United States
          Environmental Protection
Office of
Radiation Programs
Washington, D.C. 20460
EPA 520/1-84-023-1
October 1984
&EPA    Radionuclides
          Response to Comments
          for Final Rules
          Volume  I

                                                   EPA  521-84-023-1
40 CFR Part 61
National Emission Standards for
Hazardous Air Pollutants
                           RESPONSE TO COMMENTS


                                 VOLUME I
                             October 22,  1984
                       Office  of  Radiation Programs
                   U.S.  Environmental  Protection  Agency
                         Washington, D.C.  20460

                             TABLE OF CONTENTS


     2.1  Basis for the Standard  	   3
     2.2  Dose and Risk Calculations	  25
     2.3  Control Technology 	  38
     2.4  Proposed Limits 	  39
     2.5  Implementation	  41
     2.6  Cost	  43
     2.7  Other Comments 	  44


     3.1  Basis for the Standard	  46
     3.2  Dose and Risk Calculations	  47
     3.3  Control Technology 	  48
     3.4  Proposed Limits	  48
     3.5  Implementation	  49
     3.6  Costs	  49
     3.7  Other Comments	  50


     4.1  Basis for the Standard	  51
     4.2  Dose and Risk Calculations 	  54
     4.3  Control Technology	  57
     4.4  Proposed Limits	  58
     4.5  Implementation 	  60
     4.6  Costs	  62
     4.7  Other Comments 	  63


     5.1  Basis for the Standard  	  65
     5.2  Dose and Risk Calculations 	  68
     5.3  Control Technology 	  79
     5.4  Proposed Limits	  81
     5.5  Implementation	  83
     5.6  Costs 	  85
     5.7  Other Comments	  87

                            CONTENTS (Continued)

     6.1  Basis for the Standard  	  88
     6.2  Dose and Risk Calculations 	  91
     6.3  Control Technology 	  93
     6.4  Proposed Limits	  95
     6.5  Implementation	  95
     6.6  Costs	  97
     6.7  Other Comments 	  98


     7.1  Basis for the Standard  	  100
     7.2  Dose and Risk Calculations	  105
     7.3  Control Technology 	  109
     7.4  Proposed Limits	  109
     7.5  Implementation 	  110
     7.6  Costs	  110
     7.7  Other Comments	  Ill


     8.1  Basis for the Standard  	  112
     8.2  Dose and Risk Calculations	  113
     8.3  Control Technology 	  115
     8.4  Proposed Limits	  115
     8.5  Implementation 	  116
     8.6  Costs 	  116
     8.7  Other Comments 	  116


     9.1  Basis for the Standard  	  117
     9.2  Dose and Risk Calculations 	  118
     9.3  Control Technology 	  119
     9.4  Proposed Limits	  119
     9.5  Implementation	  119
     9.6  Costs 	  119
     9.7  Other Comments 	  119

                            CONTENTS (Continued)

     10.1  Basis for the standard  	 120
     10.2  Dose and Risk Calculations 	 123
     10.3  Control Technology	 123
     10.4  Proposed Limits 	 123
     10.5  Implementation 	 123
     10.6  Costs 	 123
     10.7  Other Comments 	 123


     11.1  Basis for the standard  	 124
     11.2  Dose and Risk Calculations 	 124
     11.3  Control Technology 	 124
     11.4  Proposed Limits 	 124
     11.5  Implementation 	 124
     11.6  Cost 	 124
     11.7  Other Comments 	 124

APPENDICES:  Index to Commenters

     A.    State, Local, and Federal Governments

     B.    Members of Industry

     C.    Members of the Public


     In 1977, congress amended the Clean Air Act  (the Act)  to address
airborne emissions of radioactive materials.  Before 1977,  the emissions
were either regulated under the Atomic Energy Act or unregulated.  Section
122 of the Act required the Administrator of the Environmental Protection
Agency, after providing public notice and opportunity for public hearings,
to determine whether emissions of radioactive pollutants cause or
contribute to air pollution that may reasonably be anticipated to endanger
public health.  An affirmative determination was made, and  EPA published a
Federal Register notice on December 27, 1979, listing radionuclides as
hazardous air pollutants under Section 112 of the Act.

     On April 6, 1983, EPA published in the Federal Register (48 FR 15076)
proposed standards for certain source categories of radionuclides.  At the
same time, EPA proposed not to regulate other source categories of
radionuclides.  Specific standards were proposed for sources in four
categories:  (1) Department of Energy facilities, (2) NEC-licensed
facilities and non-DOE Federal facilities, (3) underground  uranium mines,
and (4) elemental phosphorus plants.  EPA did not propose standards for
uranium fuel cycle facilities, uranium mill tailings, management of
high-level waste, low-energy accelerators, coal-fired boilers, the
phosphate industry, and other extraction industries.  At the time of
proposal, it was thought that these nine categories were all that
potentially released radionuclides to air at levels that could warrant
regulation.  In addition to giving notice of EPA's proposed actions,
comments were solicited on all aspects of the proposed rulemaking with
particular emphasis on the general methodology used to prepare the
standards, the standards themselves, and the determinations not to propose
standards.  Comments could be made in the form of written entry to Docket
No. A-79-11 opened for this express purpose and/or in the form of
testimony given at public hearings on the matter.

     Informal public hearings were held on April 28 and 29, 1983, in
Washington, D.C., and the record was held open an additional 30 days for
submission of written comments to the docket.  In response  to numerous
requests, an additional 45-day period was granted to receive comments and
a second informal hearing was held on June 14, 1983, in Denver, Colorado
(48 FR 23665).

     Copies of the written comments and transcripts of the hearings are
available for inspection and copying at EPA's Central Docket Section, West
Tower Lobby.  Gallery One, Waterside Mall, 401 M Street S.W., Washington,
D.C. 20460.

     This document summarizes major concerns and issues arising from
written and oral comments on the proposed rulemaking, as well as EPA's
response to these.   Each commenter is identified by a letter and number
after the comment.   The response to the comment then follows.  In the

interest of clarity and economy, some comments are paraphrased, and some
closely related comments are combined.  A list of the commenters and their
identification numbers is given in Appendices A, B, and c.


     2.1  Basis for the Standard

     Comment 2.1.la:  Notwithstanding Section 122 of the Act and the EPA's
draft Toxic Air Pollutants Policy, the EPA did not seriously consider
regulatory alternatives when it listed radionuclides under Section 112.
Thus, in making this proposal, the EPA failed to consider regulating
emissions from DOE and NRC-licensed facilities under its general authority
to issue Federal radiation protection guidance.  Similarly, for uranium
mines and elemental phosphorus plants, where the risks are to small,
close-in populations, the EPA fails to consider that regulation by State
or local governments might be more appropriate.  (1-53)

     Comment 2.1.1b:  The listing and the proposal are in violation of the
Clean Air Act, which requires under Section 122 that the EPA finds a
significant risk before imposing any regulations.  Sections 108, 111, and
112 must then all be considered.  Further, to invoke Section 112, the
presence of an extreme hazard to public health must be established.
Nowhere in the record can a discussion, in compliance with the principle
of reasoned decisionmaking, be found to justify the EPA's course of
action.  (I-3b, I-20a, 1-53)

     Comment 2.1.1c:  Under Section 122, the EPA is required to show that
a pollutant poses a significant risk.  Under the rule established in Ethyl
Corporation v. EPA, adopted by Congress in the 1977 amendments to the Act,
to constitute a significant risk, a source category "must make more than a
minimal contribution to total exposure" and there must be "a significant
increment to the total human burden."  All emissions of radionuclides from
industrial processes contribute less than 1 percent of the population dose
from radon.  Even emissions from uranium mining, which has the largest
emissions of any source category considered, have been shown by the EPA
and the NEC to present an insignificant incremental risk.  Thus, there is
no basis for applying any regulatory program under the Clean Air Act.
(I-3b, I-4b, I-20a)

     Comment 2.1.Id:  When determining the significance of risk under the
scope of Section 122, Congress did not intend the carcinogenicity of a
substance to be the determining factor.  It was understood that each of
the four unregulated pollutants identified in Section 122 were
carcinogenic.  The determination:

     a.  must be based on a consideration of the levels of exposure
         actually occurring or reasonably anticipated to occur;

     b.  should be made with respect to subcategories for the named

     c.   must be based on a source category basis, considering the
         incremental addition to the existing burden in the atmosphere;

     d.   should consider all available relevant information and proceed only
         after opportunity for public comment and hearing.  (I-3b, I-20a)

     Comment 2.1.1e; By use of the phrase "which may reasonably be
anticipated" in Section 122 (as well as 108, 111, and 112), Congress imposed
on the EPA the burden of reasoned decisionmaking with respect to both  its
endangerment determination and its listing decision.  Under this rule, the EPA
"must make plain its course of inquiry, its analysis, and its reasoning."  The
EPA has failed to do this:

     a.  Nowhere in the listing or the proposal does the EPA "make plain its
         course of inquiry, its analysis, and its reasoning."

     b.  Nowhere does the EPA evaluate the risks in terms of comparative
         assessment of risk, despite congress1 view that this is necessary.

     c.  The Agency has not provided a full explanation of all assumptions or
         an estimate of the margin of error in the dose and risk estimates.

     d.  The EPA has not evaluated the relative or incremental doses or risks.

     e.  The EPA gives no consideration to the risks associated with emissions
         at the levels occurring or whether the risks may be due only  to
         certain subcategories of radionuclides.  (I-3b, I-20a)

     Comment 2.1.If;  The EPA's position is contrary to law under the  Clean
Air  Act.   In Ethyl Corporation v. EPA supra, the court held that the EPA could
regulate incremental contributions to a problem only if the incremental
contributions were a significant source of overall exposure.  Thus, since the
sources  the EPA proposes to regulate are negligible with respect to the
background and unregulated sources, the EPA's listing of these under Section
112  is contrary to  law.   (1-47)

     Comment 2.1.1g;  Emissions already regulated under the Atomic Energy
Act  clearly pose an insignificant danger at current and reasonably expected
emission levels.  The permitted emission levels have been found adequate "to
protect  life, health and safety."  Crowther versus Seaborg, 312 F. Supp.
1205,  1234-35  (Denver, Colorado 1970).  Similarly, Dr. Sinclair of the
National Council on Radiation Protection and Measurements  (NCRP) has
indicated  that such levels are safe.  Thus, the EPA should not list
radionuclides emitted by Atomic Energy Act  licensees for regulation.   (1-47)

     Comment 2.1.1h;  The risk of adverse health effects from radionuclides
is insignificant when compared to the risk of lung cancer generally and the
risk from commonplace events (i.e., drowning or accidental poisoning).
     Comment 2.1.11;  The EPA has failed to establish that emissions from
uranium mining or milling sources either present a significant risk within
the meaning of Section 122 or result in the consequences specified under
Section 112.  In fact, the EPA concluded, in its recent report to Congress,
that these potential effects [of active and inactive mines] are not of
sufficient magnitude to warrant corrective measures.  The Nuclear Regulatory
Commission's Atomic Safety and Licensing Appeals Board has also concluded
that there is serious doubt as to whether fuel cycle emissions have a
significant impact upon human health, and that this doubt is reinforced by
the fact that the emissions are vanishingly small compared to the
fluctuations in natural radon from place to place.  (I-3b, 1-22, 1-23)

     Comment 2.1.1J;  Congress did not intend processors and users of
materials that might contain trace quantities of hazardous pollutants to be
regulated under the CAA.  (I-3b)

     Response (Comments 2.1. la through j):  Commenters state that when EPA
determined radionuclides are a hazardous air pollutant (44 FR 76738) it made
the following errors:  (1) did not consider properly other alternatives and
made other procedural errors, (2) considered risks to be significant when
they were not, (3) did not explain risk estimates in sufficient detail, and
(4) did not make plain its course of inquiry.  The commenters imply that if
the determination is in error, then the proposed regulation should be

     EPA believes that its determination under Section 122 of the Act that
radioactive materials will cause or contribute to air pollution which may
reasonably be anticipated to endanger public health was correctly made.  It
was based on overwhelming evidence that radionuclides are carcinogenic.
There is an increase in risk of many kinds of cancer when people are
irradiated due to the presence of radionuclides in the environment.  Also,
it was based on the observations that radionuclides are used in large
quantities and that some facilities emit significant quantities of
radionuclides into air, with a potential for much greater releases.  As EPA
explained in the original notice, these reasons are sufficient to make a
determination there is endangerment to public health.

     Comparisons with the risks due to natural background radiation are not
relevant.  (See response to Comment 2.1.11.)  Also, comparisons of potential
risks due to emissions with the risks of cancer or other kinds of risks such
as the total risk of cancer, are inappropriate.  Commenters do not
demonstrate that considerations of such high risks lead to conclusions with
respect to risk levels that are protective of public health with an ample

margin of safety.  For example, given the large societal efforts to reduce
the risk of cancer further, there is reason to judge this risk unacceptable.
The Agency cannot then argue a certain fraction of these risks are
acceptable.  Finally, the Agency has not defined a level of risk that is so
low that everyone can agree no effort is needed to reduce it even when the
cost to do so is small.  The Agency concludes that the risks due to
radionuclides, considered as a class, are significant and that large
quantities are in use so that additional significant risks in the future are
possible unless there is constant vigilance on the part of users and the
regulatory agencies.

     Section 122 requires a generic determination based on the hazard from a
specific kind of air pollutant, without regard to other statutory
authorities.  When this decision was made in the affirmative, there is a
choice as to the section of the Clean Air Act under which to regulate
radionuclides.  EPA believes that the decision to use Section 112 was
correct because the carcinogenicity of radionuclides fits the definition of
a hazardous air pollutant.

     EPA believes it has explained its risk estimates in sufficient detail.
The Background Information Document and the various Federal Register Notices
make plain EPA's computer codes, EPA's assumptions for each source of
radionuclide emissions, and the basis for calculating risk estimates.  All
procedures are documented and can and have been reproduced by others.  All
procedures have been reviewed and generally accepted in the scientific

     The Agency has made plain on numerous occasions its position that risk
estimates are not exact calculations but are presumed accurate only to an
order of magnitude.  EPA makes no pretense that its risk estimates are so
accurate that they become the sole basis for our decisions.  Rather, risk
estimates are projections which, along with other considerations, have led
to a determination of hazard with respect to radionuclides.

     Comment 2.1.2a:  There is considerable evidence that Congress intended
that non-threshold carcinogens be regulated under Section 112.  (P-3b)

     Comment 2.1.2b:  Radionuclides should not^be regulated under Section
112.  This portion of the Clean Air Act is to be used in cases where the
pollutant in question is extremely hazardous and poses a significant risk to
human health.  (I-3a, I-3b, I-4c, I-20a, I-20b, 1-31, I-4"7. 1-49)

     Comment 2.1.2c:  Section 112 of the clean Air Act is an unconstitu-
tional delegation of the legislative function to an administrative agency.
The only direction the EPA was given in Section 112 was to establish
standards that provide "an ample margin of safety."  The terra is vague and,
for pollutants with no threshold effects, does not offer the guidance that
Congress must provide.  The only way this constitutional issue may be

avoided is if Section 112 is implemented as Congress intended, that is, only
in those limited instances where extreme health hazards are shown.  (I-3b)

     Comment 2.1.2d:  The commenter strongly disagrees with any proposed
attempt to delist radionuclides.  Based on evidence that radionuclides cause
cancer and genetic damage, the EPA correctly concluded that radionuclides
are hazardous air pollutants.  Section 112(b)(l)(B) states that the
Administrator may not remove a substance from the list "unless he finds ...
that such a pollutant is clearly not a hazardous air pollutant."  Mere
assertions that risks are not significant or the presentation of limited
data will not suffice.  (P-17)

     Comment 2.1.2e:  It does not appear that the EPA followed the policy
set forth in the Federal Register of October 10, 1979 when setting the
proposed radionuclide standards.  We understand that the EPA may have based
the proposed rule on a new toxic air pollutant policy that is still under
development and has not been published.  (I-lb)

     Comment 2.1.2f:  Section 112 of the Clean Air Act does not recognize
the problems inherent in establishing standards for carcinogens, for which
no threshold exists.  The proposed EPA airborne carcinogen policy indicates
that regulation of these materials should include consideration of risk.

     Comment 2.1.2g:  It was proper of EPA to list radionuclides as
hazardous air pollutants under Section 112 of the Clean Air Act, and any
derivative standards must insure an ample margin of safety.  (P-3a)

     Comment 2.1.2h:  The Atomic Industrial Forum concurs that standards are
not needed under Section 112 of the Clean Air Act for uranium fuel cycle
facilities, uranium mill tailings, and high level wastes.  (I-la)

     Comment 2.1.2i:  The EPA has offered no rationale for treating
emissions of regulated radionuclides or any subcategory of regulated
radionuclides as "extremely hazardous;" therefore, regulation under Section
112 is not justified.  (1-47)

     Comment 2.1.2J;  The EPA's consideration of the five factors used in
evaluating each source category is consistent with the mandate of Section
112 of the Clean Air Act.  In addition, the benefits society derives from
the activity that produces the potential radiation exposure should be
evaluated.  (1-45)

     Response (Comments 2.1.2a through j):  Comments were received both for
and against the regulation of radionuclides under Section 112 of the Act.
Some commenters believe that the radionuclides are not hazardous enough to
warrant regulation under Section 112.  Other commenters agreed with listing
radionuclides under Section 112 and objected strongly to attempts to delist

     Radionuclides were listed under Section 112 of the Act in the following
way.  After providing public notice and opportunity for public hearings
(provided by 44 FR 21704, April 11, 1979), the Administrator determined
radioactive pollutants cause or contribute to air pollution that may
reasonably be anticipated to endanger public health.  On December 27, 1979,
EPA published a notice in the Federal Register listing radionuclides as
hazardous air pollutants under Section 112 of the Act (44 FR 76738, December
27, 1979).

     The listing under Section 112 of the Act was made for the following

     •  Radionuclides are considered to be carcinogenic without a
        threshold when ingested or inhaled.

     •  Radionuclides are widely used, often in very large amounts, and,
        in some cases, there is significant potential for unnecessarily
        high radionuclide emission rates into air.

     •  Radionuclide emission rates at current levels cause significant
        levels of risk in some cases.

     •  Radionuclides are emitted to air from many different facilities.

     The statutory criterion for delisting radionuclides is that the
Administrator finds that "the pollutant clearly is not a hazardous air
pollutant."  This criterion has not been met.  The record supports the
listing as correct, and EPA is not delisting radionuclides.

     The Office of Radiation Programs did not base its proposed standard
on  a new toxic air pollutant policy that has not been made public.  EPA's
proposed policy for regulating hazardous air pollutants, published in the
Federal Register on October 10, 1979, is being reconsidered but has not
yet been withdrawn or replaced.  The proposed standard for radionuclides
conforms with EPA's proposed policy.

     Some commenters stated that radionuclides should not be listed under
Section 112 because EPA has not demonstrated radionuclides are "extremely
hazardous."  The statutory criterion for listing under Section 112 is
that radionuclides, in the judgment of the Administrator, cause or
contribute to air pollution which may reasonably be anticipated to result
in an increase in serious irreversible, or incapacitating reversible,
llness.  The record strongly supports the conclusion that radionuclides
meet this criterion because of their carcinogenicity and wide use in very
large quantities.  The benefits society derives from the activities
involved are irrelevant to the use of Section 112.  Under law, the
decision to list radionuclides under Section 112 was correct.

     Comment 2.1.3;  Section 111 would provide a more appropriate
regulatory program for radionuclides.  To the extent that radionuclides
present a significant health risk, the risk would appear to be possible
only in a small area adjacent to a source.  This is the condition for
which Congress intended that controls be considered under Section
lll(d).  Senate Report No. 91-1196 in fact included radioactive
substances on a list of substances most likely to be covered under
Section 111.  (I-3b, 1-49)

     Response;  Commenters state that radionuclides should be listed
under Section 111 rather than 112, principally because the risks
associated with radionuclides are localized.

