required.
EPA has concluded that Dr. Roe's
studies with chloroform on dogs, rats
and four strains of mice at low dose
levels further strengthens the hypothesis
of chloroform carcinogenicity. In one
study, the mice fed 17 mg/kg/day
-hloroform showed no incidence of
inal carcinoma, but an excess of
imors of the renal cortex were
Jserved in the male ICI—Swiss mice,
at a dose level of 60 mg/kg/day. The
negative results observed in the dog
experiment may be attributed to the fact
that either the animals were not
exposed for a suitable length of time
(i.e., duration of life span) or that an
insufficient number of animals were
tested. The negative results of the rat
study may be attributed to the lack of
strain sensitivity.
Using a no-observed-effect-level for
chloroform of 17 mg/kg/day, Dr. Roe
recommended 300 ppb chloroform in
drinking water as an acceptable level.
According to his calculation this would
provide a margin of safety of 2000 for a
standard person drinking two liters of
water per day. The NAS Safe Drinking
Water Committee and many other
scientists now believe that the methods
at present do not exist to establish a
threshold for long-term effects of
carcinogens: thus, the safety factor of
2000 referred to in Roe's
recommendation of 300 ppb THM does
not apply to carcinogens since no
exposure can be considered to be
absolutely "safe". EPA is directed by the
DWA to reduce human exposure to
armful contaminants in drinking water
i the extent feasible. EPA's THM MCL
»f 0.10 mg/1 can be feasibly achieved.
The comment regarding the costs vs. the
benefits of the use of GAG is discussed
elsewhere in this Appendix.
54. Some commenters said that EPA's
proposed MCL of 100 ppb was
needlessly low and will require costly
additions or changes to water treatment
facilities without achieving any
corresponding benefit in water quality.
EPA has found that exposure to
TTHMs should be minimized. The level
of the MCL at 0.10 mg/1 TTHMs was
determined to be a feasible level for
achievement under the interim
regulations. Systems are encouraged to
reduce the level of TTHMs below the
MCL if technically feasible. EPA expects
that compliance with the MCL will
benefit drinking water consumers in
reduced exposure to THMs as well as
reduced exposure to other disinfection
by-products which may have adverse
health effects. For some systems the
aesthetic quality of the water will also
improve because taste and odor
producing compounds will be reduced
along with reductions in TTHM levels.
As discussed in the preamble and in
EPA's Economic Analysis accompanying
this final regulation, costs are not
considered to be significant in that most
required changes will be relatively
minor.
54. It was stated by several
commenters that there are a lack of
health effects data on THMs other than
chloroform and therefore, if an MCL is
set, it should only apply to chloroform.
EPA has found that the THMs other
than chloroform (bromoform,
dibromochloromethane,
dichlorobromomethane) are structurally
similar to chloroform, and possibly
undergo similar metabolic pathways and
exert similar bioeffects. Like chloroform,
bromoform exposure leads to fatty
degeneration and centrilobular necrosis
of the liver. Bromoform,
dibromochloromethane and
dichlorobromomethane have been
reported to be mutagenic in Ame's
bacterial test system. This test provides
information indicative of the potential of
genetic damage in biological systems.
Thus, because of the chemical
similarities in chemical structure and
biological activity, EPA's concern
regarding potential toxic effects of these
chemicals and setting the MCL for
TTHMs is reasonable.
55. Several commenters stated that
there was no hard evidence that low
level exposure to TTHMs produces
cancer.
Based on current scientific knowledge,
EPA must extrapolate from the results of
animal tests using higher dosages to
determine potential human health risks
from exposure to low levels of particular
contaminants. With chemicals such as
chloroform that have been shown to be
carcinogenic in animals, no level of
exposure can be presumed safe.
Therefore, EPA has concluded that
TTHMs in drinking water must be
reduced to the extent feasible as
required by the SDWA.
Nevertheless, recent studies using low
levels of 2-amino N-acetyl-fluprene (2-
AAF) in mice suggest that low level
exposure of animals to this compound
produces liver tumors when applied.
These adequately controlled studies
(23,000 animals) showed a no threshold
effect (liver cancer) was observed for
AAF at the 1% level. In order to be able
to measure below the 196 effect
somewhere in the order of 100,000
animals would be required.
56. Some commenters claimed that
other animal experiments have
suggested the existence of definite
threshold limits for toxic and
carcinogenic effects.
EPA's position is that available data
suggest a non-threshold response for
carcinogenesis. As an example, the
recent Acetyl Amino Fluorene
experiments were consistent with a no
threshold mechanism for liver tumor
induction. This position is supported by
the comments of Drs. Upton, Kennedy,
Bingham and King from the National
Cancer Institute (NCI), Food and Drug
Administration (FDA), Occupational
Safety and Health Administration
(OSHA) and Consumer Products Safety
Commission (CPSC), respectively, as
noted in the preamble and presented in
Appendix B. EPA's position is discussed
in both the preamble and the Statement
of Basis and Purpose. Also, the National
Academy of Sciences addressed this
issue in "Drinking Water and Health"
(NAS, 1977) as follows:
Carcii
threshold i
effect can be caused by a single hit, a single
molecule, or a single unit of exposure, then
the effect in question cannot have a threshold
in the dose-response relationship, no matter
how unlikely it is that the single hit or event
will produce the effect. Mutations in
prokaryotic and eukaryotic cells can be
caused by a single cluster of ion pairs
produced by a beam of ionizing radiation. We
would expect that mutations can be caused
by a single molecule or perhaps group of
molecules in proximity of DNA. The
necessary conclusion from this result is thai
the dose-response relationship for radiation
and chemical mutagenesis cannot have a
threshold and must be linear, at least at low
doses.
We therefore conclude that, if there is
evidence that a particular carcinogen acts by
directly causing a mutation In the DNA, it is
likely that the dose response curve for
carcinogenicity will not show a threshold and
will be linear with dose at low doses (pp. 37-
38).
cinogemc effects may well not have
lold dose-effect relationships. If an
-------
68660 Federal Register / Vol. 44, No. 231 / 'Thursday. November 29, 1979 / Rules and Regulations
Methods Do Not Now Exist to Establish a
Threshold {or Long-Term Effects of Toxic
Agents
With respect to carcinogenesis. it Beems
plausible at Tint thought, and it has often
been argued, that a threshold must exist
below which even the most toxic substance
would be harmless. Unfortunately, a
threshold cannot be established
experimentally that is applicable to a total
population. A time-honored practice of
classical toxicology is the establishment of
maximum tolerated (no-effect] doses in
humans based on finding a no-observed-
adverse-effect dose in chronic experiments in
animals, and to divide this dose by a "safety
factor" of, say, 100, to designate a "safe" dose
in humans. There is no scientific basis for
such estimations of safe doses in connection
with carcinogenesis. For example, even if no
tumors are obtained in an assay of 100
animals, this means only that at a 95%
confidence level, the true incidence of cancer
in Ihis group of animals is less than 3%. Even
if we were to carry out the formidable task of
using 1,000 animals for the assay and no
tumors appeared we could only be 95% sure
that the true incidence were less then 0.3%.
Obviously, 0.3% is a very high risk for a large
human population.
In fact, there are no valid reasons to
assume that false-negative results of
carcinogenicity tests are much less frequent
than false-positive ones. To dismiss all
compounds that did not induce tumors in one
or two mouse and rat experiments as non-
carcinogenic is wrong. Labeling as
"carcinogens" all substances that gave rise to
increased incidence of tumors is justified
only if there is conclusive evidence of a
causal relationship. The "relative risk" of
compounds that are not found to induce
tumors in animal experiments must also be
considered. But this requires evaluation of
data other than those collected in chronic
toxicity studies on rodents.
Experimental procedures of bioassay in
which even relatively large numbers of
animals are used are likely to detect only
strong carcinogens. Even when negative
results are obtained in such bioassays, it is
not certain that the agent tested is
unequivocally safe for man. Therefore, we
must accept and use possibly fallible
measures of estimating hazard to man.
57. As noted by a number of
commenters, the assumption of parallel
response between test animals and
humans does not hold for many species.
EPA believes that animal experiments
that demonstrate a carcinogenic
response are indicative of a potential
carcinogenic response in the human
population. This is supported by Drs.
Upton, Kennedy, Bingham, and King
from the NCI. FDA, OSHA, CPSC, and
NIEHS, respectively, whose testimony is
presented in the preamble and
Appendix B.
58. Some commenters stated that
EPA'a extrapolation procedure
erroneously utilized two "very
consecutive" techniques to determine
the MCL for THMs. They said that either
technique could probably be justified,
but not both.
The level of the MCL is based upon
feasibility of available treatment
technology and maintenance of
biological safety and not on an
extrapolation technique from
experimental data. The need to limit
human exposure is demonstrated by the
potential adverse health effects from
long term exposure to chloroform from
animal studies.
59. Comments were received that
alleged that EPA estimates of
environmental exposures to chloroform
appear to be erroneous and suggested
that EPA make every effort to obtain
correct values for contributions from air,
food and water. Also, they suggested the
possibility that in vivo formation of
chloroform and other THMs in the
human body might occur. The
commenters felt that the available data
suggest that more cost-effective
avenues, such as control of chloroform
in the work place, may be available for
reducing THMs in the environment than
by implementing the proposed TTHM
MCL.
• EPA's estimates of environmental
exposure to chloroform were based
upon the most recent available data and
are considered to be adequate
representations of exposure levels. The
speculation of in vivo formation of
chloroform and other THMs in the
human body contradicts what is known
concerning the fate of chloroform in a
mammalian system although this may be
occurring from ingestion of chlorine in
water. In mammalian systems,
chloroform is metabolized to carbon
dioxide and other metabolites. The rale
of metabolism will be dependent upon
the species. Therefore, there is little
chance of chloroform being
biochemically produced endogenously
in the human body.
With regard to the suggestion that
there may be more cost-effective means
for controlling chloroform in other
aspects of the environment, EPA has
found that drinking water is a significant
contributor to overall human exposure
to THMs. Moreover, control of THMs in
drinking water is not a significant
burden upon water utilities, and will
result in reduced human exposure to
other potentially harmful disinfection
by-products as well. Thus, EPA believes
that these regulations are necessary for
reducing human exposure to chloroform
from a significant source. OSHA and
FDA have likewise taken action to
reduce human exposure to chloroform
under their respective statutory
authorities.
60. Some commenters noted that the
concentrations of THMs found in public
water systems present no mutagenic,
teratogenic, acute, subchronic, or
chronic lexicological health risk to the
public.
Baaed on the evidence in EPA's
rulemaking record, EPA has concluded
thai THMs pose a carcinogenic risk at
the levels found in drinking water. No
safe level can be deemed to exist for
human exposure to carcinogens and
therefore, levels of these contaminants
should be reduced to the extent feasible.
61. Some commenters alleged that
EPA misconstrued the four general
"principles" for risk assessment stated
by the NAS in its report "Drinking
Water and Health." They argued that
EPA did not properly use these
principles and ignored the available
data. Specifically, with regard to the
first NAS principle, EPA was faulted for
not taking into account a number of
variables in extrapolation of the animal
data to humans, including differences
between species response to
carcinogens, weight, intake of food and
water, and routes of exposure. With
regard to the second NAS principle, they
argued that EPA ignored animal
experiment data that showed a
threshold level for no-effect responses
with respect to a number of suspected
carcinogens, as well as experiments
involving animals and humans
suggesting a no-effect level for
chloroform. In support of their claim that
threshold levels can be established for
carcinogens, they cited the existence of
in vivo biological processes and human
exposure to natural carcinogens without
adverse health effects. With regard to
the third NAS principle, they claimed
EPA did not consider the significance of
the detoxification and repair
mechanisms operative in animals and
humans in its health assessment of
THMs. Finally, with regard to the fourth
NAS principle, they claimed EPA
ignored the guidelines for assessing risk
for chloroform as set forth in EPA's
"Interim Guideline for Carcinogen Risk
Assessment." The comments also
faulted EPA for using only the linear
model for extrapolating the NCI animal
data to humans, while ignoring the data
presented by Roe, Eschenbrenner, and
Miller, as well as the estimates of risk
by Tardiff using the "margin of safety,"
"probit-log" and "two step"
extrapolation models.
The EPA has carefully evaluated all
available data and believes it has
properly followed the four NAS
principles. Each of the commenters1
concerns have been thoroughly
considered in determining the health
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68661
basis of the regulation. EPA has used the
>resent state-of-the-art in toxicology in
ising the NCI bioassay study on the
:arcinogenicity of chloroform for
issessing cancer risk to humans. The
studies by Roe. Eschenbrenner, and
Miller were not suited for risk
extrapolation because either the
dosages were not high enough to
observe the response or the experiments
were not performed for. long enough time
periods to observe tumorigenic
response.
The question of threshold and/or no
threshold for carcinogens is discussed
elsewhere in this Appendix, in the
preamble and in EPA's Statement of
Basis and Purpose accompanying these
regulations. The linear non-threshold
model is a conservative risk model and
consistent with the method used by the
NAS. The basis of the regulation is that
a human health risk exists even though
precise quantification of the risk cannot
be made using current lexicological
procedures. Therefore, EPA's regulatory
approach is to minimize human
exposure to these potential carcinogens
to as low a level as is feasible.
62. Some commenters said that EPA
ignored the relationship between dose
and time-to-tumor observation in
issessing the health risk of a
:arcinogenic material.
EPA does recognize the potential
relationship between dose and time-to-
tumor, but this has not been taken into
consideration in the calculation of risk
because scientific methods and data are
not currently available to adequately
perform such a computation.
63. Dr. Timothy DeRouen, representing
the Coalition for Safe Drinking Water,
critiqued the epidemiological studies
cited by EPA in the proposed
regulations. He discussed the studies for
a possible relationship between
chlorinated drinking water and cancer
mortality. His principal points and
EPA's responses are as follows:
(1) Dr. DeRouen commented that
although some consistencies exist to
support the premise of a relationship
between organic chemicals in drinking
water and cancer risk, comparable
inconsistencies exist that were not
pointed out by EPA.
EPA has concluded that in
epidemiological studies, inconsistencies
are always present, due to one or more
confounding factors. Because of this and
as noted in the preamble and EPA's
Statement of Basis and Purpose, EPA
did not rely upon the epidemiology
studies as a basis for the regulations.
Rather, they have been found to support
the hypothesis, as Dr. DeRouen noted,
that some relationship may exist
between cancer risk and chloroform in
drinking water. EPA's conclusions based
on animal studies are justified.
(2) Dr. DeRouen said that
correlational studies are the crudest
kind of epidemiology investigation and
their results should be used to suggest
more definite studies. However, they are
not considered accurate enough for
decision-making.
EPA believes that since several of the
individual correlational studies when
evaluated collectively suggest that
chloroform in water poses a risk, the
hypothesis is strengthened. Drs. Upton
and Schneiderman of the NCI supported
this conclusion and suggested that
reducing TTHM concentrations by 100
micrograms per liter could lead to a
decrease in cancer rates of up to 7.5% in
men and 10% in women for bladder
cancer and between 7.5% and 8.5% in
large intestinal cancer for women and
men, respectively, assuming the validity
of one of the studies.
(3) Dr. DeRouen also commented that
the epidemiological studies did not
adequately adjust the data for
confounding variables such as
urbanization and industrialization. He
noted that in a recent study where
additional variables were considered,
the statistical significance "dissipated"
relative to GI and urinary tract cancers.
As noted previously, taking into
account the multitude of interplaying
factors in epidemiology studies is a
complex problem. EPA has carefully
evaluated the available study results,
and taken collectively, they generally
support the hypothesis of the risk of
chloroform in drinking water. The
commenter's concerns that the impact of
several variables "dissipated" when re-
examined may be valid but these issues
do ndt vitiate the basis of the
regulations. EPA's finding that
chloroform may pose a carcinogenic risk
to humans is based primarily upon
animal toxicity studies.
(4) Dr. DeRouen noted that the
epidemiological studies would have
more credence if the health effects were
uniformly distributed over all race-sex
groups, but that this was usually not the
case in the drinking water/organics
studies.
EPA believes that it is not necessary
to have a uniformly distributed effect
over all race-sex groups, although when
this is the case conclusions can be more
strongly supported. Rarely in even well-
controlled experimental studies are the
effects uniformly distributed among sex
groups even in in-bred strains of test
animals.
(5] Dr. DeRouen stated that
unexpected and unlikely statistically
significant correlations were reported
for some cancer sites, and significant
relationships were not seen in humans
for liver or kidney cancers, which were
the effects seen in the animal tests.
EPA believes that site-specific cancers
are not necessarily observed across
species. This was supported by Drs.
Upton and Kennedy of NCI and FDA,
respectively.
(6) Dr. DeRouen commented that in
many studies, the presence of
statistically significant results would
change depending upon the statistical or
analytical model selected. In general,
therefore, the statistical methods are
usually specified in the protocol before
performing the study.
EPA agrees with this comment and it
is supported by Dr. Hoel from NIEHS.
The epidemiological studies cited were
correlational, preliminary and
hypothesis generating, rather than case-
control or prospective in nature. It is
therefore expected that further studies
could be designed based on those
already conducted which could be more
definitive. EPA has pointed out many of
these same problems in its evaluation of
the epidemiological studies in the
preamble accompanying the February 9,
1978, proposal, and EPA's Statement of
Basis and Purpose as did the NAS, Safe
Drinking Water Committee, in its review
of the studies. The primary basis for the
regulations is the animal toxicology
studies including the NCI bioassay
results demonstrating that chloroform
was an animal carcinogen under
conditions of the test. EPA has
concluded that the epidemiological
studies conducted so far are sufficient
hypothesis-generating studies, and taken
as a whole are supportive of the animal
data in pointing out the possible human
risk. The pros and cons of the studies
are discussed in more detail in the
Agency's Statement of Basis and
Purpose for these regulations.
64. Dr. F. J. C. Roe, representing the
Coalition for Safe Drinking Water,
submitted written and oral comments.
He also submitted copies of his recent
studies on chloroform carcinogenicity.
His major points and EPA's responses
are as follows:
(1) Dr. Roe stated that regulatory
contexts usually do not distinguish
between highly dangerous cancer-
causing agents and those such as
chloroform for which the evidence is
equivocal.
EPA has concluded that the SDWA
directs EPA to protect the public health
from any contaminant which "may have
any adverse effect" on human health.
Nevertheless, EPA evaluated the risk of
exposure to chloroform to the general
population based on its toxic effects,
cancer potential and exposure potential.
Chloroform has been found to be an
-------
68662
Federal Rbgister / Vol. 44, No. 231 / Thursday. November 29, 1979 / Rules and Regulations
animal carcinogen with well known
acute and chronic effects. Its presence in
treated Finished drinking water
potentially exposes over 100 million
people over their lifetime. EPA believes
this to pose a substantial risk.
<2) He stated that the NCI bioassay
was faulty because it erroneously used
corn oil as the vehicle for administering
chloroform to the test animals, not
enough control animals were used, and
concommitant exposure to other
carcinogens occurred. He urged that
prior to setting an MCL, the study
should be repeated in a wider dose
range and under better controlled
conditions.
Although additional studies taking
into account the above objections may
lead to slightly different responses one
way or the other, EPA believes that the
findings of carcinogenicity would
remain unchanged in light of previously
reported studies on other carcinogens
and the statistically significant results
obtained in the NCI chloroform
bioassay. EPA is sponsoring a study that
takes into account Dr. Roe's suggestions.
However, it would not be prudent to
delay setting an MCL for TTHMs
pending refinement of the data, given
the existence of credible data to date
demonstrating an adverse health risk.
(3) Dr. Roe stated that the Theiss
(pulmonary adenomas) study produced
erroneous statistical results.
As stated in the Statement of Basis
and Purpose, EPA did not rely on the
Theiss study to reach its conclusions.
The study was only included as
background information to the published
positive results.
(4) Dr. Roe said that the four
principles of the NAS (1977) and the
non-threshold risk concept for
carcinogenesis are not scientifically
sound.
As discussed previously, EPA relied
upon the judgment of the National
Cancer Institute who commissioned and
evaluated the bioassay of chloroform in
rats and mice and concluded that
significant rates of chloroform-related
tumors were detected in both rats and
mice under conditions of the test. The
National Academy of Sciences in
"Drinking Water and Health" (1977)
concluded that chloroform had been
shown by those and other studies to be
an animal carcinogen and, as such,
should be considered a risk to humans.
Other studies sponsored by EPA are
underway further refining our
knowledge of the toxicology and
carcinogenicity of chloroform, which
may provide more information on dose-
response relationships.
Federal health regulatory agencies
have carefully considered various
approaches for dealing with potential
human carcinogens and the possible
presence or lack of thresholds for
carcinogens. These agencies have
concluded as a matter of policy that in
the absence of evidence to the contrary
it must be assumed that substances that
have been shown to be animal
carcinogens in properly conducted tests,
must be assumed to be potential human
carcinogens, and that threshold
exposure levels below which there
would be no risk have not been
demonstrated experimentally.
Drs. Upton, Kennedy, Bingham, King
and Bates/Hoel of NCI. FDA, OSHA.
CPSC, and NIEHS, respectively,
supported EPA and these principles
enunciated by the NAS.
(5) Dr. Roe also submitted results of
three additional mouse studies that were
conducted on chloroform along with his
written comments.
In the first of these studies, the mice
of an outbred Swiss albino strain (ICI)
were given daily (six days per week)
oral doses of 17 mg/kg or 60 mg/kg
chloroform in tooth paste base for 77-80
weeks. The animals were observed for
an additional 16 weeks. Twenty-two
percent of the high dose males
developed adenomas or
hydronephromas of the kidney. In the
second study male mice of the same
strain responded similarly, with 18% of
the high dose having historically the
same tumors.
In the third mouse study, the response
of the male mice of four strains were
compared. In each of the four strains, 52
male mice were given 60 mg of
chloroform per kilogram (six days per
week) using the same experimental
design as previously outlined. As in the
previous experiments, mice of the ICI
Swiss strain developed more kidney
tumors than did the vehicle control
mice. No excess tumors were found in
the remaining three strains.
Dr. Cipriano Cueto (representing the
National Cancer Institute) stated to the
National Drinking Water Advisory
Council (1978) that Dr. Roe's results
were entirely consistent with the NCI
studies. Dr. Cueto also said that the
results of other studies relied upon by
Roe using rat and Beagle dog study were
also not surprising based on the doses
administered and the previous NCI
results.
(6) Dr. Roe calculated that a 70
kilogram man consuming one liter of
water containing 100 ppb of chloroform
would have a 7,000 fold safety factor.
Dr. Roe assumed that the mouse was the
most sensitive animal model and that 10
mg/kg was the "no effect level" for
kidney tumor enhancement.
As discussed previously, the EPA has
found that thresholds for carcinogens
have not been sufficiently demonstrated
and that this type of calculation
therefore contradicts that policy and
does not take into account many of the
principles enunciated by the NAS. Thus,
EPA has rejected Dr. Roe's approach as
unacceptable for regulating carcinogens
in drinking water.
(7) Dr. Roe also stated that it was
"reasonable to assume that none (of the
THMs) is more active than chloroform
itself," and, therefore, a level of 300 ppb
for chloroform alone would be as
protective as a similar limit for all THMs
as proposed by EPA. However, Dr. Roe
did not present any scientific facts or
principles to support his statement that
other THMs are less potent than
chloroform.
As discussed earlier, EPA has found
that in vitro mutagenicity data indicate
that the other THMs are more active
mutagens than chloroform. EPA's
regulation of total THMs has also been
based upon the similar chemical
structures and expected biological
activity, of all THMs, the availability of
analytical methods that analyze for total
THMs, and the fact that all THMs are
produced as a result of disinfection
practice.
(8) Dr. Roe stated that animal
detoxification mechanisms were
overwhelmed by the administration of
very high doses of chloroform in the
animal studies. He based his comment
on the following observations:
(1) Females of the species did not
appear at risk.
(2) Ames type assays were negative.
(3) Tumor formation was dependent
upon an indirect mechanism which
involved both sex hormone status and .a
deviation from normal metabolic
breakdown pathways.
In EPA's opinion, there are many
experimental conditions under which
one sex or the other is more sensitive to
the compound under test and therefore
this difference in the results is not
surprising. The in vitro assays of the
Ames type have been shown to be
insensitive to certain chemical classes;
simple chlorinated hydrocarbons appear
to be one of these chemical classes. Dr.
Roe, presented direct evidence to
support his third hypothesis; however,
other studies have shown a relationship
between chloroform toxicity and
testosterone levels in animals.
(9) Dr. Roe asserted that consistent
increased survival of three different
species exposed to chloroform suggested
a beneficial effect.
EPA has carefully reviewed the
available data and EPA does not believe
-------
Federal Register / Vol. 44. No. 231 / Thursday. November 29. 1979 / Rules and Regulations 68663
the evidence is sufficient Iff support this
contention.
64. Dr. Arthur Furst, representing the
Coalition for Safe Drinking Water,
submitted comments, many of which are
similar to those detailed previously. His
comments and EPA's responses are set
forth below:
(1} Dr. Furst commented that the NCI
chloroform bioassay was not definitive,
that results from animal studies using
high dosages (100,000 ppb] cannot be
extrapolated to predict human health
effects at low dosages (100 ppb], and
that human risks cannot be extrapolated
from animal data. These comments have
been responded to elsewhere in this
Appendix.
[2] He also faulted EPA's risk
assessment for not following the sigmoid
curve which he claimed should
represent the dose-response that one
would expect from biologically active
compounds. EPA has found that the
dose-response curve for carcinogens
would not be expected to be represented
by a sigmoid curve. Rather a linear non-
threshold curve is believed to be
appropriate in assessing a health risk
from carcinogens. Carcinogenic,
reversible, or non-reversible progressive
chronic response are not "all-or-none"
responses, nor do they lend themselves
to easily definable criteria for
categorizing the biological response.
Therefore, carcinogenic responses do
not satisfy the conditions upon which
use of the sigmoid curve is based.
(3) Dr. Furst also claimed that there is
a threshold for carcinogens, and that the
histological type of tumors produced in
the experimental animals was not
related to the human tumor response.
As discussed previously, EPA's policy
with respect to risk assessment for
potential carcinogens is to include the
conservative linear-dose response curve
and not a carcinogenic response
threshold level so as not to
underestimate potential risks. With
regard to the type of tumors in animals
versus human tumor responses, EPA has
concluded that the animal toxicity
studies can be related to man
irrespective to differences in tumor sites.
This is supported by Drs. Upton,
Kennedy, Bingham, King, Bates and
Hoel of NCI, FDA, OSHA, CPSC, and
NIEHS, respectively.
(4) Dr. Furst claimed that release of
benzo(a)pyrene could be a factor to be
considered when GAG treatment is
used. He questioned the use of CAC,
claiming that the treatment of water by
CAC may be replacing THMs with more
potent carcinogens such as
benzo(a]pyrene.
EPA has evaluated the available
studies involving extraction of CAC
with distilled water and the total level
of PAHs in the effluent were found to be
insignificant
(5) Dr. Furst suggested that a time to
tumor experimental design be
undertaken using multiple dose levels.
EPA is currently proceeding with
additional tests. However, regulatory
action need not await the outcome to
such studies,
[6] Dr. Furst stated that carcinogens in
the environment can interact, thus
modifying each others' responses. He
stated that there is no association
between organic chemicals in New
Orleans drinking water and cancer
rates.
EPA agrees that synergistic
interactions between toxic chemicals
can occur which is all the more reason
to consider approaches that will reduce
human exposures where feasible. The
association between New Orleans
drinking water and increased cancer
rates has been suggested by
epidemiology studies but is far from
conclusive. EPA's discussion of the
epidemiological studies is set forth
elsewhere in this Appendix, in the
preamble, and in EPA's Statement of
Basis and Purpose.
(7) Dr. Furst objected to the conditions
under which the NCI bioassay was
carried out He felt that a single massive
dose by oral gavage does not compare
with a minute fraction of the dose
ingested throughout the day. The doses
used in this bioassay overwhelmed the
ability of the liver to detoxify the THMs.
EPA has concluded that high dose
animal studies are necessary and valid
methods of determining risks from
human exposure at lower doses.
These questions are more fully
addressed elsewhere in this Appendix
and in the Statement of Basis and
Purpose.
65. Comments submitted by Dr. Frank
L. Lyman on behalf of the Coalition for
Safe Drinking Water and EPA's
responses are as follows:
(1] Dr. Lyman commented that the 100
ppb level for TTHMs is unnecessarily
restrictive.
As discussed thoroughly in the
Statement of Basis and Purpose, EPA
believes that human exposure to
carcinogenic chemicals should be
minimized to the extent feasible. The
level of 0.10 mg/1 TTHM in this interim
regulation is based upon technological
and economical feasibility in that the
level is achievable and is consistent
with the SDWA mandate to reduce
exposure to contaminants in drinking
water to the extent feasible, taking into
consideration die potential health risks.
(2] Dr. Lyman staled that the possible
benefits of GAC are unknown and GAG
itself may have harmful effects on water
quality.
The questions of benefits and release
of harmful chemicals have been
addressed previously in this Appendix.
Data to date do not support the
speculation that there are adverse
effects from GAC use.
(3) Dr. Lyman noted that chloroform
has been found in tomatoes, grapes and
milk and is also produced in food
processing. He urged that the total body
burden must be considered in regulating
chloroform.
As discussed previously in this
Appendix and in the Statement of Basis
and Purpose, EPA has examined several
exposure routes of chloroform and feels
that regulations controlling chloroform
in drinking water are necessary since
water can be the most significant source
of exposure under typical conditions.
(4) Dr. Lyman commented that, in
spite of wide-spread chronic industrial
exposure to chloroform, there is no
evidence of human carcinogenesis.
The unavailability of occupational
risk data showing a precise relationship
between exposure to chloroform in the
work place and human carcinogenesis
does not mean that chloroform poses no
risk to humans. Systematic and
scientifically sound studies have not yet
been conducted to evaluate the
possibility. However, in view of the
positive carcinogenic response in the
animal studies, EPA feels that
regulations are appropriate at this time.
This will result in reduced human
exposure to many disinfection by-
products, not only chloroform and
THMs.
(5) Dr. Lyman stated that animal
studies are useful in comparing effects
on laboratory animals to human toxicity.
EPA concurs with the use of animals in
evaluating toxic effects of chemicals.
