40CFRpart61
National Emission Standards
for Hazardous Air Pollutants
EPA-402-R-96-022
RESPONSE TO COMMENTS
NESHAPS RULEMAKING ON NUCLEAR
REGULATORY COMMISSION AND AGREEMENT STATE LICENSEES
OTHER THAN NUCLEAR POWER REACTORS
December 1996
U.S. Environmental Protection Agency
Office of Radiaiton and Indoor Air
Washington, DC 20460
-------
-------
Response to Comments Document
40 CFR Part 61, Subpart I
(NRC-Licensees Other Than Commercial Nuclear Power Reactors)
This Response to Comments Document responds to comments
received relating to the Agency's rescission of 40 CFR Part .61
Subpart I as applied to facilities other than commercial nuclear
power reactors and not addressed in the preamble to the final
rule. EPA received written comments from 14 commenters including
individuals, environmental groups, States and industry
representatives in response to EPA's proposal to rescind 40 CFR
Part 61 Subpart I for NRC licensees other than commercial nuclear
power reactors. See 57 FR 56877 (December 1, 1992). EPA
subsequently published a notice reopening the comment period for
this rulemaking. See 60 FR 50161 (September 28, 1995). This
Notice reaffirmed EPA's 1992 proposal to rescind Subpart I,
described the expected proposed revisions to the NRC program
which would support EPA's rescission, and invited additional
comment on five questions relating to the sufficiency of the
revisions to the NRC program to support the finding required by
CAA Section 112(d)(9). EPA received written comments from 34
commenters including several individuals, environmental groups,
States and industry in response to the Notice. In addition, EPA
also received comments from individuals, environmental groups and
industry representatives during public hearings conducted on
January 14, 1993 and February 29, 1996.
-------
Given events which occurred between EPA and NRC after the
1992 proposal, including NRC's proposed constraint rule, EPA
believes that the more recent comments submitted by a party in
response to the 1995 Notice, should be accorded more weight than
comments previously submitted by that same party in 1992, where
there is inconsistency between the comments.
In addition, EPA's review of the comments has been limited
to the question of whether EPA should rescind Subpart I for NRC
licensees other than commercial nuclear power reactors. This
rulemaking was not intended to reconsider and did not address
whether EPA should have promulgated Subpart I in 1989 or the
appropriateness of EPA's 10 -mrem/yr dose standard. EPA therefore
rejected as irrelevant to this rulemaking, comments addressed to
the validity or appropriateness of the promulgation of Subpart I.
1. General
In response to the 1992 proposal and 1995 Notice, 25
commenters supported rescission and 23 commenters opposed
rescission. The Agency has reviewed all comments received in
this rulemaking carefully.
2. 1995 Request for Comments
This section of the Response to Comments Document addresses
the comments received in response to the five questions that the
Agency posed in the Federal Register notice (60 FR 50161,
September 28, 1995) announcing the reopening of the comment
period on its proposed rescission of 40 CFR 61, Subpart I as it
-------
applies to NRC licensed facilities other than commercial power
reactors.
2 .1 Question 1: If NRC adopts the proposed ALARA constraint
level rule, will the resultant NRC regulatory program assure
that routine radionuclide emissions from NRC licensees other
than nuclear power reactors result in doses which are
consistently and predictably no greater than 10 mrem/year?
Comment.- Numerous commenters, both for and against
rescission, note that it is difficult to comment on the
effectiveness of a proposed regulation, and caution the EPA
to reserve its final decision until the NRC has promulgated
its final ALARA constraint rule.
Response: The Agency believes that the draft ALARA
constraint rule and the background provided by the Agency in
the Federal Register notice allow for informed public
comment on which EPA can base its decision. As the Agency
stated in the Federal Register notice announcing the
reopening of the comment period, "EPA intends to take final
action concerning its proposal to rescind subpart I for NRC
and Agreement State licensees other than nuclear power
reactors on or after the date that NRC takes final action on
the ALARA 'constraint level1 rule." 60 FR at 50165
(September 28, 1995). Further, as stated in the September
notice, the Agency did not take final action until NRC
completed the 'constraint rule'.
-------
Gonzznentr Commenters supporting the Agency's proposal to
rescind agree that adoption of an ALARA constraint level
rule will provide additional assurance that the same level
of protection provided by implementation of Subpart I will
be consistently and predictably achieved. While favoring
rescission, many of these commenters object to the
imposition of such a rule as unnecessary. Citing the
Agency's Targeted and Random Survey results, and noting that
those levels were achieved prior to the NRC's lowering its
exposure limit to 100 mrem/yr and imposing ALARA as a
mandatory part of every licensee's radiation protection
program, they believe that a constraint rule will only
result in additional compliance burdens without affecting
the already acceptable levels of emissions.
Commenters opposed to the Agency's proposal to rescind
do not agree that adoption of the ALARA constraint level
will result in the NRC's regulatory program assuring that
doses are consistently and predictably below 10 mrem/yr.
Those commenters believe that since actual measurements of
dose are not required, the NRC will continue a perceived lax
regulatory approach which they believe relies on
self-reporting and inadequate inspections and enforcement.
The NRC's initiatives to move to more risk-based and
performance-based regulations heighten the commenters'
concerns about the long-term effectiveness of the NRC's
program. These commenters are also concerned that NRC
-------
recently increased its cost limit to $2,000 per person rem
avoided.
.Response; The Agency does not find the concerns raised by
those opposed to rescission to be compelling. In developing
Subpart I, the Agency was cognizant that a large fraction of
the facilities covered by the NESHAP do not employ
continuous effluent or environmental monitoring, and it
determined that release fractions and effluent control
adjustment factors could be developed and applied for use in
compliance assessment. The Agency also finds that levels of
emissions and doses caused by these facilities, as
documented in its Targeted and Random surveys, do not lead
to the conclusion that the NRC improperly relies on self-
reporting in deference to an effective inspection and
enforcement.program.
The Agency also finds the argument that NRC has
increased the cost limit per person rem avoided not relevant
to this rulemaking. NRC has increased the cost limit person
rem avoided from $1,000 to $2,000 and this would potentially
lead to more stringent regulation. In addition, NRC has
adopted a 10 mrem/yr constraint level which assures that
radionuclide emissions by the affected licensees will be
consistently and predictably below 10 mrem/yr. If a
licensee exceeds the constraint level, NRC can require the
licensee to take corrective actions to reduce emissions.