     EPA disagrees with this comment.  Radionuclides are carcinogens for
which the Agency assumes there is no concentration threshold.  Hazardous
materials with this kind of serious effect are best listed under
Section 112 where costs are a secondary consideration.  Section 111 of
the Act is used primarily for stationary sources emitting pollutants for
which thresholds exist, an example being sulphur dioxide emissions.
Section 111 specifically requires the Agency to consider costs.  (Section

     Comment 2.1.4a:  Under the principle of in pari material, the EPA is
required to give effect to all statutes that govern a particular source.
Thus, under Section 112, the EPA is required to regulate sources that
meet its requirements, including uranium fuel cycle sources, high-level
waste disposal sites, and uranium and thorium mill tailings. (P-15a,

     Comment 2.1.4b:  In determining whether other regulatory schemes are
adequate to preclude the need for regulation under Section 112, the
Agency should consider:  health standards set with an ample margin of
safety without regard to technological feasibility and costs, standards
set as emission standards, the right of public access to emission data,
and the right of citizen enforcement.  None of these provisions, or their
equivalent, are found in the Atomic Energy Act, the Uranium Mill Tailings
Radiation Control Act, or the Nuclear Waste Policy Act.  Thus, standards
must be adopted under the CAA.  (P-15a)

     Comment 2.1.4c:  The EPA fails to explain why some facilities
emitting more radionuclides than some of the sources regulated under this
proposal are not subject to these standards.  (G-19)

     Comment 2.1.4d:  We concur with the EPA's decision to consider
standards on a source-by-source basis.  Such a rationale is clearly
supported by the Act.   The Act does not require CAA standards where other
regulations already exist that accomplish the same goals, nor does it
require regulation of insignificant sources to guard against possible
future increases.   (I-4a,  I-4b)

     Response (Comments 2.1.4a through d):  Commenters both agree and
disagree with EPA's determinations not to regulate specific groups of
sources that emit radionuclides to air.

     There are two separate issues here.  EPA considered both of them:

     1.  Should EPA propose regulations for all facilities emitting
detectable amounts of radionuclides?

     2.  Should EPA propose regulations under the Clean Air Act for
sources whose radionuclide emissions are controlled under other EPA
regulations, either regulations specifically for radionuclides or
regulations limiting particulate emissions, thus controlling radionuclide
emissions also?

     Radiation detection equipment is extremely sensitive.  Consequently,
every stack in the country discharges to air detectable quantities of
naturally-occurring radionuclides.  Given enough resources, even the most
minute quantities can be measured.  Those sources emitting extremely
small quantities of radionuclides have not been categorized or analyzed
by EPA because the risks to individuals and to population groups are

     Because of the large amounts of resources needed to propose and
promulgate an emission standard,  EPA disagrees with the comment that
major EPA standards such as apply to the uranium fuel cycle, uranium mill
tailings, and high-level wastes should be duplicated by clean Air Act
standards.  The Agency is aware that the use of the Clean Air Act confers
certain advantages that other authorities do not, such as the right of
suit by a member of the public.  However, it is EPA's judgment that the
effort and expense of issuing duplicate standards would be excessive
without any reasonable expectation of improved public health protection.
The public is better served if the Agency's resources are used elsewhere.
Comment 2.1.4c has been rendered moot by EPA's decision to withdraw the
proposed standards.

     Comment 2.1.5a;  The EPA was correct in rejecting a zero-risk
interpretation of Section 112.  it pointed out that the language of
Section 112 does not imply the absence of all risk and correctly
recognizes that to remove all risk would require the elimination of all
emissions, a move that in most cases would require the closing of major
industrial and utility facilities.  (I-4b, I-4c, 1-42, 1-53)

     Comment 2.1.5b;  The requirement that emission standards be
established with an "ample margin of safety" does not require a zero
emission limitation,  ection 112,  unlike Section 115, does not use the
term "prohibition."  This difference cannot be ignored.  In addition,
references to zero emissions in the legislative history are not
unequivocal.  (I-3b)

     Comment 2.1.5c:  Even if radionuclides are hazardous air pollutants,
a zero or near-zero emission standard is not required by the Clean Air
Act.  The legislative history provides little guidance in defining the
phrase "ample margin of safety."  Administrator Ruckelshaus noted that
for non-threshold pollutants, there is no way to establish a perfectly
"safe" level of exposure, but called for an approach comparing risks and
benefits.  In addition, Justice Stevens noted that safety is not the
equivalent of risk free.  (1-47)

     Response (Comments 2.1.5a through c):  Commenters generally agree
that EPA was correct in rejecting a zero-risk interpretation of the Clean
Air Act (Section 112).

     This comment raises an issue that is similar to 2.1.4, but more
extreme.  It has been suggested that the language of the Clean Air Act
requires EPA to prevent all emissions of radionuclides to air regardless
of the disruptions and cost to the industry and, thus, to reduce the risk
from radionuclide emissions to zero.

     As pointed out in the response to 2.1.4, probably every stack in the
country discharges into air minute quantities of radionuclides which
could be measured given enough resources.  These emissions cause,  in
theory, some risk greater than zero.  Thus, attempting to achieve an
actual zero level of risk is impossible.  Either EPA must reject the
possibility of implementing Section 112 for non-threshold carcinogenic
materials on a technicality, or EPA must proceed on the grounds that the
intent of Congress and of the Act does not mean literally a "zero risk."
EPA has adopted this second course which it believes to be the more

     EPA agrees with the comment that protection of the public with an
ample margin of safety is not the equivalent of risk free.

     Comment 2.1.6a:  The standards are inconsistent with the Clean Air
Act requirement that they provide an ample margin of safety.  The Federal
Register notice concedes that the standards will allow one death for
every 500 people exposed to radioactive radon from uranium mines.   (G-22,

     Comment 2.1.6b:  The EPA has adopted an approach to setting
standards that is in conflict with the health protection requirements of
the Clean Air Act.   The commenter can see no justification for a standard
that leaves people exposed to risks as high as 1 chance in 500 of
contracting cancer.   (P-17)

     Comment 2.1.6c:   The proposed standards are not adequate to protect
the public health and safety with an ample margin of safety as required
by Section 112.   The need to do this is clearly supported by the
legislative history.   Further, an actual emission limitation must be
issued unless there is overriding justification for not doing so.  in
Section 109, Congress has indicated that an "adequate margin of safety
must protect sensitive individuals."  By definition, an "ample margin of
safety" requires a more protective standard than an "adequate margin of
safety."  (P-3b, P-15a)

     Comment 2.1.6d;   The regulations should be built around a policy
that interprets "ample margin of safety" as allowing a risk no greater
than 1 in a million.   (P-15b)

     Comment 2.1.6e:   In deciding whether limitations beyond BAT are to
be required, EPA should focus not on whether emissions would otherwise be
significant, but rather on whether an unreasonably high health risk would
exist in the absence of more stringent control requirements.  (1-42)

     Comment 2.1.6f:   The proposed rules ignore the principles of good
safety regulation or practice.  Authorized levels should be set below
derived or estimated health risk levels.  Further, limiting exposure by
requiring ALARA or BAT is also good practice.  However, setting BAT
levels by regulation and varying levels is inconsistent and
inappropriate.  (1-32)

     Comment 2.1.6g;   The EPA's technology-based rationale makes the
proposed standards unauthorized and unlawful, since standards must be
based on health risk rather than best available control technology.  The
Administrator must prove (by comparative risk assessment) that current
regulations do not provide an ample margin of safety.  (G-lb, G-3)

     Response (Comments 2.1.6a through g):  Commenters seem divided on
the issue of whether EPA's proposed standards provide an ample margin of
safety.  Given that we cannot achieve a zero level risk, what level of
risk is equivalent to protection of public health with an ample margin?
Where should emission levels be set for carcinogenic materials with no

     Needless to say, EPA has not found an easy answer to this question.
The Agency examined the idea of a de minimus level of risk, one that was
so low that virtually everyone would agree that there is no need to
reduce it further because public health is amply protected at this
level.  Comment 2.1.6d is an example of this approach, and EPA would
agree that a fatal cancer risk of one chance in one million over a
lifetime is very small, probably on the order of a de minimus risk
level,  in practice,  such a level of risk would be associated (for Low
LET radiaion) with 0.05 mrem/y whole body exposure or 0.2 mrem/y to the
lung,  such a low emission limit would be disruptive in the extreme,
limiting the generation of electricity by both coal and nuclear power, as
an example.

     If the Agency reasonably cannot limit emissions either  to a zero
risk level or to a true de minimus  level, then  it must use judgment as  to
what is an ample margin of safety,  basing its judgment on various
factors.  These factors were previously discussed in the Federal Register
notice that proposed these regulations.  EPA emphasized at that time that
there were no formulas to balance such factors  in a quantitative way. but
that in the end, it came to the use of the Agency's reasoned judgment as
to what emission levels should be.

     EPA has withdrawn its proposed standards for elemental phosphorus
plants, Department of Energy facilities, and Nuclear Regulatory
Commission licensed facilities and  non-DOE Federal facilities, because  it
believes current practices limit risk to levels that are protective of
public health with an ample margin  of safety.  While individual risks for
a few facilities were at a level that might have led to a decision to
regulate if this were the only criterion for judgment, there are other
considerations, including low aggregate population risks, that argue
against the need for regulations.   It is the Administrator's judgment
that the present record does not support the conclusion that regulation
is necessary for these source categories.

     EPA considers the risks due to radon emissions from underground
uranium mines to be significant and believes action is necessary to
protect populations and individuals living near them.  EPA's most recent
estimates of the lifetime risks to  individuals  living near these mines
range from one in one thousand to one in one hundred.  The potential
exists for even higher risks in some situations, e.g., a person living
very close to several horizontal mine vents or  in areas influenced by
multiple mine emissions.  Lifetime  risks in these situations can be as
high as one in ten.  The fatal cancer risks to  the total population, both
regionally and nationally, is 5 fatal cancers/year.

     Because radon-222 is a noble gas and the volume of air discharged
through mine vents is very large, there is no practical method to remove
radon-222 from the mine exhaust air.  Adsorption onto activated charcoal
is the most widely used method for  removing noble gases from a low volume
air stream.  However, application of this method to the removal of
radon-222 from mine ventilation air at the volumes of air which must be
treated would require large, complex, unproven  systems which would be
extremely costly (i.e., at least $18-44/lb of 0303).  Therefore, it
is the Administrator's judgment that it is not  feasible to prescribe or
enforce an emission standard for radon-222 emissions from underground
uranium mines because radon-222 cannot be emitted through a conveyance
designed to capture the gas under current conditions.  Instead, EPA has
decided to begin development of work practice, design, equipment or
operational standards to control radon releases from underground uranium
mines.   An Advance Notice of Proposed Rulemaking announcing this decision
is published as a part of this rulemaking.

     Comment 2.1.7a:  The EPA's use of costs to industry to undercut
protection of the public is unjustified and unlawful.  (G-22, P-15a)

     Comment 2.1.7b;  The commenter does not believe that Section 112
gives the EPA any authority to perform cost-benefit analyses in order to
set standards.  The inadequate definition of BAT together with a poor
examination of the presence of an "unreasonable residual risk" does not
satisfy the mandate of the Clean Air Act.  The following approach is

     a.  All sources of significant amounts of radionuclide emissions
         should be subject to standards.

     b.  These standards should reflect the lowest emissions achievable
         using the most effective control technology currently in use or
         that has been demonstrated and is readily available.

     c.  If the remaining emissions are predicted to add to the lifetime
         risk of the most sensitive individual by an amount greater than
         one in a million, technology-forcing controls should be required.

     Comment 2.1.7c:  The EPA has relied on a legally flawed approach to
developing its proposed standards for all source categories.  Protection
of public health must be the primary criterion, not technological
feasibility and/or costs.  (P-15a)

     Comment 2.1.7d:  The EPA must abandon its BAT/unreasonable residual
risk approach since it is contrary to the mandate of the Clean Air Act.

     Comment 2.1.7e:  Economic, non-air impact, and energy requirements
should not be considered for standards issued under Section 112.  (G-24,
     Comment 2.1.7f:  The Act prohibits consideration of cost to the
polluter when considering the levels of control needed to protect public
health.  (P-15a, P-15b)

     Comment 2.1.7g:  Consistent "dollars per health effect averted"
criterion should be used when developing standards for the many sources
involved.  (I-lb,  1-50, 1-53)

     Comment 2.1.7h:  Under Section 112, cost may be considered both in
determining whether to regulate and in establishing the appropriate level
of control.  As the EPA has recognized, cost-effectiveness is the only
means by which standards for a non-threshold pollutant can be established.
This strategy has been used in previous applications of Section 112 and
was judicially recognized in Adamo Wrecking Co. v. United States.  (I-3b,

     Comment 2.1.71:  Basing  the  standards on what  the  facilities  can
achieve at reasonable cost  is grossly  inadequate under  the CAA.  EPA's
rejection of a cumulative population standard and/or direct  emission
limits for specific groups  of facilities  is unjustified.  (P-17)

     Response (Comments  2.1.7a through i):  Most commenters  have
questioned EPA's use of  control technology costs as a factor  in
developing the proposed  rules.  A few  recommend the use of cost-effective
or cost-benefit studies  as  a  means of  establishing  emission  levels.

     EPA agrees with comments that state  that the Clean Air  Act precludes
the use of a cost-effective or cost-benefit analysis as the  primary basis
for emission standards established under  Section 112, and the Agency has
not done so.  In developing these final actions, EPA has considered the
availability and practicality (cost) of control equipment to a degree.
It is not reasonable to  do  otherwise.  When making  judgments on the level
of emission standards or on the need for  emission standards, it is highly
desirable to know whether control technology exists or not, whether best
available control technology  is being  used, and whether the  cost of
control technology is reasonable  compared to the cost of the facility and
operating costs.  EPA believes this is within the intent and spirit of
Section 112 of the Clean Air  Act.

     The premise of cost-effective or  cost-benefit  analysis  is that the
risks to surrounding populations  can be accurately  calculated and  that
these risks can somehow  be  accurately  balanced against the cost of
control technology and the  benefits of the activity with sufficient
precision to establish an emission limit  or to demonstrate the lack of
need for an emission limit.   First of  all, it is the Agency's opinion
that such a procedure is not  allowed when developing standards under
Section 112 of the Clean Air  Act.  Congress clearly intended that public
health protection is primary  and  costs a  secondary  (some would say
impermissible) consideration.  Second, because of the large numbers of
different kinds of activities,  the varying benefits of these activities,
the varying costs and effectiveness of very complex types of control
technologies, and the uncertainty associated with risk estimates,  a
cost-benefit analysis would be extraordinarily complicated and not
accurate or precise.  Third,  it is the Agency's experience based on more
simple situations that cost-effectiveness analysis may lead  to results
that are not sufficiently protective of those persons living closest to
the plant.  While large  populations may be protected to a degree most
would find ample, those  nearest individuals would have unreasonably high
risks because cost-effectiveness  results  for populations may not require
the use of available technology.  This unwelcome result arises because
available control technology  for  radionuclides is usually not costly
compared to facility and operations costs, and so is installed routinely,
without a cost-effectiveness  analysis.  This use of available control
technology for carcinogenic materials  is  generally considered to be
prudent policy and protective  of  public health, while failure to require
such technology is inequitable.

     Protecting the public closest to the facilities considered in this
rulemaking with an ample margin of safety also provides ample protection
of populations.  Doing the reverse, only protecting populations, does not
always protect individuals, and EPA believes that both individuals and
populations should be protected with ample margins of safety.

     Comment 2.1.8a:  The EPA should make an effort to explain carefully
the rationale for its action, including the perspective with respect to
national and international protection guidance based on data developed by
groups such as BEIR, NCRP. ICRP, and UNSCEAR.  (I-lb, 1-38, P-lb)

     Comment 2.1.8b:  The NCRP/ICRP general population radiation dose
limits should be adopted as the basis for emissions limits on
radionuclides under Section 112 of the Clean Air Act.  (G-23, 1-45)

     Comment 2.1.8c:  The EPA should adopt the NCRP limits and ALARA for
all source categories.  For sources already regulated, the EPA should set
a regulatory cut-off limit of 10-30 mrem/y so compliance can be monitored
rather than calculated.  (1-38)

     Comment 2.1.8d:  National and international scientific organizations
who maintain continuing review of radiation bioeffects data conclude that
existing public dose standards remain generally valid.  (I-la)

     Comment 2.1.8e:  The EPA should base standards on scientific
evidence and the recommendations of advisory bodies such as the ICRP and
the NCRP.  This would promote public confidence and a stable regulatory
environment in which the benefits of nuclear science and technology could
be enjoyed.  (G-lb)

     Response  (Comments 2.1.8a through e):  Commenters request EPA to
make more use of radiation protection guidance developed by national and
international groups such as the NRCP and ICRP.  EPA is asked to adopt
the general population dose limits recommended by these groups as the
basis for radionuclide emission limits.   In particular, the Agency is
asked to base risk estimates on the recommendation of the BEIR-3
committee report, rather than the recommendations of the BEIR-1 committee.

     EPA agrees with the comment that it would be more appropriate if
BEIR-3 recommendations on risks due to radiation are used.  Accordingly,
risk estimates in the Final Background Information Document are based on
BEIR-3 information.  In most instances,  the risk estimates change very
little, less than 5 percent.  In some cases where the lung or the bone is
the primary organ being irradiated by alpha particles, total risk may be
reduced as much as a factor of three.  The change in risk estimates does
not change estimates of dose.

     The thrust of these comments seems to be that EPA should use 500 mrem/y
whole body, 1500 mrem/y to any organ to the highest exposed  individual as
the basis for emission limits for radionuclides.  These dose limit
recommendations are the same as Federal Radiation Council  (FRC) guidance.
Such high dose limit recommendations are generally recognized as maximum or
upper limits, not to be exceeded or even approached without  good reasons.  A
person receiving 500 mrem/y to the entire body for a lifetime would have an
increased potential total cancer risk of about one in one  hundred due to the
irradiation.  Both the ICRP and NCRP note that such risk limits are not
appropriate when exposure to radionuclides is continuous over long periods
of time, because their upper limits do not represent levels of exposure
where risks are of small consequence.

     ICRP, NCRP, and FRC guidance also specify that exposure to radiation
should be kept as low as practicable.  But such an admonition is not as
precise or enforceable as a numerical standard and does not necessarily
require or represent a low level of risk.  However, the Agency recognizes
that an emissions policy based on this guidance has led to generally low
emissions of radionculides from most facilities.

     Because of the efforts of various parties to keep emissions low, the
doses to individuals living near emission sources are generally far below
the maximum limits.  This has enabled the Agency to find that current
practice provides an ample margin of safety.

     Comment 2.1.9a:  We agree that the EPA should base standards on
absorbed dose equivalent rather than specific concentrations of radio-
nuclides.  (I-33b, P-lb, I-la)

     Comment 2.1.9b:  Dose-based standards set quantitative limits on
emissions on a continuous basis, thus complying with Section 302(k) of the
Act.  (I-4b, I-4c)

     Comment 2.1.9c:  We agree with the EPA's decision to  use dose type
standards wherever the nature of the emission makes this practical.  In the
case of radon emissions from underground uranium mines, the
air-concentration format appears acceptable.  (l-lb)

     Comment 2.1.9d:  EPA's rejection of a cumulative population standard
and/or direct emission limits for specific groups of facilities is
unjustified.  (P-17)

     Comment 2.1.9e:  The dose limit standard is not an acceptable form
under Section 112(b).  (P-15a, P-17)

     Comment 2.1.9f:  We support the EPA's use of a "dose  to person"
standard,  since it relates directly to public health.  (G-2a, G-2b)

     Comment 2.1.9g;  The proposed dose limit standard is not an acceptable
form under Section 112(b).  (P-15a, P-155)

     Comment 2.1.9h;  The use of both whole body and organ dose limits runs
counter to the ICRP concept of an effective whole body dose.  The current
scientific thinking in this area advocates a committed, effective dose
equivalent as the proper standard.  (G-2b, G-16, I-lb, 1-31, 1-40, 1-47)

     Response (Comments 2.1.9a through h):  Most commenters agreed that an
indirect emission limit in the form of a dose equivalent limit was a
reasonable and appropriate form for the final rules.  Other commenters
disagreed, arguing that a dose equivalent limit was not acceptable.

     EPA still believes that a dose-rate limit would be proper.  However,
since EPA is withdrawing the proposed dose-rate standards, these comments
are not applicable.

     Comment 2.1.10a;   The rationale for the proposed rulemaking is
particularly unclear and obscure and appears ill-advised and perhaps
detrimental to the broad interests of society.  (P-lb)

     Comment 2.1.10b:   Although we agree with the conclusions the EPA
reached regarding coal-fired boilers,  phosphate mining, and other extraction
industries, the EPA's  general analysis regarding radionuclides is not in
accordance with the requirements of the Clean Air Act and does not consider
all relevant information.  Cl-49, 1-53)

     Response (Comments 2.1.10a and b);  Commenters question the data base
and rationale for the  proposed rule,  implying EPA did not consider all
relevant information.

     EPA believes its  data base and rationale were sufficient to support the
proposed rules and are sufficient to support the final actions.  Considerable
additional information, in particular  for uranium mines and elemental
phosphorus plants, has been received from commenters and from additional EPA
studies.  The information is considered under responses to comments on the
appropriate specific rules.

     Commenters should be aware that EPA was directed by the Court to
propose standards based on information at hand.   The Court was of the
opinion EPA had sufficient information to propose standards.