EPA believes that carcinogenic! ty is one
of several end points of toxicity and the
statement by Dr. Lyman presented
below also applies to the carcinogenic
effect: "The toxicologist uses lower
animals to predict the effects of
chemicals on humans. Generally, the
toxicity of a compound in lower animals
is similar to that in humans on a dose
per unit of body weight, particularly if
the metabolic pathways and
detoxification mechanisms are similar."
Thus, EPA believes that cancers
produced by chemicals in animals are
evidence of human risk. Drs. Upton,
Kennedy, Bingham, King, Bates and
Hoel of NCI, FDA, OSHA. CPSC, and
NIEHS, respectively, support this belief
as presented in Appendix B.
(8] Dr. Lyman criticized EPA's use of
the results of animal studies exposing
them to high dosages to extrapolate
-------
68664 Federal Register / Vol. 44. No. 231 / Thursday, November 29, 1979 / Rules and Regulations
human health risks associated with
exposure to low dosages on the grounds
that high dose exposures were more
likely than low doses to cause tissue
damage which he claimed was a
prerequisite to cancer introduction by
chloroform. In support of his argument,
he noted that high doses of liver and
kidney toxins cause cancer to develop in
those organs. He concluded that
because lower dosages were less likely
to damage tissue, they were also less
likely to result in the development of
tumors.
EPA does not agree with Dr. Lyman's
hypothesis that tissue damage is
necessary for cancer induction. The
scientific community has hot yet
reached a consensus on this point. There
are chemcials that cause the kind of
tissue damage Dr. Lyman describes that
do not go on to cause cancer (i.e., 1,1,1-
trichloroethane). Thereby, tissue
damage does not invariably lead to a
carcinogenic response. Therefore, it is
prudent and consistent with current
scientific thought to assume that low
level exposure to carcinogens, which
may or may not cause direct tissue
damage poses a human health risk.
FDA, CPSC, NIEHS, NCI, and EPA agree
that site-specific cancers are not
necessarily found across species.
(7) Dr. Lyman also said that
thresholds for carcinogens exist.
EPA believes that thresholds for
carcinogens have not been
experimentally demonstrated to date.
This is thoroughly discussed in the
preamble and in response to previous
comments.
(8) Dr. Lyman commented that in
order to produce tumors in people It
would require drinking 15,000-30,000
gallons of water daily with a
concentration of 311 ppb to produce
tumors in humans.
EPA has evaluated this estimate and
has concluded that the direct
comparison of dosages from animals to
humans in this way neither scientifically
valid nor relevant.
(9) Dr. Lyman noted that one must
differentiate between a real and
potential risk.
EPA believes that sufficient
information has been presented to
demonstrate a risk from THM exposure
that reduction of that risk is feasible and
regulation is warranted and required by
the SOW A.
66. Comments submitted on behalf of
the Coalition for Safe Drinking Water by
Parrel R. Robinson and EPA's responses
are as follows:
(1) Dr. Robinson said that surveys of
drinking water in various cities did
demonstrate the presence of THMs but
there were no realistic historical data
with which these levels could be
compared; the available epidemiological
data are unreliable.
EPA is relying primarily on the animal
toxicity data as the basis of the
regulation. The correlational
epidemiology is not inconsistent with
this data, and assuming that similar raw
water quality and chlorine dosage have
been used over previous years which is
a reasonable assumption in most cases,
THM levels would not be significantly
different.
(2] Dr. Robinson commented that
there are significant problems in
interpreting animal data and
extrapolating their results to humans.
This has been responded to in detail
above and in the preamble and
Statement of Basis and Purpose.
(3) Dr. Robinson said NCI bioassays
are only applicable to that strain of
animals under the conditions of testing.
EPA believes that properly conducted
studies in test animals do provide
evidence of potential human risks from
those chemicals. This is thoroughly
discussed elsewhere in this Appendix,
the preamble and the Statement of Basis
and Purpose.
(4} Dr. Robinson commented that
there is a threshold for carcinogens. He
claimed that threshold cancer response
extrapolations are contrary to scientific
fact.
EPA believes that thresholds for
carcinogens have not been
demonstrated at this time. This is
discussed in detail in the preamble, this
Appendix and in the Statement of Basis
and Purpose.
(5) Principles enunciated by the NAS
are not principles but opinions.
EPA has relied on the NAS as
representing the consensus of scientific
opinion on these subjects.
67. Comments submitted by Dr.
Alexander Grendon on behalf of the
Coalition of Safe Drinking Water were
as follow, that:
(1) EPA has not balanced costs
against benefits for GAC. He stated that
the costs were enormous while the
theoretical benefits are minor.
(2) That there is a threshold for
carcinogenesis.
(3) That cancer death rates have been
declining for 25 years.
(4) That a person would have to live
74 years before a tumor would develop
due to chloroform exposure.
(5) A person would have to live 35
lifetimes before dying from chloroform
induced cancer.
Most of these comments has been
addressed previously in this appendix
and in the preamble. In regards to the
time-to-tumor question, EPA feels that
the state-of-the-art of toxicology does
not provide for estimates such as those
Dr. Grendon submitted. Rates of some
types of cancer have declined but other
types have risen in the past 25 years.
68. Comments submitted by Dr.
Richard Reitz, representing Dow
Chemical Company, and EPA's
responses are as follows:
(1) Dr. Reitz commented that the use
of GAC for organic chemical removal
may release chemicals into treated
waters that are carcinogenic. EPA has
responded to this comment elsewhere in
this Appendix.
(2) Dr. Reitz criticized EPA's use of the
most conservative model for assessing
human risk which he said greatly
overestimated the risk of trace levels of
organic chemicals in drinking water. He
said that NCI should develop two
separate risk extrapolation models, one
for direct-acting carcinogens and
another for metabolically model
activated carcinogens. He commented
that the extrapolation developed by Dr.
David Rail and used by EPA's Cancer
Assessment Group (CAG) was not
appropriate for THMs since THMs are
not direct-acting carcinogens but are
carcinogens generally "involved in the
variable drug metabolizing system," for
which that model was not designed.
In support of his argument that EPA
used an inappropriate risk model, he
cited inconsistencies between the mouse
and rat data in the NCI study. He noted
that although based on the model one
would have expected rats to be more
sensitive to chloroform than mice, even
though metabolism was required to
activate chloroform, the opposite results
were obtained. He therefore concluded
that EPA's model overestimated the risk
to rats by eleven-fold and overestimated
the risk to humans by an even greater
margin. Using pharmakokinetic data. Dr.
Reitz predicted that the "chloroform
risk" was one order of magnitude lower
than that estimated by EPA.
EPA recognizes that other risk
estimation models exist. Depending
upon various assumptions, the computed
levels can be significantly different
among models. EPA has relied on the
scientific expertise in the area of risk
assessment of the NAS and EPA's CAG
for its risk models which are considered
to be state-of-the-art. While these
models may be more conservative than
Dr. Reitz's model, EPA believes that this
was a reasonable and responsible
choice in view of the SDWA's mandate
to protect the public health.
EPA further found that the NAS-CAG
models were appropriate for use for
chloroform based on the best scientific
evidence available. The fact that the
results from the rat studies showed them
to be four times less sensitive to
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations 68665
chloroform than mice does not mean
hat the data cannot be used for human
isk extrapolation. Species variability in
ancer inductions mechanisms could be
in explanation for this apparent
inconsistency.
(3) Dr. Reitz stated that the doses of
chloroform used in the NCI study
produced gross liver damage long before
the production of tumors. Thus, he said
it was impossible to determine whether
the carcinogenicity of chloroform was
due to a genotoxic reaction or simply a
secondary reaction to the extensive liver
and kidney necrosis (i.e., epigenic).
As discussed previously, EPA feels
high dosage tests are necessary and
valid. EPA believes that large doses
over long periods of time are required to
produce effects in relatively small
populations of animals and to increase
the experimental sensitivity. The NCI.
FDA, CPSC and NIEHS have concurred
with this conclusion.
Moreover, one cannot conclude that
the use of high dosages in animal
experiments means that the resulting
carcinogenicity is attributable solely to
a toxic assault on the organ. Rather,
toxic assaults leading to organ damage
do not always evoke a carcinogenic
response. Therefore, the particular
ihemical, in this case chloroform, must
ilso be implicated as a factor when a
:arcinogenic response is found.
(4) Dr. Reitz commented that since
chloroform belongs to the class of
chemicals which require metabolic
activation for toxicity, one would expect
the incidence of oncogenicity to be
greater in those species with greater
capacities to metabolize the chemical.
Dr. Reitz assumed that the metabolic
capability of rats was greater than mice
and that of humans was greater than
rats. He also postulated that glutathione
availability was the limiting factor in the
rate of macromolecular binding (a factor
hypothesized as being a critical step in
carcinogenicity).
Since more glutathione was expected
to be available after lower dose
exposures, Dr. Reitz argued that the
chemical's carcinogenic potential at low
dosages would be lower than if
exposure had occurred at higher
dosages. Based on these assumptions,
he concluded that the human risk for
chloroform was 71 times less than that
estimated by CAC. Dr. Reitz said his
calculations would result in an MCL
between 0.01 mg/1 and 0.1 mg/1 for
incremental risk of 10" 8 and 10" ".
respectively.
EPA does not agree to with Dr. Reitz's
issumptions. His hypothesis concerning
}lutathione availability as a limiting
Factor in cancer induction has been
shown not to be valid in tests using
other similarly metabolized carcinogens
at low exposure levels. Despite the
differences between Dr. Reitz's and
EPA's risk estimates, no specific risk
value served as the basis for EPA's
TTHM MCL, which was based upon
technical feasibility factors.
(5) Dr. Reitz cited a study whereby
chronic industrial exposure (50-125
ppmj of British Confectionary workers
to chloroform for up to 10 years twenty
years ago did not produce convincing
epidemiology to link chloroform with
increased cancer risk. EPA recognizes
the difficulties involved with conducting
epidemiology studies and this subject
has been addressed previously.
(6) Dr. Reitz recommended the
following changes be incorporated into
the proposed THM regulation:
(a) That the MCL should be increased
to 1.0-10 mg/1 based on health effects
data and risk models.
The MCL was based on a positive
qualitative findings of carcinogenicity
from animal bioassays and not on any
quantitative risk extrapolation. The
MCL for chloroform is that level which
can be achieved given technological and
economic feasibility factors.
(b) That definitive interspecies
metabolism studies be carried out to
allow a rationale species/species
extrapolation. EPA agrees mat this
would provide additional information
and has additional studies underway.
However, regulatory action need not
await the outcome of such studies.
(c) That a complete evaluation of the
chloroform carcinogenicity potential
below 200 mg/1 be conducted. More
research can always be conducted. EPA
has an ongoing carcinogenicity study to
evaluate chloroform at low levels of
exposure. Again, regulatory action need
not be delayed.
69. Dr. Joseph Schlosser, of Tulane
Medical School, stated that:
(1) Bronchiogenic cancer should not
be related to the Mississippi River and
drinking water.
(2) The petrochemical industry could
be the cause of increased cancer in
Southern Louisiana.
(3) There is no consistent thinking
about what the reason is for the high
incidence of cancer in the New Orleans
area. EPA's conclusions regarding the
human epidemiology data, including that
involving New Orleans, has been
discussed elsewhere in this Appendix,
in the preamble, and in the Statement of
Basis and Purpose.
70. Three commenters said that
separate MCLs should be set for each
THM, such as chloroform, instead of for
total THMs. One of these said that
MCLs should only be established for
those specific contaminants proven to
be human or animal carcinogens. It was
argued that, while all THMs were
included in the proposed standards, only
chloroform has been shown to produce a
dose-response relationship for epithelia
tumors of the kidney and renal pelvis in
the rat and for hepatocellular
carcinomas in mice. The other
commenters felt that if standards were
set for the THMs, concentrations of all
THMs should be converted to the same
base such as milliequivalents because
grouping THMs on a weight basis and
expressing the total THMs as mg/1 was
scientifically incorrect.
EPA's rationale for establishing a
MCL for total THMs, instead of for only
chloroform or for each THM separately,
is set forth in greater detail in the
preamble to these regulations and
commenters are referred thereto.
Although less is known about the health
effects of the other THMs than about
chloroform, EPA believes that
carcinogenicity need not be proven
before regulatory action may proceed.
Based upon the similarity in chemical
structure of all the THMs and the best
available information on the health
effects of the other THMs, EPA believes
that they, as well as chloroform, pose
adverse health risks which should be
minimized to the extent feasible. It is
also reasonable to regulate total THMs
as a group because the gas
chromatographic analytical method
concurrently analyzes all four THMs;
also treatment methods that would be
employed to reduce chloroform would
simultaneously reduce all of the THMs,
since they are all formed through the use
of chlorine in the disinfection process.
On the question of the use of
milliequivalents instead of milligrams,
EPA does not believe that such an
approach would necessarily be
meaningful since insufficient
information is available to judge the
relative potency of the four THMs to
warrant that approach. Moreover,
milligrams per liter have been used as
the standard measurement for other
drinking water MCLs in the NIPDWR
and this term has become familiar to the
water utilities that must comply with
such standards.
71. In addition to those comments
previously discussed, 136 comments
were received discussing other issues
related to sampling and monitoring for
TTHMs. Of these, 43 commenters said
they supported the sampling and
monitoring requirements in the proposed
regulations and found them to be
adequate and reasonable. Many of these
commenters, however, felt that EPA or
the States should conduct or pay for the
analyses. Seven commenters opposed
-------
68666 Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations
the monitoring program because of the
added cost burden on utilities and noted
the lack of laboratory facilities and
skilled personnel. Fifty-one comments
favored the monitoring requirements but
opposed any requirement to notify the
public of such results on the grounds
that the public notification requirement
would create unnecessary, expensive
paper work as well as a "bad-feeling"
among the public. One commenter felt
that the reporting of THM monitoring
data to EPA by utilities should apply
only to States that are qualified for
primacy.
EPA has already responded in this
Appendix to those comments addressing
the cost of monitoring. Under the
SOW A. the cost of compliance with
these regulations must be borne by the
water utilities and EPA has taken this
factor into consideration in determining
minimum monitoring frequencies and
has found that such costs are
reasonable. With respect to public
notification of the results of TTHM
monitoring. Section 1414(c) of the
SDWA requires that systems notify the
public of any failure to comply with an
applicable MCL as well as any failure to
perform required monitoring. EPA does
not believe that the costs of such public
notification are unreasonable and any
public notice may include appropriate
explanation so that the public is
adequately informed, but not misled.
The results of all monitoring are
required to be reported to the States so
that compliance with the regulations can
be properly enforced and technical
assistance can be provided to correct
problems at the earliest possible time.
Systems are also required to report
results to EPA until such requirements
are adopted by the States with primacy.
72. Twenty-four additional
commenters raised questions regarding
laboratory capabilities,'quality
assurance of results, and sampling and
analytical procedures. They commented
about the lack of qualified and
experienced laboratories in the U.S. to
perform TTHM analyses and about the
fact that analytical procedures were not
very well defined. They urged that the
laboratory certification process be
expedited and the analytical procedures
be defined as soon as possible.
On the issue of the availability of
laboratory facilities and analytical
procedures, EPA has responded to those
commenters concerned about the
availability of sufficient numbers of
laboratories capable of providing
acceptable analytical data by extending
the time frame for initiation of
monitoring by systems serving more
than 75.000 people from the proposed
three months after promulgation to one
year after promulgation. The 10,000 to
75,000 size category of systems are given
3 years from promulgation to begin
monitoring. This will allow additional
time for Slate and private laboratories
to develop their capabilities and to
became certified by EPA to provide data
in support of compliance
determinations. A quality assurance and
certification program is also being
developed by EPA. to determine the
capable laboratories and to insure the
reliability of data.
73. One commenter noted that EPA
had failed to quantify the contribution of
industrial and municipal discharges to
the total concentrations of THMs and
their precursors. EPA was urged to
control THMs and precursor materials
at their source; much of the THM in
drinking water could be eliminated by
not permitting any industrial or
municipal discharges of THMs or THM
precursors.
While THMs do occur in some
drinking water sources as a result of
municipal and industrial discharges,
EPA has found that such levels are
generally significantly lower than the
levels associated with chlorination by-
products in the finished drinking water.
Most THMs in drinking water are the
result of the reaction between chlorine
and natural precursor compounds in the
treatment process. Therefore, in most
cases, control of THMs or precursor
compounds municipal or industrial
discharges would not likely have any
significant effect upon THM levels in the
drinking water.
74. One commenter noted that
because of the inaccuracy and
imprecision inherent in the analytical
procedure for measurement of THMs,
the MCL should include an allowance
for the variations in analytical results.
Although EPA has established a single
numerical value for the TTHM MCL, the
variabilities associated with the
analytical procedures have been taken
into account in determining what
laboratories will be deemed qualified for
performing TTHM analyses. EPA has
determined that 20% of 0.10 mg/1 TTHM
will be an allowable variation in the
analytical results for purposes of
laboratory approval and certification.
Recent data show variations in properly
run procedures of 10% to 20% and it is
expected that as more experience is
gained, the allowable variation will be
reduced. Thus, while it is necessary to
establish a single MCL value, quality
control of laboratories is believed to be
the most appropriate way of taking into
account analytical variability.
Appendix B—Summary of Major
Comments (for responses, see Appendix
A)
I. Coalition for Safe Drinking Water
A. Introduction
The Coalition for Safe Drinking Water
is a group of approximately 90 water
systems—both investor and municipally
owned—formed to present information
and comments concerning EPA's
proposed regulations.
The Coalition's doubts and
disagreements about the substance of
the proposed regulation centered upon
EPA's conclusions that:
[1] The trace amount of THMs
normally found in drinking water may
pose a health risk, and,
(2] The GAG treatment technique is,
at this time, required to reduce the levels
of THMs in drinking water.
The Coalition also doubted EPA's
authority to propose these new
requirements as "amendments" to the
interim primary drinking water
regulations.
B. Legal Issues
1. EPA lacks the authority to
promulgate the regulations as
amendments to the National Interim
Primary Drinking Water Regulations.
The regulations are entirely new
regulations and not modifications and to
propose these regulations requires
recommendations from NAS. The NAS
has not made this recommendation.
Further, the GAC technology was not
available in December 1974 and all
exemptions for water systems to avoid
hardship will end on January 1,1981.
C. Health Issues
1. Chloroform poses no potential
cancer risk and there are no available
data that support the premise of a
causa] relationship between the
concentrations of THMs normally found
in drinking water and cancer in humans.
2. The epidemiologicel studies that
have been conducted concerning
drinking water and a possible
connection with cancer in humans are
inconclusive.
3. EPA has relied upon animal studies
for the hypothesis that trace organics
pose a health concern. However,
extrapolation of results in animal cancer
studies to humans is fraught with its
own set of problems and uncertainties.
4. The proposed regulations are based
upon fear of the unknown using
equivocal animal data and extrapolation
models and methods which are
unreliable.
5. The study cited by EPA to support
the carcinogenicity of chloroform was "a
preliminary screening test (by the
-------
Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68667
National Cancer Institute (NCI]) and not
a definitive study." The study was not
intended to be used to extrapolate
health effects of chloroform to drinking
water levels. The NCI study was
inadequately controlled and did not
follow proper scientific protocols. A new
EPA/NCI study is underway and
corrects deficiencies of the previous
study and it is recommended that the
implementation of any regulations be
delayed until the studies are complete.
6. Dr. Roe's studies showed a no
observed effect at 595,000 (drinking
water equivalent) ppb chloroform in
drinking water. Dr. Roe recommended a
level of 300 ppb THM in drinking water
based upon his studies of chloroform.
Dr. Francis). Roe's study with
chloroform on dogs, rats and four strains
of mice at low dose levels does not
produce tumors in animals. Three
hundred ppb would provide a margin of
safety of 2,000. However, EPA uses 500
as a margin of safety.
7. EPA's proposed MCL of 100 ppb is
needlessly low and will require costly
additions or changes to water treatment
facilities without any corresponding
benefit being obtained.
8. There are no health effects data
which support carcinogenicity of the
other THMs.
9. Based upon most appropriate
statistical extrapolation model, the level
of the THM MCL should be no lower
than 0.30 mg 1 since this provides a more
than adequate margin of safety.
However, this level is still too low to be
justified on a cost-benefit basis if GAC
is required.
10. There is no hard evidence that low
level exposure to any of the chemicals
produces cancer.
11. EPA estimates of environmental
exposures to chloroform appear to be
erroneous and it is suggested that EPA
make every effort to obtain correct
values for contributions from air, food
and water. Also, there is the possibility
that in vivo formation of chloroform and
other THMs in the human body may
occur. At this point, the available data
suggest that more cost-effective
avenues, such as control of chloroform
in the work place, may be available for
reducing THMs in the environment than
by implementing the proposed THM
MCL.
12. The concentrations of THMs
detected in water systems present no
mutagenic, teratogenic, or acute,
subchronic, and chronic toxicological
health risk to the public.
13. EPA has misconstrued the four
very general "principles" stated by NA5.
EPA has not properly used these
principles and has ignored the available
data. With regard to the first principle,
EPA has not taken into account a
number of variables in extrapolation of
the animal data to humans; some of
these variables include differences in
such items as species response to
carcinogens, weight between animals
and man, intake of food and water, and
routes of exposure. With regard to the
second principle, EPA has ignored
existing scientific data that show a
threshold for no-effect responses with
respect to a number of suspected
carcinogens; there are a number of
suspected carcinogens for which animal
experiments have established a
threshold level of effects; experiments
involving humans suggest a no-effect
level exists for chloroform; in vivo
biological processes militate in favor of
a no-effect level; and human exposure to
natural carcinogens without adverse
health effects support thresholds. With
regard to the third principle, EPA has
not considered the significance of the
detoxification and repair mechanisms
operative in animals and humans in its
health assessment of THMs. With
regard to the fourth principle, EPA has
ignored the guidelines for assessing risk
for chloroform as set forth in EPA's
"Interim Guidelines for Carcinogen Risk
Assessment." EPA used only the linear
model for extrapolating the NCI data to
humans, ignored the data of Roe,
Eschenbrenner, and Miller, and ignored
the estimates of risk by Tardiff using the
"margin of safety," "probit-log" and
"two step" extrapolation models.
14. EPA has ignored the relationship
between dose and time-to-tumor
observation in assessing the health risk
of a carcinogenic material.
D. Treatment Technology and
Economic/Energy Assessments
1. GAC has never been tested or
proven on a full-scale operation in the
United States and therefore constitutes
a nationwide experiment for water
treatment.
2. The use of GAC will have
substantial financial impact upon water
supplies and actual costs are very
difficult to predict and are understated.
For example, the average capital cost for
a system serving over one million people
will exceed $106 million with annual
costs of more than $23 million. Rate
increases for residential customers
could be in the range of 40-70% and
these rates could double where there are
specific problems, such as land
acquisition. These costs may result in
insurmountable problems for some
utilities in obtaining financing for GAC
treatment facilities. EPA's assessment of
the feasibility of financing the GAC
treatment facilities is totally out ot step
with the realities of both the financing
markets and operating needs of the
public utilities.
3. The regulations will promote
substantial new consumption of energy
in operation of the treatment
technologies as well as in secondary
energy consumptions such as energy
usage for GAC regeneration or energy
associated with the manufacture and
transportation of GAC.
4. The economic impact assessment
did not take into account the costs of
treating wastewater from GAC
operations, such as backwash waters,
wet scrubbers and drainage from carbon
slurries. It is estimated that 50,000
gallons of waste water will be generated
for every one million gallons of drinking
water treated and half of that amount
will need to be discharged. This will
result in increased flows and higher
O&M costs at municipal waste water
treatment facilities on the order of four
percent.
5. The costs were underestimated
because of specific factors in the
analysis. Based upon the use of CAC,
the difference between their potential
national cost estimates and EPA's
estimates could be explained primarily
by four factors [It was not clear to what
extent these comments differentiated
between costs for GAC for TTHM
control and costs for GAC to control
other synthetic organic chemicals in the
separate treatment technique
requirement):
(a) EPA determined its estimated
capital costs for a system based upon
the capacity of the entire system;
whereas, the coalition estimated the
system capital costs as equal to the sum
of the capital costs for each treatment
plant based on the capacity of each
plant.
(b) EPA's estimates were based upon
the system capacity on the average day
of the peak month; whereas, the
coalition's estimates were based upon
the actual capacity of each treatment
plant.
(c) EPA assumed that some of the
affected systems would design facilities
for a 9-minute empty bed contact time
(EBCT)^whereas, the coalition assumed
that all GAC facilities would be
designed for an 18-minute EBCT.
(d) The coalition's estimates for
specific systems, based on the costing
out of the individual components, were
30-60% higher than EPA's proposed
estimates.
6. EPA has underestimated the costs
of implementing the regulation by
underestimating the number of impacted
systems. This is the result of basing the
analyses upon a model for the water
supply industry and using a number of
unfounded assumptions regarding the
-------
68668 Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations
number of systems that purchase water
and use alternate disinfectants. Also,
assumptions and predictions based upon
MOMS were used to determine the level
of THMs and the extent to which
systems would be impacted further.
Instead, EPA should have conducted
sampling at all systems and based its
estimates upon those results. The
estimate of the 390 systems serving
greater than 75,000 persons was not
derived from EPA's Inventory of
Systems but was based upon the TBS
Policy Testing Model which left out
numerous systems including all Federal
Systems (e.g. District of Columbia) and
the States of Hawaii and Alaska. Also,
the hypothetical results of the TBS
Model were never checked on to
compare with reality. Finally, the
number of systems using specific
treatment systems such as GAC or no
cost modifications were arbitrary
assumptions.
7. EPA should provide a cost estimate
of the stated goal of lowering the MCL
at a later time to 50 ppb or 10 ppb.
8. The financial implications on water
utilities have been underestimated*by
EPA. The financial analysis assumed
that the rate increase required to finance
the necessary revenue requirements
would be obtained easily. Also,
projections of future capital
requirements in addition to the cost of
the CAC process for various water
systems were not factored into the
analysis.
9. In order to install GAC, water
utilities will need to raise capital
through large rate increases. There are
substantial regulatory barriers which
could preclude water utilities from
obtaining the necessary rate increases.
Even if utilities are able to raise the
capital funds, the quality of their credit
and the attractiveness of their common
stock will be severely reduced; this will
reduce their ability to obtain external
financing for normal water supply
activities.
10. The GAC treatment process may
result in serious problems and these
may outweigh the alleged environmental
benefit associated with GAC treatment.
These problems include potential air
pollution from regeneration and the
waste water associated with GAC from
contactor disinfection, backwashing,
GAC quenching and transport, drainage
from carbon slurries, and the
regeneration furnace scrubbers. The
total volume of waste water resulting
from GAC facilities will be
approximately 43.000 gallons per million
gallons of water treated. Some of the
waste water can be recycled but some
will require pretreatment prior to
disposal.
11. The use of GAC may constitute a
larger health hazard than that of the
alleged improvement of water quality.
The potential health hazards associated
with GAC include desorption,
chromatographic effect (competitive
displacement), resorption (leaching) of
heavy metals and polycyclic aromatic
hydrocarbons contained in the virgin or
regenerated carbon, release of carbon
fines, promotion (catalytic reactions) on
the carbon itself of hazardous
compounds due to chemical reactions
between chlorine and organic
compounds, bacterial growth on the
carbon and air pollution from
regeneration facilities. Indirect hazards
associated with the GAC usage derive
from the manufacture of GAC and the
production of energy necessary to
operate GAC facilities. These industries,
such as the coal industry, pose a high
risk of morbidity and mortality to the
workers. Because of these concerns,
additional research and testing should
be conducted prior to implementation of
GAC in this country's major
waterworks. It is suggested that
toxicological evaluations be conducted
using concentrated effluents from GAC
to assess these potential hazards.
12. EPA is required to analyze the
costs of its actions in terms of the
benefits hoped to be obtained but EPA
has not done that.
E. Other Comments
1. EPA has failed to quantify the
contribution of industrial and municipal
discharges to the total concentrations of
THMs and their precursors. EPA should
control the THMs and precursor
materials at their source, and much of
the THM in drinking water could be
eliminated by not permitting any
industrial or municipal discharges of
THMs or THM precursors.
2. Because of the inaccuracy and
imprecision inherent in the analytical
procedure for measurement of THMs,
the MCL should include an allowance
for the variations in analytical results.
3. If there is a necessity for a MCL for
THMs, the MCL should apply to all
water systems.
4. The EPA has not addressed the
significant primary and secondary
environmental problems associated with
the use of GAC treatment facilities. Such
concerns would normally be considered
in an Environmental Impact Statement
(EIS) prepared in accordance with the
National Environmental Policy Act.
However, EPA has stated that the
supporting documentation for the
regulations is the functional equivalent
of an EIS. The EPA documents are not
the functional equivalent of an EIS as
they have not remotely analyzed the full
potential environmental impact.
n. American Water Works Association
The AWWA's recommendations
were:
1. Expanded and accelerated health-
effects research on THM and synthetic
organics as recommended by the NAS.
2. Establishment of 100 ppb level of
TTHMs as a goal for all public water
supply systems.
3. Elimination of EPA's proposed
requirement of GAC as a treatment
technique. In its place, EPA sponsorship
of at least four plant-size research
projects to gather financial and
operating, as well as scientific data.
4. Adoption of EPA's proposed
monitoring program for TTHM, except
that public notification should not be
required.
5. Establishment of an EPA financed
and operated monitoring program for
synthetic organic chemicals.
III. Environmental Defense Fund
The scientific evidence supporting the
regulations is massive and convincing.
A number of epidemiology studies have
been conducted and provide strong
support for the regulations in that taken
as a whole they show a consistent
pattern of association between drinking
water and cancer mortality rates at
certain sites.
Using the NAS model and Dr. Roe's
data, the estimated risk of ingesting 200
ppb of chloroform over a lifetime in a
community the size of one million would
be predicted to result in 20 excess
cancer deaths.
In a case study in New York State, it
was found that for urban area
populations drinking chlorinated water
had a relative risk of 2.7 compared to
populations in urban areas that do not
drink chlorinated water. This would
result in 250 excess cancer deaths per
year in a population of one million.
The benefits of the regulation far
outweigh the costs.
Because chloramines are quite
ineffective in killing viruses and because
viruses are not monitored for in drinking
water supplies, any encouragement of
chloramine usage should proceed with
great caution.
The overwhelming consensus of the
scientific community is that testing
animals with high dosages is perfectly
adequate for relating to humans.
Any delay in promulgating the
regulations would be unconscionable, in
view of the health effects data, and
improper, in view of the requirements of
the SDWA.