-------
2.2 Question. 2: If NRC adopts the proposed ALARA
constraint level rule, will NRC have sufficient authority to
require any affected facility with routine radionuclide
emissions at a level which results in doses exceeding 10
mrem/yr to reduce its emissions to a level resulting in
doses no greater than 10 mrem/yr?
Comment: Many commenters supporting the proposed rescission
believe that the NRC and the Agreement States already have
effective programs with ample authority to compel licensees
to achieve emission levels that will not result in doses
exceeding 10 mrem/yr. Promulgation of an ALARA constraint
rule will serve to enhance the existing authority.
Commenters opposed to the proposed rescission do not
believe that imposition of an ALARA constraint rule will
provide the NRC with an enforceable basis for assuring that
doses are maintained at levels below 10 mrem/yr. Guidance
and ALARA are not enforceable limits, and therefore cannot
provide the same level of protection as the Agency's NESHAP.
Response: NRC's promulgation of the constraint rule, while
not a limit per se, requires licensees to notify NRC if the
licensee exceeds 10 mrem/yr tede and implement adequate
corrective actions to ensure that doses do not exceed 10
mrem/yr tede. EPA has determined that the constraint rule
together with the recently finalized Agreement State
policies and procedures for adequacy and compatibility
-------
ensure that emissions of radionuclides will be consistently
and predictably below the Subpart I limit.
2 .3 Question 3: If NRC makes the proposed ALARA constraint
rule a matter of Division Level 2 compatibility, will this
assure that each individual Agreement State establishes an
ALARA constraint level for its licensees which is not
greater than 10 mrem/yr, and requires its licensees to
report and correct exceedances of that level?
Comment: Commenters favoring rescission contend that
Division Level 2 compatibility will provide the necessary
assurance.
Commenters opposed to rescission contend that Division
Level 2 compatibility will not provide the same level of
assurance as Division Level 1, which requires "strict"
compatibility. Several officials of State programs question
whether or not Division Level 2 compatibility would allow
them to adopt requirements more strict than those imposed by
the NRC.
Response: The Agency believes that Division Level 2
compatibility, or an equivalent level of compatibility, will
assure that emissions from facilities licensed by the
Agreement States will be maintained at NRC's constraint
level of 10 mrem/yr tede or at a more stringent level.
Adoption of Division Level 2 compatibility, which requires
each Agreement State to incorporate provisions into its
programs at least as stringent as those in the NRC's rule,
-------
does provide States the opportunity to incorporate even
stricter standards.
2.4 Question 4: Are the NRC policies establishing criteria.
to evaluate the adequacy and compatibility of Agreement
State programs, and adopting procedures to permit suspension
or termination of Agreement State programs, sufficient to
enable the NRC to take necessary action if it determines
that an Agreement State program is inadequate or
incompatible?
Comment: Commenters favoring rescission believe that the
NRC has always had the necessary authority to revoke
Agreement State status upon a showing of cause, and contend
that the NRC's policy statement will enhance all parties'
understanding of the Agreement State program. Several note
that the "problems" identified with the Agreement State
program have been issues of information flow rather than
issues affecting the protection of the public health and
safety.
Commenters opposed to rescission contend the NRC has
never shown any commitment to evaluating the adequacy of
Agreement State programs and has never suspended or revoked
Agreement State status. Given the cutbacks in the NRC's
program, they believe that a weak Agreement State program
would be tolerated by the NRC.
-------
Response: The Agency finds that the NRC authority does
provide for suspension or termination of Agreement State
status upon a finding that the state program is inadequate
or incompatible with the NRC's requirements. Adoption by
the NRC of the policy statements and procedures concerning
adequacy and compatibility provide the required assurance
that inadequate or incompatible agreement state programs
will be corrected or revoked, thus assuring protection of
public health and safety. See also letter from Martin
Malsch, NRC Deputy General Counsel to Ramona Trovato,
Director of EPA's Office of Radiation and Indoor Air,
November 19, 1996, Docket Entry A-92-50, IV-G-8.
2.5 Question 5: Do these four actions [addressed in
questions 1-4 above], in addition to other actions taken by
NRC combine to provide an ample margin of safety to protect
public health?
Comment: Commenters supporting rescission believe that the
NRC's program has consistently provided an ample margin of
safety to protect public health, and cite the EPA's Targeted
and Random surveys as proof. They believe the four actions
will only serve to enhance the degree of assurance that the
ample margin of safety is maintained in the future.
Opponents of rescission do not believe that the NRC's
program, with or without the four actions, provides the
requisite ample margin of safety because: (1) NRC's
regulations allow doses of 100 mrem/yr, and provide for
-------
exceptions to allow doses as high as 500 mrem/yr; (2)
guidance and ALARA are not enforceable regulations; (3)
NRC's mandate is not focused on protecting public health;
and (4) NRG has never revoked any Agreement State status. A
few commenters also believe that NRC's "dismal" record of
protecting the public (e.g., TMI-2, Radiation Sterilizers,
Inc., Time magazine cover story from March 4, 1996) and its
ill-conceived regulatory initiatives (e.g., BRC, elimination
of requirements marginal to safety) provide ample proof that
it does not intend to hamper licensees by making them
operate in a manner that protects the public. One commenter
also believes that in making any "ample margin"
determination under CAA section 112(d)(9), EPA should base
such a determination on the most recent information on low-
dose exposure.
Response: The legislative history of Section 112(d)(9)
provides guidance as to what is meant by "an ample margin of
safety" and what process the Administrator is to follow in
making that determination in a rulemaking conducted under
section 112(d)(9). The Conference Report accompanying S.
1630 - the Clean Air Act Amendments points out that the
"ample margin of safety" finding under section 112(d)(9) is
the same "ample margin of safety" requirement that was
contained in Section 112 of the CAA prior to its amendment
in 1990.
10
-------
EPA has already made a determination in promulgating
Subpart I that compliance with the 10 mrem/yr dose standard
protects the public health with an ample margin of safety.