     EPA believes, and its Science Advisory Board agrees,  that EPA has
gathered the appropriate scientific information needed for a risk assessment
in a technically proficient manner and that this data and information was
used to develop scientifically defensible approaches for modeling the
transport of radionuclides through the environment from airborne releases,
in calculating doses received by persons inhaling or ingesting this

radioactivity, and  in estimating the potential concern and genetic  risks of
the calculated doses.  The final Background Information Document has been
greatly modified to explain these procedures in more detail.

     Comment 2.1.1la:  The EPA has ignored the data base associated with
both the exposure to and harm from natural background radiation.  These data
provide a rational, defensible basis for establishing the "ample margin of
safety" required by Section 112.  This information shows that not only is
the proposed emission limit within the natural variation in background
levels, but that the calculated resulting harm is indistinguishable from
that occurring in the absence of emissions.  (G-24, I-lb, 1-38, 1-50)

     Comment 2.1.lib:  The risk resulting from emissions from the sources
that EPA is proposing to regulate is not significant because the dose is
small compared to natural background levels.  (I-lb, l-3b, 1-38, 1-47. 1-49,

     Comment 2.1.lie:  It is unreasonable to calculate incremental
population doses to arrive at a number of projected deaths when the
constituent individual doses are within the noise of background
variability.  (1-38)

     Comment 2.1.1Id:  The proposed limit is a small fraction of the
variation in natural background, cannot be measured, and represents a
vanishingly small level of risk (between one in six million to one in forty
million using the BEIR-3 estimates).  It makes little sense to impose
restrictions that are much smaller than the risks encountered in everyday
living.  (G-16, I-lb, 1-31, I-33a, I-33b, P-lb)

     Comment 2.1.lie:  The proposed limits are less than the variation in
natural background  radiation levels.  There is little data to support any
change in cancer risk between high and low background areas.  Thus, there is
little reason to assume the proposed regulations will have any beneficial
effect upon public  health.  (G-16, I-lb, 1-17, I-33a, I-33b, P-lb)

     Comment 2.1.1If;  The EPA did not compare risks due to radionuclides
with natural background risks.  This implies a rejection of comparative risk
analysis that is insupportable.  (1-47)

     Response (Comments 2.1.1la through f):  Commenters question the need
for the standard when dose to people is small compared to the dose due to
the natural radiation background.  They note that everyone is exposed to a
natural radiation background of about 100 mrem/y to all organs of the body,
with additional significant exposures of some lung tissue to radon decay
products; these exposures vary greatly with location.  This natural
background having always been with us, they argue such exposure must be at a
level that is protective of public health with an ample margin of safety.
and, therefore,  standards less than this are unreasonable.

     EPA disagrees with this argument.  Between 1 and 3 percent of fatal
cancers have been attributed to the natural radiation background.  This risk
is significant (approximately 2 in 1000).  EPA has concluded that the level
of risk associated with background radiation should not be judged acceptable
just because it is unavoidable.

     The natural radiation background is not a good benchmark for hazardous
material emission limits that are intended to be protective of public health
with an ample margin of safety.  Nor is an arbitrary fraction of the natural
radiation background a useful rationale.  Rather, several factors were
considered by EPA in developing its proposed and final standards.

     Comment 2.1.12a:  There is no justification for refusing to consider
the total risk posed by radionuclides and dozens of other hazardous
materials present in the same particulate emissions and for considering only
average risk when a substantial number of plants may be much more
dangerous.  (P-17)

     Comment 2.1.12b:  The EPA's failure to evaluate the impact from all
sources in relative proximity (80km) to one another is a fatal flaw in the
proposed standards.  (P-15a, P-15b)

     Comment 2.1.12c:  The words "contribute to" in the Clean Air Act
Amendments of 1977 require the EPA to address the cumulative effects of all
sources of a pollutant when setting a standard.  (P-15b)

     Comment 2.1.12d:  We can identify no provision for the assessment and
associated regulation of multiple-source contributions to the dose for
near-site individuals or populations.  (1-34)

     Response (Comments 2.1.12a through d):  Commenters want EPA to consider
the total risks posed both by radionuclides and other hazardous materials
when present in the same particulate emissions.  Also, EPA should consider
the collective impact of all sources of radionuclide emissions within a
small region.

     EPA tends to agree with the first part of this comment, which appears
directed primarily to coal-fired boilers.  In theory, it would be best to
consider the need for particulate controls for coal-fired boilers based on
the total risks caused by all the hazardous constituents in particulates,
not just radionuclides.  However, the current state of knowledge does not
enable EPA to estimate this total risk.  For other reasons, EPA is limiting
particulate emissions to low levels by requiring best available technology
on new boilers.  This limits the risks due to radionuclide emissions to low
levels also.  But it is not possible at this time to estimate risks due to
combinations of radionuclides, other chemicals, and the particulates

     The  rulemaking  record  does not contain  information on  the  frequency  of
situations  in which  an  individual  is affected  by multiple sources.   Based on
available information about major  sources, it  appears  that  such situations
are rare.   For  those rare situations which may occur,  if one  assumes that
multiple  sources  increase the  risk by a  factor of  two, EPA's  decisions would
not be different.

     Comment 2.1.13a:   The  proposed regulations should be revised because
they are  not based on the best scientific information  available and  because
they will inevitably increase  the  fears  of people  with only a little
knowledge of the  effects of radiation.   (G-23, 1-38, P-18a)

     Comment 2.1.135:   In an area  where  misinformation and misunderstanding
are prevalent,  the inconsistent application  of numerical limits, together
with the  establishment  of unnecessarily  restrictive  limits, can only magnify
the problem of  radiation exposure  in the eyes  of an  already confused
public.   (I-lb)

     Comment 2.1.13c:   If the EPA  is meant to  protect  the environment, then
it could  better spend its time abating noxious and dangerous  (and
nonradioactive) gases discharged from the burning  of fossil fuels.   The EPA
could aid the nation's  mental health by  reducing the anxiety evident  at the
mere mention of "radiation."   (P-18a)

     Comment 2.1.13d:   The EPA should consider the effect of its proposed
standards on the  public's perception of  safety.  (G-la, G-lb)

     Comment 2.1.13e:   The proposed rule is  based  on perception, rather than
on fact,  and will simply increase  public concern with no increase in public
health.   (1-31, I-43b)

     Response (Comments 2.1.13a through  e):  Commenters believe  EPA  should
not promulgate the final rules because this  action would increase the fears
of people living nearby the regulated facilities.

     EPA  believes that  this is not a valid reason  for not promulgating final
rules.  The Clean Air Act does not permit this criterion to be  used  as a
basis for emission levels.

     Comment 2.1.14a:   Application of BAT is contrary  to the accepted ALARA
principle of radiation  protection.  By accepted definition, ALARA includes  a
reasonable weighing  of  risks,  benefits, and  costs,  and is also  designed to
be used on a site-specific basis.  (I-3b)

     Comment 2.1.14b:   Incorporation of Best Available Technology, BAT,
should be required by the regulations.  Consistency between ALARA and BAT
should be required and BAT should be applied on a  case-by-case  basis.  (G-24)

     Comment 2.1.14c:  There is some ambiguity regarding application of the
BAT requirement (48 FR 15079, col 2).  The intent of the phrases "with
allowance for variation in emissions" and "once a determination is made that
additional controls are necessary" is unclear.  (1-42)

     Comment 2.1.14d:  ALARA is a philosophy of operations and should not be
quantitatively imposed as a regulation.  (l-la)

     Comment 2.1.14e;  The ALARA principle, if properly applied, leads to
substantially lower doses than the present limits of 500 mrem/y to any
individual and a suggested 170 mrem/y if the exposure is continuous.  ALARA
should not be quantified as part of the regulatory process.  (P-la, P-lb)

     Comment 2.1.14f:  10 CFR 20, coupled with the ALARA principle, is a
perfectly acceptable approach to the regulation of radionuclide emissions.
     Response (Comments 2.1.14a through f):  Commenters questioned EPA's use
of Best Available Technology (BAT) and As Low As Reasonably Achievable
(ALARA) as considerations when developing emission standards.  EPA is asked
not to "quantify" ALARA as a regulation, implying that EPA's proposed
standards for DOE and NRC facilities do this.

     EPA has withdrawn the proposed rules for reasons unrelated to these

     ALARA, as one commenter pointed out, is a philosophy of operations.  It
means that every process will be constantly scrutinized to insure that
emissions are held to as low as reasonably achievable levels.  In practice,
this means available technology by definition, because "available" implies
technology that is proven and in use, thus reasonable to use.

     EPA has not chosen to establish numerical standards at levels that
reflect ALARA principles because it does not appear necessary at this time.

     Comment 2.1.15a:  The EPA has not explained why the standards
considered adequate for uranium fuel cycle facilities are two and one-half
times higher than those justified for DOE facilities.  (G-19)

     Comment 2.1.15b:  The EPA should develop generally applicable standards
without regard to who operates the facility.  (G-23, I-lb)

     Comment 2.1.15c;  A consistent risk basis should be used for all
standards.  (G-lb, G-24, I-lb)

     Comment 2.1.15d:  Although setting different standards for different
industries is consistent both with the Clean Air Act and ALARA, it will lead
to confusion.  (P-la, P-lb)

     Comment  2.1.15e;   The proposed  standards  are not  consistent with  10  CFR
20 and  40 CFR Parts  190 and  191 proposed.   In  addition,  the  proposed 40 CFR
Part 61  standards  are  not consistent  among  the four  source categories.

     Comment  2.1.15f;   No rationale  is  given for having  three different
standards for NRC  licensees, DOE  facilities, and uranium fuel cycle
facilities.   (G-2b,  G-16, I-lb, 1-31,, 1-32, 1-40, 1-47,  P-lb)

     Response (Comments 2.1.15a through f):  Commenters  stated all
radionuclide  emission  standards should  be based on the same  level of risk.
Therefore, standards should  be generally applicable, not different, for all
sources  of emissions.

     If  standards  are  numerically different, it does not necessarily mean
that they were  based on inconsistent  principles.  For  example, EPA's
standard for  the uranium fuel cycle  promulgated under  the Atomic Energy Act
applies  to all  exposure pathways  and  to all sources  that may be irradiating
a single individual.   Therefore,  it  is  to be expected  that such a standard
would not be  identical to a  Clean Air Act standard established to limit air
emissions only.

     Comment  2.1.16:   The EPA might  consider expressing  the  standard as
effective whole body dose such as is  being  considered  by the NRC.  Since
whole body dose is normally  controlling, the EPA statement that the proposed
standards would have to be reduced to achieve  comparable organ protection
under a  risk  equivalent system does  not seem justified.   (I-lb)

     Response:  The  commenter wishes  EPA to consider using the effective
whole body dose equivalent as a unit  of measure for  the  final rule.  Under
this system all irradiations, whether to the entire  body or  to a part of  the
body, would be  expressed as  that  dose to the whole body  that would be
equivalent based on  consideration of  risk.  The Agency had specifically
asked for comments on  this question when the standards were  proposed.  There
were few objections  to this  alternative.  However, this  issue is not
relevant since  EPA is  withdrawing all its proposed standards.

     Comment  2.1.17:   The commenter agrees  that the EPA  should consider the
potential for increased emissions and risk  in  the future.  However, when  it
considered whether a Section 112  standard is required  for a  particular
pollutant, the  EPA should have also considered potential future reductions
in public health risk.   (1-42)

     Response:  EPA  agrees with this  comment and, in fact, did consider
implicitly future  reductions in public  health  risk.  EPA would not have
proposed a standard  for  sources believed to have both  low emission rates  and
expectations  that  in the near term there would be changes such that there is
no longer potential  for  risks to  increase in the future.  Such conditions
would be met,  for  example,  if a source  had a limited operational lifetime or
would no longer maintain large inventories of  radionuclides.

     Comment 2.1.18:  The EPA should repropose the standards.  The standards
and associated background information should be widely disseminated and
there should be 180 days for public comment.  (1-34)

     Response ;  EPA believes it did widely publicize its proposed standards
and associated background information documents by means of mailing lists,
Federal Register notices, and public announcements.  Furthermore, EPA
actively made contact with those organizations it believed were most
interested or most affected by the proposed standards.

     Originally, EPA allowed 60 days for comment, and then, in response to
requests, extended the comment period an additional 45 days.  EPA believes
this is a reasonable period for public comment.

     Comment 2.1.19;  As the legislative history makes clear, the Act does
not require decisions based on the "maximum exposed" individual, but to
"sensitive citizens... who in the normal course of daily activity are
exposed to the daily environment."  The "maximum individual" defined in the
EPA risk assessments is a theoretical construct,  not a "sensitive citizen."
     Response :  The term "maximum exposed individual" refers to those people
living closest to the facility at the point of highest risk.  EPA intends
that calculations of the dose equivalent to these people be made using
realistic assumptions that do not grossly overestimate the risks.  The
Agency believes this is the intent of the legislative term "sensitive
citizen."  The Agency does not intend that dose calculations always be made
at the site boundary using the most conservative assumptions.  But for most
facilities, it is reasonable to conclude that someone is living very close
to the site boundary in the downwind direction, the usual location of the
point of highest risk.

     Comment 2 . 1 . 20 ;  The Preamble to the proposal emphasizes population
dose and risks, but does not address how they are used.  Controlling
exposure to the nearest off-site resident is reasonable if secondary
consideration is given to collective population doses.  (G-23)

     Response :  In deciding to withdraw the proposed standards, EPA
considered both individual and population doses.  In some cases, individual
dose was high enough for regulation to be considered, but EPA is withdrawing
the standard because, among other reasons, population doses were low.

     Comment 2.1.21;  The NESHAP requirements of the Act do not appear to
justify considering the number of people exposed as part of the process of
selecting the level of an emission standard.  (G-24)

     Response;  EPA disagrees with this comment.  Section 112(b)(l)(B) of
the Act requires emission standards "at the level which (in the
Administration's judgment) provides an ample margin of safety to protect the
public health."  EPA believes the term "public health" indicated a

consideration of the nearby populations in addition to a consideration of
nearby individuals.  This is customary public health practice for pollutants
widely dispersed in the environment.

     Comment 2.1.22:  In describing the circumstances under which extreme
measures may be considered (following application of BAT), the EPA uses the
phrase "significant emissions" (48 FR 15079, col. 2).  It is more
appropriate to consider the risk to the exposed population.  (1-42)

     Response;  It is not necessary to consider one of these factors to the
exclusion of the other.  In making its decisions, EPA considered both.

     Comment 2.1.23:  Contrary to EPA's own proposed policy, it did not
explicitly consider risk in establishing the proposed limits.  (I-la)

     Response:  EPA disagrees with this comment.  EPA estimated risks for
both individuals and populations for a wide variety of sources, including
those for which standards were proposed.  These risk estimates were one of
the factors leading to the proposal of emission standards and the decision
not to propose standards for other sources.  However, EPA recognizes that
risk estimates are uncertain.  Risk was also the primary basis for the
decision to withdraw the proposed standards.

     Comment 2.1.24:  The EPA does not provide adequate justification for
the failure to set numerical emission limits in all cases.  (P-3a, P-3b)

     Response:  For all categories except uranium mines, this comment has
been rendered moot by EPA's decision to withdraw the proposed standards.
For uranium mines, the Administrator has made a finding that it is not
feasible to set a numerical emission limit.

     2.2  Dose and Risk Calculations

     Comment 2.2.la:  The use of extrapolated risk projections based on a
linear non-threshold model produces risk estimates that are merely
conservative upper-bound estimates.  (I-20a, 1-42, 1-47, 1-53, P-lb, P-18a)

     Comment 2.2.1b:  The linear non-threshold model was not devised to be
used without consideration of other factors.  The numbers that the model
produces should not be confused with actual risk.  (I-20a, 1-47,  1-53, P-lb)

     Comment 2.2.1c:  Presently available information indicates that it is
not reasonable to assume that the risk associated with low doses is directly
proportional to the risk that has been demonstrated at higher doses for
low-LET radiations.  (I-4b,  P-lb)

     Comment 2.2.Id:  The EPA's use of the linear non-threshold theory is
inappropriate in isolation from other factors and actual risk data.  It
represents the upper limit of risk at low doses.  (I-3b)

     Response;   This series of comments refers to the Agency's use of a
linear-non-threshold dose response model to estimate health effects for low
dose rate, low-LET, radiation exposure, rather than one preferred by the

     The linear-quadratic model was recommended in the NAS BEIR-3 report
rather than the linear which was suggested to be an upper limit.  The
linear-quadratic model in BEIR-3 and NCRP64 appears to be based upon the
initial site model for the theory of dual radiation action (TDRA).

     This initial model is only one of several microdosimetric and/or
biological models of dose-response (see Goodhead, D.T., Models of Radiation
Inactivation and Mutagenesis, pp. 231-247 in Radiation Biology in Cancer
Research, R.E.  Meyn and H.R. Withers, editors, Raven Press, New York,  1980
for 11 models).  Goodhead (op. cit. and An Assessment of the Role of
Microdosimetry in Radiobiology, Rad. Res. 91.:45-76, 1982) showed  the
inability of site model rnicrodosimetry to explain observed experimental
phenomena.  For reasons given by Goodhead, it would be difficult  to support
the initial site model.  A more general treatment of the TDRA is  required to
agree with experimental observations.

     Newer models of radiocarcinogenesis are based on biological and
carcinogenesis theory rather than physical microdosimetric theory.  (D.T.
Goodhead, An Assessment of the Role of Microdosimetry in Radiobiology, Rad.
Res. 9.1:45-76, 1982; D.T. Goodhead, Models of Radiation Inactivation and
Mutagenesis, pp. 231-247 in Radiation Biology in Cancer Research, R.E. Meyn
and H. R. Withers, editors, Raven Press, New York, 1980; C.A. Tobias,  eŁ
al., The Repair-Misrepair Model, pp. 195-230 in Radiation Biology in Cancer
Research, R.E. Meyn and H.R. Withers, editors, Raven Press, New York,  1980;
P.J. Walsh, Possible Effects of Division and Repair of Altered Cells on Dose
Response Relationships in the Context of Carcinogenesis, Intern. J. Environ.
Studies, 5:285-288, 1974; etc.).  The inducibility of repair enzymes has
been demonstrated  (H. Tuschl, et al., Effects of Low-Dose Radiation on
Repair Processes in Human Lymphocytes, Rad. Res. 81^:1-9, 1980; G. Oliver!,
et al., Adaptive Response of Human Lymphocytes to Low Concentrations of
Radioactive Thymidine, Science 223:594-597. 1984).  It has also been
demonstrated that repair may be erroneous and increase the probability of
malignant transformation (Office of Science and Technology Policy, Chemical
Carcinogens; Notice of Review of the Science and Its Associated Principles,
Federal Register 49, #100; 21594-21661, 1984; O.K. Myers, DNA Repair and  the
Assessment of Radiation Hazards, pp. 106-144 in Summary and Proceedings of a
Biology Workshop on Biological Repair Mechanisms and Exposure Standards,
ORA4/1EA 78-2(R), D. Billen, editor, Institute for Energy Analysis, oak
Ridge, 1978; M. Terzaghi and J.B. Little, Repair of Potentially Lethal
Radiation Damage in Mammalian Cells is Associated with Enhancement of
Malignant Transformation, Nature 253:548-549, 1975).  One interesting
feature is that some of the models predict extrapolation from high to  low
dose will underestimate the risk (i.e., P.J. Walsh, 1974; O.K. Myers,  1978).

     The BEIR-3 Committee considered only three models—linear,
linear-quadratic, and quadratic—from among the possible models that might
fit the dose response data.  Several analyses have shown that the
dose-response curve for breast and thyroid cancer is linear.  Recently.
Matsuura et al. (H. Matsuura et al., Pathological and Epidemiologic Study of
Gastric Cancer in Atomic Bomb Survivors, Hiroshima and Nagasaki, 1959-77, J.
Rad. Res. 25:111-129, 1984) reported that using T65 dosimetry the
dose-response for stomach cancer is linear; both linear-quadratic and
quadratic dose response are excluded.  In the absence of human data for
other radiogenic cancers, it may not be prudent to assume a less
conservative dose response model.  It should be noted, however, that both
the linear and linear-quadratic dose response models have been used to
estimate risks in the final Background Information Document
(EPA-520/1-84-022-1).  This allows a direct comparison of risk estimates
based on differing scientific opinions.

     Comment 2.2.1e:  A recent article by Hickey, et al. in Health Physics
discussed the problem of extrapolating from high dose data to risk at low
doses and the inapproprlateness of basing public policy on this practice.