It is abundantly clear that the public
wants safer drinking water since large
-------
federal Register / Vol. 44. No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68669
numbers are turning to alternative
sources of water (bottled water] or to
home water treatment devices.
Unfortunately, all of the available
evidence indicates that these
alternatives are not adequate substitutes
for municipally treated drinking water.
A regulation applicable to only half
the population is not good enough and is
inconsistent with the congressional
intent that maximum feasible protection
of public health be provided. The
coverage should be expanded.
The level of the MCL should be the
level achievable by the application of
the most effective THM reducing
technique applied to a relatively clean
water source, such as an average water
supply. A level of 50 ppb was suggested
as a possible alternative to the proposed
MCL.
IV. Supporting Comments on Health
Basis of Regulation
A. Dr. Samuel Epstein, from the
University of Illinois, endorsed the
following principles:
1. There is no safe level of exposure to
a carcinogen.
2. Animal carcinogens should be
considered as human carcinogens.
3. Chemicals found to be carcinogenic
at high doses in animals are
carcinogenic at much lower doses in
humans.
4. Chloroform is not the only chemical
of concern in contaminated drinking
water.
5. If the effects of cigarette smoking
are eliminated, cancer rates are not in
decline for many sites.
6. There have been 13 epidemiological
studies which in context demonstrate an
association between chlorinated
drinking water and gastrointestinal
urinary tract cancer.
7. GAC is a proven water treatment
technology.
Dr. Epstein summarized the scientific
basis for the regulations as follows:
1. Less than 10% of the 700 chemicals
identified have been tested "for their
toxicologic and carcinogenic effects."
2. NCI lists 23 of these as carcinogens,
30 as mutagens and 11 as promoting
agents.
3. Fish and shellfish which live in
polluted water have a high incidence of
tumors.
4. Organic extracts of drinking water
have been shown to be carcinogenic and
mutagenic in animal tests.
5. Organic chemicals in drinking water
have shown reproductive effects in one
preliminary laboratory test.
6. Epidemiologic studies suggest
association between drinking water
contaminants and cancer.
B. Susan B. King, chairperson of the
U.S. Consumer Product Safety
Commission (CPSC) testified that CPSC
concurred with the four principles for
safety and risk assessment set forth by
the NAS in its report, "Drinking Water
and Health" and that CPSC also utilized
them in their regulations of carcinogens.
CPSC also concurred in EPA's
conclusion that humans are also
susceptible to effects observed in
animals, as properly qualified. Ms. King
noted that thresholds have not been
demonstrated at which a "no effect"
level for a carcinogen could be
presumed and that varying individual
susceptibilities must be considered in a
heterogenous human population. She
endorsed testing of chemicals at high
levels in animals for assessing possible
human risks. CPSC uses factors such as
potency, extent and nature of human
exposure and human uptake factors in
evaluating risks from carcinogens.
CPSC's interim policy for regulating
carcinogens consists of prohibiting use if
a reasonable substitute exists and
prohibiting use in the absence of a
reasonable substitute unless this would
result in both unacceptable social and
economic costs. CPSC's approach is
comparable to EPA's in that the extent
of the exposure and risk are considered
as well as the availability and costs of
alternatives.
C. Dr. Donald Kennedy, Commissioner
of the Food and Drug Administration
(FDA), stated that FDA was in full
accord with the objective of protecting
public health from organic chemicals in
drinking water, and endorsed EPA's
efforts to reduce exposure to THMs.
FDA's recent actions to remove
chloroform from drug and cosmetic
products were consistent with this
position.
The FDA agreed that feeding high
doses of a carcinogen to test animals
provides the most practical way to
predict whether a chemical may cause
cancer in humans. Dr. Kennedy noted
that "the NCI study was a good one that
provided a clear demonstration that
chloroform is carcinogenic in
experimental animals." FDA concurred
with EPA's assessment that, since one
cannot conclude with certainty that
cholorform is or is not a human
carcinogen, prudent public health policy
demands that we assume the potential
for carcinogenesis in humans unless
there is strong evidence to the contrary.
Dr. Kennedy submitted as part of his
written comments a paper entitled
"What Animal Research Says About
Cancer." In summary, it noted that
testing with large doses of a chemical is
the usual, and in most instances, the
only way to determine whether it causes
cancer. Epidemiology is fraught with
unreasonable confounding factors from
retrospective designs, and therefore, the
threshold hypothesis has been rejected
on the grounds that no threshold has yet
been demonstrated for a carcinogen.
However, animal testing can be used to
confirm a cause-and-effect relationship
between dosage and the incidence of
cancer—a relationship general enough
to be applied confidently to most
hazardous chemicals used over long
periods. Moreover, the similarities
between cancer in animals and human
beings, such as the fact that cancer cells
are capable of metastasizing—breaking
away from the original cancer and
seeding themselves elsewhere—as well
as the growing evidence that career-
causing chemicals injerfere with the
biochemistry of genetic material, are
powerful arguments for the
appropriateness of using animals as
models for people.
Finally, he found persuasive the
comparison between the substances
known to cause cancer in human beings
and their effect on laboratory animals;
or 18 such substances, all but two were
also found to be carcinogenic in •
animals.
D. Dr. Eula Bingham, Assistant
Secretary of the Department of Labor
and head of the Occupational Safety
and Health Administration (OSHA),
concurred with Dr. Donald Kennedy's
testimony. Dr. Bingham stated that trace
contaminants may increase the risk of
human cancer and produce other
chronic effects. Large numbers of people
are placed at risk to chemicals if they
are present in drinking water.
Dr. Bingham supported limiting
exposure to carcinogens to the lowest
feasible level. She stated that animal
evidence provides the best qualitative
test for assessing potential human
carcinogenic risk and that there is
presently no means for determining a
safe exposure level to a carcinogen. Due
to the long latency period for chemical
carcinogenesis, it would be imprudent to
await the results of human
epidemiological studies.
Thus, OSHA's generic proposal to
regulate carcinogens relies on animal
extrapolation for the detection of
carcinogenic activity of chemicals.
Because of the statistical insensitivity of
laboratory bioassays conducted with
limited numbers of animals, she stated
that positive test results with
experimental animals should generally
supersede negative results and that it is
appropriate to test chemicals at high
exposure levels.
E. Dr. Arthur Upton, Director of the
National Cancer Institute submitted
-------
68670
Federal Register / Vol. 44. No. 231 / Thursday. November 29. 1979 / Rules and Regulations
comments. Those points not previously
included are stated below.
There are currently 32 carcinogens or
suspected carcinogens, 30 mutagens or
suspected mutagens. and 11 promoters
in drinking water identified from a 1978
list of organic compounds.
Two sets of studies have been carried
out to explore the relationship in
humans between THMs in drinking
water and possible increases in cancer.
The first set used presumed measures of
THM contamination (i.e., surface waters
likely to be chlorinated) vs. ground
water (likely to not be chlorinated]. The
second set used actual measures of
THM levels. Nine of ten indirect studies
showed a number of statistically
significant associations between water
quality and cancer.
From the three quantitative studies
one could tentatively conclude that
cancer of the urinary bladder, and
perhaps large intestine are correlated,
with THMs in water. He noted that a
decrease of 100 micrograms per liter of
chloroform in water could lead to a
decrease in cancer rates of up to 7.5% in
men and 10% in women for bladder
cancer and between 7.5 and 8.5% in
large intestinal cancer for women and
men, respectively. Although these
studies did not purport to "prove" a
cause-effect association between THMs
and cancer, Dr. Upton testified that the
weight of evidence showed a "high
index of suspicion" of such a
relationship.
The additive or more than additive
effects from multiple exposure to an
array of organic carcinogens in drinking
water are of such significance as to
warrant an appraisal of the opportunity
for modification of the total carcinogenic
burden which may be traceable or
produced by water processing to reduce
the levels of total exposure.
The fact that source carcinogens from
drinking water may persist in body
tissues makes quantification of these
effects difficult.
In the absence of conclusive and
quantitative empirical evidence, Dr.
Upton supported EPA's reliance on the
NAS principles set forth in "Drinking
Water and Health." He stated that every
dose of a demonstrated carcinogen
should be regarded as carrying some
potential or presumptive risk. Animal
studies must be used to evaluate human
carcinogenic risk and to predict the
safety of environmental chemicals if
human victims are to be spared. He
endorsed EPA's proposed TTHM MCL
of 100 ppb as a "comprehensive public
health measure" in the direction of
cancer prevention. Measures taken to
control large classes of contaminants
were deemed useful for reducing levels
of material whose carcinogenic or
mutagenic potential was still unknown.
F. Dr. Upton was accompanied by Dr.
Marvin Schneiderman and Dr. Umberto
Saffiotti, from NCI. who explained the
difficulties in predicting with any degree
of accuracy, human risk posed by
carcinogens due to low levels of
exposure, variability in such levels,
measurement problems, long latency
periods and other confounding factors.
They also endorsed EPA's approach to
regulating THMs. Those points stated by
Dr. Marvin Schneiderman of the NCI not
covered previously are outlined below.
The experimental conditions to detect
cancer in 1 in 100 or 1% of the time
requires 20,000 animals. Experiments
performed with 100 animals per dose
group can detect approximately a 3%
incidence. Three percent is an
enormously high incidence. After all,
breast cancer, the most common human
cancer has a lifetime probability of 7.5%
and lung cancer is 6%. Therefore, three
percent is in line with the most common
of cancers that cause the greatest
concern.
G. Dr. Riley Housewright, National
Academy of Sciences, provided a review
of the NAS report, "Drinking Water and
Health" and stated the following:
Drinking Water regulations have not
always been based entirely on health
considerations even though protection of
consumer health is the unqualified logical
goal. For various reasons, drinking water
standards have historically been set on the
basis of: 1. contaminant background levels. 2.
analytical detection limits, 3. technological
feasibility of treatment processes. 4. aesthetic
considerations, S. health effects, and
combinations of the above. In our report we
have attempted to summarize the current
knowledge of the health effects of
contaminants in drinking water with the
purpose of providing the scientific
information required for establishing
regulations based on health effects.
The NAS report did provide a
relatively long list of recommendations
for research but these recommendations
were not be in lieu of establishing a
standard for chloroform. He stated, _.
"there appears to be no question but
that, first of all, chloroform is found in
drinking water, and it is a carcinogen."
Dr. Housewright also stated that the
hazards of ingesting chemical pollutants
in drinking water can be assessed in
two general ways: epidemiology studies
and laboratory studies of toxicity. The
insidious effects of chronic exposure to
low doses of toxic agents are difficult to
recognize, because there are few, if any,
early warning signs, and, when signs are
ultimately observed, they often imply
irreversible effects. In evaluating the
potential effects on health of organic
compounds found in drinking water, the
NAS principal concern was to assess
their carcinogenicity. The risk
associated with the ingestion of
compounds that were identified as
carcinogenic were calculated by
extrapolation from animal data.
Chloroform was one of the compounds
that produced cancer in both rats and
mice. The NAS Safe Drinking Water
Committee believed that: "these tests
were valid and there is a hazard to man
associated with the ingestion of
chloroform," and that "chloroform and
other THMs present a health hazard and
that steps should be taken to prevent
their formation or to remove them from
drinking water." He stated that "Our
committee believed these tests were
valid and that there is a hazard to man
associated with the ingestion of
chloroform."
In addition, Dr. Housewright stated
the following:
Some early epidemiological studies
suggested an association between
THMs and cancer. Our review of ten
epidemiological studies concluded that
the association was small and that there
was a large margin of error. In most of
the studies evaluated, the THM
exposure and duration levels were
inferred and confounding factors known
to affect cancer incidence, such as
cigarette-smoking, occupation, use of
alcohol and drugs, socio-economic
status and many others, were
inadequately controlled. The failure of
these studies to clearly establish a
positive or negative cause and effect
relationship between THMs and cancer
resides to some extent in the
complexities inherent in doing such
studies.
We believe that THMs in drinking
water present a human health hazard.
The principal basis for this is that
exposure to them results in cancer in
two species of experimental animals.
This conclusion is neither confirmed nor
denied by the results of epidemiological
studies now available; confirmation
would require more sensitive
epidemiological studies than have been
conducted thus far. The examination of
currently available epidemiological
evidence gives no reason to change the
conclusion of the study Drinking Water
and Health which recommends that
"strict criteria be applied when limits for
chloroform in drinking water are
established to protect the public health."
H. Dr. Richard Bates. National
Institute of Environmental Health
Sciences, and Dr. David Hoel, National
Institute of Environmental Health
Services and National Academy of
Sciences, stated that determination of a
quantitative standard for a contaminant
in drinking water must be based upon a
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68671
judgment of the risk that is socially
acceptable and upon a scientific
estimation of the actual risk posed by
the contaminant. Scientific estimation of
risk from carcinogenic chemicals is not
yet an exact science and until that time,
regulatory agencies will have to act
according to the most likely
interpretation of scientific information
while resolving uncertainties in a way
that assures protection of the public
health.
The four principles from NAS are
consistent with what is now known
about chemical carcinogenesis. The first
principle is now widely accepted.
Because epidemiology studies have
problems of sensitivity and specificity
and harmful effects can only be noted
after the damage is done, experimental
studies must be relied upon to judge the
potential carcinogenicity of a chemical
to humans. This practice is supported by
the observation that most known human
carcinogens are also carcinogenic in
experimental animals, that generally the
same kinds of metabolic enzymes that
activate and detoxify chemical
carcinogens are present in both human
tissues and experimental animals, and
that the general process of cancer
development is similar in humans and
experimental in animals.
With regard to the second and third
principles, which discuss the inability to
establish thresholds for carcinogens and
the validity of using high doses, the
fundamental reason for testing at high
dose levels is to enhance the sensitivity
of the experimental bioassay to detect a
chemical carcinogen. A study of 100
animals can only detect the induction of
cancer in no less than one percent of the
animals. In order to detect lower levels
of risk, it would be necessary to test
much larger numbers of animals or to
use mathematical procedures to
estimate the level of risk from lower
levels of exposure. The former approach
is normally economically infeasible. The
latter approach is based upon debatable
scientific assumptions including that
there is no threshold below which
exposure to a carcinogen entails no risk.
At the present time, it cannot be
determined unequivocally whether or
not thresholds exist or to determine
which individuals in the population may
or may not be able to tolerate additional
exposure to carcinogenic chemicals.
The methods described in "Drinking
Water and Health" are the best
available to provide guidance on low
level risks. In view of the many
uncertainties, the safest action is always
to reduce exposure to a chemical
carcinogen to the lowest feasible level.
With regard to the numerical values
that are produced by the models in
terms of human risk per unit of
exposure, Dr. Hoel stated that because
of the inability to estimate the possible
biological errors, biological differences
between species and within species, and
the experiences with the empirical data,
the use of model predictions in ascribing
some certain number of deaths in a
population is not necessarily
appropriate. He stated that the models
could be used to rank carcinogens
relative to their potency.
V. Calgon Corporation
As an example of comments providing
information and data concerning the
technical basis of the regulations,
comments submitted by Calgon
Corporation are summarized below.
1. GAC has been widely used for over
18 years in potable water applications to
control taste, odor, and color in the U.S.
and presently over 60 plants in the U.S.
use GAC. In these applications, GAC
has worked effectively with minimal
problems without hazard or injury.
2. GAC is used to remove organic
chemical contaminants from potable
water in 21 cities in Europe and have
been operating for up to 10 years. Most
of these plants have on-site reactivation
and have been operating without any
adverse effects or undue difficulties.
3. GAC does not get into the water
system from the filter beds. The bulk of
the carbon lost is lost during the
periodic backwashing of the carbon
beds.
4. GAC does not add heavy metals or
polynuclear aromatics (PAH) to the
finished water. A composite sample of
four activated carbons contained 7.36%
ash of which 0.08% was soluble in water.
Analysis for inorganic compounds
showed very low levels but most
significant is that the soluble portion of
the ash is dissolved and discarded
during the backwashing operation.
During reactivation, the ash compounds
are liberated and driven off in the
furnace or the quench tank which
contains boiling hot water, extracting
any water soluble ash that is present.
Activated carbon is made by a multi-
step process which is not conducive to
the formation or retention of PAHs. The
raw material, coal, is subject to an
oxidation step, followed by a
devolatilization step, followed by a long
term high temperature (up to 2,000* F]
activation step during which time the
carbon granules are constantly turned in
a reducing atmosphere. This process
will drive off any materials with boiling
points characteristic of PAHs.
Experiments by a U.S. FDA laboratory
have not been able to extract any PAHs
from activated carbon. Activated carbon
is such a strong adsorbent that even a
small amount of polynuclear aromatics
that might exist would be strongly
adsorbed by the carbon.
5. GAC adsorbs organics and allow
bacteria naturally present in the water
to grow within the carbon bed.
However, these bacteria are removed
during backwashing and any bacteria in
the effluent are easily controlled by
disinfection following GAC. Bacterial
growth has not been a problem at the
more than 60 plants in the U.S. or in the
systems in Europe.
6. In addition to its effectiveness in
removing taste, odor and color from
potable waters, GAC provides other
advantages to the water treatment plant,
such as savings in the amount of
backwash water that is needed. Twenty
to 40% savings over conventional media
has been experienced by plants using
GAC. Also, the demand for chlorine was
reduced in these plants by 13% to 14%
because organic contaminants had been
reduced. Finally, use of GAC has
extended service life between
backwashes because of a reduction in
head loss.
7. Energy requirements, based upon
actual experiences, with reactivation of
GAC used to treat industrial waste
waters, are approximately 8,000 BTUs
per pound of reactivated carbon. It is
reasonable to expect that reactivation of
GAC used to treat drinking water would
require less energy. While the
reactivation process is relatively energy
intensive, the consumption of additional
energy for reactivation of GAC from
drinking water facilities will be
insignificant in view of national
consumption of energy.
8. Experience with furnaces
reactivating GAC from industrial waste
water facilities has shown that proper
application of air pollution control
technologies can be operated to comply
with applicable air pollution
requirements.
9. Compliance with the MCL is
feasible and use of GAC for this purpose
would most likely be for precursor
removal.
10. The allotted time for compliance
with the MCL is adequate. Systems that
elect to use GAC to reduce THMs could
be modified very quickly. For most
applications, replacement of the existing
filter media with GAC will be adequate.
For greater bed depths, the necessary
contact time can probably be achieved
with relatively simple modifications of
the existing filter systems. A few
systems may require greater bed depth
and thus additional time will probably
be needed for those systems to make the
modifications.
11. The utilities' cost estimates for
GAC are overstated in that the capital
-------
68672 Federal Register / Vol. 44, No. 231 / Thursday. November 29. 1979 / Rules and Regulations
required for reactivation is based upon a
redundant furnace. Based upon actual
experience, it has not been necessary to
have such substantial stand-by
reactivation capacity. A more
reasonable approach would be to utilize
two furnaces of equal size with a total
capacity equal to the peak flow rates
and provide for stocking of buffer
carbon to meet needs during periods of
maintenance. Also, the use of an outside
reactivation service could be used
during long down times of the furnace.
Detailed cost estimates were provided
for GAC for two system sizes.
12. In order for the demand for GAC
to be spread over a reasonable time
frame, it is recommended that the
regulations be phased in three segments
separated by three months each.
VI. National Drinking Water Advisory
Council
It is the opinion of the National
Drinking Water Advisory Council
(NDWAC) that EPA is justified in
establishing an MCL of 100 ppb for
THMs in Finished drinking water on the
basis of health hazard and feasibility.
However, the MCL should not be
restricted to utilities serving greater than
75,000 persons. The Council
recommends that an MCL of 100 ppb
THM also apply to utilities serving
between 10,000 and 75,000 persons
beginning three years after
implementation of the regulation
covering those utilities serving greater
than 75,000 persons.
The Council also recommends that the
implementation of the MCL of 100 ppb
THM for utilities serving less than 10,000
persons be at the option of the agency
having primacy in each state. The
agency having primacy will be more
familiar with the water supplies in that
state and be better able to evaluate the
potential for THM formation as a result
of chlorine disinfection. This would
serve to avoid unnecessary financial
burdens on these utilities. The decision
for compliance by those utilities should
be made within five years.
The Council believes that the THM
requirements should initially apply to all
water sources (surface and ground).
Where no THM problem is determined,
the state should have the responsibility
to determine the need for future
monitoring requirements in order to
assure that THMs do not pose a problem
in the future.
It is imperative that the EPA publicly
clarify its position relative to lowering
the MCL for THM below 100 ppb. If the
Agency believes the current health
effects data supports an MCL lower than
100 ppb a detailed justification should
be provided.
It is recommended that the EPA
reconsider its restriction on the use of
chloramines. Chloramines have been
effectively used for disinfection in
certain water systems for many years.
Consequently, the Council believes that
EPA's proposed regulation is unduly
restrictive.
As previously expressed, the NDWAC
is of the opinion that the standard plate
count, although useful to the utility
operator, should not be established as a
regulatory requirement.
The Council concurs with the
averaging method described in the
proposed regulation for determining the
level of THM in drinking water supplies.
Appendix C—Analysis of
Trihalomethanes
Part I: The Analysis of Trihalomethanes
in Drinking Water by the Purge and Trap
Method
1. Scope
1.1 This method [1] is applicable in
the determination of four
trihalomethanes, i.e. chloroform,
dichlorobromomethane,
dibromochloromethane, and bromoform
in finished drinking water, raw source
water, or drinking water in any stage of
treatment. The concentration of these
four compounds is totaled to determine
total trihalomethanes (TTHM].
1.2 For compounds other than the
above-mentioned trihalomethanes, or
for other sample sources, the analyst
must demonstrate the usefulness of the
method by collecting precision and
accuracy data on actual samples as
described (2).
1.3 Although the actual detection
limits are highly dependent upon the gas
chromatographic column and detector
employed, the method can be used over
a concentration range of approximately
0.5 to 1500 micrograms per liter.
1.4 Well in excess of 100 different
water supplies have been analyzed
using this method. Supplementary
analyses using gas chromatography
mass spectrometry (GC/MS) have
shown that there is no evidence of
interference in the determination of
trihalomethanes (3). For this reason, it is
not necessary to analyze the raw source
water as is required with the Liquid/
Liquid Extraction Method [4].
2. Summary
2.2 Trihalomethanes are extracted
by an inert gas which is bubbled through
the aqueous sample. The
trihalomethanes, along with other
organic constituents which exhibit low
water solubility and a vapor pressure
significantly greater than water, are
efficiently transferred from the aqueous
phase to the gaseous phase. These
compounds are swept from the purging
device and are trapped in a short
column containing a suitable sorbent.
After a predetermined period of time,
the trapped components are thermally
desorbed and back/lushed onto the head
of a gas chromatographic column and
separated under programmed
conditions. Measurement is
accomplished with a halogen specific
detector such as electrolytic
conductivity or microcoulometric
titration.
2.3 Confirmatory analyses are
performed using dissimilar columns, or
by mass spectrometry (5).
2.4 Aqueous standards and
unknowns are extracted and analyzed
under identical conditions in order to
compensate for extraction losses.
2.5 The total analysis time, assuming
the absence of other organohalides, is
approximately 35 minutes per sample.
3. Interferences
3.1 Impurities contained in the purge
gas and organic compounds outgasing
from the plumbing ahead of the trap
usually account for the majority of
contamination problems. The presence
of such inteferences are easily
monitored as a part of the quality
control program. Sample blanks are
normally run between each set of
samples. When a positive
trihalomethane response is noted in the
sample blank, the analyst should
analyze a method blank. Method blanks
are run by charging the purging device
with organic-free water and analyzing in
the normal manner.
If any trihalomethane is noted in the
method blank in excess of 0.4 pg/1, the
analyst should change the purge gas
source and regenerate the molecular
sieve purge gas filter. Subtracting the
blank values is not recommended. The
use of non-TFE plastic tubing, non-TFE
thread sealants, or flow controllers with
rubber components should be avoided
since such materials generally out-gas
organic compounds which will be
concentrated in the trap during the
purge operation. Such out-gasing
problems are common whenever new
equipment is put into service; as time
progresses, minor out-gasing problems
generally cure themselves.
3.2 Several instances of accidental
sample contamination have been noted
and attributed to diffusion of volatile
organics through the septum seal and
into the sample during shipment and
storage. The sample blank is used as a
monitor for this problem.
3.3 For compounds that are not
efficiently purged, such as bromoform,
small variations in sample volume,
purge time, purge flow rate, or purge
temperature can affect the analytical
-------
Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68673
result. Therefore, samples and standards
must be analyzed under identical
conditions.
3.4 Cross-contamination can occur
whenever high-level and low-level
samples are sequentially analyzed. To
reduce this likelihood, the purging
device and sample syringe should be
rinsed twice between samples with
organic-free water. Whenever an
unusually concentrated sample is
encountered, it is highly recommended
that it be followed by a sample blank
analysis to ensure that sample cross
contamination does not occur. For
samples containing large amounts of
water soluble materials, it may be
necessary to wash out the purging
device with a soap solution, rinse with
distilled water, and then dry in a 105*C
oven between analyses.
3.5 Qualitative misidentifications are
a problem in using gas chromatographic
analysis. Whenever samples whose
qualitative nature is unknown are
analyzed, the following precautionary
measures should be incorporated into
the analysis.
3.5.1 Perform duplicate analyses
using the two recommended columns
(4.2.1 and 4.2.2) which provide different
retention order and retention times for
the trihalomethanes and other
organohalides.
3.5.2 Whenever possible, use GC/MS
techniques which provide unequivocal
qualitative identifications (5).
4. Apparatus
4.1 The purge and trap equipment
consists of three separate pieces of
apparatus: the purging device, trap, and
desorber. Construction details for a
purging device and an easily automated
trap-desorber hybrid which has proven
to be exceptionally efficient and
reproducible are shown in Figures 1
through 4 and described in 4.1.1. through
4.1.3. An earlier acceptable version of
the above-mentioned equipment is
described in (1).
4.1.1 Purging Device—Construction
details are given in Figure 1 for an all-
glass 5 ml purging device. The glass frit
installed at the base of the sample
chamber allows finely divided gas
bubbles to pass through the sample
while the sample is restrained above the
frit. Gaseous volumes above the sample
are kept to a minimum to eliminate dead
volume effects, yet allowing sufficient
space for most foams to disperse. The
inlet and exit ports are constructed from
heavy-walled Vi-inch glass tubing so
that leak-free removable connections
can be made using "finger-tight"
compression fittings containing Teflon
ferrules. The removable foam trap is
used to control samples that foam.
4.1.2 Trapping Device—The trap
(Figure 2) is a short gas chromatographic
column which at <35° C retards the
flow of the compounds of interest while
venting the purge gas and, depending on
which sorbent is used, much of the
water vapor. The trap should be
constructed with a low thermal mass so
that it can be heated to 180" C in less
than 1 minute for efficient desorption,
then rapidly cooled to room temperature
for recycling. Variations in the trap ID,
wall thickness, sorbents, sorbent
packing order, and sorbent mass could
adversely affect the trapping and
desorption efficiencies for compounds
discussed in this text. For this reason, it
is important to faithfully reproduce the
trap configurations recommended in
Figure 2. Traps containing Tenax only,
or combinations of Tenax and other
sorbents are acceptable for this
analysis.
4.1.3 Desorber assembly—Details for
the desorber are shown in Figures 3, and
4. With the 6-port valve in the Purge
Sorb position (Figure 3), the effluent
from the purging device passes through
the trap where the flow rate of the
organics is retarded. The GC carrier gas
also passes through the 6-port valve and
is returned to the GC. With the 6-port
valve in the Purge-Sorb position, the
operation of the GC is in no way
impaired; therefore, routine liquid
injection analyses can be performed
using the gas chromatograph. After the
sample has been purged, the &-port
valve is turned to the desorb position
(Figure 4). In this configuration the trap
is coupled in series with the gas
chromatographic column allowing the
carrier gas to backflush the trapped
materials into the analytical column.
Just as the valve is actuated, the power
is turned on to the resistance wire
wrapped around the trap. The power is
supplied by an electronic temperature
controller. Using this device, the trap is
rapidly heated to 180° C and then
maintained at 180* C with minimal
temperature overshoot. The trapped
compounds are released as a "plug" to
the gas chromatograph. Normally,
packed columns with theoretical
efficiencies ne'ar 500 plates/foot under
programmed temperature conditions can
accept such desorb injections without
altering peak geometry. Substituting a
non-controlled power supply, such as a
manually-operated variable transformer,
will provide nonreproductible retention
times and poor quantitative data unless
Injection Procedure (8.9.2) is used.
4.1.4 Several Purge and Trap Devices
are now commercially available. It is
recommended that the following be
taken into consideration if a unit is to be
purchased:
a. Be sure that the unit is completely
compatible with the gas chromatograph
to be used for the analysis.
b. Use a 5-ml purging device similar to
that shown in Figure 1.
c. Be sure the Tenax portion of the
trap meets or exceeds the dimensions
shown in Figure 2.
d. With the exception of sample
introduction, select a unit that has as
many of the purge trap functions
automated as possible.
4.2 Gas chromatograph—The
chromatograph must be temperature
programmable and equipped with a
halide specific detector.
4.2.1 Column I is an unusually
efficient column which provides
outstanding separations for a wide
variety of organic compounds. Because
of its ability to resolve trihalomethanes
from other organochlorine compounds,
column I should be used as the primary
analytical column (see Table 1 for
retention data using this column].
4.2.1.1 Column I parameters:
Dimensions—8 feet long x 0.1 inch ID
stainless steel or glass tubing. Packing—
1% SP-1000 on Carbopack-B (60/80)
mesh. Carrier Gas—helium at 40 ml/
minute. Temperature program sequence:
45° C isothermal for 3 minutes, program
at 8° C/minute to 220° C then hold for 15
minutes or until all compounds have
eluted.
Note.—It has been found that during
handling, packing, and programming, active
sites are exposed on the Carbopack-B
packing. This results in tailing peak geometry
and poor resolution of many constituents. To
correct this, pack the first 5 cm of the column
with 3% SP-1000 on Chromosorb-W 60/80
followed by the Carbopack-B packing.
Condition the precolumn and the Carbopack
columns with carrier gas flow at 220* C
overnight. Pneumatic shocks and rough
treatment of packed columns will cause
excessive fracturing of the Carbopack. If
pressure in excess of 60 psi is required to
obtain 40 ml/minute carrier flow, then the
column should be repacked.
4.2.1.2 Acceptable column equivalent
to Column I: Dimensions—8 feet
long x 0.1 inch ID stainless steel or glass
tubing. Packing—0.2% Carbowax 1500
on Carbopack-C (80/100) mesh. Carrier
Gas—helium at 40 ml/minute.