For the purposes of present rulemaking, EPA has used the 10
mrem/yr standard found to provide an ample margin of safety
as the benchmark by which it evaluates the NEC regulatory
program and the Court of Appeals for the B.C. Circuit has
recently approved this practice in unpublished opinion.
Sierra. Club, et al., v. Environmental Protection Agency,
No.95-1562 (D.C. Cir. October 22, 1996). EPA's study of NRC
licensees not engaged in nuclear power generation,
determined that the overwhelming majority were below the 10
mrem/yr limit suggesting that the NRC program was currently
affording protection with an ample margin of safety.
As stated in the Federal Register notice, the Agency
has determined that the promulgation of a constraint rule
and adoption of the policy statements and procedures on
adequacy and compatibility of Agreement State programs,
provide the additional assurances needed to make the finding
required by Section 112(d)(9) that the NRC's program
pursuant to the Atomic Energy Act will provide an ample
margin of safety to protect the public health in the future.
The constraint level, while not a limit, is a value above
which certain actions are required. Contrary to the
opinions expressed by many commenters opposing rescission,
Section 112(d)(9) does not require the Agency to find that
11
-------
3.
the NRG regulatory program is identical to EPA's NESHAP
program. Rather, the Agency must determine that the NRC and
EPA regulatory programs are functionally equivalent. Given
the NRC's actions with respect to the constraint rule and
the policy statement concerning adequacy and compatibility,
together with EPA's study and experience implementing
Subpart I, EPA has determined that the NRC's regulatory
program provides an enforceable basis for assuring doses
caused by air emissions of radionuclides from NRC and
Agreement State licensed facilities will be consistently and
predictably below 10 mrem/yr. Thus, the EPA believes that
the NRC regulatory program will protect the public to the
same level as would be afforded by continued implementation
of Subpart I.
Adequacy of NRC Program - Incinerator Operations
Commentr Some commenters argue that NRC is lax in
implementing its regulatory program and thus, does not
protect the public health with an ample margin of safety.
Specifically, one commenter argues that NRC does not
adequately regulate the nuclear operations at the University
of Michigan and hence the EPA should not rescind Subpart I.
The following arguments were presented in support of the
view that the NRC is not properly regulating the incinerator
operations, as well as other categories of licensees.
12
-------
3.1 NRG Concentration Limit
Comment:: The NRC does not use a dose limit. Instead they
use a concentration limit, and this makes it difficult for
the public to determine the degree of risk they are being
exposed to.
.Responser This comment was also received in response to
EPA's 1992 proposal. EPA considers this issue to be
irrelevant, since NRC has adopted a 10 mrem/yr tede
constraint level. Further, EPA allows the use of
concentration tables under, Subpart I for determining
compliance with that Subpart.
3.2 NRC 100 mrem/yr Limit
Comment:: The NRC's limit of 100 mrem/yr is 10 times that of
EPA's limit of 10 mrem/yr.
Response: This comment was received in response to EPA's
1992 proposal. EPA considers the issue raised to be
irrelevant because NRC has adopted a 10 mrem/yr tede
constraint level. Also, NRC's earlier 500 mrem/yr limit in
combination with ALARA, has been found in EPA's study to
achieve the functional equivalency of the Subpart I limit.
3.3 NRC Relationship with Industry
Comment:: The NRC is trying to relax its regulations at the
urging of industry and at the expense of the public.
Response: EPA disagrees with the comment because NRC has
adopted a 10 mrem/yr tede constraint rule. In addition,
effective January 1, 1994, the NRC regulations governing
13
-------
facilities like the University of Michigan were made more
stringent in that the maximum allowable dose in an
unrestricted area was lowered from 500 to 100 mrem/yr. The
fact that some permissible concentrations have increased
does not reflect a lessening of protection. Rather, it
reflects the change in radiation protection to a methodology
that better -reflects the risks resulting from a given
exposure.
3.4 EPA Survey Results Flawed
Comment: The designated survey results are flawed in that
the EPA picked only the wealthiest and brightest facilities;
i.e., those most likely to comply with the limits. The
random survey is flawed because the respondents knew that
the EPA was conducting the survey.
Response: In the designated survey, the EPA did not pick
only the "wealthiest and brightest facilities". Rather, the
Agency spent a great deal of effort to identify those
facilities that were likely to emit the largest quantities
of radioactivity, thereby resulting in the highest radiation
dose.
The letter sent to the random survey respondents
contained the following sentence: "You should also be aware
that Section 113 [of the Clean Air Act] provides for
possible criminal sanctions for anyone who knowingly makes
any false statement, representation, or certification in a
report required by EPA. " The Agency has no reason to
14
-------
believe that any of the respondents supplied false
information.
3.5 EPA Survey Relies on Self-Reporting
Comment: It is not realistic to rely on self-reporting.
The University ran the COMPLY code and got a dose of 0.3
mrem/yr. The Michigan Department of Natural Resources got a
COMPLY dose of 10.6 mrem/yr, with 9.7 mrem/yr due to
radioiodine.
Response: Both the EPA and the NRC regulations provide for
criminal penalties in the event of false information being
supplied by a licensee. Periodic inspections of facilities
and records assure that licensees are in compliance. The
differing COMPLY results are discussed in item 3.5.1 below.
3.5.1 Specific Example
Comment: Several COMPLY studies made by the Michigan
Department of Natural Resources (DNR) and included as an
attachment to testimony suggest that the radiation dose from
the University of Michigan incinerator exceeded the Subpart
I limits of 10 mrem/yr for all radionuclides and 3 mrem/yr
for radioiodine. In addition, commenters cited another
study that had an estimated thyroid dose of 66 mrem from
radioiodine,. which they state is far higher than the 3
mrem/yr limit. They also state that the University of
Michigan's COMPLY run gave a dose of 0.3 mrem/yr, far lower
than the other studies.
15
-------
Response: The Agency has run the COMPLY code using the same
1989 emissions data, building and stack parameters, and
distance to the resident as in the DNR COMPLY runs supplied
by the commenter. The result is a dose in the worst sector
of 0.5 mrem/yr from all radionuclides and 0.3 from
radioiodine. There are two reasons for this dose being
approximately a factor of 20 lower than the DNR results
cited by the commenter.