     Response;  The article by Hickey, et al. while interesting is not an
adequate basis to overthrow the current scientific consensus on cancer risk
assessments.  (See the Office of Science and Technology Policy on "Chemical
Carcinogens:  Notice of Review of the Science and Its Associated
Principles,"  Federal Register 49: 21593-21661, 1984.)  The principle
espoused by Hickey, et al., Hormesis, has been observed to act only in
mammalian studies where there was a high intercurrent infection rate, and
the animals were sick.  Sacher reviewed "radiation hormesis" in mammals and
concluded, "These characteristics, together with those discussed above, lead
to the conclusion that hermetic longevity effects generally have the form of
a non-cumulative decrease of vulnerability that does not modify the aging
process or the accumulation of irreparable radiation injury."  (G.A. Sacher,
Life Table Modification and Life Prolongation, pp. 582-638 in "Handbook of
the Biology of Aging," C.E. Finch and L. Hayflick, editors.  Van Nostrand
Reinhold Co., N.Y. 1977) Sacher also pointed out that regardless of the
presence of a "hermetic" effect, the proportion of animals developing tumors
and the Gompeitz function ("force of mortality") slope of general mortality
both show the expected dose related response at all dose levels, including
the lowest exposure used.  Since animals (or people) in a healthy robust
state are not expected to show "hormesis" (T.D. Luckey, "Hormesis with
Ionizing Radiation," CRC Press, Inc., Boca Raton, Florida, 1980) and the
induction of cancer is not abated, it would not be prudent policy to allow
unnecessary radiation exposure at any level.

     Hickey.  et al. also support their position with some reports of
demographic epidemiology studies.  This type of study is limited by the
accuracy of estimation of exposure, lack of data on age-specific in- and
out-migration,  accuracy of diagnosis, etc.  The BEIR-3 committee reviewed

several of these studies on cancer and stated:  "We conclude that these
types of studies, depending as they do on death record data aggregated
crudely by geographic region, do not constitute a sufficient basis for
deciding whether one or another type of environmental factor, such as
background radioactivity, is related to cancer rates.  Thus, as a test of
the effect on cancer risks of low-dose-rate lifetime exposure to radiation,
this approach does not appear to be fruitful in the United states within the
framework of variations in background-radiation exposure of populations
large enough to provide data that would be statistically useful."  ("The
Effect on Populations of Exposure to Low Levels of Ionizing Radiation: 1980"
(BEIR-3), National Academy Press, Washington, 1980)

     Superimposed on the population size and exposure is the fact that one
of the basic assumptions in a regression analysis of a demographic study is
that those variables which are not included are randomly distributed with
respect to the explanatory variables studied.  Variables known to be
correlated with cancer include age, sex, marital, condition, race,
birthplace, religion, ethnicity, socioeconomic status, and occupation
(MacMahon, B. and Pugh, T.F., "Epidemiology:  Principles and Methods,"
Little, Brown, and Company, Boston, 1970;  Lilienfeld, A.M., Levin, M.L.,
and Kessler, I.I., "Cancer in the United States," Harvard University Press,
Cambridge, Mass. 1972).  Also influencing vital statistics will be in- and
out-migration, age structure of migrants, and other demographic factors for
which accuracy of data is limited or data is constantly being revised (see
"The Methods and Materials of Demography" by H.S. Shryock, J.S. Siegel and
Associates, condensed edition by E.G. Stockwell, Academic Press, N.Y., 1976),

     While age-adjusting may compensate for differences in age, sex, and
race between populations, other unidentified and unincluded variables may be
of critical importance.  As Lilienfeld, et al. pointed out (A.M. Lilienfeld,
et al. Op. Cit. 1972), "In view of the long latent period which is presumed
to be associated with the development of certain neoplasms, it is
theoretically possible that factors of etiologic importance will be masked
by the substantial migration of an 'already exposed1  population into or out
of a given geographic region."

     For these and other reasons the Agency feels that prudent radiation
protection still requires assumption of a linear-non-threshold model for
induction of cancer and genetic effects at low levels of exposure.

     Comment 2.2.If:  The EPA's risk assessment, based on BEIR-2 and
UNSCEAR, understates the risks of chronic exposure to low levels of
radiation.  The analysis ignores the work of Elkind,  MacMahon, and Morgan,
which suggests that low doses have greater impacts than the linear
extrapolations from high doses predict, and it also ignores the effects on
radiosensitive populations such as children.  Also, there is no evidence
that the EPA considered cumulative or potential synergistic effects in its
evaluation.  (P-15a)

     Response:  The Agency v~1ieves the use of a linear dose response model
without any dose rate reduction factor for low dose rate low-LET radiations
is sufficiently conservative.  While the Agency agrees with the MAS BEIR-3
Committee that a linear response is less conservative for high-LET
radiations than for low-LET radiation, it is an adequate representation of
our knowledge or numerical risk estimation at this time.

     The EPA analysis includes age sensitivity to the same extent as the MAS
BEIR-3 report.  Although the MAS analysis for solid cancer dose does not
consider the sensitivity of those of age ten or less to be as great as the
MAS BEIR-1 report, increased probability of cancer induction during
childhood is taken into account in the EPA analysis to the extent it was
identified in the NAS BEIR-3 report.  It is true that in general instances
the Agency did not consider the cumulative effect of different sources in
the same community or synergistic effects.  In all of the cases examined,
the principal source provided the vast majority of the estimated effect so
that the contribution of minor sources was negligible compared to the
uncertainty in the risk estimate.  Moreover, the Agency is unaware of any
scientific reports of synergism between radiation and toxic metals.

     Comment 2.2.2a;  In its discussion of health risk, the EPA fails to
consider the BEIR-3 Report of 1980.  The EPA has ignored the 1980 report by
the National Academy of Sciences (BEIR-3).  The EPA should have taken into
account recent publications dealing with estimating the health hazards posed
by radiation.  These include the BEIR-3 report and NCRP 64.  (G-2b, I-lb,
I-lc, 1-17, I-20a, 1-31, 1-32, P-la, P-lb. P-9, P-18a)

     Response;  The Agency has updated the risk estimate for exposure to
radionuclides given in the Background Information Document (EPA83).  In
doing so, we have used the age-dependent risk coefficients for a linear
response given in the BEIR-3 report, "Effects on Populations of Exposure to
Low-Levels of Ionizing Radiation."  Our use of the BEIR-3 report for risk
estimates is described in the Final BID where the Agency's choices between
the various models proposed by the BEIR-3 Committee are examined in more
detail than outlined in this response to comments.

     To allow for an assumed lesser response for low-LET radiations at low
doses and dose rates, the 1980 NAS BEIR Committee based its "preferred risk
estimates" on a hypothetical linear-quadratic dose response function.  The
Committee made this choice after analytically examining the cancer mortality
data and particularly the leukemia mortality data for A-bomb survivors on
the basis of three dose response functions:  linear, linear-quadratic, and
quadratic.  For reasons outlined in the final BID, the Agency believes only
the first two of these functions are compatible with data on human cancer.
Risk estimates for low-LET radiations in the final BID are based on the
BEIR-3 linear and linear quadratic dose response models.

     Although the majority of the members on the BEIR-3 Committee
"preferred" a linear-quadratic response in 1980, we believe their
quantitative basis for this judgment is considerably weaker now because of
the subsequent reassessment of the A-bomb dosimetry.  The Committee's
analysis of dose response functions assumes that most of the observed excess
leukemia (and solid cancer) among A-bomb survivors was due to neutrons
(NAS80).  Current evidence, however, is conclusive that neutrons were a
minor component of the dose in both Hiroshima and Nagasaki (Bo82, RERF83a,
RERF83b).  Therefore, it is likely that the linear response observed among
the A-bomb survivors, which the BEIR Committee largely attributed to
neutrons, was, in fact, due to their gamma dose, not a dose of high-LET
radiation (EPA84a).

     Although there is evidence for a nonlinear response to low-LET
radiations in some, but not all, studies of animal radiocarcinogenesis, the
Agency is not aware of any data on human cancers that is incompatible with a
simple linear model.  In such a case, we believe it is preferable to adopt
the simplest hypothesis that adequately models the observed radiation
effect.  Occam's razor is still a viable scientific rule for separating
necessary from ad hoc assumptions.  Moreover, EPA believes that risk
estimates for the purpose of assessing radiation impacts on public health
should be based on scientifically creditable risk models that are not likely
to understate the risk.  Given the current bias in the doses assigned to
A-bomb survivors, such an approach seems particularly reasonable, as well as

     The low dose rate effectiveness factors (DREF) developed by NCRP
Committee 40 are based on their analysis of a large body of plant and animal
data that showed reduced effects at low doses for a number of endpoints,
including to a lesser extent radiogenic cancer in animals, chiefly rodents.
However, no human data confirm these findings.  A few human studies
contradict them.  Highly fractionated small doses to human breast tissue are
apparently as carcinogenic as large acute doses (NAS80, La80).  Furthermore,
small acute doses (less than 10 rad) to the thyroid are as effective per rad
as much  larger doses in initiating thyroid cancer (UN77, NAS80)-  Moreover,
the increased breast cancer due to chronic low dose occupational gamma ray
exposures among British dial painters is comparable to, or larger, than
expected on the basis of acute high dose exposures (Ba81).  While none of
these examples are persuasive by themselves,  collectively they indicate that
it may not be too prudent to estimate cancer risks due to low doses and low
dose rates on the basis of observations at large doses.  However, as pointed
out in the final Background Information Document the BEIR-3 linear quadratic
model, which we have also used to estimate risks, is equivalent to a DREF of
2.5, the same as that used by by ICRP and UNSCEAR.
All references refer to Chapter 8, Volume I of the Background Information
Document (EPA 520/1-84-022-1).

     Comment 2.2.2b;  The risks predicted by BEIR-3 are  less  than  those
predicted by BEIR-2 by a factor of 2.  More importantly, the  report states
that reported predictions are  "average values per rad and are not  to be
taken as estimates at only  1 r^-"1 of dose."  The committee only made
predictions for  . . . continuous exposures to 1 rad per  year or more.  To go
from a 1 rad per year prediction to a 0.005 rem per year limit is  not
sensible.  (I-20a, P-18a)

     Response:  This comment assumes that persons exposed to  radionuclide
emissions receive total doses  as small as 0.005 rem per  year.  This is
clearly impossible since normal background radiation from internal and
external sources is at least 0.1 rem per year.  The risk of interest is the
incremental risk due to a change in the annual dose rate from, to  use the
commenter example, 0.1 to 0.105 rem per year.  This is within an order of
magnitude of the nominal 1  rad per year dose rates listed in  the risk table
prepared by the WAS BEIR-3  Committee.  Morever, we note  that the actual
calculations made by that Committee to prepare these tables are based on a
dose rate of 0.1 rem per year.

     The Agency does not know  of any human data on radiogenic cancer at low
doses that is appreciably more consistent with a linear-quadratic dose
response function than a simple linear response.  Some data sets such as
that for the radium dial painters do show a quadratic component at extremely
high doses.  In contrast, other studies, such as the one on U.S. uranium
miners, show a reduced effect  per unit dose in the high dose range.
However, these perturbations occur at doses too large to be of regulatory
concern.  The Agency acknowledges that the shape of the dose response
function for chronic exposures occurring in the range of environmental doses
is unknown, and it has used both the linear and linear-quadratic dose
response models.  Moreover, it should be appreciated that the difference
between the calculated risks using these two models, about 2.5, is smaller
than the uncertainty in the risk estimates due to other  factors.

     Comment 2.2.3:  The Committee should be aware that an October 1981
article in Health Physics has  cast a cloud over all risk estimates based on
the Hiroshima and Nagasaki  data.  A complete reanalysis of the dose curves
has not, to the commenter's knowledge, been published, but any rule based on
incomplete data is unconscionable.  (P-18a)

     Response;  The Agency  is  aware of the 1981 Health Physics article and
many other reports on the dose reassessment in Japan.  Changes in  the
Japanese Atomic Bomb Survivor  data base may require some adjustment of the
risk estimates.  This is discussed in the final BID.

     Comment 2.2.4:  The Federal Register does not mention Lucky's work on
the possible beneficial aspects of ionizing radiation.   (P-18a)

     Response:   See Comment 2.2.Id.

     Comment 2.2.5a;  No scientifically accepted epidemiological studies
associate any increased risk to members of the public at levels of exposure
associated with the various activities covered by the EPA's "Preliminary
Report."  (1-47)

     Response:  The commenter has confused the ability to prove or disprove
a hypothesis with the validity of the hypothesis.  Numerous reports have
reviewed the basic requirements of epidemiologic studies of background
radiation.  In general, in addition to detailed demographic, socioeconomic,
health statistics and exposure data across the time period involved, they
require a base of millions to hundreds of millions of person-years of data
to detect the incremental increase in risk projected by current risk
coefficients, even for exposure levels around 4 times average background.
Charles Land has examined this problem in some detail (C.E. Land, Science,
209:  1197-1203, 1980); see also:  E.E. Pochin, Health Physics, 31.: 148-151,
1976; S.G. Goss, Health Physics, 29:  715-721, 1975; C. Buck, Science, 129;
1357-1358, 1959; G. Hems, Brit. Med. J.,  1:  393-396, 1966.  None of the
geographic epidemiology studies published to date have been able to meet the
necessary criteria.  This includes both negative and positive reports.

     Many of the reports cited by the commenter fall into this category.
For example, Dr. Lane's study was too small to detect any difference in
health risk even if the risk were present.  The power of this study to
determine a carcinogenic effect, assuming one was present,  is only 0.38,
i.e., less than a 50-50 chance.  Even though Lane calculated (incorrectly) a
somewhat greater power, on page 34 of his thesis he states, "This is a
relatively low power, however, and,  consequently, a negative finding may be
attributed to small sample size and should not be considered strong evidence
for the null hypothesis of no association."  Moreover, we do not believe
that this study can be used to test EPA risk estimates.  Using the radon
daughter concentrations reported in his thesis, it has calculated, using the
EPA risk models, what difference in lung cancer mortality due to radon might
have occurred in his case and control groups for lifetime exposure.  This
difference was less than 0.2 cases,  not a detectable amount in any
epidemiological study, let alone one consisting of fewer than 100

     The Kerala region of India has a population of about 70,000 with an
estimated average exposure of about 342 mrem/yr, even though a small
fraction (3105 persons) is exposed to 915 mrem/yr (C.M. Sunta, et al., TLD
Survey of Natural Radiation Environmental Along the South-West Coast of
India, Bhabha Atomic Research Centre, Bombay, 1978).  While the commenter
cites Gopal-Ayengar,  et al. as showing no effects in Kerala, he does not
mention positive reports by Kochupillai,  et al. (N. Kochupillai, et al.,
Nature 262;   60-61, 1976) on increased congenital abnormalities, nor
reproductive problems in 22 couples at about 20 times background (Sunta, et
al. 1978).

     Likewise,  in the Brazilian study group, a  1970 report by Cullen,
et al., is referenced but not a 1975 report by  Costa-Ribeiro, et al.
(C. Costa-Ribeiro, et al. Health Physics, 28:   225-232,  1975) showing
increased chromosome aberrations in cytogenetic studies  of lymphocytes of
workers in a monazite plant.  No reports on cancer incidence in the exposed
Brazilian population are expected.  A recent report by the Brazilian
investigators  (T. L. Cullen, et al. pp. 805-808 in Radiation Protection A
Systematic Approach to Safety, 5th IRPA Congress, Pergamon Press, NY, 1980)
concluded, "The  lack of reliable medical practice and records renders an
epidemiological  study impractical."

     Many of the other references cited involve demographic epidemiology or
hormesis which were discussed in the response to Comment 2.2.Id.

     Comment 2.2.5b:  In a study conducted in China, a population of 74,000
people exposed  to a high natural level of radiation was  compared with a
similar group whose dose was roughly 63 percent less (196 millirems versus
72 millirems).  No significant differences in mortality, morbidity, or
genetic damage were found.  (I-3b, 1-49)

     Response:  The commenter referenced the 1980 report of the High
Background Radiation Research Group in China.   This study was updated in
1981 (J. Radiat. Res. 22:  88-100, 1981).  This update reported 383,653
person years and 418,265 person years in the cancer mortality study in the
high-background and control groups, respectively.  Since exposure in the
high-background area was only about 3 times that in the control area, it
should not be expected that an increase in radiation-related cancer would be
demonstrated if present.  Cytogenetic studies showed an  increase in two-hit
aberrations, the type one might expect to be associated with high-LET
radiation, in the high-background study group.  However, the total number of
cases is small and the significance is not known.

     About all  that can be said of the high-background area studies is,  they
can't answer our questions.  They do show EPA is not underestimating risk by
orders of magnitude.

     Comment 2.2.5c:  It was stated earlier in  the Denver hearings that the
epidemiological studies done in Grand Junction  by the Colorado Department of
Health have not indicated that there is a bioeffect from exposure to the
radiation levels observed.  This is not a conclusion of  the study.  The
latent period for the exposure levels experienced has not been exceeded.

     Response:  The Agency agrees.

     Comment 2.2.6;  In considering such factors as radionuclide uptake by
vegetation,  and consumption of locally-produced crops and milk, has the EPA
used conservative assumptions such as those developed by the Heidelberg
Institute for Energy?  (P-13b)

     Response;   EPA attempts to model environmental transport of
radioactivity as realistically as possible; we make a best estimate of the
expected environmental concentrations, all things considered.  The
Heidelberg Institute of Energy's assumptions, as indicated in the comment,
are extremely conservative and serve to provide estimates which presume the
maximum potential transfer at each step through the environment.  One would
expect the resulting estimates to far exceed those in an actual situation
and hence be inappropriate to EPA assessments.

     EPA used realistic assumptions for food pathways and uptake factors.
The Agency did not use the values cited by the commenter because they are
overly conservative.

     Comment 2.2.7:  Total body burden has been ignored in the assessment of
risk (due to radon progeny).  The reported risks are based on lung cancer
alone.  Sites such as bone and liver must also be included.  In addition,
pathways such as ingestion must be considered.  (P~15b)

     Response;  The Agency considers all fatal cancers and hereditary risks
in the assessment of radionuclide emission.  In the case of inhaled
radon-222, almost all of the risk, approximately 99 percent, is of
radiogenic lung cancer.  In such cases EPA does not the dose and risk to
other organs, since their contribution to the total list risk is
considerably less than the uncertainty in the estimated risk of lung
cancer.  It is true that radon decay progeny, principally lead-210, are
deposited on vegetation and can reach men by the ingestion pathway,
ultimately causing a dose to bone and liver.  Again, this risk is
unimportant relative to risk due to inhaling the daughters.  The Agency
explored this exposure route in detail in EPA 520/1-83-008-1, Final BIS for
the Control of Byproduct Materials from Uranium Ore Processing.  Typically
the lifetime risk from the ingestion of radon products is one five
hundredths of the risk due to the inhalation of radon progeny.

     Comment 2.2.8:  The EPA risk assessment understates risks by relying on
1970 census data.  (P-15a)

     Response;  Increases in population since 1970 are small compared to the
uncertainty in the risk estimates.  More recent census data, not yet
available, could show increased or decreased risks for a larger population,
since estimates of population risk depend critically on the exact location
of individuals within the region.  The Agency believes it is sufficiently
prudent to use currently available data.

     Comment 2.2.9:  The EPA's risk assessments could be made more realistic
by including an estimate of the actual number of people exposed to various
sources and by estimating risks for less than a 70 year continuous
exposure.  A series of assessments, using different assumptions about key
uncertainties,  would be preferable to the very conservative assessment made
by the EPA.  (1-53)

     Response:  Since atmospheric dispersion is not geographically bounded,
the number of people actually exposed by releases from a source is only
limited by the assessment area considered.  For a regional population
exposure estimate, EPA customarily considers exposures to individuals at
distances within 80 km of a facility.  The lifetime risk to an individual is
a consistent starting point for the analysis.  EPA is not suggesting that
individuals customarily never move from a given spot for their entire lives
but rather that such considerations are to be applied ex post facto to the
extent that the expected risk to an individual may be made more realistic.
Values of factors and parameters used in the assessments are intended to be
reasonable estimates and are not values chosen to maximize the impact of a

     Comment 2.2.10:  The proposed EPA standards are overly restrictive
because they are based on hypothetical rather than real people.  EPA's risk
assessment is a worst case analysis, which is contrary to the requirements
of the Clean Air Act.  It deals with hypothetical individuals and upper
limit estimates of carcinogenic risks.  It should have included an analysis
of comparative risks and margins of error.  (I-la, I-3a)

     Response:  The Agency has used reasonable parameters in the models used
to assess risk.  However, the Agency has not assumed that the current
locations of housing relative to a given source will remain constant over
time.  New housing will continue to be built in areas where it is
permitted.  The magnitude of other risks does not affect the risk due to
radionuclides.  Therefore, the Agency's use of comparative risks is confined
to the risk management process where a number of factors are considered in
the decisions leading to an appropriate regulatory control level.

     Comment 2.2.11:  If EPA does not set standards based upon zero
emissions, then standards should be based on worst case assumptions in the
analysis.  Although EPA appears to have used worst case assumptions in part
of the analysis, they were not used consistently throughout the analysis.