Temperature program sequence—60* C
isothermal for 3 minutes, program at 8*
C /minute to 160° C, then hold for 2
minutes or until all compounds have
eluted.
Note.—It has been found that during
handling, packing, and programming, active
sites are exposed on the Carbopack-C
packing. This results in poor resolution of
constituents and poor peak geometry. To
correct this, place a 1 ft. 0.125 in. OO x 0.1 in.
-------
68674 Federal Register / Vol. 44, No. 231 / Thursday. November 29, 1979 / Rules and Regulations
ID stainless steel column packed with 3%
Carbowax 1500 on Chromosorb-W 60/80
mesh in series before the Carbopack-C
column. Condition the precolumn and the
Carbopack columns with carrier gas flow at
190* C overnight. The two columns may be
retained in series for routine analyses.
Trihalomethane retention times arc listed in
Table 1.
4.2.2 Column II provides unique
organohalide-trihalomethane
separations when compared to those
obtained from Column I (see Figures 5
and 6). However, since the resolution
between various compounds is generally
not as good as those with Column I, it is
recommended that Column Q be used as
a qualitative confirmatory column for
unknown samples when GC/MS
confirmation is not possible.
4.2.2.1 Column II parameters:
Dimensions—6 feet long x 0.1 inch ID
stainless steel or glass. Packing—n-
octane on Porisil-C (100/120 mesh).
Carrier Gas—helium at 40 cc/minute.
Temperature program sequence—50° C
isothermal for 3 minutes, program at 6°/
minute to 170* C. then hold for 4 minutes
or until all compounds have eluted.
Trihalomethane retention times are
listed in Table 1.
5.8 Organic-free water is defined as
water free of interference when
employed in the purge and trap analysis.
5.8.1 Organic-free water is generated
by passing tap water through a carbon
filter bed containing about 1 Ib. of
activated carbon. Change the activated
carbon bed whenever the concentration
of any trihalomethane exceeds 0.4 u.g/1.
5.8.2 A Milliporf Super-Q Water
System or its equivalent may be used to
generate organic-free water.
5.8.3 Organic-free water may also be
prepared by boiling water for 15
minutes. Subsequently, while
maintaining the temperature at 90° C,
bubble a contaminant-free inert gas
through the water for one hour. While
still hot, transfer the water to a narrow-
mouth screw-cap bottle with a Teflon
seal.
5.8.4 Test organic free water each
day it is used by analyzing according to
Section 8.
5.9 Standards.1
5.9.1 Bromoform—98%—available
from Aldrich Chemical Company.
5.9.2 Bromodichloromethane 97%—
available from Aldrich Chemical
Company.
5.9.3 Chlorodibromomethane—
available from Columbia Chemical Inc.,
Columbia. S.C.
5.9.4 Chloroform—99%—available
from Aldrich Chemical Company.
1 As a precautionary measure, all standards must
be checked for punly by boiling point
determinations or CC/MS assays (S).
5.10 Standard Stock Solutions
5.10.1 Place about 9.8 ml of methyl
alcohol into a ground glass stoppered 10
ml volumetric flask.
5.10.2 Allow the flask to stand
unstoppered about 10 minutes or until
all alcohol wetted surfaces have dried.
5.10.3 Weigh the flask to the nearest
0.1 mg.
5.10.4 Using a 100 jil syringe,
immediately add 2 drops of the
reference standard to the flask, then
reweigh. Be sure that the 2 drops fall
directly into the alcohol without
contacting the neck of the flask.
5.10.5 Dilute to volume, stopper, their
mix by inverting the flask several times.
5.10.6 Transfer the solution to a
dated and labeled 15 ml screw cap
bottle with a Teflon cap liner.
Note.—Because of the toxicity of
trihalomethanes, it is necessary to prepare
primary dilutions in a hood. It is further
recommended that a NIOSH/MESA
approved toxic gas respirator be used when
the analyst handles high concentrations of
such materials.
5.10.7 Calculate the concentration in
micrograms per microliter from the net
gain in weight.
5.10.8 Store the solution at 4" C.
Note.—All standard solutions prepared in
methyl alcohol are stable up to 4 weeks when
stored under these conditions. They should
be discarded after that time has elapsed.
5.11 Aqueous Calibration Standard
Precautions.
5.11.1 In order to prepare accurate
aqueous standard solutions, the
following precautions must be observed.
a. Do not inject more than 20 u.1 of
alcoholic standards into 100 ml of
organic-free water.
b. Use of 25 fil Hamilton 702N
microsyringe or equivalent. (Variations
in needle geometry will adversely affect
the ability to deliver reproducible
volumes of methanolic standards into
water.)
c. Rapidly inject the alcoholic
standard into the expanded area of the
filled volumetric flask. Remove the
needle as fast as possible after injection
d. Mix aqueous standards by inverting
the flask three times only.
e. Discard the contents contained in
the neck of the flask. Fill the sample
syringe from the standard solution
contained in the expanded area of the
flask as directed in Section 8.5.
f. Never use pipets to dilute or transfer
samples or aqueous standards.
g. Aqueous standards when stored
with a headspace are not stable and
should be discarded after one hour.
h. Aqueous standards can be stored
according to Sections 6.4 and 8.6.
5.11.2 Prepare, from the standard
stock solutions, secondary dilution
mixtures in methyl alcohol so that a 20
uA injection into 100 ml or organic-free
water will generate a calibration
standard which produces a response
close (±10%) to that of the sample (See
9.1).
5.11.3 Purge and analyze the
aqueous calibration standards in the
same manner as the samples.
5.11.4 Other calibration procedures
(3] which require the delivery of less
than 20 \i\ of a methanolic standard into
a 5.0 ml volume of water already
contained in the sample syringe ore
acceptable only if the methanolic
standard is delivered by the solvent
flush technique (6).
5.12 Quality Check Standard (2.0 ug/
D
5.12.1 From the standard stock
solutions, prepare a secondary dilution
in methyl alcohol containing 10 ng//il of
each trihalomethane (See Section 5.10.8
Note).
5.12.2 Daily, inject 20.0 fil of this
mixture into 100.0 ml of organic-free
water ana analyze according to Section
8.
6. Sample Collection and Handling
6.1. The sample containers should
have a total volume of at least 25 ml.
6.1.1 Narrow mouth screw cap
bottles with the TFE fluorocarbon face
silicone sepata cap liners are strongly
recommended.
6.2 Sample Bottle Preparation
6.2.1 Wash all sample bottles and
TFE seals in detergent. Rinse with tap
water and finally with distilled water.
6.2.2 Allow the bottles and seals to
air dry at room temperature, then place
in a 105* C oven for one hour, then allow
to cool in a area known to be free of
organics.
Note.—Do not heat the TFE seals for
extended period of time (>1 hour) because
the silicone layer slowly degrades at 105* C.
6.2.3 When cool, seal the bottles
using the TFE seals that will be used for
sealing the samples.
6.3 Sample Stabilization—A
chemical reducing agent (Section 5.6) is
added to the sample In order to arrest
the formation of trihalo-methanes after
sample collection (3, 7). Do not add the
reducing agent to samples when data on
maximum trihalomethane formation is
desired. If chemical stabilization is
employed, the reagent is'also added to
the blanks. The chemical agent (2.5 to 3
mg/40 ml) is added to the empty sample
bottles just prior to shipping to the
sampling site.
6.4 Sample Collection
6.4.1 Collect all samples in duplicate.
8.4.2 Fill the sample bottles in such a
manner that no air bubbles pass through
the sample as the bottle is filled.
-------
Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68675
6.4.3 Seal the bottles so that no air
bubbles are entrapped in it.
6.4.4 Maintain the hermetic seal on
the sample bottle until analysis.
6.4.5 Sampling from a water tap.
6.4.5.1 Turn on water and allow the
system to flush until the temperature of
the water has stabilized. Adjust the flow
to about 500 ml/minute and collect
duplicate samples from the flowing
stream.
6.4.6 Sampling from an open body of
water.
6.4.6.1 Fill a 1-quart wide-mouth
bottle with sample from a representative
area. Carefully fill duplicate sample
bottles from the 1-quart bottle as noted
in 6.4.2.
6.4.7 If a chemical reducing agent
has been added to the sample bottles,
fill with sample just to overflowing, seal
the bottle, and shake vigorously for 1
minute.
6.4.8 Sealing practice for septum seal
screw cap bottles.
6.4.8.1 Open the bottle and fill to
overflowing, place on a level surface,
position the TFE side of the septum seal
upon the convex sample meniscus and
seal the bottle by screwing the cap on
tightly.
6.4.8.2 Invert the sample and lightly
tap the cap on a solid surface. The
absence of entrapped air indicates a
successful seal. If bubbles are present,
open the bottle, add a few additional
drops of sample and reseal the bottle as
above.
6.4.9 Blanks.
6.4.9.1 Prepare blanks in duplicate at
the laboratory by filling and sealing
sample bottles with organic-free water
just prior to shipping the sample bottles
to the sampling site.
6.4.9.2 If the sample is to be
stabilized, add an identical amount of
stabilization reagent to the blanks.
6.4.9.3 Ship the blanks to and from
the sampling site along with the sample
bottles.
6.4.9.4 Store the blanks and the
samples collected at a given site (sample
set) together. A sample set is defined as
all the samples collected at a given site
(i.e., at a water treatment plant, the
duplicate raw source waters, the
duplicate finished waters and the
duplicate blank samples comprise the
sample set).
6.5 When samples have been
collected according to Section 6, no
measurable loss of trihalomethanes has
been detected over extended periods of
storage time (3). It is recommended that
all samples be analyzed within 14 days
of collection.
7. Conditioning Traps
7.1 Condition newly packed traps
overnight at 180* C with an inert gas
flow of at least 20 ml/min.
7.1.1 Vent the trap effluent to the
room, not to the analytical column.
7.2 Prior to daily use, condition traps
10 minutes while backflushing at 180° C.
It may be beneficial to routinely
condition traps overnight while
backflushing at 180* C.
7.2.1 The trap may be vented to the
analytical column; however, after
conditioning, the column must be
programmed prior to use.
8. Extraction and Analysis
8.1 Adjust the purge gas [nitrogen or
helium) flow rate to 40 ml/min.
8.2 Attach the trap inlet to the
purging device. Turn the valve to the
purge-sorb position (Figure 3).
8.3 Open the syringe valve located
on the purging device sample
introduction needle.
8.4 Remove the plungers from two 5
ml syringes and attach a closed syringe
valve to each.
8.5 Open the sample bottle and
carefully pour the sample into one of the
syringe barrels until it overflows.
Replace the syringe plunger and
compress the sample. Open the syringe
valve and vent any residual air while
adjusting the sample volume to 5.0 ml.
Close the valve.
8.6 Fill the second syringe in an
identical manner from the same sample
bottle. This second syringe is reserved
for a duplicate analysis, if necessary
(See Sections 9.3 and 9.4).
8.7 Attach the syringe-valve
assembly to the syringe valve on the
purging device.
8.8 Open the syringe valve and inject
the sample into the purging chamber.
Close both valves. Purge the sample for
11.0±.05 minutes.
8.9 After the 11-minute purge time,
attach the trap to the chromatograph
(turn the valve to the desorb position)
and introduce the trapped materials to
the GC column by rapidly heating the
trap to 180°C while backflushing the trap
with an inert gas between 20 and 60 ml/
min for 4 minutes.
8.9.1 If the trap can be rapidly
heated to 180*C and maintained at this
temperature, the GC analysis can begin
as the sample is desorbed, i.e., the
column is at the initial 45*C operating
temperature. The equipment described
in Figure 4 will perform accordingly.
8.9.2 With other types of equipment
(see Section 4.1.4 and Reference 1)
where the trap is not rapidly heated or is
not heated in a reproducible manner, it
may be necessary to transfer the
contents of the trap into the analytical
column at <30*C where it is once again
trapped. Once the transfer is complete (4
minutes], the column is rapidly heate'd to
the initial operating temperature for
analysis.
8.9.3 If injection procedure 8.9.1 is
used and the early eluting peaks in the
resulting chromatogram have poor
geometry or variable retention times,
then Section 8.9.2 should be used.
8.10 After the extracted sample is
introduced into the gas chromatograph,
empty the gas purging device using the
sample introduction syringe, followed
by two 5-ml flushes of organic-free
water. When the purging device is
emptied, leave the syringe valve open
allowing the purge gas to vent through
the sample introduction needle.
8.11 Analyze each sample and
sample blank from the sample set in an
identical manner (see Section 6.4.9.4) on
the same day.
8.12 Prepare calibration standards
from the standard stock solutions
(Section 5.10) in organic-free water that
are close to the unknown in
trihalomethane composition and
concentration (Section 9.1). The
concentrations should be such that only
20 jil or less of the secondary dilution
need be added to 100 ml of organic-free
water to produce a standard at the same
level as the unknown.
8.13 As an alternative to Section
8.12, prepare a calibration curve for
each trihalomethane containing at least
3 points, two of which must bracket the
unknown.
9. Analytical Quality Control
9.1 Analyze the 2 p.g/1 check sample
daily before any samples are analyzed.
Instrument status checks and lower limit
of detection estimations based upon
response factor calculations at five
times the noise level are obtained from
these data. In addition, response factor
data obtained from the 2 fig/1 check
standard can be used to estimate the
concentration of the unknowns. From
this information, the appropriate
standard dilutions can be determined.
9.2 Analyze the sample blank to
monitor for potential interferences as
described in Sections 3.1, 3.2, and 3.4.
9.3 Spiked Samples
9.3.1 For laboratories analyzing more
than 10 samples a day, each 10th sample
should be a laboratory generated spike
which closely duplicates the average
finished drinking water in
trihalomethane composition and
concentration. Prepare the spiked
sample in organic-free water as
described in Section 5.11.
9.3.2 For laboratories analyzing less
than 10 samples daily, each time the
analysis is performed, analyze at least 1
laboratory generated spike sample
which closely duplicates the average
finished drinking water in
-------
68676 Federal Register / Vol. 44, No. 231 / Thursday, November 29. 1979 / Rules and Regulations
tnhalomethane composition and
concentration. Prepare the spiked
sample in organic-free water as
described in Section 5.11.
9.4 Randomly select and analyze
10% of all samples in duplicate.
9.4.1 Analyze all samples in
duplicate which appear to deviate more
than 30% from any established norm.
9.5 Maintain an up-to-date log on the
accuracy and precision data collected in
Sections 9.3 and 9.4. If results are
significantly different than those cited in
Section 11.1, the analyst should check
out the entire analyses scheme to
determine why the laboratory's
precision and accuracy limits are
greater.
9.6 Quarterly, spike an EMSL-
Cincinnati tnhalomethane quality
control sample into organic-free water
and analyze.
9 6.1 The results of the EMSL
tnhalomethane quality control sample
should agree within 20% of the true
value for each tnhalomethane. If they do
not then the analyst must check each
step in the standard generation
procedure to solve the problem (Section
5 9, 510. and 5.11).
9.7 Maintain a record of the
retention times for each tnhalomethane
using data gathered from spiked
samples and standards.
9.7.1 Daily calculate the average
retention time for each tnhalomethane
and the variance encountered for the
analyses.
9.7.2 If individual trihalomethane
retention time varies by more than 10%
over an eight hour period or does not fall
with 10% of an established norm, the
system is "out of control." The source of
retention data variation must be
corrected before acceptable data can be
generated.
10. Calculations
10.1 Locate each trihalomethane in
the sample chromatogram by comparing
the retention time of the suspect peak to
the data gathered in 9.7.1. The retention
time of the suspect peak must fall within
the limits established in 9.7.1 for single
column identification
10 2 Calculate the concentration of
the samples by comparing the peak
height or peak areas of the samples to
the standard peak height (8.12). Round
off the data to the nearest u.g/1 or two
significant figures.
10.3 Report the results obtained from
the lower limit of detection estimates
along with the data for the samples.
10.4 Calculate the total
trihalomethane concentration (TTHM)
by summing the 4 individual
trihalomethane concentrations in /ig/1.
TTHM (jig/l) = (Conc. CHCl,) + (Conc.
CHBrCl,) + (Conc. CHBr,Cl)+(Conc.
CHBr).
10.5 Calculate the limit of detection
(LOD) for each trihalomethane not
detected using the following criteria:
LOO (ug/l) -
/ A^ATT \
\ BXATT /
' peak height sample
peak height standard
(cone sld pg/l)
where B=pcak height (mm) of 2 /ig/l quality
check standard
A=5 times the noise level in (mm) al the
exact retention time of the
trihalomethane or Ihe baseline
displacement in (mm) from Ihe
theoretical zero at the exact retention
time of the trihalomethane.
ATT=Attenuation factor
11. Accuracy and Precision
11.1 One liter of organic-free water
was spiked with the tnhalomethanes
and used to fill septum seal vials which
were stored under ambient conditions.
The spiked samples were randomly
analyzed over a 2-week period of time.
The single laboratory data listed in
Table II reflect the errors due to the
analytical procedure and storage.
References
1. Bcllar. T A . ] ] Lichlcnberg.
Determining Volatile Organics at the
Mtcrogram per Litre Levels by Gas
Chromatography. journal AWWA. 6B. 739
(December 1974).
2. "Handbook for Analytical Quality
Control in Water and Wastcvvalcr
Laboratories," Analytical Quality Control
Laboratory, National Environment^
Research Center. Cincinnati, Ohio, June 1972.
3 Brass, H.)., et al., "National Organic
Monitoring Survey Sampling and Purgenblc
Organic Compounds, Drinking Water Quality
Through Source Protection," R B Pojasek.
Editor, Ann Arbor Science, p. 398.1977
4 "The Analysis of Trihalomcthnncs in
Finished Water by the Liquid/Liquid
Extraction Method, Method 501 2"
Environmental Monitoring and Support
Labortory, Environmental Research Center,
Cincinnati, Ohio, 45268, May 15,1979.
5 Budde. W L. and J W. Eichelberger,
"Organics Analysis Using Cas
Chromntogrnphy-Mass Spectrometry." Ann
Arbor Science. Ann Arbor. Michigan, 1979
6. White, L. D. et al.. "Convenient
Optimized Method for the Analysis of
Selected Solvent Vapors in the Industrial
Atmosphere," AIHA Journal, Vol 31, p 225.
1970
7 Kopfler, F. C.. et al. "GC/MS
Determination of Volatiles for the National
Organics Reconnaissance Survey (NORS) or
Drinking Water, Identification and Analysis
of Organic Pollutants in Water," L. H Keith.
Editor. Ann Arbor Science, p. 87,1976
Table I—Retention Data for Tnhalomethanes
Retention time minutes
Tnhalomelhanc
Acceptable
Alternative)
Column I lo
I'. splOOO column!
CarbopackB 04".
Carbowax
Carbopack
Column II
n-octano
Porasil C
Chloroform . . 107 82 122
Bromodchloromelhano 137 108 147
ChlofOdibrornomothflnc
(DitxomochForomothanol 165 132 166
Bromoform 192 157 102
Tablell—Single Laboratory Accuracy and Precision
tor Tnhalomeinanes
Precision Accuracy
Spiko Number Mean standard percent
jig/1 samples ng/l deviation recovery
12
120
1190
16 ..._
160
1600
20
200
1960
23
230
2310
Chloroform
12 12
8 11
II IDS
Brcxnodichloronxiihsno
12 15
8 15
11 145
Chlorodibromomolhano
12 ID
8 19
11 185
Bronioforni
12 23
8 23
11 223
014
018
79
005
039
102
009
070
106
016
138
163
too
92
88
94
94
91
95
95
04
100
100
97
BILLING CODE 6560-01-M
-------
Federal Register / Vol 44. No. 231 / Thursday, November 29.1979 / Rules and Regulations 66677
OPTIONAL
FOAM TRAP
\A IN. O.D. EXIT
EXIT 1/4
IN. .O.D.
|— MMM. O.D.
INLET JM
IN. O.D.
TOMM. GLASS FRIT
MEDIUM POROSITY
SAMPLE INLET
2.-WAY SYRINGE VALVE
T7CM. 20 GAUGE SYRINGE NHDLE
6MM.O.D. RUBBER SEPTUM
10MM. O.D.
INLET
1/4 IN. O.D.
1/16 IN. O.D.
STAINLESS STEEL
]3X MOLECULAR
SIEVE PURGE
GAS FILTER
PURGE GAS
FLOW CONTROL
FIGURE 1. PURGING DEVICE
-------
PACKING PROCEDURE
CONSTRUCTION
MUllirURrOif IRAP
GENfRAl PURrOSHRAP
GlAli WOOl 5MM
CRAOI It
SIUCA Gil
• CM
UNA*
3% OV.I ICM
OlASS WOOl
SMM
CIA5J WOOl I«\M
ItNAX 73CM
37. OV.I ICM
OlASS WOOl
SMM
IRAr INK!
m
AND rfRRUUS
ritnun nui
WIRI WBAPPfD SOIIO
IMfRMOCOUrif /COrilROUfH
stmoR
flfCIROrilC
ItMPfRAIUKf
COMtROl
AMD
PfROMtttR
IUBINO 75CM 0103 IM ID
OI7J IN OO SIAItlltSS Slffl
IRAP INlIf
FIGURE 2 TRAP
-------
CARRIER GAS FLOW CONTROL
PRESSURE REGULATOR
PURGE GAS x
FLOW CONTROL X
MOLECULAR
SIEVE FILTER
LIQUID INJECTION PORTS
MKnjnju4
COLUMN OVEN
CONFIRMATORY COLUMN
TO DETECTOR
"^ANALYTICAL COLUMN
OPTIONAL 4-PORT COLUMN
SELECTION VALVE
TRAP INLET (TENAX END)
6-PORT VALVE / RESISTANCE WIRE
HEATER CONTROL
TRAP
100°C
NOTE: ALL LINES BETWEEN
TRAP AND GC
SHOULD DE HEATED
TO 80°C
PURGING DEVICE
•n
(D
n.
n
n
oo_
35'
5T
<
o
p
to
u
c
CO
C.
ffi
Z
o
ro
3
cr
Q
FIGURE 4 PURGE-TRAP SYSTEM (DE5ORB MODE)
c_
re
CA
o;
0.
o
tn
-------
CARRIER GAS FLOW CONTROL
PRESSURE REGULATOR \ 1 fckf
LIQUID INJECTION PORTS
PURGE GAS
FLOW CONTROL
13X MOLECULAR
SIEVE FILTER
OPTIONAL 4-PORT COLUMN
SELECTION VALVE
TRAP INLET (TENTAX END)
COLUMN OVEN
CONFIRMATORY COLUMN
TO DETECTOR
ANALYTICAL COLUMN
z
n
70
n
oo
HEATER CONTROL
NOTE: ALL LINES BETWEEN
TRAP AND GC
SHOULD BE HEATED
PURGING DEVICE TO 8O°C
FIGURE 3 PURGE-TRAP SYSTEM (PURGE-SORB MODE)
p
is
o.
ffi
I
CD
3
cr
CD
T
to
CO
50
5"
en
m
S
a.
70
(0
00
c.
o
09
-------
'COIUMN: 017. CARBOWAX ISOO ON CARBOPACK-C
PROGRAM, ftO'C-3 MINUIIf I'/MINUlf IO I6O'C
DlflCfOH: CUCIROlVtlC CONDUCIIVIIT je
at
o
o
oc
O
_J
• 10 17 14
RETENTION TIME MINUTES
S CHROMATOGRAM OF ORGANOHALIDES
JO
I
a
90
to-
oa.
en
5?
"
2
p
I
o.
03
>jr_
2
o
g-
I-
fD
CD
•»]
CO
50
ra"
CO
EJ
Q..
50
*
§•
w
-------
COIUMM- ..OCIANI ONFOtASIl C
riOORAM- 10-C-l MINUIIf 6-/MINUII IO I7O~C
01 ifcio»• tticrtoiriic coMoucnvirr
a.
5
0)
JO
IB
aa
o
Z
p
to
CO
3
a.
D3
2
o
re
I
ro
a.
BILLING CODE 6560-OI-C
c
C
o
3
ce
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations
68683
Part II: Analysis of Trihalomethanes in
Drinking Water by Liquid/Liquid
Extraction
1. Scope.
1.1 This method (1.2) is applicable
only to the determination of four
trihalomethanes, i.e., chloroform.
bromodichloromethane,
chlorodibromomethane, and bromoform
in finished drinking water, drinking
water during intermediate stages of
treatment, and the raw source water.
1.2 For compounds other than the
above-mentioned trihalomethanes. or
for other sample sources, the analyst
must demonstrate the usefulness of the
method by collecting precision and
accuracy data on actual samples as
described in (3) and provide qualitative
confirmation of results by Gas
Chroma tography/Mass Spectrometry
(CC/MS) (4).
1.3 Qualitative analyses using CC/
MS or the purge and trap method (5)
must be performed to characterize each
raw source water if peaks appear as
interferences in the raw source analysis.
1.4 The method has been shown to
be useful for the trihalomethanes over a
concentration range from approximately
0.5 to 200 jig/1. Actual detection limits
are highly dependent upon the
characteristics of the gas
chromatographic system used.
2. Summary
2.1 Ten milliliters of sample are
extracted one time with 2 ml of solvent.
Three fil of the extract are then injected
into a gas chromatograph equipped with
a linearized electron capture detector
for separation and analysis.
2.2 The extraction and analysis time
is 10 to 50 minutes per sample
depending upon the analytical
conditions chosen. (See Table 1 and
Figures 1, 2, and 3.]
2.3 Confirmatory evidence is
obtained using dissimilar columns and
temperature programming. When
component concentrations are
sufficiently high (>50 /jg/1), halogen
specific detectors may be employed for
improved specificity.
2.4 Unequivocal confirmatory
analyses at high levels (>50 pg/1) can
be performed using GC/MS in place of
the electron capture detector. At levels
below 50 /ig/1, unequivocal confirmation
can only be performed by the purge and
trap technique using GC/MS (4. 5).
2.5 Standards dosed into organic
free water and the samples are
extracted and analyzed in an identical
manner in order to compensate for
possible extraction losses.
2.6 The concentration of each
trihalomethane is summed and reported
as total trihalomethanes in fig/1.
3. Interferences
3.1 Impurities contained in the
extracting solvent usually account for
the majority of the analytical problems.
Solvent blanks should be analyzed
before a new bottle of solvent is used to
extract samples. Indirect daily checks
on the extracting solvent are obtained
by monitoring the sample blanks (6.4.10).
Whenever an interference is noted in
the sample blank, the analyst should
reanalyze the extracting solvent. The
extraction solvent should be discarded
whenever a high level (>10 fig/1) of
interfering compounds are traced to it.
Low level interferences generally can be
removed by distillation or column
chromatography (6); however, it is
generally more economical to obtain a
new source of solvent or select one of
the approved alternative solvents listed
in Section 5.1. Interference free solvent
is defined as a solvent containing less
than 0.4 fig/1 individual trihalomethane
interference. Protect interference-free
solvents by storing in a non-laboratory
area known to be free of organochlorine
solvents. Subtracting blank values is not
recommended. <
3.2 Several instances of accidental
sample contamination have been
attributed to diffusion of volatile
organics through the septum seal on the
sample bottle during shipment and
storage. The sample blank (6.4.10) is
used to monitor for this problem.
3.3 This liquid/liquid extraction
technique efficiently extracts a wide
boiling range of non-polar organic
compounds and. in addition, extracts the
polar organic components of the sample
with varying efficiencies. In order to
perform the trihalomethane analysis as
rapidly as possible with sensitivities in
the low jig/1 range, it is necessary to use
the semi-specific electron capture
detector and chromatographic columns
which have relatively poor resolving
power. Because of these concessions.
the probability of experiencing
chromatographic interferences is high.
Tnhalomethanes are primarily products
of the chlorination process and
generally do not appear in the raw
source water. The absence of peaks in
the raw source water analysis with
retention times similar to the
trihalomethanes is generally adequate
evidence of an interference-free finished
drinking water analysis. Because of
these possible interferences, in addition
to each finished drinking water analysis,
a representative raw source water (6.4.5)
must be analyzed. When potential
interferences are noted in the raw
source water analysis, the alternate
chromatographic columns must be used
to reanalyze the sample set. If
interferences are still noted, qualitative
identifications should be performed
according to Sections 2.3 and 2.4. If t!
peaks are confirmed to be other than _
trihalomethanes and add significantly to
the total trihalomethane value in the
finished drinking water analysis, then
the sample set must be analyzed by the
purge and trap method (5).
4. Apparatus
4.1 Extraction vessel—A15 ml total
volume glass vessel with a Teflon lined
screw-cap is required to efficiently
extract the samples.
4.1.1 For samples that do not form
emulsions 10 ml screw-cap flasks with a
Teflon faced septum (total volume is ml)
are recommended. Flasks and caps—
Pierce—#13310 or equivalent. Septa—
Teflon silicone—Pierce #12718 or
equivalent.
4.1.2 For samples that form
emulsions (turbid source water) 15 ml
screw cap centrifuge tubes with a Teflon
cap liner are recommended. Centrifuge
tube—Corning 8062-15 or equivalent.
4.2 Sampling containers—40 ml
screw cap sealed with Teflon faced
silicone septa. Vials and caps—Pierce
#13075 or equivalent. Septa—Pierce
#12722 or equivalent.
4.3 Micro syringes—10,100 fil.
4.4 Micro syringe—25 fil with a 2-
inch by 0.006-inch needle—Hamilton
702N or equivalent.
4.5 Syringes—10 ml glass
hypodermic with luerlok tip (2 each).
4.6 Syringe valve—2-way with luer
ends (2 each)—Hamilton #86570—1FM1
or equivalent.
4.7 Pipette—2.0 ml transfer.
4.8 Glass stoppered volumetric
flasks—10 and 100 ml.
4.9 Gas chromatograph with
linearized electron capture detector.
(Recommended option—temperature
programmable. See Section 4.12.)
4.10 Column A—4 mm ID x 2m long
glass packed with 3% SP-1000 on
Supelcoport (100/120 mesh) operated at
50°C with 60 ml/min flow. (See Figure 1
for a sample chromatogram and Table 1
for retention data.)
4.11 Column B—2 mm ID x 2m long
glass packed with 10% squalane on
Chromosorb WAW (80/100 mesh)
operated at 67*C with 25 ml/min flow.
This column is recommended as the
primary analytical column.
Trichloroethylene, a common raw
source water contaminate, coelutes with
bromodichloromethane. (See Figure 2 for
a sample chromatogram and Table 1 for
retention data.)