The first reason is that the DNR runs used the COMPLY
default annual average wind speed and frequency. The Agency
used a wind rose (which accounts for the actual annual
average frequency and speed with which the wind blows toward
a given sector). This 'reduced the estimated doses by about
a factor of five to 2.2 for all radionuclides and 2.0
mrem/yr for iodine. This is because the actual frequency
the wind blows toward the maximum sector is 0.11 versus the
default value of 0.25 and the actual wind speed is 4.2
meters/sec versus the default value of 2.0 meters/sec.
The second reason is that in the DNR COMPLY runs all
the person's food (vegetables, milk, and meat) was assumed
to be raised at home. The Agency put the farms supplying
dairy and meat products one mile away (approximately -the
distance to the city limits; cattle and other farm animals
cannot be raised within the city limits of Ann Arbor).
Having the cattle one mile away but with all the produce
raised at home reduced the doses by about another factor of
16
-------
four to 0.5 for all radionuclides and 0.3 mrem/yr for
iodine.
The Agency notes that both the use of a wind rose, and
placing farms in realistic locations are allowed by the
compliance procedures.
The Agency does not know what assumptions and
parameters were made to obtain the 66 mrem/yr thyroid dose
in the other study. However, note that a 66 mrem/yr thyroid
dose corresponds to an effective dose equivalent (ede) of
2.0 mrem/yr (the weighing factor is 0.03), so that it would
still be within the limit of 3 mrem/yr ede for radioiodine.
The Agency does not know the reason for the difference
between the 0.5 mrem/yr from its COMPLY run and the
University's 0.3 mrem/yr. The difference is not
significant, however, and could be caused, for example, by
the maximum sector versus the actual sector.
3.6 Basis for 40 CFR Part 61, Subpart I Standard
3.6.1 "Standard Male"
, Comment.- The EPA standards are based on the "standard
male". Why were population groups like women and children
not considered?
Response: This comment relates to the assumptions used in
the 1989 •rulemaking. As such it is not relevant to the
limited scope of this rulemaking. Notwithstanding the
above, for the sake of clarity EPA makes the following
observation:
17
-------
The risk is that from a lifetime exposure to 10
mrem/yr. Because the exposure, dose and risk are
calculated for the whole lifetime, the child's
risk is inherent. Additionally, the dose range
resulting from age dependence appears to be small
for lifetime exposures. See "Radionuclides:
Background Information Document for Final Rules",
EPA 520/1-84-022-1. The "standard man" is the
nomenclature used by the International Commission
on Radiation Protection (ICRP). It does not
presume a sex; rather, factors like body weight
and metabolism are standard.
3.6.2 Multiple Sources Not Considered
Comment: Why does the EPA not consider multiple sources?
For example, there may be a hospital, a university, and a
radiopharmaceutical company all in the same town.
Response: This comment 'relates to the assumptions used in
the 1989 rulemaking. As such it is not relevant to the
limited scope of this rescission rulemaking. Notwithstanding
the above, for the sake of clarity EPA makes the following
observation:
Because the dose drops off with approximately the
square of the distance between the source and receptor,
the effects of other, more distant sources on persons
located close to one source, are usually trivial.
18
-------
3.7 CAA Section 112(d)(9) Ample Margin Determination
Comment: The determination of safe and acceptable should be
made under the guidelines of the Vinyl Chloride decision;
that is, only health factors should be considered, not cost.
Does the NRC's ALARA principle (as low as reasonably
achievable) violate the Vinyl Chloride guidelines by taking
into account costs and benefits?
.Response; Section 112(d)(9) authorizes EPA to decline to
regulate radionuclide emissions from NRC licensees under the
CAA, provided the Agency determines that NRC's regulatory
program protects the public health with an ample margin of
safety. The legislative history of Section 112(d)(9)
provides additional guidance as to what is meant by "an
ample margin of safety" and what process the Administrator
is to follow in making that determination in a rulemaking
conducted under Section 112(d)(9). The Conference Report
accompanying S. 1630 points out that the "ample margin of
safety" finding under Section 112(d)(9) is the same "ample
margin of safety" requirement that was contained in Section
112 of the CAA prior to its amendment in 1990.
In 1989 EPA promulgated NESHAPs for radionuclides
utilizing the two step Vinyl Chloride analysis. In the
first step EPA determined an acceptable level of risk based
only on health factors and then in the second step the
Agency considered health, cost and other factors to
determine the "ample margin of safety". EPA has already
19
-------
made a determination in promulgating Subpart I that
compliance with the 10 mrem/yr dose standard protects the
public health with an ample margin of safety. See 54 FR
51654 (December 15, 1989). EPA conducted a risk assessment
in promulgating Subpart I in 1989. EPA is not revisiting
either the risk analysis or decision methodology that
supported the promulgation of Subpart I; rather, EPA is only
considering whether NRC's regulatory program will result in
meeting the 10 mrem/yr dose standard established in Subpart
I as being the level that provides an ample margin of safety
thereby rendering Subpart I unnecessarily duplicative. EPA
has determined that NRC's regulatory program, including the
10 mrem/yr constraint rule, will protect the public to the
same level as would continued implementation of Subpart I.
The determination that 10 mrem/yr ede constitutes an
ample margin of safety was made in the context of the
rulemaking promulgating Subpart I. The appropriate place to
raise concerns regarding the adequacy of the standard or
seek review of the appropriate methodology was in the
context of that rulemaking and not in this rescission
rulemaking. For purposes of the present rulemaking EPA has
used 10 mrem/yr ede, the standard found to provide an ample
margin of safety to protect the public health, as the
benchmark by which EPA evaluates the NRC regulatory program
and in an unpublished opinion the Court of Appeals for the
20
-------
B.C. Circuit approved this practice. Sierra Club, et al., v.
Environmental Protection Agency, No.95-1562 (B.C. Cir. October
22, 1996) .
4. Adequacy of NEC Program General
4.1 EPA Limit Too High
Comment: The EPA limit of 10 mrem/yr is too high. Thus, EPA
does not protect public health with an ample margin of
safety.
Response: This question is outside the scope of the present
regulatory proceedings since the Agency is not revisiting
the basis for the standard established in the 1989
rulemaking. The commenter's assertion that the EPA limit of
10 mrem/yr is too high to protect the public health with an
ample margin of safety is not relevant to the issue of
whether or not Subpart I should be rescinded. However, the
Agency does direct the commenter's attention to the language
of Section 112 of the Clean Air Act which grants the
Administrator the authority to establish emission limits at
the level which he or she determines will protect the public
health and safety with an ample margin.