     Response:  EPA tries to estimate the risk associated with emissions of
radionuclides to air as accurately as it can.  Worst case assumptions are
not used throughout the analysis; rather, the Agency tries to use the most
realistic ones.  EPA has contracted with the Oak Ridge National Laboratory
to develop the computer codes used to make risk estimates, committing very
significant resources over long periods of time to this task.  Nevertheless,
it is recognized that the accuracy of these estimates is limited to about an
order of magnitude, and this uncertainty is considered when proposing

     If worst case assumptions were used throughout an analysis, the results
would be wildly inaccurate by many orders of magnitude.  EPA would not have
confidence in such results and believe most people would find them

     Comment 2.2.12;  The EPA should calculate doses based on an integration
of the whole-body or organ dose to an individual over 80 rather than 70
years.  Seventy years is based on average life expectancy, and is
inappropriate, given current life expectancies, to protect the maximally
exposed individual.  (P-17)

     Response;  EPA believes that the use of a life table analysis
yielding an average life expectancy of 70.7 years is appropriate for
estimating the risk to large populations and to the maximum individual.
The life table approach used to estimate dose and risk considers life
spans out to 110 years.

     Comment 2.2.13:  The EPA's overly conservative assumptions and
overly stringent standards are not warranted for regulating
scientifically well-known substances.  The EPA should explain why its
methodology differs from radiation protection recommendations provided by
the ICRP and NCRP.  (I-3b)

     Response;  EPA does not believe it has used overly conservative
assumptions (see response to Comment 2.2.9) or that it has proposed
overly stringent standards.

     The proposed standards were not overly strict just because they were
lower than recommendations of the ICRP and NCRP of 500 mrem/y to the
whole body and 1,500 mrem/y to an organ.  Such levels of radiation
exposure are generally recognized to represent a significant level of
risk, especially if the exposure lasts for many years.  Therefore, they
are maximum permitted levels that should not be approached without good
reasons, and people should not be exposed to such levels for any length
of time.  They do not represent an exposure level that provides, over a
long term, an ample margin of safety for the public.  Furthermore, they
apply to the sum total of exposure to an individual such that occurs
through all pathways and from all sources.  They are not appropriate as
limits for a single facility and a single pathway.

     Comment 2.2.14a:  The numerous assumptions involved in the
AIRDOS-EPA and RADRISK models introduce uncertainties in the dose
estimates that are larger than those arising from the errors of
measurements.   It would not be meaningful to propose a standard of 10
mrem per year to the whole body with the levels of uncertainties involved
in the calculations of this value.  (G-lb, P-lb)

     Comment 2.2.14b:  The codes specified are not site-specific and are
extremely conservative, probably by at least an order of magnitude.
Their technical validity should be validated by someone outside of EPA.
(G-la, G-lb)

     Comment  2.2.14c:   The  mathematical codes  that  are  required  for
calculation of  the  dose-equivalent values  to the members of  the  public
are controversial and  not generally  accepted by the scientific
community.  (G-lb,  G-3)

     Comment  2.2.14d:   The  EPA model does  not  account for wide variations
in wind  direction,  velocity,  and  atmospheric stability  caused by terrain
and other  factors affecting surface  roughness.  Nor does it  account  for
the effect of particle size on transport dynamics.   It  therefore produces
significant errors  in  estimated concentrations.  (1-17, 1-26)

     Comment  2.2.14e:   The  AIRDOS-EPA model is inadequate because it
appears  to provide  no  means for calculating resuspended air
concentrations  or subsequent  deposition to the ground surface.   (P-2a,

     Response (Comments  2.2.14a through e):  The original AIRDOS code was
developed  for the Department  of Energy by  the Oak Ridge National
Laboratory who  then made minor modifications to produce the  version  known
as AIRDOS-EPA.   The AIRDOS-EPA code can be made site-specific by using
site-specific meteorology,  population distributions, and food pathways
factors.   It  has been  reviewed by the Nuclear Regulatory Commission
(NUREG-CR-3209,  March  1983) who found no substantial deficiencies.   The
EPA considers that  the codes  represent the state of the art.

     While it is true  that  the AIRDOS-EPA  code does not directly account
for the  influences  of  terrain, the user-supplied meteorological  data for
his site will tend  to  reflect the influence of terrain upon  the  wind
direction, velocity, and atmospheric stability.  Moreover, the
calculation of  the  dose  to  the maximum individual will not normally  be
influenced very  much by  the  terrain, assuming he is reasonably close to
the facility.

     While AIRDOS-EPA  does  not explicitly  include calculation of  the
resuspension  of  materials which have been  dispersed and deposited on the
ground surface,  the effect  of resuspension on inhalation may be
approximated  by  simply ignoring deposition.  A source of windblown
activity (e.g.,  a gypsum tailings pile) may be assessed by determining
the annual release  and using  the area source model  in AIRDOS-EPA.

     Comment  2.2.15;   EPA,  throughout its  proposal,  refers to risks of
health effects as if they were proven or readily observable  events.  Such
language is wrong and  could be used to promote unwarranted legal
actions.  (G-2b, G-16, I-lb,  1-31, 1-32, 1-40, 1-47)

     Response:   There  is some merit to the comment  in that radiation
effects at low doses are not  readily observable.  However, the BEIR
Committee addressed this problem as follows:  "Contrary to widespread
belief, evaluation  of  the causal nature of an observed association is a
statistical problem, and does not involve  the concept of 'proof   in any

definitive sense.  The evaluation requires the assembly of information
and concepts from many different sources and their integration into an
overall estimate of the likelihood that an association is or is not
causal.  Where controversy exists, attention should primarily be focused
not on whether the association is 'proven,' but on the limitations, the
adequacy of data, the lack of design and corresponding artifacts, and
finally the choice of the appropriate 'control1 for the situation.  In
the process of assembly and integration of data, there may come a point
at which it becomes - for the purpose of making decisions or taking
actions of practical import - more prudent to act as though the
association were causal, than to continue to regard it as non-causal.
Where controversy exists, it should be focused on whether or not
currently available data lead us to this point, rather than on the
unanswerable question of whether the causal nature of the association is
or is not 'proven.1  It is important to recognize that both the
evaluation of individual pieces of evidence and the relative weights
assigned to evidence of different kinds contain substantial elements of
subjectivity, and that there are few biological issues on which belief in
the strength of the evidence of causality does not vary widely among
experts in the field."  (BEIR-1, pp. 93-94.)

     In implementing the recommendations of the BEIR Committee, the "EPA
Policy Statement on Relationship Between Radiation Dose and Effect, March
3, 1975" (pp. 143-145 in National Interim Primary Drinking Water
Regulations, EPA-570/9-76-003), the Agency noted, "It is to be emphasized
that this policy has been established for the purpose of estimating the
potential human health impact of Agency actions regarding radiation
protection, and that such estimates do not necessarily constitute
identifiable health consequences."

     The International Commission on Radiological Protection developed
risk estimates similar to those in BEIR-1 and noted that the risk factors
were intended to be realistic estimates of the effects of irradiation at
low annual dose-equivalents (ICRP Publication No. 28, Report of the
Stockholm Meeting, Ann. ICRP. 2, No. 1,  1979).

     The current scientific consensus treats the estimated risks or
potential health effects as real even though they cannot be
demonstrated.  While the uncertainty in the estimate is recognized, it
would not be prudent to treat the estimated risk as unreal.

     2.3  Control Technology

     Comment 2.3.1:  where only a few individuals are at risk, EPA should
consider whether the public health can be protected with an ample margin
of safety through means other than the installation of extremely
expensive emission controls.  (1-42, 1-53)

     Response:  The principal alternative  to control  technology  is  land
use controls—keeping people away  from  the point of emission.  As a last
resort, EPA suggested land controls  for uranium mines to  reduce  risks
from radon because control technology was  not available.  Commenters
stated that land use controls are  impractical for uranium mines, and it
was partially for this reason the  proposed standard was withdrawn.

     Comment 2.3.2;  In  identifying  best available technology  (BAT). EPA
should focus on technology that has  been adequately demonstrated at a
commercial scale plant for the source category on which the control
requirement is to be imposed.  (1-42)

     Response:  In determining available technology for a source
category, the adequate demonstration of the technology at a commercial
scale facility with similar emission characteristics  is a major criterion
in the EPA's evaluation  of applicable controls.

     2.4  Proposed Limits

     Comment 2.4.la:  The application of ALARA at 10  mrem/yr is
troublesome because it is close to the  levels most people consider  de
minimis.  (P-lb)

     Comment 2.4.1b:  In view of the estimates of dose and risk and the
projected limited growth for some  types of facilities, it hardly seems
necessary at this time to drastically reduce radiation dose limits.

     Comment 2.4.1c;  The 10 mrem/yr limit is too restrictive, borders on
insignificance, and cannot be measured  against existing background
levels.  (I-36a, I-36b,  1-50, P-lb)

     Comment 2.4.Id:  The proposed standard of 10 mrem/yr is not
practical, reasonable, or enforceable.  It is about 1.1 microrads per
hour, very close to the  detection  limit.   This includes the ability to
discriminate by measurement from the background.  (G-lb,  G-24, P-lb)

     Comment 2.4.1e:  EPA proposed 10 mrem/yr standards are far below any
reasonable limit for the maximally exposed individual.  (l-3a)

     Response (Comments  2.4.la through  e):  For various reasons, many
commenters consider the  proposed standards too strict; in particular, a
10 mrem/y unit for any organ for all licensed facilities, and a 10  mrem/y
unit for whole body for  DOE facilities  is  considered  by some too
restrictive.  These comments are no  longer applicable because the
proposed standard is withdrawn.

     Responses to other parts of these comments are given elsewhere:

         a.   Need for the standards - See response to Comment 2.1.2.

         b.   Evaluation by comparison with Natural Background - See
             response to Comment 2.1.11.

         c.   Enforcement and Detection Limits - See response to Comments
             in Section 2.5.

     Comment 2.4.2a;   The proposed indirect standards should be upper
limits that  also consider other sources and background concentrations.
(G-24, P-15a)

     Comment 2.4.2b:   The proposed standards do not provide the level of
health protection required by the Clean Air Act.  (P-15a)

     Comment 2.4.2c:   We recommend that the EPA establish a generic dose
limit of 10  mrem/yr to any individual from any type of facility, with the
continued use of the  ALARA concept to keep dose as far below this limit
as possible.  (G-17,  G-21)

     Comment 2.4.2d:   The intent of Congress was that EPA establish zero
emissions standards.   (P-3a)

     Comment 2.4.2e:   The EPA should propose limits that reduce the risk
to as close  to zero as possible.  (P-2a, P-2b)

     Response (Comments 2.4.2a through e);  Many commenters do not
consider the proposed standards to provide enough protection.

     EPA has concluded present practices are protective of public health
with an ample margin  of safety.  For Federally licensed facilities and
Federally owned facilities,  "ALARA" is required by Federal Guidance
established  by the Federal Radiation Council and still in effect.

     For reply to the question of whether the standards are not strict
enough unless they require zero emissions, see response to Comment 2.1.5.

     Comment 2.4.3a:   If indirect standards are to be used, they should
be met at the facility property line.  (G-24, P-36, P-17)

     Comment 2.4.3b:   The standard should apply to normally occupied
areas, not necessarily the site boundary.  (I-8a, I~8b)

     Comment 2.4.3c:   Under the CAA, risk must be based on reasonably
probable risk to average persons, accounting for sensitive population
subsets, rather than  hypothetically maximally exposed individuals.  (l-3b)

     Response  (Comments 2.4.3a  through c);  Commenters question  the
 location where  the standards should apply.

     EPA carefully considered the question of where the standards should
 apply, that is, to individuals  assumed to live at the site boundary or  to
 individuals where they normally live, the actual residence.  Most
 facilities have people living so near the property lines that whether
 they are assumed to  live directly on the line or a few hundred feet away
makes no practical difference.

     A facility in a truly remote area may have no one living near its
 boundaries.  If there is no reason to prevent people living there, it is
 somewhat foolhardy to assume no one will move into this area, and
 controls should be based on this assumption.  For this reason, facilities
usually use emission controls designed under the assumption that someone
 is living very near  their site  boundary.

     Comment 2.4.4a:  DOD, DOE,  and other facilities have some potential
 for accidental  releases that exceed the proposed standards.  Because
 these are not specifically exempted, litigation seriously affecting the
nuclear weapons program without benefiting the public could result.

     Comment 2.4.4b;  The proposed limit is so low for NEC facilities it
does not allow  for any accidental releases due to personnel or process
 errors.  (I-37a)

     Response  (Comments 2.4.4a  and b):  Commenters imply that
 "accidental" releases should be exempted from any standards.  EPA
carefully considered this question but did not to exempt accidental
 releases from the proposed standards for the following reasons:

     a.  The difference between an "accidental" release and releases due
         to bad planning is often difficult to determine.  An exemption
         thus would make implementation difficult and more complex,
         giving the  impression  of a "loophole."

     b.  Accidental releases of a magnitude sufficient to violate the
         proposed standards are extremely rare.  There seems to be no
         need to make specific  allowance for them.

     c.  Including "accidental"  releases with the emission limit
         encourages good practice to prevent "accidental" releases.

     2.5  Imp1ementat ion

     EPA received the following comments with respect to implementation
of final rules.  They are no longer applicable because the proposed
standards have been withdrawn.

     Comment 2.5.1:   It is not clear:

     a.  how EPA proposes that the standards be enforced;

     b.  what the enforcement costs would be; and

     c.  how enforcement by the states and particularly the radiation
         control programs would be funded.

(G-23, 1-34)

     Comment 2.5.2;   It is essential that EPA conduct surveillance of
releases of radionuclides from the DOE and NRC plants.  The DOE and its
predecessor, the AEC,  have histories of under-reporting releases.  The
EPA must be able to survey, conduct investigations, and carry out
enforcement actions without interference.  (P-13b)

     Comment 2.5.3;   We object to measuring emission compliance at a
fence line by using a model that fails to take into account local terrain
and meteorology and is not based on actual measurements.  (P-15b)

     Comment 2.5.4a:  The required use of the EPA computer code is
troublesome because many sites have their own codes that incorporate what
they know of their area.  Uncertainty should be taken into account, not
avoided.  (1-50, P-lb)

     Comment 2.5.4b:  The proposed EPA standards may require substantial
revisions to existing models used by licensees to assure compliance with
existing standards.   (I-la)

     Comment 2.5.5:   Controlling dose levels by reliance on a single code
to determine dose is not desirable.  Means should be found to permit
compliance via environmental measurements.  (G-23, P-lb)

     Comment 2.5.6:   No indication is given concerning the type of
quality control programs (for AIRDOS-EPA and RADRISK) required to assure
reliable dose calculations on the basis of emissions and environmental
conditions for many different facilities and locales.  (P-lb)

     Comment 2.5.7;   The proposed limits are so low that there is
essentially no practical way to measure whether facilities are in
compliance,  state-of-the-art monitoring techniques cannot distinguish
between normal fluctuations in natural background radiation and the
extremely low levels permitted by the proposed standards.  (G-la, G-lb,
G-2b, P-lb, P-9, 1-17, 1-19, 1-49)

     Comment 2.5.8;   AIRDOS-EPA incorrectly calculates doses over hilly
terrain.  The model also does not take into account releases from short
stacks or vents; a majority of the NRC and Agreement States licensees
release in this manner.  Due to these shortcomings the model's use should
not be mandatory.  (P-la, P-lb)

     Comment 2.5.9:  The Act requires the EPA to mandate source
monitoring to demonstrate that the standards are being met.  (P-15a)

     Comment 2.5.10;  We believe that the proposed limits should provide,
at a minimum, adequate provisions for monitoring radionuclide burdens in
carnivorous teleosts and avian raptors within 80 km of an identified
point source, with the 10 mrem/yr. limit applied to these species until a
wildlife standard is developed.  (G-19)

     Comment 2.5.11:  A sensitivity and robustness analysis, as
recommended by ICRP-29, should be made of the models the EPA is relying
on to demonstrate compliance.  (G-20)

     2.6  Cost

     EPA received the following comments with respect to implementation
of final rules.  They are no longer applicable because the proposed
standards have been withdrawn.

     Comment 2.6.1:  Attempts to unnecessarily restrict emissions,
establish ALARA standards, or impose additional standards on already
regulated facilities, when doses are already minimal, can only result in
unnecessary expense in attempting to demonstrate compliance.  (I-lb)

     Comment 2.6.2:  An emission standard reflecting BAT should not be
imposed if the existing health risk is insignificant.  High cost
incremental emission reductions that do not significantly reduce health
risks should not be required.  (1-42, 1-53, P-lb)

     Comment 2.6.3a:  It is not clear that the total cost of the proposed
regulations will be less than $100 million.  Therefore, a cost-benefit
analysis must be performed as required by Executive Order 12291,
February 19, 1981.  (I-2a, I-2d, P-la, P-lb)

     Comment 2.6.3b:  The proposed standards do not require an economic
analysis under Executive Order 12291 for the following reasons:

     1.  The total annual cost will be less than $100 million.

     2.  The proposed standards were issued under a court-ordered
         deadline and are thus exempt under Section 8(a)(2) of the
         Executive Order.

     3.  The Executive Order can apply only to regulations "to the extent
         permitted by law."  For Section 112, congress has explicitly
         rejected the notion that economic considerations must be given
         highest priority.

     Comment 2.6.4;  Unnecessarily restrictive practices are inhibitory
and set the United States at a competitive disadvantage.  (P-lb)

     Comment 2.6.5;  While the EPA clearly has the discretion to consider
cost, it has failed to adequately consider economic cost in the proposed
rulemaking.  The cost for each health effect averted is $250 million for
elemental phosphorus and $10 million for underground uranium mines.  This
is unreasonable.  In its development of radiation protection standards,
the EPA has indicated that between $250 and $500 thousand per health
effect averted is a reasonable regulatory cost.  (I-3b, 1-53)

     Comment 2.6.6;  The EPA's task is to reduce, not eliminate risk.  If
affected industries can not economically survive under the proposed
rules, the rules should be relaxed.  (P-19)

     Comment 2.6.7:  A cost-benefit analysis, including an evaluation of
the improvement in public health resulting from the proposed standards,
should be performed.  (I-la)

     2.7  Other Comments

     Comment 2.7.1;  The definition of "dose to an individual" in terms
of a dose rate (48 FR 15076, 15077) is incorrect.  The Agency avoids this
error in the proposed regulations, and should correct the supplementary
information accompanying the final rule.  (P-3b)

     Response;  EPA agrees.

     Comment 2.7.2:  The EPA should consider developing standards for
indoor radon exposure.  (G-23)

     Response:  Standards for indoor radon exposure would be outside the
scope of this rulemaking because virtually all of this radon results from
naturally occurring sources.

     Comment 2.7.3:  The proposed NRC standards do not cover naturally
occurring or accelerator produced materials licensed by state agencies.
These materials should be covered.  (G-23)

     Response;  EPA examined naturally occurring radionuclides in a
number of source categories and made decisions on the need for a standard
for each category.  EPA did not consider the users of accelerator
produced materials; consequently, no standard was proposed for this
group, and no decision was made not to propose a standard.

     Comment 2.7.4;  The proposed rule states that industrial gauges,
static eliminators, radiographic devices, self-illuminating watches, and
smoke detectors involving production of sealed sources do not emit
radionuclides.  At the end of their useful lives these devices,
particularly smoke detectors containing americium or radium, will find
their way to landfills or incinerators.  (P-21)

     Response:  These sealed source devices do not present  the potential
for airborne  release during their operating lives.  At  the  end of  their
useful  lives,  licensed sources such as  industrial gauges and  radiographic
devices will  be disposed of at licensed disposal sites  with no potential
for airborne  release.  Smoke detectors  are and are likely to  continue to
be disposed of in commercial landfills  and incinerators even  though  the
directions specify that they be returned to the manufacturer  for disposal
at a licensed  facility-  A typical detector for home use contains
approximately  1 microcurie of 241 americium.  Commercial detectors
contain about  15 microcuries.  Tests of such detectors  show a few
hundredths to a few tenths of percent of the americium  could  be released
after incineration at 2000°F for four hours.  Thus, the EPA concludes
that such devices do not pose a significant threat of airborne release
even if incinerated at the end of their useful lives.   (For more details
see NUREG/CP-001, Radioactivity in Consumer Products, U.S.  Nuclear
Regulatory Commission, Washington, D.C., August 1978, pp 434-440.)

     Comment  2.7.5;  Page 15078, second column, E-3, the 1st  line  is a
misquote of the FRC recommendations.  The RPG is 500 mrem/yr.  As an
operational technique, it can be assumed that the average of  a critical
group is one-third of the maximum.  Thus, if the average of the critical
group is 170 mrem/yr, it can be assumed that the RPG of 500 mrem/yr is
met.  (P-lb)

     Response:  The commenter has slightly misstated the FRC  guidance.
The guide is, as stated, 0.5 rem/year when the individual whole body
doses are known.  Where individual whole body doses are not known, a
suitable sample of the exposed population should be developed.  The guide
for the sample population is 0.17 rem/capita/year.