4.12 Column C—2 mm ID x 3m lor
glass packed with 6% OV-11/49S SP-
2100 on Supelcoport (100/120 mesh)
temperature program 45*C for 12
-------
68684 Federal Register / Vol. 44. No. 231 / Thursday, November 29, 1979 / Rules and Regulations
minutes, then program at l°/minute to
°C with a 25 ml/min flow. (See Figure
[or a sample chromatogram and Table
or retention data.)
4.13 Standard storage containers—15
ml amber screw-cap septum bottles with
Teflon faced silicone septa. Bottles and
caps—Pierce #19830 or equivalent.
Septa—Pierce #12716 or equivalent.
5. Reagents
5.1 Extraction solvent—(See 3.1).
Recommended—Pentane". Alternative—
hexane, methylcyclohexane or 2,2,4-
trimethylpentane.
5.2 Methyl alcohol—ACS Reagent
Grade.
5.3 Free and combined chlorine
reducing agents—Sodium thiosulfate
ACS Reagent Grade—sodium sulfite
ACS Reagent Grade.
5.4 Activated carbon—Filtrasorb—
200, available from Calgon Corporation,
Pittsburgh, PA, or equivalent.
5.5 Standards."
5.5.1 Bromoform 96%—available
from Aldrich Chemical Company.
5.5.2 Bromodichloromethane 97%—
available from Aldnch Chemical
Company.
5.5.3 Chlorodibromomethane—
available from Columbia Chemical.
1 orporated, Columbia, S.C.
i.5.4 Chloroform 99%—available
m Aldrich Chemical Company.
j.6 Organic-free water—Organic-
free water is defined as water free of
interference when employed in the
procedure described herein.
5.6.1 Organic-free water is generated
by passing tap water through a carbon
filter bed containing carbon. Change the
activated carbon whenever the
concentration of any trihalomethane
exceeds 0.4 fig/1.
5.6.2 A Millipore Super-Q Water
System or its equivalent may be used to
generate organic-free deionized water.
5.6.3 Organic-free water may also be
prepared by boiling water for 15
minutes. Subsequently, while
maintaining the temperature at 90° C.
bubble a contaminant free inert gas
through the water at 100 ml/minute for
• Pentane has been selected as the best solvent
for this analysis because it elutes. on all or the
columns, well before any of the Inhalomelhones.
High altitudes or laboratory temperatures in excess
of 75"F may make the use of this solvent
impractical For these reasons, alternative solvents
are acceptable, however, the analyst may
experience baseline variances In the elution areas
of the tnhalomethanes due to coelulion of these
solvents The degree of difficulty appears to be
dependent upon the design and condition of the
-'—Iron capture detector Such problems should be
;nificanl when concentration! of the coeluling
ilomclhnne are in excess of 5 jig/L
\s a precautionary measure, all standards must
„ Checked for punty by boiling paint
determinations or CC/MS assays
one hour. While still hot, transfer the
water to a narrow mouth screw cap
bottle with a Teflon seal.
5.6.4 Test organic free water each
day it is used by analyzing it according
to Section 7.
5.7 Standard stock solutions.
5.7.1 Fill a 10.0 ml ground glass
stoppered volumetric flask with
approximately 9.8 ml of methyl alcohol.
5.7.2 Allow the flask to stand
unstoppered about 10 minutes or until
all alcohol wetted surfaces dry.
5.7.3 Weigh the unstoppered flask to
the nearest 0.1 mg.
5.7.4 Using a 100 p.1 syringe,
immediately add 2 to 3 drops of the
reference standard to the flask, then
reweigh. Be sure that the reference
standard falls directly into the alcohol
without contacting the neck of the flask.
5.7.5 Dilute to volume, stopper, then
mix by inverting the flask several times.
5.7.6 Transfer the standard solution
to a dated and labeled 15 ml screw-cap
bottle with a Teflon cap liner.
Note.—Because of the toxicity of
trihalomethanes, it is necessary to prepare
primary dilutions in a hood It is further
recommended that a NIOSH/MESA-
approved toxic gas respirator be used when
the analyst handles high concentrations of
such materials.
5.7.7 Calculate the concentration in
mtcrograms per microliter from the net
gain in weight.
5.7.8 Store the solution at 4* C.
Note.—All standard solutions prepared in
methyl alcohol are stable up to 4 weeks when
stored under these conditions. They should
be discarded after that time has elapsed.
5.8 Aqueous calibration standard
precautions.
5.8.1 In order to prepare accurate
aqueous standard solutions, the
following precautions must be observed:
a. Do not inject more than 20 \i.\ of
alcoholic standards into 100 ml of
organic-free water.
b. Use a 25 pi Hamilton 702N
microsyringe or equivalent. (Variations
in needle geometry will adversely affect
the ability to deliver reproducible
volumes of methanolic standards into
water.)
c. Rapidly inject the aloholic standard
into the expanded area of the filled
volumetric flask. Remove the needle as
fast as possible after injection.
d. Mix aqueous standards by inverting
the flask three times only.
e. Discard the contents contained in
the neck of the flask. Fill the sample
syringe from the standard solution
contained in the expanded area of the
flask as directed in Section 7.
f. Never use pipets to dilute or transfer
samples and aqueous standards.
g. Aqueous standards, when stored
with a headspace, are not stable and
should be discarded after one hour.
Aqueous standards can be stored
according to Sections 6.4.9 and 7.2.
5.9 Calibration standards.
5.9.1 Prepare, from the standard
stock solutions, a multicomponent
secondary dilution mixture in methyl
alcohol so that a 20 uJ injection into 100
ml of organic-free water will generate a
calibration standard which produces a
response close (± 25%) to that of the
unknown. (See 8.1.)
5.9.2 Alternative calibration
procedure.
5.9.2.1 Construct a calibration curve
for each trihalomethane containing a
minimum of 3 different concentrations.
Two of the concentrations must bracket
each unknown.
5.9.3 Extract and analyze the
aqueous calibration standards in the
same manner as the unknowns.
5.9.4 Other calibration procedures
(7) which require the delivery of less
than 20 /il of methanolic standards to
10.0 ml volumes of water contained in
the sample syringe are acceptable only
if the methanolic standard is delivered
by the solvent flush technique (8).
5.10 Quality Check Standard
Mixture.
5.10.1 Prepare, from the standard
stock solutions, a secondary dilution
mixture in methyl alcohol that contains
10.0 ng/ul of each compound. (See 5.7.6
and 5.7.8.)
5 10 2 Daily, prepare and analyze a
2.0 u.g/1 aqueous dilution from this
mixture by dosing 20.0 pi into 100 ml of
organic-free water (See Section 8.1).
6. Sample Collection and Handling.
6.1 The sample containers should
have a total volume of at least 25 ml.
6.1.1 Narrow-mouth screw-cap
bottles with the TFE fluorocarbon faced
silicone septa cap liners are strongly
recommended.
6.2 Glassware Preparation.
6.2.1 Wash all sample bottles, TFE
seals, and extraction flasks in detergent.
Rinse with tap water and finally with
distilled water.
6.2.2 Allow the bottles and seals to
air dry, then place in an 105° C oven for
1 hour, then allow to cool in an area
known to be free of organics.
Note.—Do not heat the TFE seals for
extended periods of time (>1 hour) because
the silicone layer slowly degrades at 105° C
6.2.3 When cool, seal the bottles
using the TFE seals that will be used for
sealing the samples.
6.3 Sample stabilization—A
chemical reducing agent (Section 5.3) is
added to all samples in order to arrest
the formation of additional
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations 68685
tnhalomethanes after sample collection
(7.9) and to eliminate the possibility of
free chlorine reacting with impurities in
the extraction solvent to form interfering
organohalides. DO NOT ADD THE
REDUCING A GENT TO SAMPLES A T
COLLECT/ON TIME WHEN DATA
FOR MAXIMUM TRIHALOMETHANE
FORMA TION IS DESIRED. If chemical
stabilization is employed, then the
reagent is also added to the blanks. The
chemical agent (2.5 to 3 mg/40 ml) is
added in crystalline form to the empty
sample bottle just prior to shipping to
the sampling site. If chemical
stabilization is not employed at
sampling time then the reducing agent is
added just before extraction.
6.4 Sample Collection.
6 4.1 Collect all samples in duplicate.
6.4.2 Fill the sample bottles in such a
manner that no air bubbles pass through
the sample as the bottle is Tilled.
6.4.3 Seal the bottle so that no air
bubbles are entrapped in it.
6.4.4 Maintain the hermetic seal dn
the sample bottle until analysis.
6.4.5 The raw source water sample
history should resemble the finished
drinking water. The average retention
time of the finished drinking water
within the water plant should be taken
into account when sampling the raw
source water.
6.4.6 Sampling from a water tap.
6.4.6.1 Turn on the water and allow
the system to flush until the temperature
of the water has stabilized. Adjust the
flow to about 500 ml/minute and collect
duplicate samples from the flowing
stream.
6.4.7 Sampling from an open body of
water.
6.4.7.1 Fill a 1-quart wide-mouth
bottle with sample from a representative
area. Carefully fill duplicate sample
bottles from the 1-quart bottle as in 6.4.
6.4.8 If a chemical reducing agent
has been added to the sample bottles.
fill with sample just to overflowing, seal
the bottle, and shake vigorously for 1
minute.
6.4.9 Sealing practice for septum seal
screw cap bottles.
6.4.9.1 Open the bottle and fill to
overflowing. Place on a level surface.
Position the TFE side of the septum seal
upon the convex sample meniscus and
seal the bottle by screwing the cap on
tightly.
6.4.9.2 Invert the sample and lightly
tap the cap on a solid surface. The
absence of entrapped air indicates a
successful seal. If bubbles are present,
open the bottle, add a few additional
drops of sample, then reseal bottle as
above.
6.4.10 Sample blanks.
6.4.10.1 Prepare blanks in duplicate
at the laboratory by filling and sealing
sample bottles with organic-free water
just prior to shipping the sample bottles
to the sampling site.
6.4.10.2 If the sample is to be
stabilized, add an identical amount of
reducing agent to the blanks.
6.4.10.3 Ship the blanks to and from
the sampling site along with the sample
bottles.
6.4.10.4 Store the blanks and the
samples, collected at a given site
(sample set), together in a protected
area known to be free from
contamination. A sample set is defined
as all the samples collected at.a given
site (i.e., at a water treatment plant,
duplicate raw source water, duplicate
finished water and the duplicate sample
blanks comprise the sample set).
6.5 When samples are collected and
stored under these conditions, no
measurable loss of trihalomethanes has
been detected over extended periods of
time (7). It is recommended that the
samples be analyzed within 14 days of
collection.
7. Extraction and Analysis.
7.1 Remove the plungers from two
10-ml syringes and attach a closed
syringe valve to each.
7.2 Open the sample bottlec (or
standard) and carefully pour the sample
into one of the syringe barrels until it
overflows. Replace the plunger and
compress the sample. Open the syringe
valve and vent any residue air while
adjusting fhe sample volume to 10.0 ml.
Close the valve.
7.3 Fill the second syringe in an
identical manner from the same sample
bottle. This syringe is reserved for a
replicate analysis (see 8.3 and 8.4).
7.4 Pipette 2.0 ml of extraction
solvent into a clean extraction flask.
7.5 Carefully inject the contents of
the syringe into the extraction flask.
7.6 Seal with a Teflon faced septum.
7.7 Shake vigorously for 1 minute.
7.8 Let stand until the phases
separate (/60 seconds).
7.8.1 If the phases do not separate on
standing then centrifugation can be used
to facilitate separation.
7.9 Analyze the sample by injecting
3.0 n\ (solvent flush technique, (8)) of the
upper (organic) phase into the gas
chromatograph.
8. Analytical Quality Control.
8.1 A 2 ug/1 quality check standard
(See 5.10} should be extracted and
analyzed each day before any samples
are analyzed. Instrument status checks
• If for any reason the chemical reducing agent
has not been added to the sample, then it must be
added just prior lo analyses at the rale of 2 5 to 3
mg/40 ml or by adding 1 mg directly to the sample
in the extraction flask.
and lower limit of detection estimations
based upon response factor calculations
at 5 times the noise level are obtained
from these data. In addition, the data
obtained from the quality check
standard can be used to estimate the
concentration of the unknowns. From
this information the appropriate
standards can be determined.
8.2 Analyze the sample blank and
the raw source water to monitor for
potential interferences as described in
Sections 3.1, 3.2, and 3.3.
8.3 Spiked samples.
8.3.1 For those laboratories
analyzing more than 10 samples a day,
each 10th sample analyzed should be a
laboratory-generated spike which
closely duplicates the average finished
drinking water in tnhalomethane
composition and concentration. Prepare
the spiked sample in organic-free water
as described in section 5.9.
8.3.2 In those laboratories analyzing
less than 10 samples daily, each time the
analysis is performed, analyze at least
one laboratory generated spike sample
which closely duplicates the average
finished drinking water in
tnhalomethane composition and
concentration. Prepare the spiked
sample in organic-free water as
described in section 5.9.
8.3.3 Maintain an up-to-date log on
the accuracy and precision data
collected in Sections 8.3 and 8.4. If
results are significantly different than
those cited in Section 10.1, the analyst
should check out the entire analysis
scheme to determine why the
laboratory's precision and accuracy
limits are greater.
8.4 Randomly select and analyze
10% of all samples in duplicate.
8.5 Analyze all samples in duplicate
which appear to deviate more than 30%
from any established norm.
8.8 Quarterly, spike an EMSL-
Cincinnati tnhalomethane quality
control sample into organic-free water
and analyze.
8.8.1 The results of the EMSL
trihalomethane quality control sample
should agree within 20% of the true
value for each trihalomethane. If they do
not, the analyst must check each step in
the standard generation procedure to
solve the problem.
8.7 It is important that the analyst be
aware of the linear response
characteristics of the electron capture
system that is utilized. Calibration
curves should be generated and
rechecked quarterly for each
trihalomethane over the concentration
range encountered in the samples in
order to confirm the linear response
range of the system. Quantitative data
cannot be calculated from non-linear
-------
68686 Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations
responses. Whenever non-linear
responses are noted, the analyst must
dilute the sample for reanalysis.
8.8 Maintain a record of the
retention times for each trihalomelhane
using data gathered from spiked
samples and standards.
8.8.1 Daily calculate the average
retention time for each trihalomethane
and the variance encountered for the
analyses.
8.8.2 If individual trihalomethane
retention time varies by more than 10%
over an eight hour period or does not fall
within 10% of an established norm, the
system is "out of control." The source of
retention data variation must be
corrected before acceptable data can be
generated.
9. Calculations
9.1 Locate each trihalomethane in
the sample chromatogram by comparing
the retention time of the suspect peak to
the data gathered in 8.8.1. The retention
time of the suspect peak must fall within
the limits established m 8.8.1 for a single
column idpntification.
9.2 Calculate the concentration of
each trihalomethane by comparing the
peak heights or peak areas of the
samples to those of the standards.
Round off the data to the nearest jxg/1 or
two significant Figures.
Concentration, ug/1 = sample peak height/
standard peak height X standard
concentration, jig/I.
9.3 Calculate the total
trihalomethane concentration (TTHM)
by summing the 4 individual
trihalomethane concentrations in ug/I:
TTHM Oig/1) = (cone. CHCUJ+fconc.
CHBrCU) + (conc. CHBr,Cl) + (conc.
CHBr,)
9.4 Calculate the limit of detection
(LOD) for each trihalomethane not
detected using the following criteria:
/lAXATTA
LOD O.n/1) - f 1 « P C9/D
V'BXATn/
Where
B = peak height (mm) of 2 fig/1 quality check
standard
A = 5 limes the noise level in mm at the
exact retention time of the
tnha omethane or the base line
displacement in mm from theoretical
zero at the exact retention time for the
Inhalomethane.
ATT = attenuation factor.
9.5 Report the results obtained from
the lower limit of detection estimates
along with the data for the samples.
10. Precision and Accuracy
10.1 Single lab precision and
accuracy. The data in Table II were
generated by spiking organic-free water
with tnhalomethanes as described in
5 9. The mixtures were analyzed by the
analyst as true unknowns.
Table 1.—Retention Times for Tnhalomethanes
T'lha'o-nelfiare
Retention lirre nmutos
Column
A
Chloroform. ... ^ .. to
Bromodichloromethane IS
Cnioroaibromomeinane 26
(Dibromochloromeinane)
broffloform _ — _ _ 55
Column
B
13
••25
56
109
Column
C
49
110
23 I
394
•On (his column, uichlorocthylcne a common raw soured
water contaminate, cooljtes with bromod.cliloromclhano
Table \\.-Smglg Laboratory Accuracy and Precision
Dose lovel Numbei ol
H9/1 samples
Compound
CHCIj _ .. . . .
CHCI, . _
CHBrCL _ _
CHBiO, _
CHBr,CI . . . _ ....
CHBr,d
CHBr, . . . _ _
CHBr, . _
9 1
69
12
12
27
17
29
14
5
3
5
2
5
3
5
3
Precision
relative Accuracy
Mean pg/l standard percent
deviation. recovery
percent
73
13
15
20
16
22
16
S3 II
98 II
1 4 i:
17 ;
90 1
10 ;
12 1
-References
1. Mieure. |. P, "A Rapid and Sensitive
Method for Determining Volatile
Organohalides in Water." Journal A WWA.
69. 60,1977.
2 Reding. R. et al. "THM's in Drinking
Water Analysis by LLE and Comparison to
Purge and Trap". Organics Analysis in Water
and Wastewater. STP 686 ASTM. 1979.
3 "Handbook for Analytical Quality
Control in Water and Waste water
Laboratories," Analytical Quality Control
Laboratory. National Environmental
Research Center, Cincinnati, Ohio, June 1972.
4. Budde, W. L, J. W. Eichelberger,
"Organic Analysis Using Gas
Chromalography-Mass Spectromelry," Ann
Arbor Science, Ann Arbor. Michigan, 1979.
5. "The Analysis of Trihalomelhanes in
Finished Water by the Purge and Trap
Method," Environmental Monitoring and
Support Laboratory, Environmental Research
Center. Cincinnati, Ohio. 45268. May 15.1979.
6 Richard ] ]; C. A. Junk. "Liquid
Extraction for Rapid Determination of
Halomelhancs in Vfalcr. fourna/A WWA. HU
62, January 1977.
7 Brass. H )., et al.. "National Organic
Monitoring Survey: Sampling and Purgeablc
Organic Compounds, Drinking Water Quahtj
Through Source Protection." R. B. Pojnsek.
Editor, Ann Arbor Science, p. 398.1977.
8. White, L D.. et al. "Convenient
Optimized Method for the Analysis of
Selected Solvent Vapors in Industrial
Atmosphere," AIHA Journal. Vol 31, p. 225.
1970.
9 Kopfler, F. C.. et al "CC/MS
Determination of Volatiles for the National
Organics Reconnaissance Survey (NORS) or
Drinking Water. Identification and Analysis
of Organic Pollutants in Water." L. H Keith.
Editor. Ann Arbor Science, p. 87.1976.
BILLING CODE 6S60-01-M
-------
Federal Register / Vol. 44. No. 231 / Thursday. November 29.1979 / Rules and Regulations 68687
z
LU
>
_l
O
V/l
U
LU
^
COLUMN PACKING: 37.SP-1000
CARRIER GAS: 5% CH4 IN ARGON
CARRIER FLOW: 60 ML/MIN.
COLUMN TEMPERATURE: 50°C
DETECTOR: ELECTRON CAPTURE
CN
O
RETENTION TIME IN MINUTES
FIGURE 1. FINISHED WATER EXTRACT
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29.1979 / Rules and Regulations
LU
z
<
I
H-
LL4
COLUMN PACKING: 10%
SQUALANE CARRIER
FLOW: 25ml/min COLUMN
TEMPERATURE: 67
512X
1024X
256X
67
(min)
8
10
11 12
FIGURE 2. EXTRACT OF STANDARD
-------
Federal Register / Vol. 44. No. 231 / Thursday. November 29.1979 / Rules and Regulations_Jj868g
COLUMN PACKING: 6% OV-1H4% SP-2100
ui CARRIER FLOW: 25 ml/min
- TEMPERATURE PROGRAM: 45°C-12 MINUTES
O 1°/MINUTE TO 70°C
UJ
O
2
O
at
63
5 10 15 20 25 30 35 40 45
"TIME (min)
FIGURE 3. EXTRACT OF STANDARD
BILLING CODE 6560-01-C
-------
68690 Federal Register / Vol. 44. No. 231 / Thursday. November 29. 1979 / Rules and Regulations
Part III—Determination of Maximum
Total Trihalomethane Potential (MTP)
The water sample used for this
determination is taken from a point in
the distribution system that reflects
maximum residence time. Procedures for
sample collection and handling are
given in EMSL Methods 501.1 and 501.2.
No reducing agent is added to "quench"
the chemical reaction producing THMs
at the time of sample collection. The
intent is to permit the level of THM
precursors to be depleted and the
concentration of the THMs to be
maximized for the supply being tested.
Four experimental parameters
affecting maximum THM production are
pH, temperature, reaction time and the
presence of a disinfectant residual.
These parameters are dealt with as
follows:
Measure the disinfectant residual at
the selected sampling point. Proceed
only if a measurable disinfectant
residual is present. Collect triplicate 40
ml water samples at the pH prevailing at
the time of sampling, and prepare a
method blank according to the EMSL
methods. Seal and store these samples
together for 7 days at 25°C or above.
After this time period, open one of the
sample containers and check for
disinfectant residual. Absence of a
disinfectant residual invalidates the
sample for further analyses. Once a
disinfectant residual has been
demonstrated, open another of the
sealed samples and determine total
THM concentration using either of the
EMSL analytical methods.
Attachment 7.—Statement of Basis and
Purpose for an Amendment to the
National Interim Primary Drinking
Water Regulations on Trihalomethanes,
August 1979
Office of Drinking Water Criteria and
Standards Division. Environmental
Protection Agency. Washington. D.C.
20460.
Table of Contents
I. Summary
II. Introduction
III. The Role of Chlorine and Other
Disinfectants
IV. Sources of Trihalomethanes Exposure
V. Metabolism
VI. Acute and Chronic Health Effects in
Animals
A. Hepatotoxicity
B. Nephrotoxicity
C. Central Nervous System
D. Teratogenicity
E. Mutagenicity
F. Carcinogenicity
VII. Human Health Effects
A. NAS Principles of Toxicological
Evaluation
B. Epidemiologic Studies
VIII. Mechanisms of Toxicity
IX. Risk Assessment
X. Maximum Contaminant Levels
XI. References
I. Summary
The trihalomethanes (THMs) are a
family of organic compounds, named as
derivatives of methane, where three of
the four hydrogen atoms are substituted
by a halogen atom. Although halogens
can include fluorine, chlorine, bromine
and iodine, only chlorine and bromine
substituents are now considered for the
purpose of this regulation. THMs in
drinking water are produced by the
action of the chlorine added for
disinfection or oxidation, with the
naturally occurring organic precursors
(e.g., humic or fulvic acids) commonly
found in source waters.
THMs are commonly found in
drinking water supplies throughout the
United States. Chloroform has been
found at concentrations ranging from
0.001-0.540 mg/1 and (TTHM) potential
concentrations as high as 0.784 mg/1
have been detected. The concentrations
of TTHM increase when raw water
supplies are treated with chlorine for
disinfection and other purposes. TTHM
concentrations are indicative of the
presence of other halogenated and
oxidized organic chemicals that are
produced in water during chlorination.
People are also exposed to chloroform
in the air they breathe and the food they
eat. Analyses of the relative
contribution of chloroform in drinking I
water, air and food exposures assumed \
various levels of exposure based on
monitoring studies. Drinking water may
contribute from zero to more than 90% of
the total body burden.
Chloroform has been shown to be
rapidly absorbed on oral and
intraperitoneal administration and
subsequently metabolized to carbon
dioxide, chloride ion, phosgene, and
other unidentified metabolites. The
metabolic profile of chloroform in
animal species such as mice, rats, and
monkeys is indicated in Table 4 and is
qualitatively similar to that in man.
Mammalian responses to chloroform
exposure include: central nervous
system depression, hepatotoxicity,
nephrotoxicity, teratogenicity, and
carcinogenicity. These responses are
discernible in mammals after oral and
inhalation exposures to high levels of
chloroform ranging from 30-350 mg/kg;
the intensity of response is dependent
upon the dose. Although less
toxicological information is available for
the brominated THMs. mutagenicity and
carcinogenicity have been detected in
some test systems. Physiological
chemical activity should be greater for
the brominated THMs than for
chloroform.
Although short-term toxic responses
to THMs in drinking water are not
documented, the potential effects of
chronic exposures to THMs should be a
matter of concern. Prolonged
administration of chloroform at
relatively high dose levels (100-138 mg/
kg) to rats and mice, manifested
oncogenic effects. Oncogenic effects
were not observed at the lowest dose
level (17 mg/kg) in three experiments.
Since methods do not now exist to
establish a threshold no effect level of
exposure to carcinogens, the preceding
data do not imply that a "safe" level of
exposure can be established for humans.
Human epidemiological evidence is
inconclusive, although positive
correlations with some sites have been
found in several studies. There have
been 18 retrospective studies shown in
Table 7 that have investigated some
aspect of a relationship between cancer
mortality or morbidity and drinking
water variables. Due to various
limitations in the epidemiological
methods, in the water quality data, and
problems with the individual studies, the
present evidence cannot lead to a firm
conclusion that there is an association
between contaminants in drinking water
and cancer mortality/morbidity. Causal
relationships cannot be proven on the
basis of results from epidemiological
studies. The evidence from these studies
thus far is incomplete and the trends
and patterns of association have not
been fully developed. When viewed
collectively, however, the
epidemiological studies provide
sufficient evidence for maintaining the
hypotheses that there may be a potential
health risk, and that the positive
correlations may be reflecting a causal
association between constituents of
drinking water and cancer mortality.
Preliminary risk assessments made by
the Science Advisory Board (SAB), the
National Academy of Sciences (NAS),
Tardiff, and EPA's Carcinogen
Assessment Group (CAG) using
different models have estimated the
incremental risks associated with the
exposure from chloroform in drinking
water. The exposure to THMs from air
and food have not been included in
these computations. The risk estimates
associated with the MCL at the 0.10
mg/1 level are essentially the same from
the NAS and CAG computations (3.4 x
10~4 and 4 x 10"") assuming two liters of
water at 0.10 mg/1 chloroform consumed
daily for 70 years.
On the basis of the available
toxicological data summarized in the
following report, chloroform has been
shown to be a carcinogen in rodents
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations 68691
(mice and rats) at high dose levels. Since
its metabolic pattern in animals is
[qualitatively similar to that in man, it
should be suspected of being a human
carcinogen. Epidemiological studies also
suggest a human risk. Therefore,
because a potential human health risk
does exist, levels of chloroform in
drinking water should be reduced as
much as is technologically and
economically feasible using methods
that will not compromise protection
from waterborne infectious disease
transmission.
Although documentation of their
toxicity is not so well established, other
THMs should be suspected of posing
similar risks. Because the treatment
process that can reduce drinking water
levels of chloroform have about the
same effectiveness in reducing levels of
the other THMs, the proposed regulation
is addressed to these substances, as
well.
II. Introduction
The extent and significance of organic
chemical contamination of drinking
water or drinking water sources first
came to public attention in 1972, when a
report, "Industrial Pollution of the Lower
Mississippi River in Louisiana" was
published (EPA, 1972). While this report
(did not include quantification of the
pollutants found, and was directed
toward locating industrial discharges
responsible for the pollution, the report
did include analyses of finished
(treated) drinking water and provided
evidence of the presence of THMs.
Subsequently, a more thorough
examination of Finished drinking water
in the New Orleans area was carried
out, using the most sophisticated
analytical methods available (EPA,
1974). This latter study confirmed the
presence of THMs and many other
organic chemicals in finished drinking
water, and furthermore it demonstrated
that one of them, chloroform, was
present in high relative concentrations.
The findings in New Orleans
promoted other studies, primarily for the
purpose of determining how widespread
and serious the organic chemical
contamination of drinking water was.
Impetus was added by the passage of
the Safe Drinking Water Act (Pub. L. 93-
523), which directed the EPA to conduct
a comprehensive study of public water
supplies and drinking water sources to
determine the nature, extent, sources,
and means of control of contamination
by substances suspected of being
carcinogenic. The National Organics
Reconnaissance Survey of Halogenated
Organics (NORS) (Symons, et al, 1975),
or "80 City Study", was aimed primarily
at determining the extent of the
presence of four THMs, chloroform.
bromodichloromethane,
dibromochhromethane and bromoform.
along with carbon tetrachloride and 1,2-
dichloroethane, and at determining what
effect raw water source and water
treatment practices had on the
formation of these compounds (Table 1).
The presence of THMs in finished
drinking water was confirmed, and
some trend relating non-volatile total
organic carbon (NVTOC) of the raw
water and the total tnhalomethane
(TTHM) was postulated. Chloroform
occurred invariably in water which had
been chlorinated, while it was absent or
present at much lower concentrations in
the raw water. Water samples were
collected at the treatment plant in
winter and iced for shipment but not
dechlonnated. Thus, those values might
approximate minima for human
exposure in the areas selected. Of the
various THMs. chloroform was found at
the highest concentrations (averaging
'approximately 75 percent of the TTHM),
with progressively less
bromodichloromethane,
dibromochloromethane and bromoform
being detected. In some cases
chloroform was found at concentrations
greater than 0.300 mg/1; [the highest
value found was 0.540 mg/1). Carbon
tetrachloride and 1,2-dichloroethane
were found at very low concentrations
The concentration of these two
components did not increase after
chlorination; therefore, it can be
assumed that these compounds are not
related to the chlorination process.