4.2 NRC Limit Greater Than EPA's Limit
Comment: The NEC's limits are much greater than the EPA
limit of 10 mrem/yr, and the ALARA procedures rely on hidden
negotiations between the facility and the regulators to
establish actual emission levels. The NRC uses a different
21
-------
approach than the EPA's approach of formal rulemaking
coupled with open verification of compliance.
Response: EPA rejects this comment in light of NRC's
recent promulgation of a 10 mrem/yr constraint rule
applicable to materials licensees. Additionally, EPA does
not believe the NRC approach prior to the promulgation of
the constraint rule was ineffective. The effect of the
NRC's limits and ALARA procedure was to limit the dose to
less than 10 mrem/yr as evidenced by the results from both
EPA's Random and Targeted surveys, as well as EPA's
implementation of Subpart I.
4.3 Memorandum of Understanding and Penalties for Non-Compliance
Comment: Several commenters to the 1992 proposal commented
on the Memorandum of Understanding (MOU) between EPA and
NRG. Commenters were concerned that the 10 mrem/yr goal of
ALARA was not protective of public health. Those commenters
believed that 10 mrem/yr should be the limit and ALARA
should reduce the dose below that. Some commenters believe
that the MOU does not provide for rigorous emissions
testing. Commenters argue that the MOU does not define
penalties for non-compliance. They believe that penalties
for non-compliance are -more strict under the Clean Air Act
than they are under NRC's imprecise, negotiated, undefined
penalty procedures.
Response: EPA is not responding to the comments concerning
the MOU because the Agency believes such comments were
22
-------
addressed by NRC's promulgation of a constraint rule. The
constraint rule requires a licensee- which estimates or
measures a dose to the nearest resident from air emissions
greater than 10 mrem/yr, to report the dose to NEC in
writing within 30 days. The report must also describe the
corrective steps the licensee has taken or proposes to take
to ensure that the .10 mrem/yr constraint is not exceeded
again. A notice of violation will be issued if a licensee
fails to report exceeding the constraint. In addition,
provisions for penalties are contained in the NRG
regulations governing licensees. In general, the penalty
for a violation is in proportion to the severity of the
violation. The Commission does provide for the licensee to
dispute the facts and/or provide evidence of mitigating
circumstances, and considers these in its final
determination of the penalty. These decisions are a matter
of public record.
4.4 EPA Failure to Justify Proposal to Rescind
Comment: The EPA has failed to justify the proposed
rescission for the following reasons:
4.4.1 EPA Statistical Analysis of Survey Results
Comment: The EPA's statistical analysis refutes the
Agency's contention that all facilities presently comply
with the NESHAP limits.
Response: The Agency disagrees. None of the doses caused
by the facilities that the Agency has evaluated is greater
23
-------
than the limits established by the NESHAP. The statistical
analysis, which is based on the deliberately conservative
assessments of the facilities included in the random survey,
indicates that it is probable that some small number of
facilities would be found to exceed the standard using the
same conservative assumptions to evaluate the doses they
cause.
Comment: The COMPLY computer model used to calculate
compliance with Subpart I conservatively overestimates
doses, particularly where the source and the receptor are in
the same building. A more realistic dose calculation and
statistical extrapolation to the entire population would
show that no facility causes a dose that exceeds 10 mrem/yr.
Response: EPA believes that the commenter does not
understand the purpose of EPA's survey. Apparently the
commenter believes that the survey was intended to calculate
the actual dose. However, the purpose of the survey is to
determine doses in accordance with the Agency's framework
for compliance under Subpart I, i.e., COMPLY computer code.
4.4.2 EPA Survey
Comment: The EPA's assessment of the facilities does not
account for possible errors in the inputs used to estimate
doses. Given the number of facilities estimated to have
emissions that result in doses approaching the NESHAPs
limits, the EPA's failure to assure that all estimates are
24
-------
accurate simply underscores the conclusion that numerous
facilities are causing unacceptable doses.
Response: The Agency agrees that at facilities where
estimated doses approach the standard, great care must be
exercised to assure that the impacts are not underestimated.
However, the Agency does not agree that it has failed to
exercise this care. Analysts who performed the assessments
scrutinized the values assigned to input parameters prior to
running the COMPLY code, and the code itself has been
designed to flag parameter values that are outside expected
ranges. Finally, an independent quality assurance check was
made of each evaluation. In instances of uncertainty, the
Agency's policy is to assign the more conservative value
(i.e., the higher dose) to assure that the resulting
estimate does not underestimate dose.
Comment: The conclusions that the EPA has reached
concerning the current levels of exposure rely on data
generated primarily by licensees, and the EPA cannot rely on
licensees to provide unbiased data. Neither can the EPA
rely on the MOU that establishes a goal of 10 mrem/yr ede,
as licensees will simple reclassify the maximally exposed
individuals as radiation workers to meet the goal. The EPA
must recognize that Dr. Gofman's work demonstrates that low
doses of radiation are not protective of public health.
Finally, the EPA should revise its risk assessment to
reflect the findings of Dr. Gofman.
25
-------
Response: The letter sent to the random survey respondents
contained the following sentence: "You should also be aware
that Section 113 [of the Clean Air Act] provides for
possible criminal sanctions for anyone who knowingly makes
any false statement, representations, or certification in a
report required by EPA. " The Agency has no reason to
believe that any of the respondents supplied false
information.
The Agency does not find the argument that persons will
simply be reclassified to circumvent the regulations
compelling.
With respect to Dr. Gofman's findings, the Agency has
reviewed his work, and does not find that his positions are
proven by the evidence. The Agency constantly reviews the
bases for its radiation risk assessments, and has made
appropriate revisions as new data and/or analyses become
available.
Comment:: The EPA's study of NRC licensed sites and the
conclusion that the facilities are below the NESHAP limit is
unconvincing. No direct measurements were made by the EPA,
it relies entirely on data provided by the licensees, and no
effort was made in the BID to evaluate whether or not the
sample size was sufficient for drawing valid statistical
inferences.