     Comment  2.7.6:  Page 15079, first column, last paragraph.  It is not
possible to compare impacts of sources at different places  from the
maximum individual dose.  The total impact depends upon the
characteristics of the source and the distribution of the population
around the source.  (P-lb)

     Response;  The location of the maximum exposed individual in EPA's
assessments is the point of maximum off-site risk for the particular
source.   Thus, the risk to maximally exposed individuals is directly
comparable from source to source.  The total impact of  the  source does,
as noted, depend on the characteristics of the source and the
distribution of the population around the source.  EPA's analyses account
for population distribution, and the selection of the site  parameters
used in each assessment is made to assure that the health risk assessment
is representative of the source category.


     3.1  Basis for the Standard

     Comment 3.1.1;  To provide protection from long-lived radionuclides,
the EPA should promulgate a cumulative population dose standard as well
as an individual dose standard.  Contrary to the EPA's assertion that
setting emission limits to the general population would serve no useful
purpose, its analysis of population doses and risks from krypton-85,
carbon-14. and iodine-129 shows such doses can be significant from even a
single year's release.  Had the cumulative impacts over the thousands of
years long-lived radionuclides remain hazardous been considered, the need
for a cumulative population dose would be even more readily apparent.
(P-3b, P-17)

     Response:  EPA considered proposing a population as well as an
individual dose limit, but decided against it.  Doses and risks to
populations within 80 km of all DOE facilities due to long halflife
radionuclides are low.  Additional control technology is not practical or
effective.  Emissions are not likely to increase in the future.  EPA did
analyze cumulative impacts over very long periods of time and worldwide;
these impacts were also small.  (See BID page 2.26-1)-  The Agency
concluded that a limit on population dose is not needed.

     Comment 3.1.2;  New radionuclide emission standards under Section
112 of the Clean Air Act are not needed.  The DOE's implementation of the
current standard of 500 mrem/yr.,  together with the ALARA principle,
protects the public health with an adequate margin of safety.  This is
demonstrated by the EPA's own estimates of radiation dose and health
effects from operation of DOE facilities.  (G-la,  G-lb)

     Response:  EPA does not believe that current Federal Radiation
Council guidance and NRC policy of limiting exposure to individuals to
500 mrem/y whole body and 1500 mrem/y to any organ protects public health
with an ample margin of safety, as required by the Clean Air Act.  EPA
estimates that a person receiving 500 mrem/y to the whole body over a
lifetime would have an added potential fatal cancer risk of about 1 in
100 due to the radiation exposure.  In addition, that same person would
face an approximately equal level of risk of nonfatal cancer and of
passing on nonfatal genetic effects to succeeding generations.

     However, EPA recognizes that an "as low as reasonably achievable"
(ALARA) emissions policy has led to generally low emissions of
radionuclides from most facilities.  The risks associated with these
emissions are often insignificant.  The Agency expects that this current
policy will continue in the future and does not anticipate an increase in
the emission levels or the associated risks.  Therefore, the Agency
believes that where a vigorous and well-implemented ALARA program has
achieved low emissions, such practice can provide an ample margin of
safety for public health protection.

     Comment 3.1.3:  The EPA's estimate of the level of risk associated
with existing operations at DOE facilities is lower than its estimate of
the risk associated with uranium fuel cycle facilities.  Yet the EPA does
not propose to regulate uranium fuel cycle facilities under Section 112
of the Clean Air Act, apparently concluding that 40 CFR 190 provides an
ample margin of safety.  In effect, the DOE is being punished for an
effective ALARA program.  (G-la, G-lb)

     Response:  This comment is no longer applicable in that EPA does not
intend to regulate DOE facilities or uranium fuel cycle facilities under
the Clean Air Act.

     Comment 3.1.4:  The EPA's rationale for the proposed standards for
DOE facilities, that they can be met at a cost that the EPA deems
justifiable, is arbitrary.  It is not based on new scientific evidence
and is inconsistent with the EPA's rationale for excluding other sources
from standards, that is, that the benefits do not justify the costs.
(G-la, G-lb)

     Response:  EPA did not propose the emission limits for DOE
facilities primarily on the basis of cost.  EPA considered individual and
population risks, potential for significant emissions, cost and
availability of control technology, and the applicability of other
emission limits already in force.  Costs were considered to a limited
extent (see response to Comment 2.1.7) because it would not be reasonable
to do otherwise.

     A strict cost-benefit rationale would require the risks to large
population groups to be balanced against the cost of control technology.
This approach was not used because Section 112 of the Clean Air Act
requires protection of public health with an ample margin of safety and
for other reasons (see response to Comment 2.1.7).  A cost-benefit
rationale was not used by EPA to justify its determinations against
proposing standards for certain sources (48 FR 15096).

     EPA has decided to withdraw the proposed standards (48 F.R. 15076)
based on its determination that current practice provides an ample margin
to protect the public health from the hazards associated with exposure to
airborne radionuclides from this source category.

     3.2  Dose and Risk Calculations

     Comment 3.2.1:  At the fifteen DOE facilities considered to provide
a dose less than 1 mrem/yr, was the surveillance carried out by EPA?  If
it was not carried out by EPA, I consider it unacceptable.  (P-13b)

     Response:  The source terms for all of our assessments of DOE
facilities are based on monitoring conducted by the facilities.  The EPA
has a high degree of confidence in the reliability of these data and the
integrity of the personnel responsible for them.

     3.3  Control Technology

     Comment 3.3.1:  The EPA has not even considered the availability of
control technologies to reduce such long-lived radionuclides as tritium,
carbon-14, etc.  Such controls are available and should be required.

     Response;   The EPA did consider the availability of control
technologies to reduce long-lived radionuclides emitted from DOE
facilities.  ("Control Technology for Radioactive Emissions to the
Atmosphere at U.S. Department of Energy Facilities," PNL 4621, prepared
for the U.S. Environmental Protection Agency by Pacific Northwest
Laboratory, Richland, Washington).  As stated in 48 FR 15076, the Agency
considered as an alternative to the proposed standard, emission limits
for long-lived radionuclides.  Such an approach was rejected due to the
small population doses caused by such emissions and the unavailability of
effective control technologies for tritium which is responsible for the
highest of these small population doses.

     3.4  Proposed Limits

     Comment 3.4.1:  The proposed limits are twice as high as the NRC
standard for airborne emissions from nuclear power plants (10 CFR 50,
Appendix 1), and are grossly inadequate to meet the requirements of the
CAA which requires technology-forcing standards.  (P-17)

     Response;   The Clean Air Act does not necessarily require
technology-forcing standards.  EPA considers requiring technology-forcing
control technology when the risks remaining after the application of best
available technology are significantly high.  This is not the case for
DOE facilities.  When available technology is used, emissions from DOE
facilities are small and the residual risks both to individuals and
populations are also small.  The control technology currently used and
the application of other ALAP procedures at DOE facilities have resulted
in emissions such that an ample margin of safety is provided.  The NRC
requirement cited by the comment (10 CFR 50, Appendix I) is not an
emission standard but rather a design goal.

     Comment 3.4.2a:  In 48 FR 15076, p. 15081, the EPA requests comments
on setting different limits for different DOE facilities.  We oppose any
such distinction; the limit for the highest DOE facility should be the
limit for all DOE facilities.  (P-16).

     Comment 3.4.2b:  The EPA should propose separate limits for each
type of DOE facility.  This is consistent with what was done under 40 CFR
190.  (P-17)

     Response (Comments 3.4.2a and b):  EPA carefully reconsidered the
need for separate limits for different groups of DOE facilities.

However, EPA has decided to withdraw the proposed standards based on its
determination  that current practice provides an ample margin to protect
the pubic health.

     Comment 3.4.3:  The radionuclide emission standards proposed by EPA
are too stringent, and are not consistent with EPA's standards for other
radiation-releasing activities in the United States.  (G-la, G-lb)

     Response:  EPA has decided to withdraw the proposed standards based
on its determination that current practice provides an ample margin to
protect the public health from the risks associated with exposure to
airborne radionuclides from this source category.

     Comment 3.4.4:  The EPA's proposed 10 mrem/yr whole-body dose limit
for DOE facilities, based on available and reasonable control
technologies,  is laudable.  (G-21)

     Response:  See response to Comment 3.4.3.

     3.5.  Imp1emen t at ion

     EPA received the following comments with respect to implementation
of the final rules.  They are no longer applicable because the proposed
standards have been withdrawn.

     Comment 3.5.1:  The permitting and paperwork requirements of 40 CFR
Part 61, Subpart K, are excessive.  They will be time consuming, costly,
and will hamper research programs.   (G-la, G-lb, G-3)

     Comment 3.5.2:  The requirement that prior permits be obtained
before any modification of the source is totally impractical for the
experimentally flexible, distributed service nature of the DOE
facilities.  (G-3)

     Comment 3.5.3:  For locations where uranium is the critical exposure
radionuclide,  the EPA must specify the calculational approach and
sampling requirements for determining particle sizes and solubility.

     Comment 3.5.4:  The EPA must specify compliance monitoring as
required by the Act.  (P-15a)

     Comment 3.5.5:  The DOE will need four years to complete
modifications at one or more facilities; thus, it cannot meet the fall
1985 implementation deadline.  (G-la, G-lb)

     3.6  Costs

     EPA received the following comments with respect to implementation
of the final rules.  Most are no longer applicable because the proposed
standards have been withdrawn.

     Comment 3.6.1:  The reduction of an already rainiscule risk achieved
by expending $25 million (EPA's estimate) at DOE facilities is clearly
cost ineffective and directly contravenes the requirements of Executive
Order 12291.  (G-lb)

     Response;   Before withdrawing the proposed standards, EPA considered
the costs of other alternatives.  We estimate the annualized cost of
promulgating the proposed standards to be about $500,000.  To promulgate
standards 2.5 times higher than the proposed standards, the annualized
cost was about  $50,000.  These costs are based on recent studies for EPA
by Pacific Northwest Laboratories ("Control Technology for Radioactive
Emissions to the Atmosphere at U.S. Department of Energy Facilities,"
Battelle, PNL-4620 (Final), October 1984).

     Comment 3.6.2:  The $25 million cost estimate made by EPA does not
include such direct costs as complying if increased production activities
are required, costs of shutdowns if it is not possible to comply, or the
costs of assuring compliance without exception for an unusual occurrence
or an accidental release.  (G-lb)

     Comment 3.6.3:  The reductions in risks obtained by the proposed
standard, assuming we can measure incremental doses of 10 mrem/yr, are in
no way justified by the costs.  Using the EPA's estimates, population
doses would be  reduced by 4 person-rems/yr at a cost of $25 million, or
$1.7 billion/health effect averted.  (P-lb)

     Comment 3.6.4:  Preliminary DOE estimates of the costs for improved
monitoring, facility modifications, and increased manpower for monitoring
and reporting greatly exceed the $25 million that the EPA has estimated.
(G-la, G-lb)

     Comment 3.6.5:  The EPA has failed to perform a cost-benefit
analysis for the proposed standards, or the alternative regulatory
actions.  (G-la, G-lb)

     3.7  Other Comments

     Comment 3.7.1:  The EPA has not provided for accidental releases in
the proposed standards, making it virtually impossible to guarantee that
the very low limits will never be exceeded.  (G-la,  G-lb)

     Response:   EPA did not exclude accidental releases from the proposed
rules.  However, this point is now moot since EPA has withdrawn the
proposed rules.

     Comment 3.7.2:  Remedial action sites, accidental releases, and
releases regulated under 40 CFR 191 should be exempt.  (G-3)

     Response:   See response to Comment 3.4.3.


     4.1  Basis for  the Standard

     Comment 4.1.1;   It would be preferable  to  regulate  only  certain
categories of NRC  and Agreement State  licensees under  Section 112.  The
remainder should be  regulated under  section  111.   in addition,  it  is  not
clear if facilities  which are licensed to only  use sealed  sources  are
included in 40 CFR 61.130.  (I-49a)

     Response;  EPA  has decided to withdraw  the proposed standards based
on its determination that current practice protects the  public  health
with an ample margin from risks associated with airborne radionuclides
from this source category.

     Comment 4.1.2:   The EPA  should  consider  the  risks posed  by emissions
from NRC facilities  in relationship  to the risks  from  natural background
radiation and other  risks normally experienced  in life in  developing  its
standards.  (G-2a. G-2b)

     Response:  EPA  recognized that  the risks associated with emissions
of radionuclides from NRC facilities were less  than the  risk  associated
with background radiation when standards  were proposed.  Also,  see the
response to Comment  2.1.11.

     Comment 4.1.3a:  The public health rationale for  setting additional
airborne emission  limits on NRC-licensed  facilities is absent.  If there
is no demonstrable public health benefit, then  there is  no need for the
standard.  (G-2b)

     Comment 4.1.3b:  The EPA has not  demonstrated  that  the current NRC
regulations do not adequately protect  the public;  therefore,  there is no
need for the EPA rule.  Moreover, dual  regulation by the EPA and the NRC
is not warranted and  is discouraged  under Section 122(e)(2) of  the Clean
Air Act.  (G~2a, G-2b, G-3, G-16, 1-17, 1-32, I-33a, P-9)

     Comment 4.1.3c;  The EPA evaluation  indicates  that  emissions from
NRC-regulated activities are  already as low as  reasonably  achievable.
The EPA should therefore find that the  regulations  in  10 CFR Part 20 are
adequate and not promulgate National Emission Limitations  for
NRC-licensed activities.  (G-2b, I-3b)

     Response (Comments 4.1.3a through  c):  See response to Comment 4.1.1.

     Comment 4.1.4;   The EPA  has rejected a more  liberal standard because
the proposed standard is currently being  met by all facilities  in this
group.   If this is actually the case,  then under  the EPA's "factor
analysis" there is no need for the proposed standards.   (G-3, I-49a)

     Response;     See response to Comment 4.1.1.

     Comment 4.1.5a:  The EPA has demonstrated that current levels of
emissions are so low that additional control is not warranted but has not
demonstrated that there will be future potential for large releases.  Nor
is there any reason to expect that NRC controls will not prevent an
increase in risk over the present level.  (G-2b)

     Comment 4.1.5b:  The fact that all licensees appear to be meeting
the standard is not of itself justification for the standard, and the
standard could pose unnecessary restrictions on future operations.
(1-31, 1-40)

     Response (Comments 4.1.5a and b):  The fact that several comments
state that the standard could restrict future operations indicates that
there is a potential for emissions to increase in the absence of a
standard, even with NRC controls in place.  However, EPA does not believe
this will occur, given NRC's ALARA programs.

     Comment 4.1.6:  The origin of the 10 mrem/year limit to any organ is
not given; nor is the term "living nearby" defined.  (T-8a)

     Response:  The proposed numerical limit has been withdrawn.  The
proposed rule was intended to apply to the dose equivalent received by
any member of the public (regardless of whether they "live nearby").

     Comment 4.1.7a:  The EPA's assessment of NRC-licensed facilities
examines only a small number of facilities in only two of the many
categories of facilities licensed and regulated by the NRC.  The EPA
should analyze other classes of licensees, such as plutonium fuel
fabricators, research facilities, and scrap processors.  (G-2a, G-2b)

     Comment 4.1.7b:  The EPA did not mention radiochemical producers in
its background document.  Larger releases from these operations are
likely.  Nor did it note that raw materials licensees may not be able to
meet the proposed standard.   (1-31, 1-40)

     Comment 4.1.7c:  The EPA has unlawfully deprived raw materials
licensees of their right to supply meaningful comments on the proposed
limits by either failing to disclose the relevant data used to arrive at
its decision or by overlooking those licensees and thereby proceeding on
the basis of inadequate data.  (1-47)

     Response (Comments 4.1.7a through c):  Section 3 of the Background
Information Document includes risk evaluations of four categories of
facilities licensed and regulated by the NRC and/or the Agreement
States:  research and test reactors, radiopharmaceutical manufacturers,
radiopharmaceutical users (hospitals), and radiation source
manufacturers.  Previous screening analyses indicate that the facilities
in these four categories have the greatest potential radiological impacts
of all of the facilities mentioned by the commenter.  Research
facilities, given the quantities of material handled and the effluent

control systems used, do not have the potential for releasing substantial
quantities of radionuclides into the air.  Previous analyses also
indicate that emissions from plutonium fuel fabricators  (during periods
when such facilities were operated) were adequately controlled.  The EPA
believes that future activities at these facilities can  be conducted
within an ample margin of safety for protection of the public health.
The third category identified by the commenter, scrap processing, is
carried out by fuel fabricators as part of the uranium fuel fabrication
process.  Thus, these scrap processing facilities are part of the uranium
fuel cycle, and their emissions are already regulated under the limits
established by 40 CFR 190.

     The number of facilities examined in the analyses in Section 3,
while small, is sufficient to assure that the models used to estimate the
airborne emissions and risks for each of the source categories are
indicative of actual risks from real facilities.  Examining larger
numbers of facilities would not have added materially to the analyses,
particularly as most of the roughly 7500 NRC-licensed and 12,500
Agreement States-licensed facilities covered by the proposed rule use
only small quantities of licensed material, much of it in sealed
sources.  The radioactive materials possessed by such licensees do not
have the potential for causing substantial airborne effluents or impacts.

     EPA is not aware of any substantial differences between radiochemical
producers and radiopharmaceutical producers.  EPA's analysis of the
industry covered the production of both.  See Table 3.3-1 of the
Background Information Document.

     It is not clear what the term "raw materials licensees" means.  EPA
assumes it refers to facilities which process or use "source material"
(material which has a uranium or thorium content greater than 0.05
percent by weight).  Facilities which process source material as a part
of the uranium fuel cycle are already regulated by the EPA under Title 40
of the Code of Federal Regulations, Part 190.  There are other processors
of source materials (notably the producers of rare earths and metals such
as columbium and tantalum) which were not covered in the original
Background Information Document.

     All of the data on which the proposed rule was based was in the
public record at the time of proposal.  The record did not contain
emissions data from every single NRC licensee or on every conceivable
subcategory of licensees.  EPA does not agree that this constitutes
inadequate data or that anyone subject to the proposed rule has been
denied the right to supply meaningful comments.  Indeed, the person
submitting this comment also submitted extensive additional comments on
the proposal.

     Comment 4.1.8a;   The NRC limit of 500 mrera/yr (170 mrem/yr) when
sizable populations are involved, is being met by the NRC and Agreement
State licensees.  This limit conforms with the recommendations of
national and international scientific organizations and has been
determined to be adequate to protect the public health and safety.
(1-19, I-37a, 1-38)

     Comment 4.1.8b:   The proposed 10 mrem/yr dose to any organ is a
300-fold decrease from the current standard.  Yet, no data supporting
such a reduction are presented in either the FR notice or the BID.
(1-19, 1-30, I-37a)

     Response (Comments 4.1.8a and b);  See response to Comment 3.1.2.

     Comment 4.1.9:  There is no health justification given for the 10
mrem/yr organ dose limit.  The EPA does not cite any study which detects
any increase in adverse health effects in populations exposed to levels
of radiation above average background, and we do not believe any credible
evidence exists that indicates increased cancer incidence at exposures
below 10 mrem/yr whole body.  (1-19)

     Response;  EPA is not aware of any study which conclusively proves
that increased cancer incidence does or does not occur at exposures of
about 10 mrem/y.  In the absence of such proof, EPA has chosen to assume
a linear, non-threshold relationship to predict the risk of fatal cancer
associated with exposure to radiation.  Under this assumption, any
radiation dose is assumed to pose some risk of damage to health.  EPA
believes this assumption is reasonable in light of presently available
information.  See also Comment 2.25.

     Comment 4.1.10:   The EPA appears to have selected the proposed limit
on the basis that it would be easy to meet and would not require
excessive expenditures for controls.  This has not been documented, and
licensees already use controls to meet the NRC's 500 mrem/yr limit.

     Response:  Nowhere in the record has any facility that could not
meet the proposed standard been identified.  Facilities that use
currently available control technology are operating very far below NRC's
500 mrem/y limit.

     4.2  Dose and Risk Calculations

     Comment 4.2.1;  It is not clear what dose conversion factors were
used in the analysis.  (1-18)

     Response:  Individual dose conversion factors were not inserted into
the docket because of the large volume of data involved.  Single valued
dose conversion factors are not used by the RADRISK code which requires
time dependent doses for the life table calculation.   Dose models used

are essentially those of ICRP-30.  While the Agency can supply
intermediate data for specific nuclides, a complete set of risk estimates
for lifetime exposure to approximately  150 of the most commonly occurring
radionuclides are found in reference (Su81) of Volume 1 of the Background
Information Document.