BILLING CODE 6560-01-M
-------
68692 Federal Register / Vol. 44. No. 231 / Thursday. November 29.1979 / Rules and Regulations
TABLE I - Analytical results of chloroform, bromoform,
bromodichloromethane, and dibromochloromethane and total
trihalomethanes in water supplies from NORS and NOMS
(Concentrations in milligrams per liter)
NORS
Chloroform
Median 0.021
Mean
Range NF-0.311
Bromoform
Median 0.005
Mean
Range NF-0.092
NOMS
Phase I
Phase
II Phase III
Dechlorinated Terminal
0.027
0.043
NF-0.271
LD
0.003
NF-0.039
0.059
0.083
NF-0. 47
LD
0.004
NF-0.280
0/022
0.035
NF-0. 20
LD
0.002
NF-0. 137
0.044
0.069
NF-0. 540
LD
0.004
NF-0. 190
Dibromochloromethane
Median 0.001
Mean
Range NF-0.100
LD
0.008
NF-0.19
0.004
0.012
NF-0.290
0.002
* 0.006
NF-0. 114
0.003
0.011
NF-0. 250
Bromodichloromethane
Median 0.006
Mean
Range NF-0.116
0.010
. 0.018
NF-0.183
0.014
0.018
NF-0. 180
0.006
0.009
NF-0. 072
0.01 1
0.017
NF-0. 125
Total Trihalomethanes
Median 0.027
Mean 0.067
Range NF-0. 482
0.015
0.068
NF-0. 457
0.087
0.117
NF-0. 784
0.037
0.053
NF-0. 295
0.074
0.100
NF-0. 695
NF = not found
LD = less than detection limit
BT J.ING CODE 6560-OI-C
-------
Federal Register / Vol. 44, No. 231 / Thursday, November 29. 1979 / Rules and Regulations 68693
A Joint Federal/State Survey of
Organics and Inorganics in 83 Selected
Drinking Water Supplies, carried out by
EPA's Region V (Chicago) provided
additional evidence of the ubiquitous
nature of chloroform and other THMs in
chlorinated drinking water (EPA. 1975).
Two conclusions reached in that study
were that raw water relatively free of
organic matter results in finished water
that is relatively free of chloroform and
related halogenated compounds, and
that there is a correlation in some
instances between the concentrations of
chloroform, bromodichloromethane,
dibromochloromethane and bromoform
in finished water and the amount of
organic matter found in raw water.
The National Organics Monitoring
Survey (NOMS). directed by § 141.40 of
the National Interim Primary Drinking
Water Regulations (40 FR 59574,
December 24,1975), was aimed not only
at determining the presence of THMs in
additional water supplies, but also at
determining the seasonal variations in
concentration of these substances.
The NOMS sampling included 113
public water systems designated by the
Administrator, and also included
analyses for approximately 20 specific
synthetic organic chemicals deemed to
be candidates of particular concern as
well as analyses of several surrogate
group chemical parameters which are
indicators of the total amount of organic
contamination. Three phases of this
study were completed and the mean,
minimum, and maximum values of
chloroform and THMs in drinking water
are reported in Table 1. Phase I analyses
in the NOMS were conducted similarly
to the NORS. Phase II analyses were
performed after the THM-producing
reactions were allowed to run to
completion. Phase III analyses were
conducted on both declorinated samples
and on samples that were allowed to
run to completion (terminal). Again
chloroform was found at the highest
concentrations in most cases, however,
in a few cases bromoform was found to
be the highest concentration of the
THMs (0.280mg/l). The mean
concentrations of chloroform were 0.043
mg/1, 0.083 mg/1. 0.035 mg/1. and 0.069
mg/1 for Phase I, II, III (dechlorinated)
and III (terminal), respectively; the mean
concentrations for TTHMs were 0.068
mg/1, 0.117mg/l, 0.053 mg/1 and 0.100
mg/1 for Phase I, II, III (dechlorinated]
and III (terminal), respectively.
III. The Role of Chlorine and Other
Disinfectants
All available evidence indicates that
chlorination of drinking water
containing naturally occurring organic
chemicals is the major factor in the
formation of halogenated organic
chemicals, particularly the THMs in
finished drinking water. Chlorinated
organic compounds, however, can also
be introduced into drinking water from
industrial outfalls, urban and rural
runoff, rainfall, through polluted air, or
from the chlorination in sewage and
industrial wastewater.
Several studies in addition to those
mentioned above, have demonstrated
increased THM concentrations in
drinking water. Work by J. J. Rook (1974)
in the Netherlands, and the studies by
Bellar, Lichtenberg and Kroner (1974),
showed that chloroform and other
halogenated methanes are formed
during the water chlorination process. It
should be noted that these findings
came as a result of the development and
application of more sensitive and
refined analytical techniques. Recent
work by Rook (1974,1977) has provided
some insight into the organic precursors
which might be responsible for the
formation of the THMs. Studies by
Sontheimer and Kuhn (1977) indicate
that the THMs may represent only a
portion of the total halogenated
products of chlorination of water. Bunn
et al. (1975), have demonstrated that
hypochlonte in the presence of bromide
and iodide ions but not fluoride will
react with natural organic matter to
produce all ten possible trihalogenated
methanes.
It can be concluded from the above
studies and others that the THMs occur
in chlorinated drinking waters, and that
the concentrations of the various THMs
are dependent on the type and quality of
organic precursor substances, the
amount of chlorine used, and the
presence of other halogen ions as well
as contact time, temperature and pH.
A number of methods are available
for reducing levels of THMs in drinking
water. These options include
modifications of current treatment
practices, such as moving the point of
chlorination, the use of alternative
disinfectants such as chlorine dioxide,
chloramines, or ozone, and various
methods that will reduce organic
precursor concentrations such as use of
adsorbents like granular activated
carbon (GAG).
Two chemicals often mentioned as
alternative disinfectants, chlorine
dioxide and ozone, are both well known
as effective disinfectants and chemical
oxidents, and some history of their
practical use in water treatment has
been accumulated particularly in
Europe, but also in the United States.
Chlorine dioxide is usually prepared
at the water plant by the reaction of
chlorine (either as gas or as sodium
hypochlorite) with sodium chlorite.
Unless an excess of chlorine is used,
there will be unreacted sodium chlorite
left over from the reaction. When
chlorine dioxide reacts with organic
matter in the water, one of the reaction
products is the chlorite ion. Thus,
whenever chlorine dioxide is used to
treat water, the presence of chlorite ion
in the treated water can be expected.
EPA is studying the health effects of
chlorine dioxide in water, utilizing
several animal species as well as human
volunteers. Studies of the toxicology of
chlorine dioxide and chlorite ion in
drinking water reveal considerable
variations. These compounds have been
reported to affect the hematopoietic
sysfems such as oxidative changes in
hemoglobins and hemolysis of red blood
cells. Other bioeffects observed include
gastrointestinal disturbances. The
preliminary results indicate species
variability in biological manifestations.
Cats and African green monkeys appear
to lie at the extreme ends of the
spectrum from among the species
studied; cats are very sensitive to
hematopoietic effects whereas monkeys
were apparently insensitive even at
levels as high as 400 mg/1 (Bull, 1979).
An upper limit for chlorine dioxide by-
product exposure is being considered
primarily because of the lack of data
concerning the safety of this material,
and particularly its decomposition
products, at higher concentrations
(Musil et al., 1963 and Fridyland and
Kagan, 1971. Studies with cats have
shown that chlorite, which is oxidant
that can cause anemias, has a
deleterious effect on red blood cell
survival rate at chlorine dioxide
concentrations above 10 mg/1.
Preliminary studies in a small human
population did not demonstrate
substantial blood chemistry changes,
except possibly in one person known to
be deficient in glucose-6-phosphotase
dehydrogenase. Lack of sufficient health
effects data on human toxicity for CIO,
and its by-products prevents
establishment of an MCL at this time,
however, work in progress is expected
to provide much additional information
within the coming year. In the meantime,
EPA recommends that monitoring be
conducted when chlorine dioxide is
used, and that residual oxidant should
not exceed 0.5 mg/1 as C1O>.
A preliminary study concerning
ozonation of 29 organic compounds
potentially present in water supply
sources indicated the formation of a
number of products (Cotruvo, Simmon,
Spanggord, 1976,1977). These reaction
mixtures were assayed for mutagenic
activity employing 1) five strains of
Salmonella typhimurium (Ames
-------
68694 Federal Register / Vol. 44, No. 231 / Thursday, November 29. 1979 / Rules and Regulations
Salmonella/microsome assay); and 2)
mitotic recombination in the yeast
Saccharomyces cerevisiae D3. After
very extensive ozonation in water some
of the organic compounds exhibited
mutagenic activity in these systems.
Similar more recent studies under
extreme conditions with chlorine
dioxide by-products did not exhibit
mutagenic activity (SRI Report).
Combining ammonia with chlorine to
form chloramines has been called the
chloramine process, chloramination, and
combined residual chlorination. The
products of this process are
monochloramine, dichloramme or
trichloramines (nitrogen trichloride)
depending on the pH and the chlorine to
ammonia ratio. The production of the
latter species may contribute to taste
and odor problems in the finished water;
however, chloramination does not
reduce the formation of THMs.
Based on the results of numerous
investigations, the comparative
disinfectant efficiency of chloramines
ranks last when compared to ozone,
chlorine dioxide, hypochlorous acid
(HOC1). and hypochlorite ion (OC1~)
(NAS. 1977,1979). Early studies by
Butterfleld and Waties (1944,1946,1948)
demonstrated that chloramines required
approximately a 100-fold increase in
contact time to inactivate coliform
bacteria and enteric pathogens as
compared to free available chlorine at
pH 9 5. This work was later confirmed
by Kabler (1953) and by Clarke et al.,
(1962).
Results with cysts of Entamoeba
histolytica and viruses also confirm the
decreased effectiveness of chloramines
as disinfectants. Studies by Fair, et al..
(1947) showed that additional
dichloratnine is about 60 percent and
monochloroamine about 22 percent as
effective as hypochlorous acid at pH 4.5
against cysts of E. histolytica. Kelly and
Sanderson (1960) found that chloramines
in the concentration of 1 mg/1 at 25* C
required 3 hours at pH 6, or 6 to 8 hours
at pH 10 to achieve 99.7 percent
inactivation of polio virus. With 0.5
mg/1 free chlorine at pH 7.8, by
comparison, inactivation of 99.99
percent of polio virus can be achieved in
approximately 15 minutes (Liu and
McGrowan, 1973). Chloramine treatment
finds its widest application in
maintenance of chlorine residuals in the
distributing systems. The human health
effects of consuming water treated with
chloramine have not been studied in
detail.
Although all of these disinfectants can
reduce THM formation, questions have
been raised on both their toxicity and
the toxicity of their by-products. Studies
are underway to clarify these matters.
and could result in the designation of
maximum permissible levels for certain
disinfectants applied to drinking water.
The use of adsorbents for THM
removal has also introduced some
unknown factors. Assuming that the
adsorption process is effective for its
intended purpose, there is the possibility
that a breakthrough of some of the
adsorbed chemicals may occur, that
these substances will be adsorbed and
subsequently slough off to produce
intermittent contamination, or that
bacteria and/or toxins will be added to
the water from growth on the adsorbent.
All of these potential effects are
controllable in practice, and EPA
encourages the use of GAC to purify
contaminated waters and to control
THM precursors.
Thus, THM concentrations should be
reduced, but without compromising
public health from either increased risk
of infectious disease transmission or
from the chemicals that are used.
Outbreaks of infectious waterborne
disease have been noted when
chlorination systems have been
improperly operated. The alternative
control methods outlined previously are
effective, and are also being studied for
their possible side effects. As soon as
data become available, EPA will make
specific recommendations regarding
their use. At the present time, the best
approach to reduce THMs in finished
water is to reduce precursors prior to
chlorination, such as with CAC. This
approach has the benefit of reducing the
concentration of many other organic
chemicals in the water as well as to the
precursors to THM and other
chlorinated organics. Thus, once the
organic chemical concentrations in the
water have been reduced, the chemical
demand for applied disinfectant will be
reduced. Thus, human exposure to all
disinfectant chemicals and their
degradation products and by-products
will be minimized. This is the intent of
the regulation controlling THMs.
IV. Sources of Trihalomethane Exposure
McConnell et al. (1975), have reported
that chloroform occurs in many common
foods and that while some halogenated
compounds in food may result from
manufacturing, canning and pest control
practices, chloroform may be introduced
as the result of geochemical processes.
Chlorinated compounds are the
halogenated species most prevalent in
food, but at least one food, Limu Kohu, a
seaweed or algae eaten in Hawaii,
contains an essential oil which is
composed largely of bromoform
(Burreson, et al 1975).
Chloroform was widely used as an
anesthetic in the past, and, until
recently, was a common ingredient in
dentifrices and cough preparations. The
Food and Drug Administration has
taken action to halt the use of
chloroform in drug products, cosmetic
products, and food-contact articles (41
FR 145026. April 9.1976). EPA has issued
a notice of "rebuttable presumption
against registration" of chloroform-
containing pesticides (41 FR 14588, April
6,1976). Thus, in addition to drinking
water, exposure to some or all of the
THMs is complicated by other
environmental sources, however,
exposure from some of these sources is
being, reduced.
The relative human chloroform
exposures can be estimated for three
major sources of human exposure:
atmosphere, drinking water, and the
food supply. The uptake calculations are
based on the fluid intake, respiratory
volume, and food consumption data for
"reference man" as compiled by the
International Commission on
Radiological Protection. The combined
uptake for adults from all three sources
was derived by multiplying estimated
exposure levels by the estimated annual
intakes and combining the results |ODW
protocol].
Human uptake of chloroform from air,
food and drinking water is given in
Table 2. Chloroform and TTHM uptake
from drinking water was estimated by
multiplying the chloroform and THM
concentrations from NOMS data (Table
1) by the average consumption of 2 liters
of water per day for the 70 kg adult
male, by 365. One hundred per cent
absorption of the amount of chloroform
in drinking water is assumed for these
calculations. The total chloroform
uptake from water was estimated as a
mean value of 64 mg per year. The
maximum uptake value may be 394 mg
per year.
To determine uptake of chloroform
from foods, the concentration of
chloroform in each food item in North
American diets was multiplied by the
average annual consumption of that
food item by adults in the United States
(NAS. 1977), and the results were
combined again; one hundred per cent
absorption of ingested chloroform was
assumed. A calculated maximum value
of about 16 mg of chloroform uptake per
year from total food an a mean value of
9 mg based on OOW assumptions was
obtained.
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68695
Table 2—Human Uptake of Chloroform and
Tnhalomethanes from Drinking Water. Food, and Air
Exposure levels mg/year
Chemical Mean (range)
OrmKing Food Air •
water
Max Water Mm-Air
Chloroform
Tnhalomethanes
64 9 20
(0 73-343) (2-15 97) (0 41-204)
85
(0 73-572)
'Calculated from data supplied by Strateg.es ard Air
Standards Division. Office ol Air Quality Planning and Stand-
ards Environmental Protection Agency Research Triangle
Park The air samples were collected both from the rural and
mduslnal areas during the years 1974-76 The mean value
was derived from the concentrations obtained from urban in-
dustrialized areas. the minimum value from the rural area and
the maximum value from an urban industrialized area
The calculation for the uptake of
chloroform by humans from ambient air
was based upon the assumptions that 63
percent of inhaled choloroform is
absorbed, (NAS, 1977); the volume of air
inhaled by an average adult is 8.1 X 10s
liters per year; and 0.02 and 10 ppb (by
volume) are the respective minimum and
maximum chloroform concentrations in
urban air. The minimum and maximum
values for the annual uptake of
chloroform by an adult were estimated
it 041 and 204 mg, respectively.
\ssuming minimum exposures from all
iources, the atmosphere contributes 12
lercent of the total chloroform, the
drinking water contributes 23 percent,
and food is most significant (65%).
Assuming maximum exposures from all
sources, drinking water is the major
contributor at 61 percent, with air at 36
percent. Thus, the relative contribution
of drinking water to the total body
burden of chloroform may range from a
moderate to a maximum contributor as
the annual exposure from water ranges
from nil to 394 mg/year, and from 204 to
0.73 mg/year in ambient air (Table 3).
Table 3.—Uptake of Chloroform tor the Adult Human
from Air. Water, and Food
Source
Adult Percent
mg/yr uptake
Maximum Conditions
Atmosphere
Water
Food supply
Total
Atmosphere
Water
Food supply
Total
204
343
16
36
61
563
100
Minimum Conditions
041 13
073 23
200 64
Atmosphere
Water
Food supply
Total
041 1
343 00 97
900 2
35241
100
B. Metabolism
Several reports (Brown, et dl., 1974;
Labigne & Marchand, 1974, Fry et al.,
1972; Paul and Rubenstein, 1963; Taylor
et al., 1974) have indicated that
chloroform is rapidly absorbed on oral
and intraperitoneal administration and
subsequently metabolized to carbon
dioxide and unidentified metabolites in
urine. Species variation in the
metabolism of chloroform has been
summarized in Table 4. It is noteworthy
that the mouse, a species which shows
greater sensitivity to the oncogenic
effect of chloroform (Eschenbrenner &
Miller. 1945; Brown et al. 1974)
metabolized chloroform extensively to
carbon dioxide (80%) and unidentified
metabolites (3%) from an oral dose of 60
. mg/kg. Rats also metabolize chloroform
to carbon dioxide but to a lesser extent
(66%). In another report, Paul and
Rubinstein (1963) recovered 4 percent
carbon dioxide after administering 1484
mg/kg chloroform intradoudenally to
rats. The discrepancy in these two
results may be dose related.
Dose related differences in the
metabolism of compounds are known
and have recently been reported for the
carcinogen vinyl chloride. Squirrel
monkeys, when given 60 mg/kg of
chloroform orally, excreted 97 percent of
the dose, with 17 percent as carbon
dioxide and 78 percent as chloroform.
Fry, et al. (1972), recovered
unmetabolized chloroform ranging from
17.8-66.6 percent of a 500 mg dose of
chloroform given to human volunteers
during an 8 hour time period (equivalent
to about 7 mg/kg). Since the metabolism
of chemicals is also dependent on age
and sex, the widespread variation in the
quantitative disposition of chloroform in
human subjects may be due to the
experimental protocols wherein subjects
ranging from 18-50 years of age were
used Individual variability in the non-
homogenous human population is a
major factor.
Metabolic similarities between carbon
tetrachloride and chloroform include the
appearance of hahde ions in urine and
carbon dioxide in breath. A related
chemical, carbon tetrachloride, is a
common contaminant of the chlorine
used in water disinfection. Carbon
tetrachloride also is metabolized to
chloroform in trace amounts, which may
in turn, be biotransformed to carbon
dioxide. Both chloroform and carbon
tetrachloride are proven animal
carcinogens (see below). However, this
is mentioned because of possible
metabolic production of proximal
carcinogens. Toxicity of carbon
tetrachloride, however, has been
attributed to a free radical (CCI3) which
is postulated as a metabolic
intermediate. Chloroform appears to be
metabolized to form phosgene (Krishna,
1979).
Table 4.—Disposition of Chloroform—Species Variation
Metabolism (percent)
Animal species
Mouse
Rat
Rat
Rat
Monkey s
Sex Strain Dose
mg/kg
M CBA CF/ 60 po -
LPC57
M Sprague 60 po i
Dawtey
1.484 id
M Sprague 4,710 ip
Dawley
M Squirrel 60 po .
CHCL,
6
20
70
78
CO,
60
66
039
17
Urine Total
reces excretion
3 '93 Browne/ a/ (1974).
7 93 Brown el al (1974)
Paul & Hubstem
(1963)
2 97 Browne/ a/ ( 1974)
314
100
1 Includes radioactivity in carcas
Po Orally
id intradeudenelry
ip - intraperitoneal
Many carcinogens have been reported
to form complexes with proteins, DNA
and RNA (Miller & Miller, 1966). In the
case of chloroform, llelt et al., (1373)
reported covalent bonding of chloroform
metabolite(s) to tissue macromolecules
in mice. The covalent bonding increased
or decreased when the animals were
pretreated with phenobarbitai or
piperonyl butoxide, agents which
stimulate or inhibit the metabolism of
foreign compounds by mixed function
-------
68696 Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations
oxidase enzymes. This is suggestive of
the involvement of chloroform
metabolism in these processes. These
results may be interpreted to mean that
the potency of an ingested chemical will
be dependent upon its rate of
metabolism to the active form.
Information regarding the metabolism
of bromoform and other haloforms is not
available. However, the structural
similarities of these haloforms with
chloroform indicate that they should
also be absorbed by the oral and
inhalation routes of exposure and then
metabolized into carbon dioxide.and
halide ions. Related halogenated
^hydrocarbons of the dihalomethane
\erjes (e.g.. dichloromethane,
dibromomethane and
bromochloromethane) have been
reported (Kubic et al. 1974) to be
metabolized to carbon monoxide; the
rate of metabolism of dibromomethane
was higher than that of the
dichloromethane.
VI. Acute and Chronic Health Effects in
Animals
Mammalian responses to chloroform
include effects on: the central nervous
system, hepatotoxicity, nephrotoxicity,
teratogenicity, and carcinogenicity.
Reported oral LD50 values are as
follows: for rats. 300 mg/kg (DHEW.
1978); and for mice, 705 mg/kg (Plaa, et
al.. 1958).
Jones, et al. (1958), reported the effect
of various oral doses of chloroform on
mice 72 hours after exposure:
35 mg/kg—threshold hepatotoxic effect-
minimal midzonal fatty changes
70 mg/kg—minimal hepatic central fatly
infiltration
140 mg/kg—massive hepatic fatty infiltration
350 mg/kg—hepatic centrilobular necrosis
1.100 mg/kg—minimum lethal dose
Acute effects of exposure to
chloroform and bromoform vary among
species. Reported lethal doses for
chloroform and bromoform are:
Subcutaneous
lethal doso
Values in mg/kg
Mouse ...
Rabbet
_. .. ID* .
.. LDg.. .. ..
.. 704 (Chloroform)
1820 (Bromolorm)
800 (Chloroform)
4 10 (Bromolorm.
Data on the acute toxicity of
dibromochloromethane and
dichlorobromomethane are not
available.
A. Hepatotoxicity
Plaa, et al. (1968) established a dose-
response relationship in mice, measuring
parameters indicative of hepatotoxicity.
Median effective dose (ED») values of
1.4 mM/kg (166 mg/kg) were found in
mice exposed to chloroform by
subcutaneous injection. The inhalation
exposure of chloroform by mice for 4
hours at concentrations ranging from
100-800 ppm resulted in fatty infiltration
of the liver at all dose levels. These
changes were observed at necropsy 1-3
days after exposure.
Like chloroform, bromoform exposure
leads to fatty degeneration and
centrilobular necrosis of the liver (von
Oettingen. 1950).
Dibromochloromethane and
dichlorobromomethane may bring about
similar responses, although no
experiments have been reported.
B. Nephrotoxicity
Nephrotoxic effects of chloroform
were studied by Plaa and Larson (1965).
The EDSO for orally administered
chloroform in mice was 178 mg/kg as
measured by phenolsulfo-phthalein
excretion. Increases in urinary protein
and glucose excretion, indices of kidney
damage, indicated an ED50 of 104 mg/kg
chloroform. Data concerning the
nephrotoxic effect of other THMs are
not available.
C. Centra/ Nervous System Effects
Chloroform was used extensively as
an anesthetic because of its effect on the
central nervous system. Lehmann and
Hasegawa (1910) reported dizziness and
light intoxication during 20-minute
exposures to chloroform concentrations
of 4300-5100 ppm. Repeated exposures
up to six days to concentrations as low
as 920 ppm for 7 minutes resulted in
symptoms of central nervous system
depression (Lehman & Schmidt-Kehn,
1936). Additional important information
has been submitted to EPA and is
discussed below.
Effects of acute and subchronic
chloroform exposure on cholinergic
parameters in mouse brain were studied
by Vocci, et al., (1977). Male Swiss
Webster ICR mice were gavaged with
single doses of chloroform (30 and 300
mg/kg) and sacrificed 15 minutes after
administration of chloroform. In another
experiment, the mice were gavaged with
14 or 90 daily doses of chloroform (3 or
30 mg/kg) and sacrificed 18 hours after
the last administration. Neither of the.
above dosage regimens had any effect
on in vitro [SH] choline uptake in
synaptosomes. In another study (ibid) of
biosynthesis of acetylcholine in mouse
brain, chloroform (30 mg/kg)
significantly decreased the [3H]
acetylcholine synthesis (57% of control).
Administration of chloroform (3 mg/kg)
for 14 days produced a reduction in [9H]
acetylcholine (57% of control) (Vocci,
Personal Communication, April 1979).
Chloroform, dichlorobromomethane,
chlorodibromomethane and bromoform.
at concentrations of 8X10~"M did not
alter the update of norepinephrine or
dopamine into brain synaptosomes in
vitro (Vocci, Personal Communication.
April 1979).
D. Teratogenicity
Teratogenic responses to oral dosing
of animals with chloroform were
investigated. Rats and rabbits were
administered chloroform at 126 and 50
mg/kg respectively. No significant fetal
deformities were observed (Thompson
et al. 1973). Inhalation of chloroform by
Sprague Dawley rats at 30,100 and 300
ppm for 7 hours a day, on days 6 through
15 of gestation revealed significant fetal
abnormalities including: acaudia,
imperforate anus, subcutaneous edema.
missing ribs and delayed skull
ossification (Schwetz et al. 1974).
In an attempt to explain reproductive
failure in laboratory animals, i.e.. mice
and rabbits. McKinney et al. (1976)
conducted a study using CD-I mice
wherein groups of mice were given tap
water and purified tap water (passed
through a Corning 3508 ORC and a
Corning 3508 B demineralizer).
respectively. Analysis indicated reduced
amounts of chlorinated compounds in
the purified water. The study could not
relate chloroform and other chlorinated]
organics in tap water to reproductive
failures in laboratory animals, since the
concentrations of chlorinated organics
in water were lowest in those months
that reproductive failure was highest,
although there did appear lo be small,
non-significant differences in this
parameter between the highly purified
and tap water. In a revaluation
involving the effect of Durham tap water
and purified tap water as in the above
study, Chernoff (1977) did not find
striking differences in the reproductive
success of CD-I mice. No teratogcnic
studies on haloforms other than
chloroform were available.
E. Mutagenicity
The THMs (chloroform.
bromodichloromethane.
dibromochloromethane,
dibromochlorometha'ne and bromoform)
were assayed in vitro for mutagenic
activity using strains of Salmonella
typhimurium (TA 100 & TA1535). The
assays were conducted in desiccators to
allow each compound to volatilize so
that only the vapor phase came in
contact with bacteria on the petri
dishes. The activation system was
. tested and found not to be required for
the bromohalomethanes since they wei
positive in the absence of activation.
The results'obtained were as follows: (a)
-------
Federal Register / Vol. 44, No. 231 / Thursday. November 29, 1979 / Rules and Regulations 68697
Chloroform was not mutagenic in TA
00 with or without activation, nor in TA
535 without activation; (b)
romodichloromethane was mutagenic
i TA 100 without activation, with a
doubling dose of approximately 25
microliters; (c) dibromochloromethane
was mutagenic in TA 100 without
metabolic activation, with a doubling
dose of approximately 3.5 microliters;
(d) bromoform was mutagenic in TA 100
without metabolic activation, with a
doubling dose of approximately 25
microliters, and was also mutagenic in
TA 1535 with metabolic activation, with
a doubling dose of approximately 100
microliters (Tardiff, 1976). All three
compounds demonstrating mutagenic
activity did so in a dose-response mode.
For certain classes of compounds,
except for many chlorinated
hydrocarbons (Ames, 1973) the Ames
test which utilizes Salmonella
typhimurium bacteria correlates highly
(90 percent) with the in vivo
carcmogenicity bioassay.
F. Carcinogenicity
Prolonged administration of
chloroform at relatively high dose levels
to animals, specifically mice and rats,
manifested oncogenic effects. The
nvestigation conducted by
ischenbrenner and Miller (1945)
iroduced hepatomas in female mice
'(strain A) given repeated dosages
ranging from 0145 to 2.32 mg of
chloroform for a period of only four
months. Minimum doses of 593 mg/kg
chloroform per day (total of 30 doses)
produced tumors in all of the surviving
animals.
In a recent bioassay (NCI, 1976)
linking chloroform with oncogemcity,
rats and mice of both sexes were fed
doses of chloroform ranging from 90 to
200 (rats], and 138-477 (mice) mg/kg. In
this study, the lowest dose for observed
carcinogenic effect (kidney epithelial
tumors) in male rats was 100 mg/kg and
for mice 138 mg/kg administered to the
animals for a total period of 78 weeks. A
related halogenated hydrocarbon,
carbon tetrachloride, was carcinogenic
in Osborne Mendel rats and in B6C3F1
mice at dosages ranging from 57 to 160
mg/kg and 1250 to 2500 mg/kg,
respectively. The incidence of
hepatocellular tumors formed in these
animals at both dose levels almost
approached one hundred percent (Table
5). The percent survival in mice treated
with chloroform and carbon
tetrachloride is depicted in Table 8.
Mmost all the animals on treatment
rith carbon tetrachloride died between
which was toothpaste. The only finding
of neoplasia was an excess of tumors of
the renal cortex in the male JCl-Swiss
mice at a dose level of 60 mg/kg/day.
However, animals fed 17 mg/kg/day of
chloroform showed no incidence of
renal carcinoma.
Table 5.—Comparison ol Hepatocellular Carcinoma
Incidence in Chloroform and Carbon Telrachlonda-
Treated Mice
91-92 weeks whereas with chloroform
treatment at both dose levels, 73 and 46
percent of the animals survived.
Miklashevskii et al. (1966) fed
chloroform to rats at 0.4 mg/kg
apparently for 5 months and detected no
histopathological abnormalities after
this treatment. A recent study on the
carcinogenic effect of chloroform at dose
levels of 17 mg/kg/day and 60 mg/kg/
day was conducted by Roe (1976),
utilizing the rat (Sprague-Dawley), the
beagle dog and four strains of mice (ICC
Swiss. C57B1, CVA and CF/1).
Comparison with the NCI study (1976)
indicates that the number of animals
and the duration of the experiment were
essentially similar; the major differences
were the dosages, which were lower
than in the NCI study, and the vehicle.
Table 6.—Comparison of Survival of Chloroform and Carbon Tetrachloride- Treated Mice
Animal group
^Carbon
Chloroform tetra*
chlonde
Males
Controls
Low Dose
H.gh Dose
Females
Controls
Low Dose
HghDosa
5/77
18/50
44/45
1/80
36/45
39/41
5/77
49/49
47/48
1/80
40/40
43/45
Animal group
Chloroform
Caitaon letrachlondo
Initial NO
78 weeks 90 weeks IniiaJNo 78 weeds 91-92»eeks
Males
Controls
Low Dose
High Dose
Females
Controls
Low Dose
High Dose
77
50
50
80
50
50
53
43
41
71
43
36
38
37
35
65
36
11
77
50
50
80
SO
50
S3
11
2
71
10
4
38
0
0
65
0
1
Some renal tumors were also seen in
control animals in a later study. The
negative results observed in the dog
experiment may be explained on the
basis that either the animals were not
exposed for a suitable length of time (i e.
duration of life span) or that an
insufficient number of animals were
tested, or that this species may not have
been responsive to the oncogenic effect
of chloroform. The negative results of
the rat study may be explained on the
basis of lack of strain sensitivity. Based
on the extrapolation from the NCI study.
the dose was too low to produce an
effect in so few animate (Cueto, NCI,
1979).