26
-------
.Response.- The EPA has already determined in the 1989 NESHAP
rulemaking that measuring emissions is not necessary in
order to determine compliance with the limits. Therefore,
it does not.believe that relying on data provided by the
licensees was inappropriate, particularly as the letter sent
to the random survey respondents contained the following
sentence: "You should also be aware that Section 113 [of the
Clean Air Act] provides for possible criminal, sanctions for
anyone who knowingly makes any false EPA." The Agency has
no reason to believe that any of the respondents supplied
false information.
The discussion in the BID noes not dwell upon the
question of whether or not the size of the random sample is
sufficient to drawing valid statistical inferences for the
simple reason that the sample size was determined a priori
on the basis of the size needed to provide 95% confidence
that the largest value represented the 99th percentile of
the surveyed facilities.
4.4.3 ALARA Is a Future Requirement
Comment.- The EPA has relied, in part, on the fact that the
NRC will impose ALARA on all its licensees by 1994. Relying
on a requirement that is not currently effective as a basis
for rescinding the NESHAP is improper.
.Response: The comment was received in response to the 1992
proposal, as such it is no longer appropriate. In its 1991
s
revisions to 10 CFR part 20, NRC codified the ALARA
27
-------
principle which had previously been guidance. Those
revisions have been effective since 1994.
4.4.4 ALARA Does Not Support Rescission
Comment: Even when the NEC's ALARA requirement goes into
effect, it cannot be found to support rescission. The NRC's
ALARA requirement explicitly places a cost-benefit
limitation on control measures that a facility must
implement, which is in direct conflict with the strictures
of the court under NRDC vs. EPA 824 F.2d 1146 (B.C. Cir.
1987) (en bane) that "safe" be determined irrespective of
cost considerations. There is no enforceable basis in the
NRC's regulations to guarantee that emissions do not result
in doses exceeding 10 mrem/yr.
Response: The issue is now irrelevant, since NRC
promulgated a constraint level rule requiring NRC licensed
facilities other than nuclear power reactors to constrain
emissions of radionuclides to the environment to 10 mrem/yr.
EPA has determined that the constraint rule provides the
assurances needed to make the finding required by CAA
Section 112(d)(9). EPA has determined that the NRC's
regulatory program provides an enforceable basis for
assuring doses caused by air emissions of radionuclides from
NRC and Agreement State licensed facilities are consistently
and predictably below 10 mrem/yr.
28
-------
4.4.5 Ample Margin Determination
Comment:: The EPA has failed to fulfill the second step of
the process for establishing the "ample margin of safety"
from these facilities. Nothing in the rulemaking record
indicates that the EPA considered requiring additional
emissions reductions beyond those associated with the 10
mrem/yr standard. As the discussion of the M.D. Anderson
Medical Center and the Washington University Medical Center
indicates, reducing iodine releases by installation of
charcoal filters can be accomplished. Yet, the EPA did not
address whether or not facilities such as the NIH or Johns
Hopkins should employ such systems. Without performing such
an analysis, the EPA is in no position to determine whether
or not the NRC's program provides the requisite level of
protection.
Response: The determination that 10 mrem/yr ede constitutes
an ample margin of safety was made in the context of the
rulemaking promulgating Subpart I. The appropriate place to
raise concerns regarding the adequacy of the standard or
seek review of the appropriate methodology was in the
context of that rulemaking and not in this rescission
rulemaking. For purposes of the present rulemaking EPA has
used 10 mrem/yr ede, the standard found to provide an ample
margin of safety to protect the public health, as the
benchmark by which EPA evaluates the NRC regulatory program
and in an unpublished opinion the Court of Appeals for the
29
-------
B.C. Circuit approved this practice. Sierra. Club, et al., v.
Environmental Protection Agency, No.95-1562 (B.C. Cir.
October 22, 1996).
Commentr NRC is not a competent regulatory body. In
support of this premise-, a commenter in 1992 noted NRC's
oversight of Nucrar Metals, Inc., where air emissions data
are not required to be submitted is an example of the NRC's
laxity. Commenters in 1995 noted the potential for high
emissions at low-level waste sites, the UCLA reactor and
some other facilities discussed in the BIB. These
commenters believe that Subpart I is required to assure the
public that the licensees are in compliance with the
standards.
Response: EPA conducted an exhaustive study which found
that the overwhelming number of NRC licensees subject to
Subpart I were below the standard. The Agency has no reason
to believe these results have changed. COMPLY is available
to the public and no member of the public has provided data
suggesting that the 10 mrem/yr limit has been exceeded.
Comment: EPA's implementation shows that a few licensees
reported emissions above EPA's standard, therefore EPA's
conclusion that virtually all facilities were below the
limit and that those few above would be in compliance with
adjustments to the COMPLY code, is not a sound basis on
which to rescind.
30
-------
Response: EPA's survey predicted that 14 facilities would
exceed the standard. The method of prediction was
deliberately designed to be conservative, i.e. designed to
over estimate the number of facilities that would exceed the
standard. EPA believes that its experience implementing
Subpart I demonstrates the validity of EPA's prediction.
Those facilities that are out of compliance are not brought
into compliance by adjustments in the COMPLY code. In most
cases, it was found that these facilities were, in fact, in
compliance, but that they had made procedural errors in
using COMPLY, or had incorrectly compiled the data entered
in COMPLY. Facilities found to be out of compliance must
undertake measures to lower their emissions sufficiently to
come into compliance. EPA believes that the few facilities
exceeding 10 mrem/yr out of the 6,000 licensees tends to
confirm the Agency's conclusions. Additionally, no facility
reported exceeding EPA's standard for calendar year 1995.
Thus, the Agency believes it is appropriate to rescind
Subpart I.
4.4.6 EPA Benchmark
Comment.- The EPA's selection of "1 in 10,000" as the
benchmark for safety has been challenged on the grounds that
it fails to satisfy the court's demand that the level chosen
must be considered "safe" regardless of cost. Until that
litigation is resolved, the EPA cannot logically determine
whether or not the NRC's program is adequate.