     The dose conversion factors were not given explicitly in the
original analysis, but their basis is now described in detail in Volume 1
of the Background Information Document.

     Comment 4.2.2:  The EPA has not provided sufficiently detailed
information on its derivation of risk, particularly with respect to
organs exposed due to internal deposition of radionuclides.  (G-2a, G-2b)

     Response:  The Agency has revised its estimates of radiation induced
risks in light of the 1980 BEIR-3 report.  As part of this revision,
risks to particular organs have been explicitly considered.

     Comment 4.2.3:  The EPA uses risk coefficients (e.g., the pulmonary
weighting factor for the lung) which differ from the coefficients
recommended by ICRP-26, and also from those used in its own proposed
guidance for occupational exposure.  (G-2a, G-2b)

     Response:  As the commenter noted, risk coefficients used by EPA
differ from those of ICRP-26 and from those proposed for EPA occupational
guidance.  Both EPA and ICRP risk coefficients for occupational guidance
are for that purpose - not for the general population.

     In the Background Information Document, the Agency noted it was
calculating dose-equivalent rates and risks for maximum individuals and
populations.  Dose rates for individuals were calculated as 70-year
committed dose-equivalents.

     Concerning the public, the ICRP has recently (ICRP Publication No.
39, Ann. ICRP ]A No. 1, 1984) stated the "more rigorous" 70-year
committed dose-equivalent could be applied.  They also noted it would be
inappropriate "... for the commission to recommend average or typical
values of the various parameters as it has been able to do for workers,
and each situation must therefore be dealt with on its own."

     It is quite expected, therefore, that estimates for the general
population will be different from occupational parameters of EPA or

     Comment 4.2.4:  The AIRDOS-EPA computer code is inadequate for
evaluating the dose from gamma-emitting noble gases at locations close to
the release point.  (G-2b)

     Response:   EPA recognizes that calculating external doses due to
sources such as a university research reactor poses special problems.
The potential receptor can be close to the point of release and
dispersion may be dominated by such considerations as downwash or
building wake effects.  Furthermore, the mean free path of photons in air
from emitted noble gases such as argon-41 is of a magnitude comparable to
the scale of distances affecting dispersion.  The doses calculated by
AIRDOS-EPA in such circumstances can either underestimate or overestimate
the expected dose to nearby individuals.  In such circumstances, a dose
assessment which considers such effects more completely than AIRDOS-EPA
would be given consideration by EPA.

     Comment 4.2.5:  There is no indication in the Background Information
Document as to whether or not shielding was accounted for in calculating
population dose from noble gas emissions.  A 0.4 shielding factor for
buildings is appropriate.  (I-8a, I-8b)

     Response;   While EPA did not include a building shielding factor in
its calculations, such a factor would be appropriate for external photon

     Comment 4.2.6;  The EPA's analysis of research reactors is based on
a source term that overestimates emissions from the reference facility,
uses a generic site that overestimates the size of the affected
population, and neglects shielding provided by buildings.  Correction of
the EPA analysis for recent emissions data (0.79), population (0.69), and
shielding (0.4), would result in a regional population dose of 75 person-
rem/year for the reference reactor, less than 25 percent of the 343
person-rem/year estimated in the BID.  (G-2a, G-2b, I-8a, I-8b)

     Response:   The source term and the generic site used in the
assessment were not intended to represent an actual facility.  Rather,
they were chosen to provide a reasonable upper bound of the doses and
risks that could be caused by this class of facilities.  EPA agrees that
the actual collective dose to the population surrounding the MIT facility
is probably closer to the 75 person-rem/year than the 340 person-rem/year
estimated for the referenced facility.  Using a lower source term and
accounting for building shielding would also have resulted in lower
estimates of dose and risk to nearby individuals.  (See also the response
to Comment 4.4.6.)

     Comment 4.2.7;  The AIRDOS-EPA model does not account for building
wake dilution.   Further, as the NRC and NCRP have testified, the code has
not been validated for the way EPA has used it.  (1-30)

     The AIRDOS model was designed to model dispersion from a single tall
st;ck.   Thus, when applied to multiple short vent sources, it
overestimates close-in concentrations by a factor of 5 at 0.5
kUr \-ters.  (1-17)

     Response;  While AIRDOS-EPA does not account for the  increase  in
ground level air concentration due to entrapment of the plume  in  the
building wake, the effect can be approximated by reducing  the  effective
height of the release based on supplementary calculations.  The
dispersion model is not limited to tall stacks — it can be used  for area
sources or short stacks as well.  While it is true that the formulation
of this buoyant rise calculation presumes a stack release, one seldom
encounters releases with a substantial buoyancy flux which are not
associated with a stack.  Studies have been made to compare the
dispersion calculated by AIRDOS-EPA with measured data.  Typically, the
dispersion estimate of AIRDOS-EPA is within a factor of 2 of the measured
value.  (Ref:  D.E. Fields, C.W. Miller, and S.J. Cotter,  "The AIRDOS-EPA
Computer Code and its Application to the Intermediate Range Transport of
B^Kr from the Savannah River Plant," Proceedings Symposium on
Intermediate Range Atmospheric Transport Processes and Technology
Assessment, CONF-801064, October 1981.)

     Comment 4.2.8:  The analysis of the radiopharmaceutical industry in
the BID is based on out-of-date information, omits a number of
radiopharmaceutical manufacturers (such as Union Carbide Corporation),
and greatly underestimates the releases from such facilities.  (I-la,
I-lb, I-lc)

     Response:  The analysis of the radiopharmaceutical industry in the
BID is based on the best information available to the Agency at the time
the analysis was prepared.  Much of the information in the analysis was
drawn from a study performed for the Agency by a contractor (TRI79), and
Union Carbide Corporation's radiopharmaceutical facility was among the
facilities examined by the contractor.  While not included specifically
in the draft BID, recent emissions data from Union Carbide's facility
were also reviewed.  EPA is not aware of any additional information that
would affect the analysis, and requests to the commenter to provide any
such information in its possession, while acknowledged, have not resulted
in any specific information.

     4.3  Control Technology

     Comment 4.3.1:  There are no controls available to limit releases
from research reactors except lowering of power levels or reducing
operating schedules.  (G-2a, G-2b)

     Response:  Releases from research reactors are principally the
result of activation of air by neutrons.  Such releases can be minimized
by the use of carbon dioxide instead of air in beam parts, sealing up
compartments to reduce the escape of activation products (most of which
are fairly short-lived), and replacement of air by carbon dioxide in
compartments to which frequent access is not required.

     Comment 4.3.2:  Control of tritium from reactors using heavy water
is already being carried out with catalytic recombiners.  The small
amounts released by leakage or evaporation cannot be controlled further.

     Response:   EPA agrees that the amounts of tritium released by
leakage or evaporation are small and should require no further control.

     4.4  Proposed Limits

     Comment 4.4.1:  A principal concern with the proposed standards is
the incorrect perception on the part of the public and radiation workers
that the standards establish thresholds above which public health and
safety are jeopardized.  (G-16, 1-32)

     Response;   The policy of the Agency is to accept the linear
non-threshold hypothesis as a prudent basis for regulation.  Under this
hypothesis there is no threshold below which exposures can be said to be
risk-free.  The proposed limit was never intended to be construed as a
threshold above which exposure is unsafe.

     Comment 4.4.2a;  The EPA's failure to consider organ risk
differences is inconsistent with accepted radiation protection practice
(including the EPA's usual approach).  Even the proposed standard for DOE
facilities allows for organ doses three times higher than the whole body
limit.  (G-2a,  G-2b)

     Comment 4.4.2b: The ICRP recommendation to assign a lower weight to
the thyroid dose promotes use of limited resources to provide the
greatest reduction  in risk.  (1-6)

     Response (Comments 4.4.2a and b):   In the proposed rule, EPA allowed
a higher dose for individual organs than for the whole body.  This
comment is now moot since EPA has withdrawn the proposed rule.

     Comment 4.4.3:  The standard should be raised and treated as an
upper limit so as to provide a margin for accidental releases and to
permit the application of the "as low as reasonably achievable" principle
or "best available" control technology.  To facilitate this, a de minimus
level based on the order of the variation in natural background should be
set.  (G-16, I-lb, 1-32, 1-40)

     Response;   EPA has decided to withdraw its proposed standard based
on its determination that current practice protects public health with an
ample margin of safety.  Establishment of a de_ minimus level is

     Comment 4.4.4;  It is not clear whether the limit applies to any
member of the public or to populations residing near facilities.  The
distinction is important because controls required to prevent exposure to
a population can be different from those required to protect an
individual.  (G-16)

     Response:  The proposed limit was intended to be applied to any
member of the public (see 48 CFR 15089, April 6, 1983).  Thus, the
proposed standard was to protect nearby individuals, not the collective
population.  Although controls used to protect the population at large
can differ from those employed to protect the most exposed individual,
the EPA believes that decreasing emissions protects both the nearby
individuals and the members of the general population.

     Comment 4.4.5:  To avoid imposing an unnecessary burden on small
facilities, the regulation should include a small facility relief clause
that would establish a control level below which no dose estimates would
be required.  The control level should account for location (rural vs.
urban), stack height, and mixtures of radionuclides.  (P-18b)

     Response:  The comment does not apply because the standard has been

     Comment 4.4.6:  The proposed limit of 10 mrem/yr is so restrictive
that it could well restrict future power levels, operating times, or
worthwhile experimental programs.  (l-8a, l-8b)

     Response;  The comment does not apply because the standard has been

     Comment 4.4.7:  The construction of new buildings at research
reactor facilities with small sites could result in some doses in excess
of the proposed limits.  (I-8b)

     Response:  The siting of new buildings should be controlled by use
of an effective ALARA program.

     Comment 4.4.8a:  If the standard is set below the current
500 mrem/yr, it should not be set at what is being achieved now,  but
should make allowance for contingencies that might require a level
several times higher.  (I-8b)

     Comment 4.4.8b:  The proposed 10 mrem/yr limit could have a negative
impact upon programs to reduce the volume of low-level waste,
particularly the use of incinerators.  This is contrary to the intent of
the Federal Low-Level Waste Policy Act of 1980.  The EPA has not
considered possible additional costs to further reduce the emissions from
these volume reduction processes in determining whether Executive Order
12291 must be implemented.  (G-2b, G-7, G-18, 1-32, P-lb)

     Comment 4.4.8c:  The EPA's proposed 10 mrem/yr whole body dose limit
for NRC-licensed and non-DOE Federal facilities, based on available and
reasonable control technologies,  is laudable.  (G-21)

     Response (Comments 8a through 8c):  The standard has been
withdrawn.  See response to Comment 4.1.1.

     4.5  Implementation

     Comment 4.5.la:  If a reasonable limit is set, compliance should be
accomplished by environmental monitoring.  (I-33b)

     Comment 4.5.1b;  Compliance for NRC licensees should be managed by
the NRC.  (P-lb)

     Comment 4.5.1c:  Unilateral specification of the means of
demonstrating compliance is not consistent with the memorandum of
understanding between the EPA and the NRC.  (G-2b)

     Comment 4.5.Id;  Demonstration of compliance by means of the AIRDOS
and RADRISK computer codes is neither feasible nor valid because:  the
bulk of the NRC and Agreement State licensees are small institutions and
do not have the capability to run these large, complicated computer
codes, the codes are not applicable to urban sites with surrounding large
buildings, no exclusion areas, varying population, and sources having
multiple stacks, and the results are highly sensitive to changes in input
parameters, so that each licensee would have to develop his own model
(based on long-term study of weather patterns and other local parameters)
and obtain approval from the NRC.  (G-2a, G-2b, G-16, I-lb, I-lc, 1-32,
I-33b, 1-40)

     Comment 4.5.1e:  If it is decided to determine compliance through
the use of computer codes, then the complete code must be supplied to the
users.  That is, all radionuclide-specific data should be supplied by the
EPA, including dose conversion factors, scavenging coefficients, pathways
factors, and so on.  (P-18b)

     Comment 4.5.If;  Many of the NRC's smaller licensees lack the
ability to demonstrate compliance, regardless of whether a computational
or measurement approach to compliance is required.  (G-2a, G-2b)

     Comment 4.5.1g:  It is not clear whether compliance is based upon a
dose equivalent rate, a committed dose equivalent, or a dose equivalent
commitment.  A requirement to compute the actual internal dose rate,
rather than the committed dose equivalent rate, could be cumbersome,
complex, and costly.  (G-2b)

     Comment 4.5.1h:  The EPA must specify compliance monitoring as
required by the Act.  (P-15a)

     Comment 4.5.11;  The proposed standard requires implementation using
calculations based on the EPA's AIRDOS-EPA and RADRISK codes or modeling
techniques which, in the EPA's judgment, are as suitable for a particular
application as the EPA codes.  However, in the FR notice, the discussion

states that implementation will follow established NRC practice, which is
based on review of control measures used by licensees.  Thus, there is
uncertainty as to whether our control measures, as reviewed and approved
by the NRC, will be judged by the EPA to constitute acceptable modeling
techniques.  (I-8a, I-8b)

     Response (Comments 4.5.la through 4.5.1J);  See response to Comment

     Comment 4.5.2a:  It is not clear that all research reactors can
comply with the proposed standard, and the Irwin, Tennessee, plutoniura
fuel facility may not be able to comply.  (G-2a, G-2b)

     Comment 4.5.2b:  If medical facilities are not able to comply or
cannot determine that they are in compliance, health care will be
impacted.  (1-17, P-9)

     Response (Comments 4.5.2a and b);  The analysis of research and test
reactors included a review of airborne emissions from every such facility
required to submit effluent data to the NRC.  Based on this review and
the results of our analysis of the reference facility, we believe all
existing research reactor facilities do comply with the proposed limit.
The proposed limit provided an adequate margin for changes in power
levels, operating schedules, and programmatic changes.  Compliance or
noncorapliance of the Irwin, Tennessee, plutonium fuel fabrication
facility is moot, as the plant has not processed any plutonium in more
than a decade, and is being decommissioned.

     The record strongly supports the fact that all medical facilities
were in compliance with the proposed rules.

     Comment 4.5.3:  The Land Management Bureaus within this department
do not have the responsibility to establish, as suggested by the EPA,
radionuclide standards for public lands.  Indeed, we cannot acquire the
skilled personnel required to develop such standards, nor even assure
compliance with the EPA's proposed standard given our existing skill
mix.  (G-19)

     Response:  See response to Comment 4.1.1.

     Comment 4.5.4:  The doses calculated by the EPA for the reference
research reactor differ from the doses we calculate by a factor of 7.5.
This illustrates the uncertainties that may exist in demonstrating
compliance.  (I-8b)

     Response:  The person submitting this comment is apparently under
the belief that the reference reactor is identical to an existing
facility (one owned by the commenter).  This belief is incorrect.  The
reference reactor is a hypothetical facility assigned a set of parameters

(emissions, meteorology, etc.) for purposes of analysis.  The difference
of a factor of 7.5 comes about by comparing the reference reactor to a
real facility.

     4.6  Costs

     Comment 4.6.1:  The EPA has failed in most instances to consider
costs either for controls or for compliance.  In the few instances where
costs for controls were addressed, the analysis clearly demonstrates that
the benefits do not justify the costs.  (G-2a, G-2b)

     Response:  The Agency believes that giving equal weight to costs and
benefits is inappropriate in developing standards under Section 112 of
the clean Air Act.  Congress intended that public health protection
considerations be primary and that cost be a secondary consideration.
The Agency considered costs to a limited degree consistent with this
overall perspective in reaching its decisions on coal-fired boilers and
elemental phosphorus plants, but the decision to withdraw the standard
for NRC licensees and non-DOE Federal facilities does not rest on cost

     Comment 4.6.2:  The EPA indicates that the paperwork burden of
keeping additional records would be very large and unjustified.  On the
other hand, if licensees are not required to document compliance with the
limitations, then issuing the regulations is futile.  (G-2b)

     Response:  EPA believes the record indicates that all facilities
subject to the proposed rules were in compliance.  The proposed rules
relieved most small facilities of the paperwork burden of demonstrating
compliance.  Since EPA has withdrawn the proposed standard, there will be
no paperwork burden.

     Comment 4.6.3a:  Adoption of this rule could seriously affect
Department of Defense programs in the areas of research reactor
operations, use of depleted uranium shells, nuclear medicine, and nuclear
weapons testing.  Under Section 112, it may not be possible to obtain
Presidential exemption, and therefore the government may be forced to pay
exorbitant costs to pursue essential objectives if the standards are to
be met, even though there is little or no benefit in compliance.  (G-16)

     Comment 4.6.3b:  The high costs of implementing the proposed rule
and demonstrating compliance at such infinitesimally low annual exposure
levels will seriously affect radiopharmaceutical manufacturers,
bio-medical facilities, and medical institutions.  The costs could result
in reduced availability of radiopharmaceuticals, reduced funds available
for bio-medical research, and higher medical costs.  This could
negatively impact public health care.  (1-16,  1-17, 1-32, I-33a, I-33b,
I-37a, 1-41, P-9).

     Response  (Comments 4.6.3a and b):  EPA disagrees  that  the costs of
implementing the proposed rule are high or that DOD programs would have
been seriously affected.  However, these comments have been rendered moot
by EPA's decision to withdraw the proposed rules.

     Comment 4.6.4a:  The EPA has greatly underestimated  the costs of
complying with the proposed rule.  First, the estimate in the BID of the
number of affected licensees is far too low.  NRC testimony cites 7,500
NRC licensees  and 12,500 or more Agreement State licensees.  Second,  the
EPA's estimates of the costs are too  low.  The EPA has assumed that the
NRC's enforcement of the proposed rule would not require  annual proof  of
compliance by  each licensee.  Because of variations from  one facility  to
another, generic methods could not be used.  Proof of  compliance would
cause the expenditure of hundreds of  thousands of dollars for each
radiopharmaceutical manufacturer and  tens of thousands of dollars per
hospital.  The costs will exceed $100,000,000 per year and  therefore a
regulatory impact analysis is required under Executive Order 12291.
(G-2a, G-2b, 1-16, 1-17, 1-32, P-9).

     Response:  See response to Comment 4.1.1.

     Comment 4.6.4b:  The costs of complying with the  proposed standard
have not been  determined, but one licensee estimates capital costs alone
would approach one million dollars at its facility.  (1-19)

     Response;  EPA disagrees that any capital costs would  have been
required to comply with the proposed standard.  EPA believes that all
facilities covered by the proposed rule were in compliance with it.

     Comment 4.6.4c:  Compelling each licensee to monitor or model
emissions to demonstrate compliance with this inordinately  low level of
exposure will  increase costs for the  facilities and the NRC as the
regulating agency without any tangible health benefit.  (1-19)

     Response:  Compliance is not necessary since no numerical standard
is now proposed.

     4.7  Other Comments

     Comment 4.7.1:  The EPA's statement that research and  test reactors
are not required to submit data on air emissions to the NRC is
incorrect.  (I-8a)

     Response;  In the past, not all research and test  reactors were
required to submit data on air emissions to the NRC.   A requirement to
include information on air emissions in the annual report to the NRC is
being added to the technical specifications as each license is renewed.

     Comment 4.7.2;  The EPA should clarify the applicability of Its
proposed standard to licensees using only sealed sources and/or
conducting enclosed activities that result in low level fugitive
emissions into enclosed areas.  (1-49)

     Response:  Licensees using only sealed sources were excluded from
the proposed standards.


     5.1  Basis for the Standard

     Comment 5.1.la:  Because radon emissions from uranium mines pose
insignificant health risks, regulation under Section 112 of the Clean Air
Act is inappropriate.  (I-3b, 47)

     Comment S.l.lb:  Has  it been shown that there is a significant
increase in radon near mine vents?  The Salt Lake City tailings pile
releases over 5000 curies  per year; yet it cannot be detected over
natural radon one-half mile away.  (P-lb)

     Response (Comments 5.1.la and b):  EPA has assessed the risks from
radon emissions from underground uranium mines and has concluded that the
risk of fatal cancer to individuals living near large underground uranium
mines and to the total population both regional and national is
significant.  Data collected in the State of New Mexico shows that
radon-222 concentrations in air near underground uranium mines are
significantly elevated above naturally occurring levels.

     Comment 5.1.2a;  The  proposed standard is wholly unacceptable.  The
residual risks are far too high.  (P-15a)

     Comment 5.1.2b:  The  high background radon concentrations in the
vicinity of mines increase the need to regulate the emissions of radon
from mines.  (P-3b)

     Comment 5.1.2c:  People living in areas close to uranium mines
should be protected and should receive the same level of protection as
people living in more populated areas receive from emission of
radioactive material.  (1-52)

     Response (Comments 5.1.2a through c);  EPA agrees that individuals
living near underground uranium mines should be protected from risks
caused by radon-222 emissions from these mines.