Much less information is available on
the carcmogenicity of
bromohalomethanes. Preliminary results
from the strain A mouse pulmonary
tumor induction technique (Theiss et al,
1977] indicated that bromoform
produced a positive pulmonary
adenoma response while chloroform did
not. Other studies (Poirier, et al.. 1975)
indicated that in several instances
brominated compounds exhibited more
carcinogenic activity than their
chlorinated analogs in the pulmonary
adenoma bioassay.
VII. Human Health Effects
A. NAS Principles of Toxicological
Evaluation
The recent NAS (1977) report entitled
"Drinking Water and Health" identified
several principles for assessing the
irreversible human effects of long and
continued low dose exposure to
carcinogenic substances.
Principle 1- Effects in animals,
properly qualified, are applicable to
man.
Principle 2 Methods do not now exist
to establish a threshold for long-term
effects of toxic agents.
Principle 3: The exposure of
experimental animals to toxic agents in
high doses is a necessary and valid
method of discovering possible
carcinogenic hazards in man.
Principle 4. Materials should be
assessed in terms of human risk, rather
than as "safe" or "unsafe".
On the basis of studies in animals and
human toxicological data the NAS (1977)
has recommended that strict criteria
should be applied for establishing
exposure limits to chloroform.
The National Institute for
Occupational Safety and Health has
recommended that the occupational
exposure to chloroform should not
-------
68698 Federal Register / Vol. 44, No. 231 / Thursday. November 29. 1979 / Rules and Regulations
exceed 2 ppm determined as time-
weighted average exposure for up to a
10 hour work day.
The human health effects as observed
in accidental, habitual, and occupational
exposures appear to indicate that the
effects produced by exposure to
chloroform are similar to those found in
experimental animals. These include
effects on the central nervous system,
liver, and kidney.
The symptoms observed (Storms,
1973] in a 14 year old patient following
an accidental exposure to an unknown
amount of chloroform included cyanosis,
difficulty in breathing and
unconsciousness. Liver function tests
measured by serum enzyme levels four
days after ingestion indicated high
levels of SCOT. SGJ*T. and LDH. The
authors also noted damage to the
cerebellum characterized by an
instability of gait and a slight tremor on
finger-to-nose testing. The symptoms
disappeared in two weeks.
Several cases of habitual chloroform
use have also been recorded by
Heilbrunn et al. (1945). A case study of
interest was a 33 year old male who had
habitually inhaled chloroform for 12
years. The subject showed psychiatric
and neurological symptoms including
restlessness, hallucinations,
convulsions, dysarthria, ataxia, and
tremors of the tongue and fingers.
Lunt (1953) reported that delayed
chloroform poisoning in obstetric
patients, anaesthetized with chloroform
is characterized by renal dysfunction as
indicated by: Albumin, red blood cells,
and pus in the urine. Chloroform
exposure of humans by inhalation was
studied by Lehman and Schmidt-Kehl
(1936). Ten different concentrations of
chloroform were used and the
chloroform concentrations were
determined by the alkaline hydrolysis
method. Exposure at concentrations of 7
ppm for 7 minutes and at all higher
levels up to 3000 ppm caused symptoms
of central nervous system depression.
Desalva et al. (1975) studied the
effects of chloroform in humans; the
subjects were given dentifrice
containing 3.4% chloroform and
mouthwash with 0.43% chloroform for 1
to 5 years. No hepatotoxic effects were
observed at estimated daily ingestion of
0.3 to 0.96 mg/kg chloroform. Reversible
hepatotoxic effects were manifested at
23 to 27 mg/kg/day chloroform ingested
for 10 years in a study conducted by
Wallace (1959).
B. Epidemiologic Studies
By August 1979,18 epidemiological
studies, and additional unpublished
reports discussed possible relationships
between cancer mortality and morbidity
and drinking water supplies. The results
of the studies are shown in Table 7 in
the approximate chronological order of
completion. The table shows the
statistically significant results of
analysis by anatomical site. The
statistically significant positive results
are denoted by "M" for males and "F" to
females and the statistically significant
negative results are denoted by "—"
before the "M" or "F".
BILLING CODE 6560-01-M
-------
Table 7. Stalls Significant Results of Gpldenlologlcal
Cancer Studies on Drinking Water and Cancer Sites
No. . Author
\^
.l\ P»J*. H.
\
i Vr.
:. 74
76
2 Tarone 75
3 VAailenko 75
I)
5
6
7
B
9
10
It.
12
13
14
15
16
17
18
Oetouen
Harris/
Relckes
Buncker
i
Kuima
i
HcCate
Cantor .
i
Hogan| .
M, i
Krvse ;
75
76
75
77
. 75
78
79
• 77
77
i
gaY/'.J'i 77
|
Alavanja
Wllklns
77
78
78
Rafferty, 79
Brenntoan 78
Tuthllll 78
Tuthlll* 79
1
1
(to
-H
(it
(ll<
,
1C
Isophagus 1
•
-
H
al c
-H
H
idled
it'll
H
mpli
H
1
I
F
F
H
F
N
F
H
F
ncer
-F
-N
-F
-F
live
tes i
H
f
ted
H,F
i
•J t*
r-t t
(st
?•
and
H
F
H
and
Ludii
H
but
F
i
dlec
H
otal
F
H
F
ktdn
1 ••
H
•esu
Contained G.I. )'
M
F
J
ret
H
mort
N
y si
I0~»t
H
F
ts
P
liver and 1
tallblaoder \
ales
-N
F
lily
-H
es o
itlst
F
Ot 1
F
u
e
a
onl
H
F
F
only
-N
ily)
ICtV
seal
Respiratory (not
lung) '
r)
'
r*
-
le)
1
Lung, trachae.
bronchi 1
H
F
.
F
H
IIP
H
(Fen
-F
1
F
F
F
Isii
alee
K
3
.
nil
Onl
t
I
F
mf/i
J
•a
*i
,
H
"
BOfln
Bladder
H
H
N
F
H
H
F
H
f
H
»Mc
M
H
r
•o
sncTbi
.
Combined
Urinary
H
•
H
tra
H
F
(1
F
u
V
-M
F
ilys
Ivei
Leukemia
s pn
• kl
S
blem
".2SJT
Hodgktns |
H
F
an
1
-H
-F
I bli
e
H
ddei
K
M
F
-OOJ
.5-
M
H
F
F
M
F
H
F
M
F
H
H
F
r)
--
Of
3
9
E
E
E
E
E
E
E
E
E
E
E
C-C
E
C-C
C-C
E
c=c
E
1
L
•
I
n
B
S"
P
i
Q.
S3
o
re
o-
(D
i
1-1
53
CO
70
5"
en
01
O.
I
E"
r+
O
at
C • Case-Control
R • Retrospective
BILUNQ CODE 6SCO-01-C
-------
68700 Federal Register / Vol. 44, No. 231 / Thursday. November 29. 1979 / Rules and Regulations
Five of the studies were published
through August 1979. All of the studies
were retrospective in design; sixteen
were correlation studies, and four used
a case-control approach. Four studies
utilized cancer morbidity or incidence
rather than mortality as a measure of
disease frequency. The studies vary in
sample size, cancer sites considered,
factors selected as possible explanatory
variables, parameters selected as
indicators of water quality, and in the
statistical techniques used for analysis,
so caution must be used in comparing
the results of one study with the results
of another study.
There are several problems which
make the results difficult to interpret: (1)
There is limited water quality data on
organics and other contaminants in the
finished drinking water, and the data
which exist cover less than five years;
and (2) the water quality data are often
from geographic areas other than those
(usually counties) reporting cancer
mortality data.
The water quality data are recent, and
it is not known to what extent they
reflect past exposure to THMs. This is
important, since the latent period for
most types of cancer is measured in
decades. Comparison of the various
study results is difficult also because of
the different approaches used.
In general, retrospective
epidemiological studies are a useful
methodological tool in hypothesis
generation. The results from these
studies, when viewed collectively, can
provide some insight into the postulation
of causal relationships which then need
to be tested further, using
epidemiological designs such as case-
control or cohort studies, for
documentation.
When the evidence from all studies is
weighed, an emphasis can be placed not
only on the statistical significance of
single correlation coefficients but on
their consistency and patterns. When
more than one independent study shows
positive associations for site-specific
cancers, then the association may not be
due to chance alone. When the
association is verified by consistent
results across all four sex-race groups
(white male, non-white male, white
female, non-white female), the
association is more likely to be used due
to the variable considered and the
evidence should be viewed more
seriously. The studies done so far
suggest the appropriateness of concern.
There is much evidence (both
epidemiological and experimental) that
most human cancers result from a
combination of causes (Weisburger,
1977). Etiologic factors (e.g. smoking as a
cause of lung cancer, soot as a cause of
scrotal cancer in chimney sweeps) that
result in increased relative risk greater
than 5, were among the first lo be
discovered. The etiologic factors
associated with cancers of
gastrointestinal and urinary tract arc
more difficult to isolate from
epidemiological studies because of the
lower incidence and mortality rates, the
interaction of environmental causes, and
site-specific differences. The increased
relative risk of populations exposed to
most factors suspected of being
associated with gastrointestinal and
urinary cancers are less than three.
Effects as small as, or smaller than
these, are difficult to detect or quantify.
A number of the epidemiologic studies
relating "water quality" to cancer did
not define the water quality parameter
by chemical constituents but instead
compared cancers in persons who used
water from different sources. Among the
first of these was an investigation by
Page. Talbot. and Harris (1974). The
study considered Louisiana parish
(county) cancer mortality rates for 1950-
69, for total cancers and various
selected cancer sites, and related these
to the percentage of the parish
populations drinking water from the
Mississippi River, which is known to be
contaminated by many organic
chemicals (Laseter, 1972). The variables
controlled were the rural-urban
character of the parish, median income,
population density, and proportion of
population employed in the petroleum,
chemical, and mining industries. An
unweighted regression analysis showed
a positive correlation between drinking
water and total cancer (excluding
cancer of the lung, urinary tract, GI
tract, and liver), and then separately for
cancer of the gastrointestinal organs and
lung cancer. These investigations
suggested an association between
cancer mortality rates and use of
drinking water from the Mississippi.
Meinhardt, et al. (1975), commenting
on the Page-Harris report, looked at the
cancer mortality gradient by apparent
"dose" of river water and concluded
that there was a random distribution of
high and low cancer mortality rates
among the river water consumers along
the lengths of the Missouri and
Mississippi River systems.
Subsequent reports by Page and
Harris (1975.1976) on the "Relation
Between Cancer Mortality and Drinking
Water in Louisiana" utilized
explanatory variables and cancer sites
similar to those in the first study;
relationships for all four sex-race groups
were considered. Positive regression
coefficients for the water variable that
were found statistically significant were:
Total cancer sites: WM. NWM. NWF.
All other than lung: WM.
Urinary Tract. WM. NWF.
Gastrointestinal. WM, NWM. WK. NWF.
Tarone and Gart (1975) reviewed the
Page-Harris work and included an
additional variable, elevation above sea
level. By using a weighted regression
analysis for four race-sex groups,
statistically significant, positive
correlations were found between the
water variable and total cancer and lung
cancer mortality for white males (WM),
non-white males (NWM), and non-white
females (NWF). The correlations were
not statistically significant for white
females (WF) for the same sites. Thus,
there was a lack of consistency across
the four sex-race groups for the
aforementioned cancer sites.
Vasilenko and Magno (1975)
conducted an ecological study in New
Jersey and determined the relation
between water source and age-adjusted
cancer mortality from lung, stomach and
urinary tract cancer of white females.
Water quality was estimated from the
ratio of the number of households
served by public systems and private
water companies to the number served
by individual wells. Positive
associations were found for lung and
stomach cancer.
DeRouen and Diem (1975) also
reviewed the relationship of cancer
mortality in Louisiana and the
Mississippi River as the drinking water
source looking at ethnic variables as a
possible confounding factor. By dividing
Louisiana into a northern and southern
section, they were able to mimic an
ethnic division of the population. Many
of the variables (urban-rural
characteristics, median income,
employment characteristics, and
elevation above sea level) included in
the previous studies were omitted. The
water variable was handled differently
by the investigators. Population groups
were dichotomized into those who
obtained none of the water from the
Mississippi River, and those who
obtained some or all from the river. The
results show a positive relationship
between cancer mortality and drinking
water, for gastrointestinal cancer. The
cancer mortality rates for southern
parishes of Louisiana whose source of
drinking water is the Mississippi River
are higher than in the southern parishes
whose source of drinking water is not
the Mississippi River for the following:
Stomach- NWF.
Rectum: WM.
Large Intestine: WF, NWF.
Cervix' NWF.
Lung- NWF.
Total Cancer: NWF.
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29. 1979 / Rules and Regulations 68701
The cancer mortality rates tend to be
higher for the southern parishes with
/er water use than northern for river
liter parishes for cancer of the urinary
act, gastrointestinal tract, and the lung.
In another set of analyses and
comments, DeRouen and Diem (1975)
discuss the problems associated with
interpretation of regression coefficients
as they relate to the Page and Harris
Report, particularly the problem of
making interferences from correlational
studies. They concluded that
inconsistencies such as the failure to see
the same relationships for all sex-race
groups reduces the credibility of the
hypothesis of a causal relationship
between water source and cancer risk.
An analysis was done by McCabe
(1975) of EPA using the 50 (of a total of
80) NORS cities with a 1950 population
greater than 25,000 and 70 percent or
more of the city's population receiving
water comparable to that sampled by
EPA. McCabe showed a statistically
significant correlation between the
chloroform concentrations in the
drinking water and the cancer mortality
rate by city for all cancers combined.
In a second analysis by McCabe using
water quality data from Region V,
correlations between chloroform and
TTHMs and total cancer mortality were
>t positive. When the same
^relations were done using Region V
us NORS data for chloroform and total
trihalogenated methane concentration
levels, a positive statistically significant
result was obtained.
Several epidemiological studies have
been conducted in the Ohio River area.
Buncher (1975) conducted a study of 88
counties (in Ohio, bordering the Ohio
River) of which 14 used the Ohio River
as a drinking water source. Buncher
reports no significant relationship with
drinking water from the Ohio River and
the higher cancer mortality rates. There
was a weak positive correlation
between the chloroform concentration in
23 cities and the cancer mortality rate
for all cancer sites in white males.
Similar results were found in 77 cities
(59 with surface water supplies)
between chloroform concentrations and
pancreatic cancer mortality in white
females. For cities that accounted for
more than 70 percent of the county
population, there was a significant
correlation between chloroform
concentration and bladder cancer
mortality rates for both white males and
white females.
As a follow up on the Buncher study,
a study by Kuzma, et al. (1977),
insidered the 88 Ohio counties,
assified as either ground water or
irface water counties based on the
oource of the drinking water used by a
majority of the county residents. A two-
stage analysis was performed and no
statistically significant results were
shown between the drinking water from
the Ohio River and cancer mortality
rates. However, rates for stomach,
bladder, and total cancers were higher
for white males in counties served by
surface water supplies (probably
chlorinated) than in counties served by
ground water supplies (probably not
chlorinated).
Reiches, et al. (1976), re-examined the
Ohio data using a different
methodology. Correlations between the
surface drinking water variable and
cancer mortality rates for stomach
cancer and total cancers for both white
males and females were statistically
significant. The correlations between
the drinking water variable and cancer
mortality rates of the pancreas, bladder,
esophagus, gastrointestinal tract, and
urinary organs were significant for white
males only.
Although several studies defined the
water quality parameter by chlorination
or levels of chloroform, only one study
has considered the relationships of
cancer with all THMs, both collectively
and separately. Cantor et al. (1978)
studied the correlation of cancer
mortality at sixteen anatomical sites
with the presence of concentration
levels for each THM and TTHM in
drinking water for whites. Counties
were grouped according to the
percentage of the county population
served by the sampled water supply. In
both sexes, there was a positive dose-
response gradient of increasing
correlation between trihalomethane
concentration and bladder cancer. The
correlation was stronger for bromoform
than with chloroform. There was a
negative correlation in white females of
stomach cancer with total THM levels.
Kidney cancer in white males showed a
positive correlation with chloroform
levels. Lung cancer in white females
showed a positive correlation with THM
levels. Among white majes non-
Hodgkins* lymphoma showed a positive
correlation with bromoform. A positive
dose-response was observed between
brain cancer mortality (in both sexes)
with increasing use of water containing
chloroform, but the associations were
not strong.
Alavanja, et al. (1976) conducted a
retrospective, case-control study of
female cancer mortality and its
relationship to drinking water
chlorination in seven selected New York
counties. A statistically significant
association was found between a region
being served from a chlorinated drinking
water supply and combined
gastrointestinal and urinary tract cancer
mortality rates in that region. There was
also a higher mortality for the summed
gastrointestinal and urinary cancer in
urban areas served by chlorinated
surface or ground drinking water
supplies than in urban areas served by
nonchlorinated supplies, however, the
results should be viewed cautiously due
to the small numbers in the sample.
Alavanja (1977) expanded this study
and included gastrointestinal and
urinary cancer deaths. Results showed
that males living in the chlorinated
water areas of three counties and
females living in the chlonnated water
areas of two counties were at greater
risk of gastrointestinal and urinary tract
cancer mortality than individuals living
in the non-chlorinated areas. Alavanja
(1978) did a second study (shown on
Table 7), which expanded the first to
nineteen counties in New York and
several specific cancer sites.
Statistically significant positive
associations were found for males and
lung cancer and for females and
pancreatic cancer. Statistically
significant positive associations were
found for both males and females and
cancer of the large intestine, combined
gastrointestinal, and all cancers.
Kruse (1977) conducted a
retrospective, case control study of
white males and females in Washington
County, Maryland. The relationship
between mortality and morbidity from
liver (including biliary passages) and
kidney cancer in areas supplied by
chlorinated public water supplies was
analyzed. While there was a higher
incidence of liver cancer among the
exposed group; i.e., the group which
consumed chlorinated drinking water,
the correlations were not statistically
significant. It should be noted that the
sample size was small and that fewer
than 50 cases each of liver cancer and
kidney cancer were counted.
Salg (1977) also conducted a
retrospective study of various cancer
mortality rates and drinking water from
a variety of sources and receiving
different types of treatment in 346
counties in seven states in the Ohio
River Valley Basin. She compared
mortality rates for white and non-white
males and females using weighted
regression analyses, surface water usage
showed weak but statistically
significant associations between
chlorinated water supplies (regardless of
source) and the following cancers: For
white males—esophagus, respiratory
organs, large intestine, rectum, bladder,
other urinary organs and
lymphosarcoma and reticulosarcoma;
for white females—breast and rectum,
-------
68702 Federal Register / Vol. 44, No. 231 / Thursday, November 29, 1979 / Rules and Regulations
and for non-white females—esophagus
and larynx. Rectal cancer showed
positive correlations across all race-sex
groups. It should be noted that the test
of significance utilized for this study
was p < 0.10, which is less stringent
than that used in other studies.
Mah, et al. (1977), conducted a
retrospective study of the white
population in the Los Angeles County
area of the relationship between cancer
mortality and morbidity and the
chlorinated drinking water supply. They
did not reveal any trends and showed
no significant relationships for either
cancer mortality or morbidity. The
authors pointed out several
methodological problems, including the
diluting effect of migration into the area
covered by this study.
Hogan, et al. (1979) also utilized the
NORS and Region V data sets and
applied various statistical procedures to
the data in order to determine the
effects of using different statistical
models. Their results were similar to
previous studies showing a positive
correlation between rectal-intestinal
and bladder cancer mortality rates and
chloroform levels in drinking water
when weighted regression analysis were
applied. However, as the authors
pointed out, "the marked extent to
which these results were dependent on
(1) the weighting scheme adopted in the
analysis, (2) the presumed
appropriateness of the data, and (3) the
characteristics of the statistical model,
was also clearly illustrated."
Wilkins (1978) conducted a case-
control study in Washington County,
Maryland and investigated the
association between liver, kidney and
bladder cancer and chlorinated water
source. A positive correlation was found
for female liver cancer and male bladder
cancer and the chlorinated drinking
water source. Due to small numbers of
cases the outcome of this study should
be viewed with suspicion.
Eafferty (1979) studied associations
between drinking water quality in North
Carolina communities and cancer
mortality rates. The drinking water
supplies were characterized by domestic
and/or industrial contribution. No
significant positive association were
found.
Tuthill and Moore (1978) investigated
the association between cancer
mortality rates and parameters of water
quality for Massachusetts community
public water supplies. The average
annual chlorine dose was one of the
independent water characteristics.
Simple correlations showed that the
average chlorine dose level in the water
was negatively associated with female
buccal cancer, and positively associated
with female esophageal and male
respiratory cancers. Occupation,
population mobility, and other
demographic variables were controlled.
In summary, many but not all of the
studies have found positive correlations
between some characteristics of
drinking water and various cancer
mortality/morbidity rates. However,
these correlations are dependent upon
the selection and appropriateness of the
data, the weighting scheme and
extrapolation in the analysis, and the
characteristics of the statistical model.
Because of these dependencies the
quantitative, causal interpretation of
results generated from an indirect or
ecological study should be viewed as
tenuous for the primary purpose of
generating hypotheses and even
questionable in most cases.
It is important in the evaluation
process to consider the results from
other epidemiological studies as they
develop hypothesis of potential causal
associations between cancer mortality
and other agents. For example, the
confounding factors of diet, occupation,
and smoking all have been suggested as
potential causative agents of bladder
cancer, Cole (1972). Therefore, any
epidemiological study that investigates
the possible association between
bladder cancer and drinking water
should be designed to avoid the
problems that result in confounding of
the data. None of the studies completed
thus far have obtained data on or
controlled for diet; several studies have
attempted to control for occupational
exposure (Page and Harris. 1974 and
1975; Cantor, et al., 1978: Tuthill and
Moore, 1978); only the studies by Kruse
(1977) and Wilkins (1978) obtained
smoking data. Only a few studies
considered four sex-race groups (the
number of non-whites is too small in
some of the geographic areas) and of
those studies only a few showed
consistent patterns of association of
specific cancer sites, e.g.. Salg (1977)-
rectum. Several studies which
considered only white populations
found positive correlation coefficients
for both sexes: Kuzma (1977)—stomach;
De Rouen (1975)— intestine, stomach
and bladder Buncher (1975)—bladder.
Reiches (1976)—stomach; Cantor
(1978)—bladder Hogan (1979)—intestine
and bladder, and Alavanja (1978)—
intestine. Only a few studies defined the
water quality variable by the chloroform
concentrations (McCabe, 1975; Buncher,
1975; Cantor et al., 1977; Hogan et al.,
1977; Alvanja, 1978), and by the THM
concentrations (Cantor et al. 1977).
Of particular interest are possible
correlations of liver and kidney cancer
rates with drinking water, since the
animal exposure data indicate that
hepatocellular carcinomas and hepatici
modular hyperplasias have been
observed in B6C3F1 strains of mice aftO
life time exposure to chloroform. Several
of the preliminary studies grouped the
cancer sites for the anatomical systems,
e.g., gastrointestinal and urinary organs,
in order to increase the sample size. One
of the studies (Cantor, 1978) which
considered site-specific cancer mortality
showed a positive association between
drinking water and cancer of the
kidneys in white males. The absence of
any positive association between
drinking water and liver cancer
mortality may be due in part to small
sample sizes, very low incidence of the
disease, or because the exposure levels
of contaminants in trace amounts over a
lifetime may be below a no-effect level
(Weisburger. 1977). The incremental
increase may be too small to measure
for statistical significance. On the other
hand, many scientists believe that the
specific site in which cancer appears in
animal tests need not necessarily be the
same site in which the cancer is likely to
appear in humans.
Thus, the evidence is incomplete and
the trends and patterns of association
have not been fully developed. As
stated previously, a causal relationship
cannot be established by correlation
studies. When viewed collectively, the
epidemiological studies completed thus
far provide evidence for maintaining a
hypothesis that there may be a health
risk and that the positive correlations
may be due to an association between
some constituents of drinking water and
cancer mortality. The animal test data
alone provide a firm basis for policy
decisionmaking. Additional
epidemiological studies may provide
evidence regarding the strength of the
associations and the possibility of a
causal relationship between drinking
water and cancer mortality, and thus
provide a stronger basis for further
regulatory action.
The NAS Epidemiology Subcommittee
of the Safe Drinking Water Committee
reviewed the first thirteen of the
aforementioned eighteen studies. In the
report, "Epidemiological Studies of
Cancer Frequency and Certain Organic
Constituents of Drinking Water—A
Review of Recent Literature Published
and Unpublished," September 1978, the
Committee reached the following
conclusions, which are consistent with
EPA. Among the group of studies that
characterized water quality by actual
measurements, the results suggest:
TtiRt higher concentrations of THMs in
drinking water may be associated with an
-------
Federal Register / Vol. 44, No. 231 / Thursday, November 29. 1979 / Rules and Regulations 68703
increased frequency of cancer of the bladder.
The results do not establish causality, and
Ihe quantitative estimates of increased or
(lecreased risk are extremely crude. The
positive association found for bladder cancer
was small and had a large margin of error;
not only statistical, but much more
importantly, because of the very nature of the
studies.
Further research is being conducted
with more definitive analytical studies.
A large case-control bladder cancer
study with 3,000 cases and 6,000 controls
is being conducted by the National
Cancer Institute (NCI). Three other case-
control colon cancer studies are being
conducted in Louisiana, Pennsylvania,
and Utah. The results of these studies
may provide more solid evidence to
answer the question of possible
associations between water quality and
increased incidence of bladder and
colon cancer.
VIII. Mechanism of Toxicity
Biologic responses upon exposure of
mammals to chloroform include effects
on the central nervous system resulting
in narcosis, hepatotoxicity,
nephrotoxicity, teratogenicity and
carcinogenicity. Elucidation of the
mechanism of toxicity of chloroform and
related compounds has been attempted
by several researchers.
| Scholler (1968) and McLean (1970)
observed that phenobarbital
pretreatment of rats caused an increase
in Ifter necrosis after administration of
chloroform. Later, Brown, et al. (1974)
reported that exposure of rats to an
atmosphere containing chloroform (0.5%)
for 2 hour markedly decreases
glutathione (CSH) concentration in the
liver when the animals-have been
pretreated with phenobarbital. In an
attempt to further elucidate the role of
GSH in chloroform-induced
hepatotoxicity, Docks and Krishna
(1976) injected chloroform into rats
pretreated with microsomal enzyme
inducers-phenobarbital, 3-
methlcholanthrene, acetone and
isopropanal. A dose of chloroform as
little as 0.2 mg/kg decreased liver GSH
levels and caused centrilobular necrosis
within 24 hours in phenobarbital pre-
treated rats. At a dose of 0.05 ml/kg,
chloroform did not decrease liver GSH
or cause liver necrosis. When the rats
were not pretreated with phenobarbital,
a chloroform dose of 0.2 ml/kg caused
neither GSH depletion nor necrosis. In
this connection, it is interesting to note
that cysteine, which is a precursor of
1SH and a common amino acid in one's
liet, protected the liver from the
lepatotoxicity produced by chloroform.
'he animals were also protected from
he hepatotoxic effect by pretreatment
with cystamine, not a precursor of GSH,
thus suggestive of a mechanism other
than of GSH depletion in the
hepatotoxicity of CHC13.
Earlier reports by Ilett, et al. (1973)
suggested the possibility of another
mechanism involving the formation of
an active metabolite of chloroform
responsible for the chloroform-induced
hepatotoxicity. This study correlated the
renal and hepatic necrosis with covalent
binding of chloroform metabolites to
tissue macromolecule. Bioactivation of
xenobiotics including chloroform,
involves mixed function enzymes; the
NADPH cytochrome reductase-
cytochrome P-450 coupled systems.
Sipes, et al. (1972) studied-the
bioactivation of carbon tetrachloride,
chloroform and bromotrichloromelhane
utilizing 14c-labled compounds and rat
liver microsomes. The covalent binding
of radiolabel to microsomal protein was
used as a measure of conversion of the
compounds to reactive intermediates.
The authors concluded that cytochrome
P-450 is the site of bioactivation of these
three compounds rather than NADPH
cytochrome C reductase. CC14
bioactivation proceeds by cytochrome
P-450 dependent reductive pathways,
while CHCla activation, proceeds by
cytochrome P-450 dependent oxidative
pathways.
The isolation and identification of an
active metabolite of chloroform
supposedly responsible for toxicity was
attempted by Pohl and his co-workers
(1977). 2-oxithiazolidine-4-carboxylic-
acid, an in vitro metabolite of
chloroform, and presumably formed by
the reaction of cysteine and phosgene
(COCU), was isolated and characterized.
When the incubation was conducted in
an atmosphere of [18O] Os, the trapped
COCU contained [I8O]. These findings
suggest that C-H bond of CHC13 is
oxidized by a cytochrome P-450 mono-
oxygenase to produce trichloromethanol
which spontaneously
dehydrochlorinates to phosgene. The
electrophilic phosgene could react with
water to form carbon dioxide, a known
metabolite of CHCli in vitro and in vivo
or with microsomes to yield a covalently
bound product. The in vitro oxidation of
chloroform and its relationship to
chloroform toxicity has been further
substantiated by the studies wherein
deuterated chloroform was used. Pohl
and Krishna (1978) reported that CDC13
was metabolized slower than
chloroform suggesting that the cleavage
of C-H bond of chloroform is the rate
determining step in the enzymatic
process: The observation that CDC13 is
less hepatotoxic than CHC13 indicates
that the cleavage of the C-H bond is
also the critical step in the process
leading to CHC13 induced
hepatotoxicity. The finding that CDC13
depletes less glutathione in the liver of
rats than CHC13 suggests the active
metabolite phosgene is responsible for
the depletion of glutathione.
In the experiments involving the
isolation and characterization of «
metabolites of chloroform, the evidence
for the metabolism of chloroform to
phosgene in vitro, by the oxidative
pathway was present. Recent research
has indicated the possibility of
formation of phosgene in vivo. Pohl, et
al. (1979), isolated and characterized 2-
oxo-thiazohdine-4-carboxylic acid from
the liver of rats pretreated with cysteine
carboxylic acid after a dose of
chloroform and/or deuterated
chloroform. In these experiments,
deuterated chloroform yielded less
amount of metabolite, confirming once
again the specificity of the cytochrome
P-450 dependent enzymes in the
mediation of oxidative dehalogenation
of chloroform and its toxicity.