31
-------
Response: The Agency disagrees with both contentions. With
respect to the issue of the appropriate benchmark, the
determination that 10 mrem/yr ede constitutes an ample
margin of safety was made in the context of the rulemaking
promulgating Subpart I. The appropriate place to raise
concerns regarding the adequacy of the standard or seek
review of the appropriate methodology was in the context of
that rulemaking and not in this rescission rulemaking. For
purposes of the present rulemaking EPA has used 10 mrem/yr
ede, the standard found to provide an ample margin of safety
to protect the public health, as the benchmark by which EPA
evaluates the NRC regulatory program and in an unpublished
opinion the Court of Appeals for the B.C. Circuit approved
this practice. Sierra Club, et al., v. Environmental
Protection Agency, No.95-1562 (D.C. Cir. October 22, 1996).
With respect to determining whether NRC's program is
adequate, CAA section 112(d)(9) provides the authority for
EPA to rescind Subpart I. Congress was clear that EPA's
regulatory program, in this case Subpart I, provides the
benchmark against which the Agency determines whether the
NRC program protects the public with an ample margin of
safety. EPA is rescinding Subpart I because the Agency has
determined that the NRC regulatory program protects the
public health with an ample margin of safety.
Comment.: The EPA's "dose standards" are illegal under
Section 112. Since they will not survive judicial review,
32
-------
the EPA cannot rely on the NRC's dose standards to provide
equivalent protection. ^
.Response: Again, the determination that 10 mrem/yr ede
constitutes an ample margin of safety was made in the
context of the rulemaking promulgating Subpart I. The
appropriate place to raise concerns regarding the adequacy
of the standard or seek review of the appropriate
methodology was in the context of that rulemaking and not in
this rescission rulemaking.
Comment: The limit on air emissions of radionuclides should
be reduced to 5 mrem/yr.
Response: This question is outside the scope of the present
regulatory proceedings since the Agency is not revisiting
the basis for the standard set in the 1989 rulemaking. EPA
believes that it is apparent from the language of Section
112(d)(9) that where the EPA has already specifically
determined what level of emissions must be achieved to
provide an "ample margin of safety" that level is the
benchmark by which EPA must evaluate the adequacy of the NRG
program. EPA specifically found when it promulgated Subpart
I in 1989, that 10 mrem/yr would provide the requisite
"ample margin of safety" to protect the public health. The
Court of Appeals for the B.C. Circuit recently approved this
interpretation in an unpublished opinion. Sierra Club, et
al., v. Environmental Protection Agency, No.95-1562 (B.C.
Cir. October 22, 1996).
33
-------
5. EPA's Background Information Document
5.1 Source Terms
Comment: The Background Information Document (BID) should
make allowance for the source terms used to evaluate the
dose from the various facilities having been based upon
information supplied by the facilities in the surveys and
not on emission monitoring that is known to be accurate and
precise.
Response: If the contention is that some of the respondents
in the survey supplied false information, then the Agency
notes that the letter requesting the information under
Section 114 of the Clean Air Act points out that there are
criminal sanctions for those knowingly supplying false
information. The Agency has no reason to believe that any
respondents supplied false information.
If the contention is that all facilities should have
emissions monitoring, the Agency notes that only a small
portion of the licensees are able to monitor their
emissions. Most of the facilities (e.g., hospitals) do not
have definite emission points. In developing Subpart I, the
Agency expended considerable time and effort to develop
release fractions and emission control reduction factors for
estimating airborne emissions. •
5.2 Procedures
Comment:: The EPA did not follow its own procedures in
calculating the dose from the Johns Hopkins incinerator;
34
-------
that is, assume 100 percent release when the temperature is
above 100 C.
Response; The assumption of 100 percent release at elevated
temperatures is the pre-approved EPA method for estimating
the source term. In establishing 100 percent as the pre-
approved release fraction, the Agency was constrained by the
need to establish a value that would encompass the physical
properties of all radionuclides used by NRC licensed
facilities, even though most radionuclides are not volatile
at 100 C. In evaluating likely doses from the John Hopkins
incinerator, the EPA relied on measurement data obtained
from other facilities to estimate the likely release
fractions. This methodology is fully appropriate for the
EPA purposes. Had the facility itself used such a
methodology, it would have required the prior approval of
the EPA.
5.3 Estimate of Number of Facilities Expected to Exceed NESHAP
Comment: Two commenters disagree with the statement in the
BID that "—EPA expects that 16 facilities out of
approximately 6,000 cause doses in excess of the NESHAP
standard." These commenters are confident that had the EPA
surveyed all of the facilities, they would have found that
none have doses greater than the standard.
Response: The estimate in the BID was 14, not 16,
facilities above the standard. The Agency agrees, however,
that the estimate of 14 is based upon a statistical
35
-------
extrapolation and does not represent a certainty that there
are 14.
5.4 Radiation Dose at the Fence Line
Comment:: The BID should include estimates of the radiation
dose at the fence line. In the future persons could move
closer than the current closest resident, school, or
business.
Response: This question is outside the scope of the present
regulatory proceedings since the Agency is not revisiting
the basis for the standard established in the 1989
rulemaking. Notwithstanding the above, for the sake of
clarity EPA makes the following observation:
the estimate of .the radiation dose to the
closest resident, school, or business was an
attempt to determine the dose to any member
of the public as specified by Subpart I.
Subpart I does not require fence line doses,
only doses at currently-occupied locations.
If someone should move closer, then he or she
would then constitute the closest resident,
school, or business.
5.5 NE.C Inspection and Enforcement Procedures
Comment; The BID should describe in detail the NRG
inspection and enforcement procedure.
Response: See Appendix A of the Draft and Final Background
Information Documents. Copies are included in the docket
36
-------
for this rulemaking (A-92-50, II.-B-1, V-B-1) . See also
NRC's Enforcement Policy (A-92-50, IV-B-5).
5.6 Recommended Changes to the BID
Comment: The NRC supports EPA's decision to rescind
Subpart I for NRC and Agreement State licensed facilities.
The Commission did provide extensive specific comments on
the 1992 (draft) BID seeking minor corrections or
clarifications.
Response: These comments have been closely reviewed and
appropriate changes made to the final BID.
6. Prevention of Future Increases in Emissions
Comment: In the 1992 proposal, EPA cites two regulations
governing radionuclide emissions: 40 CFR 190 and 10 CFR 20.