IX. Risk Assessment
The establishment of chloroform as an
animal carcinogen, plus the
epidemiological data and mutagenesis
data on THMs, show that a potential
human risk exists from the consumption
of THMs, but these data do not quantify
the risk. Methods have been developed
to estimate the level of risk, based on an
assumption that there is no threshold
level for the action of a carcinogen. The
state-of-the-art at the present time is
such that no experimental tools can
accurately define the absolute numbers
of excess cancer deaths attributable to
chloroform in drinking water. Due to the
biological variability and a number of
assumptions required, each of the risk-
estimating procedures leads to a
different value. There is wide variation
among these estimates and their
interpretation.
The EPA Science Advisory Board
(SAB) (1975). using the highest levels of
chloroform then reported in drinking
water by the NORS data (0.300 mg/l)
and assuming a maximum daily intake
of 4 liters of water for a 70 kg man,
attempted to estimate the risk. The
estimates were based on the
Eschenbrenner and Miller (1945) animal
data, which themselves are subject to
great variability since the experiments
used only 5 animals per sex per dose.
Using a linear extrapolation of the
animal data over more than 2 orders of
magnitude dose from mice to humans at
the 0.300 mg/l concentration level, the
lifetime incidence for liver tumors in
man were estimated to range from 0 to
.001 (95% confidence limits) or 0 to 100 r.
-------
68704 Federal Register / Vol. 44, No. 231 / Thursday. November 29. 1979 / Rules and Regulations
10 "5 in a lifetime. This rate may be
compared with the lifetime incidence of
260 x 10 *6 for malignancy of liver
derived from the data of the Third
National Cancer Survey (1976). This
estimate would range from zero to
approximately 40% of the observed
incidence of liver cancer in the United
States that may be attributable to
exposure to chloroform in drinking
water at the 0.300 mg/1 level. It should
be noted that this value is at the upper
limit of the confidence interval and the
linear non-threshold dose-effect model
allows an estimate of maximal risk
where a risk has actually been
observed. Most other models would
yield lower estimates. The SAB,
however, also stated that a more
reasonable assumption would yield
lower estimates of the risk.
Tardiff (1976) using four different
models, calculated the maximum risks
from chloroform ingestion via tap water
Using a margin of safety of 5000 applied
to the minimum effect animal dose, i.e.,
the Weil conjecture, the "safe" level
was calculated to be 0.2 mg/kg/day.
Using the logprobit model and the slope
recommended by Mantel and Bryan, the
conclusion reached was that at a
maximum daily dose of 0.01 mg/kg the
risk would be between 0.016 and 0.683
cancers per million exposed population
per year. Using the identical data, but
with the experimental slope of the dose
response curve as found in the mice as
opposed to the slope of the one in the
previous calculation, the conclusion
reached was that a maximum daily dose
of 0.01 mg/kg would produce less than
one tumor per billion population per
lifetime. Using the linear, or one hit
model, usually considered to be the
most conservative, a risk estimate of
between 0.42 and 0.84 cancers per
million population per year was
calculated to result from a dosage level
of 0.01 mg/kg/day. The two step model
produced an estimated maximum risk of
between 0.267 and 0.283 cancers per
million population per year at a dose
level of 0.01 mg/kg/day.
In the National Academy of Sciences
(1977) report on "Drinking Water and
Health," lifetime risks were estimated
from the more recent, and much more
extensive NCI animal data using a
multi-stage model.
For a concentration of chloroform at 1
ug/liter the estimated incremental
lifetime cancer risk would fall at
approximately 1.7 x 10 ~e per microgram
per liter at the upper 95% confidence
limit, assuming 70 year daily
consumption of water at that level.
Assuming lifetime exposure at the
standard of 0.10 mg/1 level in drinking
water the incremental risk would be 3.4
x 10 ~* assuming two liters of water at
0.10 mg/1 consumed daily for 70 years.
In evaluating the risk estimates, it is
important to compare the calculated
maximum risk with the current cancer
mortality data. Both liver and Kidney
cancer are rare diseases in the U.S. (< 5
per 100,000 population per year). The
standardized mortality rates in the U.S.
for white males and females combined
are 52.5 per million per year for liver
cancer and 29.2 per million per year for
kidney carcinoma.
Based on the various nsk estimates.
Tardiff (1976) calculated that the percent
of the annual cancer mortality
attributable to chloroform in drinking
water could be 1.60% and 1.44% for liver
and kidney cancer respectively
assuming the maximum exposure levels.
Applying these percentages to the actual
cancer mortality rates, the number of
cancer deaths per year would be 168
from liver carcinoma or 84 from kidney
carcinoma; an estimated maximum of
252 cancer deaths per year attributable
to chloroform in drinking water.
Reitz. Gehring. and Park (1978)
discussed EPA's procedures in
estimating risk. They stated that EPA
"seriously overestimates the actual
potential of chloroform * * ' (for) two
major reasons," These are: (1) The
mechanism through which chloroform
exerts its toxicity, and (2) reliance on
the NCI bioassay protocols which call
for high doses of chloroform, and by not
conducting studies at lower doses which
usually induce relatively less
carcinogenicity, there is a likelihood of
ignoring a possible detoxification
mechanism which protects test animals
until they are overwhelmed by very
large doses. They also suggest that an
experiment to evaluate the
carcinogenicity of chloroform at lower
doses must be performed before high/
low dose extrapolations can be
performed. Definitive data do not exist
to prove or to disprove the above claims.
The authors indicated that EPA's
proposed standard forTHMs of 0.10
mg/1 in drinking water supplies was
based on the carcinogenic risk
estimates. It snould be pointed out the
EPA's proposed standard for THM was
based upon that feasibility of achieving
the TTHM concentration in drinking
water, as well as the potential adverse
health effects.
EPA's Office of Water Planning and
Standards and Office of Research and
Development with EPA's Carcinogen
Assessment Group, developed a risk
estimate in the draft document,
"Chloroform—The Consent Decree
Ambient Water Quality Criteria
Document" (1979). The method used
assumed consumption of 2 liters/per day
of drinking water and 18.7 gm/per day
of fish and shellfish. The lifetime risk
estimates for excess cancers range frou
10~', 10"", and 10~7 with corresponding
consumption of 2.1 ug/1. 0.21 ug/l and
0.021 ug/l. respectively. The difference
in these risk estimations may be
explained by the assumption of daily
fish consumption as well as other
exposure sources. Without the fish
consumption, the equivalent
concentrations are 4.8 ug/l and 0.48 ug/1
for estimated cancer risk of 1 X 10~*
and 1 X 10"B, respectively. When this
estimate is computed for the
concentration of 0.10 mg/1 for levels in
drinking water, the incremental risk
would be 4.0 X 10"4 assuming two liters
of water at 0.10 mg/1 was consumed
daily for 70 years.
At an assumed lifetime exposure of 2
liters of water per day at 0.10 mg/1
chloroform the risk reduction to the
impacted population was estimated as H
range of approximately 200-500 total
cases. It snould be noted however, that
these average exposure levels in the
impacted population may result in
overestimates of the risk in light of the
facts that: (1) The computations arc
based upon lifetime exposures. In
actuality the proposed interim standard
will likely be reduced in the future as
technologically feasible, and, therefore, i
the lifetime exposure values will be less.
(2) The interim standard encourages
maximum reduction obtainable using
current technology. A much lower
average exposure is likely in the futuru
because technology will most likely
improve and result in greater exposure
reductions. On the other hand, these
may be underestimated because they
are based upon toxicity exposure data
from chloroform, which is only a portion
of the TTHMs. which are only a portion
of the by-products of the chlorination
process; therefore, the magnitude of the
contribution to the risk of the other
THMs, which in some cases contribute
significantly to TTHMs, is unknown.
The exposure to THMs from air and
food have not been included in these
computations.
X. Selected Maximum Contaminant
Levels (MCLs)
Since a risk to the public exists from
exposure to TTHMs and other
chlorination by-products in drinking
water, the potential for that risk should
be reduced as much as is
technologically and economically
feasible without increasing the risk of
microbiological contamination. This can
be accomplished by several means, and
the Safe Drinking Water Act (Pub. L 93-
523) provides two major regulatory
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68705
avenues—(1) the establishment of an
MCL, or (2) the institution of a treatment
equirement.
EPA has determined that the
establishment of an MCL in the Interim
Primary Drinking Water Regulations,
along with monitoring requirements, is
the most effective and immediate
approach to reducing the levels of THMs
in drinking water. The Administrator
has determined that monitoring is both
technically and economically feasible
(refer to "Economic Impact Analysis of a
Trihalomethane Regulation for Drinking
Water." EPA, 1977). Measures taken to
reduce the THM concentrations will
concurrently provide the additional
benefit of reducing human exposure to
the other undefined by-products of
chlorination and possibly other
synthetic organic contaminants.
Since it is known that chlorination of
water is primarily responsible for the
relatively high levels of THMs in
drinking water, modifications in the
chlorination process, the substitution of
other disinfectants, and the use of
adsorbents and other technologies to
remove precursor chemicals are possible
approaches to control. The optimal
•approach would be to reduce organic
precursor concentrations prior to
Addition of a disinfectant in order to
•educe disinfectant demand and
ninimize all by-products.
Use of a chlorine residual in a less
active form such as combined chlorine
or chloramine will significantly reduce
THM formation; however, chloramines
are much less potent disinfectants than
free chlorine, and therefore, this
approach must only be used after
careful consideration, and assurance of
maintenance of excellent biological
quality. The two chemicals most often
mentioned as substitute disinfectants,
ozone and chlorine dioxide, are both
well known as effective disinfectants
and chemical oxidants. The issues of the
biological effects and toxicity of these
disinfectants and their by-products are
being clarified by studies underway. In
the meantime, EPA recommended that
the residual total oxidant levels after
application of chlorine dioxide should
be limited to 0.5 milligram per liter. .
The National Organics Monitoring
Survey found that the mean total
trihalomethane (TTHM) concentrations
in the drinking water systems evaluated
were approximately 0.068, 0.117, 0 053
and 0.100 mg/1 for Phase I, II, III
(dechlorinated) and III (terminal]
respectively, with the highest levels of
0.784 mg/1 in Phase II (refer to Table 1).
It is reasonable to assume that the
various calculated risk estimates for
ihloroform indicate a potential risk to
public health. It is possible that a
percentage of the total number of liver
and/or kidney cancers are attributable
to exposure of chloroform in drinking
water, although it is most likely that
drinking water exposure would interact
with a number of other variables such as
smoking and diet as effect modifiers in a
multifactorial manner. It is also likely
that the other by-products of
chlorination also present a potential
risk.
Thus, based upon a number of risk
extrapolations assuming various levels
of exposure to chloroform in drinking
water, it has been estimated that such
exposures may cause an annual excess
of cancers in the U.S. population
(ranging from 0 to several hundred). At
higher levels of exposure of chloroform
(> 0.300 mg/1) the cancer risk estimates
are even higher.
The reduction of TTHMs to an MCL
level of 0.10 mg/1 would reduce the
unnecessary and excessive exposure to
these potential human carcinogens,
mutagens, and chronic toxicants, and
other effects. At the same time,
measures taken to reduce THM levels
(such as the use of adsorbents) will
concurrently result in reduction of
human exposure to other contaminants
in drinking water.
Since it is economically and
technologically feasible to reduce the
THM levels in drinking water, and since
benefits are achieved by reducing the
health risks of exposure, EPA has
decided to establish the MCL at 0.10
mg/1 as the initial feasible step in a
phased, regulatory approach. As more
data become available from
implementation experience, and
toxicology and epidemiology, standards
are expected to become more restrictive.
In the meantime, EPA and the States
should continue to take steps as
necessary on a case-by-case basis to
provide adequate protection for the
delivery of safe drinking water to the
public, by minimizing the amounts of
toxic chemicals in the water.
XI. References
Alavanja, M, et al., 1978. "An
Epidemiologies! Study of Cancer Mortality
and Trihalomethanes in Drinking Water in 19
New York State Counties "US EPA. Office
of Research and Development, Health Effects
Research Laboratory, Cincinnati, Ohio,
Unpublished.
Alavanja, M.. Goldstein, I., and Susser, M.,
1977. "Case Control Study of Gastrointestinal
Cancer Mortality in Seven Selected New
York Counties in Relation to Drinking Water
Chlorination " U.S. EPA. Office of Research
and Development, Health Effects Research
Laboratory. Cincinnati, Ohio, Unpublished.
Alavanja, M. Goldstein, I., and Susser, M.,
1976 "Report of Control Study of Cancer
Deaths in Four Selected New York Counties
in Relation of Drinking Water Chlorination."
U.S. EPA, Office of Research and
Development, Health Effects Research
Laboratory, Cincinnati, Ohio, Unpublished
Draft.
American Conference on Industrial
Government Hygienists, 1975. "Threshold
Limit Values for Chemical Substance in
Workroom Air " Cincinnati, Ohio.
Ames. B. N., Vi. E. Durston, E. Yamaski,
and F D. Lee, 1973. "Carcinogens are
Mutagens: A Simple Test System Combining
Liver Homogenates for Activation and
Bacteria for Detection." Proc. Nat. Academy
of Science. USA, 70:2281-2286.
Bellar, R. A., Uchienberg,) ), and Kroner,
R. C, 1974. "The Occurrence of
Organohahdes in Finished Drinking Waters."
Journal of American Water Works
Association. 68:
Brenmman, G, et al.. 1979. "Relationship
Between Cancer Mortality and Chlorinated
Drinking Water " University of Illinois,
Unpublished Draft.
Brown, D. M.. Langley. P. F.. Smith, D.. and
Taylor, D. C., 1974. "Metabolism of
Chloroform I. The Metabolism of [C]
Chloroform by Different Species."
Xenobiolica 4.155-163.
Buncher, C. R, 1975. "Cincinnati Drinking
Water—An Epidemiologic Study of Cancer
Rates." University of Cincinnati Medical
Center, Cincinnati, Ohio.
Bunn, W. W., Haas. B B. Deane, E. R., and
Kleopfer, R D., 1975. "Formation of
Trihalomethanes by Chlorination of Surface
Water " Environmental Letters W 205-213
Burreson, C. J., Moore, R. E., and Roller, P.
P., 1975. "Volatile Halogen Compounds in the
Algae Aspargopis Taxiformis."/. Agric. Food
Chem. 24:856-861
Butterfield. C. T. and Wattie, S, 1944.
"Influence of pH and Temperature on the
Survival of Cohforms and Enteric Pathogens
When Exposed to Free Chlorine." Public
Health Rep. 50-1837-1866. .
Butterfield, C. T and Wattie. S.. 1946.
"Influence of pH and Temperature on the
Survival of Cohforms and Enteric Pathogens
When Exposed to Chloramine." Public
Health Rep. 81-157-192.
Canter, K , Hoover, R., Mason, T, and
McCabc, L., 1978 "Association of Cancer
Mortality with Halomethanes."/. of National
Cancer Inst 67-979-985.
Chernoff, 1977. "Personal Communication."
Clarke. N. A. and Kabler. P K., 1954.
"Inactivation of Purified Coxsackie Virus in
Water by Chlorine " Am ]. Hyg. 59 119-127
Cotruvo, ] A, Simmon, V F. and
Spanggord, R ), 1977. "Investigation of
Mutagenic Effects of Products of Ozonation
Reactions in Water." Annals of the New York
Academy of Sciences, in press, and ''Personal
Communications."
Cole, P C.. Hoover. R, and Friedell, G. H.,
1972. "Occupation and Cancer of the Lower
Urinary Tract." Cancer. 29:1250-60
DeRouen. T. A., and Diem, ]. E, 1975.
"Ethnic. Geographical Difference in Cancer
Mortality in Louisiana." Tulane University,
School of Public Health and Tropical
Medicine, Unpublished.
DeRouen. T. A and Diem. ]. E.. 1975. "The
New Orleans Drinking Water Controversy A
Statistical Perspective." American journal of
Public Health. 63. (No 10) 1060
-------
68706 Federal Register / Vol. 44. No. 231 / Thursday. November 29. 1979 / Rules and Regulations
DeSalva. S.. Volpe, A.. Leigh, G., and
Regan. T., 1975. "Long-Term Safety Studies of
a Chloroform-Containing Dentrifice and
Mouth Rinse in Man." Food Cosmet. Toxicol.
13- 529-532.
Dewey. W.. Personal Communication.
Medical College of Virginia.
Docks. E.L. and Krishna. G.. 1978. "The
Role of Glutathione in Chloroform Induced
Hepatotoxicity." Experimental and
Molecular Pathology 24-13-22.
EPA, 1979. "Chloroform—The Consent
Decree Ambient Water Quality Criteria
Document." EPA. Office of Water Planning
and Standards. Unpublished Draft.
EPA. April 1976. "Industrial Pollution of the
Lower Mississippi River in Louisiana."
Region VI. Dallas, Texas.
EPA. 1974. "Analytical Report. New
Orleans Water Supply Study." Region VI,
Dallas, Texas.
EPA. 1975. "A Report—Assessment of
Health Risk from Orgamcs in Dnnking Water
by an Ad Hoc Study Group to the Hazardous
Materials Advisory Committee." Science
Advisory Board. Unpublished.
EPA. June 1975. "Region V Joint Federal/
State Survey of Organics and Inorganics in
Selected Dnnking Water Supplies." Region V.
Chicago. Illinois
EPA, 1976. "Risk Evaluation of
Chloroform." Office of Pesticides Programs,
Criteria and Evaluation Division.
Unpublished.
EPA, 1977. "Chloroform Risk Assessment in
Drinking Water." Cancer Assessment Group.
Unpublished.
EPA. August 1977. "Economic Impact
Analysis of a Trihalomethane Regulation for
Drinking Water." prepared by Temple.
Barker, end-Sloan, Inc. for EPA, Office of
Drinking Water.
EPA, June 1976. "Interim Treatment Guide
for the Chloroform and other
Tnhalomethanes." EPA, Water Supply
Research Division, Municipal Environmental
Research Laboratory, Cincinnati. Ohio.
Eschenbrenner, A.B. and Miller, E.. 1945.
"Introduction of Hepatomas in Mice by
Repeated Oral Administration of Chloroform
with Observation on Sex Differences."/.
Nal'l. Cancer last. 5 251-255.
Fry, F.). Taylor. T.. and Hathaway. E.D..
1972. "Pulmonary Elimination of Chloroform
and Its Metabolites in Man." Arch. Int.
Pharmacodyn 196 98-111.
Hefferman. P., 1977, Personal
Communication.
Heilbrunn. G, Liebert. E. and Szanto. P.B.,
1945. "Chronic Chloroform Poisoning—
Clinical and Pathological Report of a Case."
Arch. Neural. Psych 53:68-72.
Hogan. M.D.. Chi. P.. Hoel. D G. and
Mitchell. T.J.. 1979 "Association Between
Chloroform Levels in Finished Drinking
Water Supplies and Various Site-Specific
Cancer Mortality Rates."/. ofEnv. Path. &
Tax. 2:873-887.
Ilett. K.F.. Reid. W.P.. Sipes, I.G.. and
Krishna. G.. 1973. "Chloroform Toxicity in
Mice: Correlation of Renal and Hepatic
Necrosis with Covalent Binding of
Metabilities to Tissue Macromolecules."
Exptl. Molec. Pathol. 19:215-229.
Jones. W.M.. Marguis. G. and Stephen,
C.R.. 1958. "Hepatoxicity of Inhalation
Anesthetic Drugs." Anesthesiology 19:715-
23.
Kelly, S.M. and Sanderson, W.W.. 1960.
"The Effect of Chlorine in Water on Enteric
Viruses. II. The Effect of Combined Chlorine
on Poliomyelitis and Coxsackie Viruses."
Am. J. Public Health. 50:14-20.
Kubic. V.L. Anders. M.W.. Engel. R.R..
Barlow. C H., and Caughey, W.S.. 1974.
"Metabolism of Dihalomethanes to Carbon
Monoxide In Vivo Studies." Drug
Metabolism and Disposition 2- 53-57.
Kruse. C.W.. 1977. "Chlorination of Public
Water Supplies and Cancer—Washington
County. Maryland Experience." A Final
Report. John Hopkins University, School of
Hygiene and Public Health to the Office of
Research and Development, Health Effects
Research Laboratory, Cincinnati. Ohio,
Unpublished Draft.
Kuzma. R J. Kuzma, C J. and Buncher. C.R.,
1977. "Ohio Dunking Water Source and
Cancer Rates." American Journal of Public
Health 67. 725-729.
Lehman, K B. and Hasegawa, 1910.
"Studies of the Absorption of Chlorinated
Hydrocarbons in Animals and Humans."
Archives of Hygiene 72- 327.
Mah, R. A., Spivey. G. H.. and Sloss. E..
1977 "Cancer and Chlorinated Drinking
Water." A Final Report. University of
California. Los Angeles. School of Public
Health to the Office of Research and
Development, Health Effects Research
Laboratory, Cincinnati. Ohio, Unpublished.
McCabe, L. J.. 1975. "Association Between
Halogenated Methanes in Drinking Water
and Mortality." EPA. Health Effects Research
Laboratory. Cincinnati, Ohio, Unpublished.
McLean. A E. M., 1970. "The Effect of
Protein Defiency and Microsomal Enzyme
Induction by DDT and Phenobarbitone on the
Acute Toxicity of Chloroform and a
Pryolizione Alkaloid. Retrosine." British
Journal of Experimental Pathology 51' 317-21.
Meinhardt. T. J. Mareinfield. C. J. Miller.
R. S.. and Wright. H T. 1975. "River Water
and Cancer Mortality." Environmental Health
Surveillance Center, University of Missouri,
Unpublished.
Michael. C E.. et al.. 1979. "Chlorine
Dioxide Water Disinfection1 A Prospective
Epidemiologic Study " U.S. EPA. Office of
Research and Development. HERL.
Cincinnati, Ohio.
NAS. 1978. "Epidemiological Studies of
Cancer Frequency and Certain Organic
Constituents of Drinking Water—A Review
of Recent Literature Published and
Unpublished." Washington. D.C.
NAS. 1977. "Drinking Water and Health."
Washington. D.C.
NAS. 1977. "Non-Fluonnated
Halomethanes in the Environment."
Washington, D C.
NAS. 1979. "The Disinfection of Drinking
Water." Washington. D.C. Unpublished Draft.
NAS. 1979. 'The Chemistry of Disinfectants
in Water: Reactions and Products."
Washington. D.C. Unpublished Draft.
National Cancer Institute, 1976. "Report on
Carcinogenesis Bioassay of Chloroform."
National Institute of Occupational Safety
and Health. 1974. "Criteria for a
Recommended Standard, Occupational
Exposure to Chloroform."
Page, T. and Harris, R. H., 1975. "Relation
Between Cancer Mortality and Drinking
Water in Louisiana " Unpublished.
Page. T., Harris. R. H., and Estein. S. S..
1976. "Drinking Water and Cancer Mortality
in Louisiana." Science 193:55-57.
Page. T.. Talbot. E.. and Harris. R. H., 1974
"The Implications of Cancer Causing
Substances in Mississippi River Water."
Environmental Defense Fund. Washington.
DC.
Paul, B B. and Rubinstein. D, 1963.
"Metabolism of Carbon Tetrachloride and
Chloroform by the Tat."/. Pharm Exp.
Thcrap. 141-141-148.
Plaa. C. L. Evancs. E. A., and Mine. G. II..
1963. "Relative Hepatotoxicily of Seven
Halogenated Hydrocarbons."/. Pharmacol.
Exp. Therap. 123: 224-229.
Plaa. C. L and Larson. R. E. 19C5. "Relative
Nephrotoxic Properties of Chlorinated
Methane. Ethane and Ethylene Derivatives in
Mice." Toxicol. Appl. Pharmacol. 7. 37-44.
Pohl. L R.. et al.. 1977. "Phosgene: A
Metabolite of Chloroform." Biochemical and
Biophysical Research Communications.
Bcthcsda, Maryland.
Pohl. L. R. and Krishna, G.. 1978.
"Deuterium Isotope Effect in Bioactivnlion
and Hepatotoxicity of Chloroform." Life
Sciences. Bethcsda, Maryland.
Pahl, et al., 1979. "Deuterium Isotope Effect
in Vivo Bioactivation of Chloroform and
Phosgene."
Poiner. L. A. Stoner, G. D.. and Shimkin.
M. B.. 1975. "Bioassay of Alkyl Halides and
Nucleotidcs Base Analogs by Pulmonary
Tumor Response in Strain A Mice." Cancer
Research 35:1411-1415.
Rafferty, P.. 1979. "Drinking Water Quality
and Cancer Mortality Patterns." University of
North Carolina. Chapel Hill. North Carolina,
Unpublished Masters Thesis.
Reiches, N A., Page T, Talbol. P., and
Harris, R H, 1976. "Carcinogenic Hazards of
Organic Chemicals in Drinking Water."
Unpublished.
Reitz. R H. Gehring. P.]. and Park. C.N..
1978. "Carcinogenic Risk Estimation for
Chloroform—An Alternative to EPA's
Procedures." /. Food & Cosmet. Toxicol. It?
511-514.
Roe, F.] C.. 1976. "Preliminary Report of
Long-Term Tests of Chloroform in Rats, Mice
and Dogs." Hazelton Laboratories, Vienna,
Virginia.
Rook, |.|. 1S7B. "Formation of Holoforms
During Chlorination of Natural Waters."
Water Treatment B- Examination 23- 234-243.
Rook. J.J., 1977. "Chlorination Reactions of
Fulvic Acids in Natural Waters." Env. Sci.
Techno!. 11. 478-482.
Salg. ].. 1977. "Cancer Mortality Rates and
Drinking Water in 346 Counties of the Ohio
River Valley'Basin." Final Report, University
of North Carolina, Department of
Epidemiology to the Office of Research and
Development, Health Effects Research
Laboratory. Cincinnati, Ohio, Unpublished.
Scholler. K.L., 1968. "Electron-Microscopic
and Autoradiographic Studies on the Effect of
Halomethane and Chloroform on Liver
Cells." Acto Anaesthesioly of Scandinavia
32:1-62.
Schwelz, B.A., Leong. B K.).. and Cchring,
P.J.. 1974. "Embryo and Fetotoxicity of
-------
Federal Register / Vol. 44. No. 231 / Thursday, November 29, 1979 / Rules and Regulations 68707
Inhaled Chloroform in Rats." Toxicol. Appl.
armacol. 2& 442-451.
pipes, G.I.. Krishna, G., and Gillette, ] R.,
|2. "Bioactivation of Carbon Tetrachlonde.
loroform, and Bromotrichloromethane:
Kole of Cytochrome P-450." Life Sciences,
Bethesda, Maryland.
Southeimer, H., and Kuhn. W., 1977. The
Engler-Bunte Institute, University of
Karlsruhe, Karlsruhe, Germany. Personal
Communication.
SRI International. March 1979. "The Effects
of Ozonation Reactions in Water." Volume 1.
SRI International. March 1979. "The Effects
of Reactions of Chlorine Dioxide in Water."
Volume 2.
Storms. W.W.. 1973. "Chloroform Parties."
JAMA, 225:160.
Symons. J.M., Bellar, T.A., Carswell. J.K.,
DeMarco, ], Kropp, K L. Robeck. G.G..
Seeger. D.R., Slocum, C.J., Smith. B.L., and
Stevens, A.A., 1975. National Organics
Reconnaissance Survey for Halogenated
Organics. /. Am. Water Works Assn. 67:634-
646.
Tardiff, R., 1976 Personal Communication.
Tardiff, R.G, 1976. "Health Effects of
Organics. Risk ft Hazard Assessment of
Ingested Chloroform." The 96th Annual
Conference of the American Water Works
Association New Orleans, Louisiana.
Tarone. R.E. and Cart, J.J.. 1975. "The
Implications of Cancer-Causing Substances in
Mississippi Rain Water." An unpublished
review of the study by R H Harris.
Taylor. D.C. Brown. D M , Keeble. R. and
-' -ngley, P.P., 1974. "Metabolism of
loroform II. A Sex Difference in the
tabolism of |C| Chloroform in Mice."
wbiotica 4:165-174.
Theiss, J.C., Stoner, C.D, Shimkin. M fi-
end Weisburger, E K., 1977. "Test of
Carcinogenicily of Organic Contaminants of
United States Drinking Waters by Pulmonary
Tumor Response in Strain A Mice." Cancer
Research 37. 2717-2720.
Tuthill. R W. and G Moore. 197B.
"Chlonnation of Public Drinking Water
Supplies and Subsequent Cancer Mortality:
An Ecological Time-Lag Study." Division of
Public Health, School of Health Sciences,
University of Massachusetts, Unpublished.
United States, Department of Health.
Education and Welfare, National Institute of
Occupational Safety and Health, 1973. Toxic
Substance List.
Vasilenko. P. and Magno, L., 1975. "Factors
Relating to the Incidence of Cancer Mortality
in New jersey " Princeton University,
Princeton, New Jersey, Unpublished
Vocci. F).. Martin, B R, Petty. S., and
Dewey, W L, 1977. "Effect of Acute and
Subchronic Chloroform Exposure on
Chlonergic Parameters in Mouse Brain."
Journal of Toxicology and Applied
Pharmacology.
Wallace, C ]. 1959 "Hepatitis and
Nephrosis Due to Cough Syrup Containing
Chloroform." Calif Med. 73 442.
Watanabe, P G. McGowan. G R., Madrid,
E O.. and Gehring. P). 1976 "Fate of |C]
Vinyl Chlonde Following Inhalation
Dosure in Rats." Toxicol. Appl. Pharmacol.
49-59.
Veisburger. J.H., 1977. "Social and Ethical
plications of Claims for Cancer Hazards."
Medical and Pediatric Oncology 3; 137-140.
Wilkins III, J.R., and C.W. Kruse, 1978.
"Chlorination of Public Water Supplies and
Cancer." School of Public Health, Johns
Hopkins University, Baltimore, Maryland,
Unpublished Doctoral Dissertation.
Von Oettingen, W.F., 1955. "The
Halogenated Hydrocarbons: Toxicity and
Potential Dangers." Public Health Service No.
414. Washington, D.C., U.S. Government
Printing Office
|FK Doc. 79-38442 Filed 11-28-79:845 om|
BILLING CODE 6560-01-M
------- |