Neither of these rules, either singly or in combination, is
sufficient to prevent future increases in radionuclide
emissions.
Response: This comment was received in response to EPA's
1992 proposal. The issue raised is now irrelevant since NRC
has adopted a 10 mrem/yr constraint level rule.
Furthermore, EPA has concluded that the constraint level
rule supports the requisite determination for rescission
under CAA Section 112(d)(9). The final promulgation of the
constraint level rule assures that radionuclide emissions by
the affected licensees will be consistently and predictably
below a level which would result in doses exceeding 10
mrem/yr, and that NRC can require an individual licensee who
37
-------
exceeds the 10 mrem/yr level to take corrective actions to
reduce emissions.
Establishment of a Medical Advisory Board
Comment: Two commenters proposed the establishment of a
medical advisory board within the EPA.
Response: The EPA does not believe that the establishment
of such an advisory board is necessary at this time.
Impact of Rescission on 40 CFR Part 61 Subpart H
Comment:: Rescission of Subpart I could lead to the
rescission of Subpart H.
Response: The Agency fails to appreciate the basis for this
concern. Promulgation of Subpart H was, in part, premised
on the understanding that the DOE was self-regulated. In
addition, EPA does not have the statutory authority to
rescind regulations which apply to DOE facilities under
Section 112(d)(9) of the CAA.
Comment; One commenter requests that gaseous diffusion
plants be included under Subpart I rather than Subpart H.
Response: This request is outside the scope of the present
regulatory proceeding since the Agency is not revisiting the
standards established in the 1989 rulemaking. Irrespective
of the above, for the sake of clarity EPA makes the
following observation:
the Agency notes that the Energy Policy Act
of 1992 (42 U.S.C., Sections 1201-1608)
provides for the formation of the United
38
-------
9.
States Enrichment Corporation (USEC), for
purposes of transforming the Federal uranium
enrichment program into a commercial venture.
The USEC will be a wholly owned Government
corporation and an agency of the United
States, subject to all Federal environmental
laws. The Energy Policy Act further provides
that the USEC will lease but not take
ownership of the Paducah and Portsmouth
Gaseous Diffusion plants. 40 CFR Part 61,
Subpart H applies to "any facility owned or
operated by the Department of Energy..." 40
CFR section 61.90. Thus, the Agency believes
both plants remain subject to Subpart H,
since the Department of Energy will retain
ownership of these plants.
Major Sources
Comment:: It is not clear whether facilities that emit
radionuclides will be considered major sources that require
operating permits. Until the EPA establishes the definition
of major source for radionuclides (required by Title V of
the CAA and 40 CFR Part 70) it is premature for EPA to
rescind Subpart I.
Response: The comment is not clear and is beyond the scope
of the present rulemaking. While the Agency has
contemplated promulgating a definition of major source for
39
-------
radionuclides under the CAA by a separate rulemaking, it has
not done so. The Agency disagrees with the assertion that
such a definition is required by Title V of the CAA or the
Part 70 regulations.
10. 1990 CAA Amendments Require EPA/Surgeon General Report
Comment; In the 1990 amendments to the CAA, Congress
required the EPA and the Surgeon General to report by 1996
on their findings on health effects and risks of emissions
and background concentrations of hazardous air pollutants
(radionuclides are considered hazardous). Therefore, the
Agency's rescission of Subpart I is premature.
Response: The Agency does not find the commenter's assertion
compelling. While it is true that the CAA as amended in
1990 requires EPA "to investigate and report, after
consultation with the Surgeon General and after opportunity
for public comment, to Congress on remaining risk," those
provisions apply to technology based standards promulgated
under revised CAA Section 112(d). See CAA Section 112 (f).
Thus, it is not appropriate for the Agency to consider the
report in the rescission of Subpart I since the radionuclide
NESHAPs were expressly "saved" as health-based standards.
CAA Section 112(q). EPA is rescinding Subpart I because the
Agency has determined pursuant to Section 112(d)(9) of the
CAA as amended, that the NRC regulatory program, including a
10 mrem/yr constraint level, protects the public health with
an ample margin of safety.
40
-------
11. Radioactive Waste
Comment: A few commenters are concerned with transporting
radioactive waste by rail and truck because it could lead to
accidents and no training or funds are being provided to
emergency personnel. Some commenters are also concerned
about the siting of radioactive waste disposal facilities.
Response: The Agency understands that transport and
disposal of radioactive waste are concerns. However, the
concerns as related to rescission of Subpart I are neither
clear nor compelling. Additionally, EPA believes that the
comments are beyond the scope of the present rulemaking.
Subpart I, limits radionuclide emissions from NRC licensed
facilities (stationary sources) to the ambient air to that
amount which would cause any member of the public to receive
in any year an effective dose equivalent no greater than 10
mrem. EPA is rescinding Subpart I because the Agency has
determined that the NRC program, which includes the 10
mrem/yr constraint level, protects the public health with an
ample margin of safety.
With respect to accidents which occur during the
transport of radioactive waste by rail or truck, the Agency
notes that during the transport of radioactive waste by rail
or truck, Department of Transportation (DOT) radioactive
material packaging and transportation regulations diminishes
the probability of an airborne release of radioactive
material exceeding the 10 mrem/yr level. DOT does provide
41
-------
training for emergency personnel with the publication and
distribution of the North American Emergency Response
Guidebook. It is unlikely that a rail or truck accident
involving radioactive waste would result in an airborne
release of radioactive material.
12. Chernobyl Data
Comment: Given the recent data from Belarus indicating an
increase in thyroid cancer in children resulting from
radiodine releases by the Chernobyl accident, EPA should not
rescind Subpart I and should possibly consider lowering the
dose limit.
Response: This comment is beyond the scope of the present
rulemaking, since the Agency is not revisiting the standard
established in the 1989 rulemaking. The limit in Subpart I
was established pursuant to an EPA policy for Section 112
pollutants first announced in the benzene NESHAP (54 FR
38044, September 14, 1989), utilizing the two-step process
outlined in the vinyl chloride decision. Natural Resources
Defense Council v. EPA, 824 F.2d 1146 (D.C. Cir. 1987). In
addition, the Agency believes it is premature to respond or
react to the Chernobyl data at this time. EPA would not act
until such time as the data has undergone the international
review process.
42
------- |