     The Agency orginally  proposed a standard to limit radon
concentrations in air due  to emissions from underground mines.  Based on
public comment, it is the  EPA's judgment that it is not feasible to
prescribe or enforce an emission standard for radon-222 emissions from
underground mines because  radon-222 cannot be emitted through a
conveyance designed to capture the gas under current conditions.  The
Agency considers the risks from underground uranium mines to be
significant and believes action is needed to protect populations and
individuals living near underground uranium mines.  Therefore, EPA has
decided to begin development of work practice, design, equipment, or
operational standards to control radon release from underground mines.

     Comment 5.1.3:  The EPA should include abandoned mines in the
proposed standards.  Battelle Pacific Northwest Laboratories has informed
EPA that radon emissions from abandoned mines constitute a hazard.
(P-3a, P-3b)

     Response:  EPA has assessed the radon emissions from inactive
underground uranium mines (EPA 520/1-83-007).  These emissions were found
to be quite small.  The radon-222 emissions from an inactive underground
mine were estimated to be about 1/500 of the radon-222 emission from an
average large operational underground mine.  Based on this assessment,
EPA has concluded that a standard for inactive underground uranium mines
is not needed.

     Comment 5.1.4;  The EPA should set an emission limit, the standard
preferred by Congress, rather than restricting the annual average
concentration of radon in the vicinity of mines to 0.2 picocuries per
year above background.  (P-3b, P-15a)

     Response:  See response to Comment 5.1.2.

     Comment 5.1.5;  The draft study of radon done by the New Mexico
Environmental Improvement Division should be considered by the EPA in
developing the radon standard for uranium mines.  (1-52)

     Response;  This study has been considered by EPA in evaluating
underground uranium mines.

     Comment 5.1.6:  Using the philosophy of comparable risks, ICRP-26
recommends a risk to critical groups of 1 x 10~Vyear.  For radon, this
is equivalent to approximately 0.75 WLM/year, which corresponds to a
radon concentration of 3 pCi/1.  The 0.2 pCi/1 standard originally
proposed by the EPA results in a lifetime lung cancer risk of 0.03
percent, or 5 x 10~6/year.  The lifetime risk of 0.03 percent, when
added to the average total risk of all cancer deaths of 16.7 percent,
results in an increased risk that is probably epidemiologically
undetectable.  (1-17)

     Response;  The commenter has apparently misread ICRP-26.  ICRP-26,
paragraph 122, states, relative to a critical group, "In these cases, the
dose-equivalent limit to individual members of the public, referred to in
paragraph 120, would still adequately restrict the average dose
equivalent, but the few individuals exposed to the dose-equivalent limit
could run a risk in the range of 10~5 to 10~4 per year.  This annual
risk would then be one order of magnitude higher than the risk range
quoted in paragraph 118.  Since exposures at the dose-equivalent limit
are not likely to be repeated over many years, however, an adequate
restriction of the lifetime dose is still likely to be achieved.  In rare
cases where the doses to a few individuals were actually found to be
received at high rates over prolonged periods, it would be prudent to
take measures to restrict their lifetime dose as implied in paragraph

     Paragraph  119 restricts the  lifetime dose to a value corresponding
to 1 mSv per year of life-long whole body exposure, an average risk of
approximately 1 x 10   per year.

     Based only on epidemiology,  and not including estimated dosimetry
and weighting factors, the ICRP estimated a total lifetime risk of 1.5 to
4.5 x 10~VwLM.  This estimate was based on selected data and a 30-year
mean manifestation period.  (ICRP-32 Ann. ICRP 9, No. 1, par 8, 1981.)
While this might be reasonable for an occupational exposure in miners, it
is not valid for a population exposure and hence would not be used in any

     The commenter appears to have lost a factor of 10 in translating the
Agency's risk estimate for 0.01 WL to that for 0.2 pCi/1.  The Agency
estimated a lifetime risk of .017 for fatal lung cancers for continuous
exposure at 0.01 WL or about 2.4  deaths per year per 100 person WL (BID
p B-12).  This would be 2.4 x 10~2 deaths/year per person WL or 2.4 x
10~5 deaths per year per person at 0.001 WL.  Continuous exposure at
0.2 pCi radon per liter would give an average individual risk of 2.4 x
10~5 if equilibrium is 0.5; or 3.3 x 10~5 if equilibrium is 0.7. not
the 5 x 10~6 risk/year calculated by the commenter.  Likewise, the
lifetime risk would be 0.17 percent or 0.24 percent depending on
equilibrium factor used, not 0.03 percent.

     While it is probably true that the increased risk is probably
epidemiologically undetectable, that does not make it less real.

     Comment 5.1.7;  The reliance on land control rather than setting an
emission standard is contrary to  the Act.  (P-15a)

     Response;  See response to Comments 5.2.la through c.

     Comment 5.1.8a:  The EPA should follow the system employed in 40 CFR
190, setting the standard for the nearest inhabitant.  (1-26)

     Comment 5.1.8b;  The standard should be established to protect real
people from unreasonable risk.  (I-3b)

     Response (Comments 5.1.8a and b):  See response to Comment 5.1.2.

     Comment 5.1.9;  We strongly  support the need for standards, but the
proposed standard fails to consider regional impacts from multiple
sources.  Such an approach would  allow radon to accumulate and exceed the
proposed standard based on a single source.  (G-20, P-15a)

     Response;  EPA will consider regional impacts from multiple sources
in developing any design, equipment, work practices, or operational
standard for underground uranium mines (see Section 5.6 of BID).

     Comment 5.1.10:  The EPA should consider setting a standard that
includes background, as this would make monitoring easier and less
costly.  (G-20)

     Response:  See response to Comment 5.1.2.

     Comment 5.1.11:  In a 1972 report, the EPA stated that uranium
mining and milling were insignificant sources of radioactivity when
compared with natural background levels.  Nothing has been developed
since that time to change that conclusion.  In fact, this was reinforced
in the NRC's most recent report on the subject (NUREG-0757).  (I-3b)

     Response;  While emissions from uranium mining and milling are only
a tiny fraction of natural background radiation,  they can have a
significant impact on nearby individuals and the total population.  The
elevated concentrations in the vicinity of a mine vent can result in
relatively high risks to exposed individuals.

     Comment 5.1.12;  Natural radionuclides are not the type of
pollutants that were intended to be regulated under Section 112 of the
Clean Air Act.  (I-3a)

     Response:  Section 112 of the Act does not distinguish between
man-made and naturally occurring pollutants.  In absence of such a
distinction, we can only conclude that Congress intended EPA to regulate
all sources of radionuclides presenting a significant risk to the public

     5.2  Dose and Risk Calculations

     Comment 5.2.la:  The assumptions and parameters used in the EPA's
radon-222 risk analysis do not conform to either  accepted values
recognized by the radiation health community or available evidence.  In
its analysis, the EPA:

     a.  Overestimates the rate of radon release.

     b.  Neglects the effect of plume rise.

     c.  Assumes a one micron particle size instead of using the known

     d.  Uses an inappropriate model for estimation of dispersion and

     e.  Overestimates the equilibrium factor between radon and its

      f.  Neglects  the  latency  period  for  the  development  of lung cancer.

      g.  Assumes an  exposure period of  70 years  to  a  mine whose average
         life  is only  15  years.

      h.  Uses  estimates that are  higher than  ICRP and NCRP recommended

      i.  Fails to  justify the  use of  relative instead of  absolute risk.

 (I-lb, I-3b, I-3c, 1-22,  1-23, 1-26,  1-31,  I-37b, 1-47, P-lb)

      Response  to a;  The  Agency did not estimate the  rate of radon
 release from underground  mines.   Rather,  the  Agency used  radon  emission
 data  from actual measurements  performed by the Battelle Pacific Northwest
 Laboratory  (PLN) (Ja80).   These data  were then used to select the
 parameters  for the reference underground  uranium mine which, in turn,  was
 subjected to further analysis.

      Response  to b;  The  comment  is correct.  The Agency  did not consider
 the effect  of  plume  rise  when  calculating radon  exposures and risks.
 Plume rise  is  considered  in the final BID where  estimates of radon
 exposures and  risks  for both ground level releases  and releases with
 plume rise  are presented.  The plume  rise estimates apply to vertical
 discharge vents and  the ground level  estimates to horizontal discharge
 vents.  Currently  the  Agency has  no reliable  information  regarding the
 relative fractions of  horizontal  and  vertical discharge vents.

      Response  to c;  The  calculations for radon-222 concentration do not
 use particle size  as a parameter.

      Response  to d;  The  dispersion method  used  by  the Agency has  been
 the basic workhorse  of local dispersion estimation  for years.   In 1977,
 the participants of  an expert  group assessing atmospheric  transport of
 radionuclides  concluded that,  for distances out  to  10  km  in reasonably
 flat  terrain,  and  given good local wind observations:  "Accuracy for the
 usual annual average concentration is about a factor  of +  2."   (Ho78)
 Furthermore, these dispersion estimates are based on  an empirical
 approach that  is inherently unbiased  and  that should  therefore  be  as
 likely to overpredict  as  to underpredict.

      The model  used  to predict concentrations near  mine vents is  also a
 proven model used  by EPA  for many years.   The Industrial  Source Complex
model has been  used  by the EPA air pollution program  for  industrial

     Response  to e:  The  commenter is questioning the  use  of an
equilibrium fraction of 70 percent for  radon daughters in  structures.
Observations of this fraction are highly  variable.  The degree  of

equilibrium depends on many variables such as ventilation and plate out
but is also strongly dependent on the degree of equilibrium which exists
in the incoming air.  For example, a calculation using a ventilation rate
of 1 h"1 and an effective plate out rate of 1 h"1 yields an
equilibrium fraction of 0.36 for an initial equilibrium fraction of 0 and
0.68 for an initial equilibrium fraction of 1.0.  Since the degree of
equilibrium in the air entering structures within areas adjacent to a
source would be expected to be low under normal meteorological
conditions, EPA has reduced its estimate of the equilibrium fraction by
one half for the calculation of risk to maximum exposed individuals.
Although some have suggested lower values, these appear to be based
primarily on studies of structures in which the radon entered only by
diffusion so that the initial equilibrium fraction was zero.

     At some distance from a source, the equilibrium of radon in air may
approach 1.0, so for structures at some distance from a source, an
equilibrium value of 0.7 will still be correct for population risk

     Response to f;  EPA uses a life table approach to calculate
stochastic risk.  The latency period, i.e., the minimum induction period,
and the risk plateaus, are an integral part of these calculations.
Latency is accounted for on an annual basis for each successive year of
life.  The methodology is described in Appendix B of the Background
Information Document.

     Response to g:  The comment refers more to individual risk estimates
than those for the region or nation.  A worked out mine is quite likely
to be replaced by another mine in that region so that regional and
national exposures will continue.  If the distribution of individuals
near the new mine is similar to that around the worked out mine, the
individual risks will, on the average, be the same as if the original
mine continued across the life span of both mines.  So it is not certain
that even the individual risk estimates are excessive.

     Response to h:  The Agency reviewed ICRP and NCRP estimates in the
"Final Environmental Impact Statement for Standards for the Control of
Byproduct Materials from Uranium Ore Processing (40 CFR 192) Volume II"
(EPA 520/1-83-008-2).  The response to this comment is adapted from that
     The current Agency estimate for radon-222 exposure is 760 fatal
     rs per 106 per:
expression of risk.
cancers per 106 person WLM for lifetime exposure and lifetime
     A comparison of the Agency estimate and assumptions compared with
those from ICRP and NCRP is given below:

     I.    EPA 760/106/WLM
           a.  lifetime exposure  from birth  (0-110 yrs. with  actuarial
               probability of death due  to all causes  in  this interval)

           b.  a relative risk of 3%/WLM

     II.   ICRP - 150-450/106/WLM (ICRP-Report 32)
           a.  risk of 5-15/106/yr/WLM
           b.  mean manifestation period of  30 years
           c.  exposure from age  18 to 65

     This epidemiology based estimate of the ICRP is not  directly
comparable to EPA's.  The periods of opportunity for expression  in the
EPA model are greater since exposure is  considered to  start at birth and
the mean life expectancy is 70 years.  While the numbers  cannot  be
compared because of differences in age structures and  competing  risks in
the two populations, a first approximation could be made  by doubling the
ICRP estimate to account for the more extended period  for expression.  In
this case, the ICRP estimate would be equivalent to 300-900/10*VWLM.

     It should also be noted that the ICRP assumption  of  an annual risk
5-15/10^/yr/WLM is an estimate based on  values ranging from 2 to
20/10^/yr/WLM, which were averaged over  all  age periods during
occupational exposure.  Their 2-20/10^/yr/WLM is the range of estimates
they took from epidemiologic studies.  The true range  is  greater since
Archer(16> documented a range of  1.4 to  35.0/106/yr/WLM and BEIR-3
documents a range of 6-47/10^/yr/WLM.  The ICRP estimate  is perhaps a
factor of 2 too low for occupational exposure and a factor of 4  too low
for valid comparison with the EPA estimate.

     III.  NCRP - 80-200/106/WLM  (unpublished, referred to by the
           a.  10/106/yr/WLM
           b.  no lung cancers before age 40
           c.  induced cancers disappear exponentially with a halftime of
               20 years
           d.  lifetime exposure  from birth  (0-85 + years)

     While EPA cannot use data from an unpublished draft  report  in
support of rulemaking, this estimate can be  compared to EPA's in the
discussion of comments.  The NCRP model  has  a unique feature  not found in
other lung cancer risk models, i.e., radiation induced lung cancers
disappear or become unavailable for expression with time.  Mathematically
a function is introduced which removes cancers exponentially  with a
halflife of 20 years.  It is not clear why this function  is introduced or
where it is supported by valid observations  and analysis.  If this

function is removed,  the NCRP estimate increases from about 80-200/106/WLM
to 200-500/106 WLM.

     Moreover, the NCRP estimates an average of 10/106/yr/WLM lung
cancers could occur.   As noted above, Archer(^) documented a range of
1.4 to 35/106/yr/WLM and BEIR-3,  6 to 47/106yr/WLM.  It is not certain
to what extent NCRP considered the entire range of risk estimates but
their risk coefficient 10/yr/WLM is about half of the BEIR-80 risk
coefficient (averaged for all ages).  In addition, absolute risk
coefficients should be weighted by the length of the follow-up in the
study, since for less than lifetime follow-up the absolute risk
coefficients increase with increasing length of observation,
(17).  For example,  a weighted average for exposures to less than 500
WLM in Archer's paper yields about 17 cases /106/yr/WLM.  It is likely
that the NCRP estimate is another factor of 2 lower than it should be.

     IV.   ICRP - 45-138/106/WLM (ICRP-32)

     The ICRP dosimetric approach is subtle.  As ICRP Report 32 points
out, their risk concept assumes a proportional relationship without
threshold between the dose to relevant target tissues and the associated
excess probability for the induction of cancer.  ICRP 32 then continues,
"On the basis of this concept the risk-relevant dosimetric quantities for
radon daughters in the lung are the mean dose or dose equivalent to the
two target tissues mentioned above, the basal cell layer in the tracheo-
bronchial (TB) region and the mean dose to the epithelium in the
pulmonary (P) region."  ICRP 32 then uses dosimetric models to calculate
the dose equivalents of interest; these models contain assumptions that
introduce considerable uncertainties.  These uncertainties coupled with
ICRP assumptions on weighting factors and quality factors lead to very
uncertain conclusions.

     However, even if these models are correct, ICRP's use of the mean
dose-equivalent for bronchial tissue reduces the REM/WLM by about a
factor of 3 since the dose equivalent in the region of the lung where
most cancers develop (lobar, segmental, and subsegmental bronchi) is
about 3 times higher than the mean dose calculated in the models
referenced by ICRP 32.  Likewise, ICRP split the risk weighting factor
for lung cancer between bronchi and pulmonary lung, so that each has 50
percent of the risk originally calculated by ICRP for the lung.  This
reduces the REM/WLM by a factor of 2.  Since no radon daughter-related
cancers have ever been observed in the pulmonary lung, this approach
introduces a likely error in the estimate for bronchi cancer by a factor
of 2.  Correcting for these ICRP assumptions would increase their risk
estimate to about 270-828/106/WLM.  Adjustments for lack of follow-up
in the studies from which ICRP derived its estimate of 1.2 x 10~2
cancers/Sv (ICRP 26)  would increase the risk estimates still further.

     Response to i;   The absolute risk and relative risk models were
introduced in BEIR-1 and used in BEIR-3.

     As currently defined:   (BEIR-3)

     Absolute Risk -  "Expression of excess risk due to exposure as the
arithmetic difference between the risk among those exposed and that
obtaining in the absence of  exposure."

     Relative Risk -  "Expression of risk due to exposure as the ratio of
the risk among the exposed to that obtaining in the absence of exposures."

     The risk projection models are also defined in BEIR-3:  Absolute
Risk Projection Model which  states, "According to this model, if a
population was irradiated at a particular dose either all at once or over
some period, expressions of  excess cancer risk in that population would
begin at some time after exposure (the latent period) and continue at a
rate in excess of the expected rate for an additional period, the
'plateau' or expression period, which may exceed the period of followup.
In this model, the absolute  risk is defined as the number of excess
cancer cases per unit of population per unit of time and per unit of
radiation dose, and,  though  it may depend on age at exposure, it does not
otherwise depend on age at observation for risk."

     Relative Risk Projection Model - "In the second model adopted in
BEIR-1, the so called relative-risk model, the excess cancer for the
interval after the latent period was expressed as a multiple of the
natural age-specific  cancer  risk for that population.  The chief
difference between the two models is that the relative-risk model took
account of the differing susceptibility to cancer related to age at
observation for risk."

     "If the relative-risk model applies, then the age of the exposed
groups, both at the time of  exposure and as they move through life,
becomes very important.  There is now considerable evidence in nearly all
the adult human populations  studied that persons irradiated at higher
ages have in general  a greater excess risk of cancer than those
irradiated at lower ages, or at least they develop cancer sooner.
Furthermore, if they  are irradiated at a particular age, the excess risk
tends to rise pari passu with the risk of the population at large.  In
other words, the relative-risk model with respect to cancer susceptibility
as a function of age  evidently applies to some kinds of cancer that have
been observed to result from radiation exposure."

     The 1980 BEIR Committee also makes some caveats regarding the
certainty of our knowledge of the two models.  However, they later
commented (p. 137), "If risks are given in absolute form - i.e., number
of cancers induced per unit of population and per unit of radiation
exposure - then a single value independent of age may be inappropriate."
For lung cancer the 1980 BEIR Committee used no single value but a risk
coefficient which varied with age.

     Since BEIR-3 provides an extensive analysis of lung cancer induction
due to radon progeny,  the projection model they used - age specific
absolute risk - was compared to the relative risk model used by EPA.
Since there was very little numerical difference, EPA has concluded that
its relative-risk model and the BEIR 1980 age-dependent absolute risk
model are essentially equivalent.

     A study by Land and Norman supported a relative risk model for lung
cancer.  In the discussion of a Smith and Doll paper, on page 216, Smith
said, "The statement I made was based on our finding that the risk of a
radiation-induced cancer seems to increase with age at exposure in direct
proportion to the expected number of deaths from cancer that would be
suffered by persons first treated at a particular age.  This suggests to
us the radiation is interacting with whatever other factors are inducing
cancer."  This statement is comparable to the definition of relative-risk
given by BEIR-3.  Since BEIR-3 (p. 312-313) reported a significant
increase in lung cancer in the population studied by Smith and Doll, it
is reasonable to consider lung cancer one of the cancers described by
Smith's statement.

     A Kato and Schull report (H. Kato and W. J. Schull, Radiation Res.,
9J):  395-432, 1982) concluded, "Thus, though in the recent BEIR report
two different models (relative risk and absolute risk models) have been
used for projection of risk beyond the period of observation, the present
data support the relative risk model projection more strongly.  The
excess deaths from cancers other than leukemia increase with age at death
for the same age cohort in proportion to the age-specific death rate from
cancers in the population of all Japan and do not show a constant excess
value by age at death for the same age cohort."

     A report by Shi-quan and Xiao-ou shows that even at ages less than
10, exposure to radon does not lead to lung cancer before the age at
which "spontaneous" lung cancer develops.  This pattern is consistent
only with relative risk and age specific absolute risk lung cancer
induction models.  It is not consistent with simple absolute risk; thus,
it lends additional support to EPA's use of the relative risk model.

     In an extensive review of the health effects of alpha radiation for
the Canadian Atomic Energy Control Board, the authors selected a relative
risk projection model and concluded the best estimate of excess relative
risk for the radon exposure data was 2.28 percent, +0.35 percent, and
that this was unlikely to underestimate the excess risk at low doses by
more than a factor of 1.5.  They further concluded the risks from radon
daughters and smoking were intermediate between multiplicative and
additive but on the balance chose multiplicative.

     Most recently Prentice, e_t al.. (R.L. Prentice, et al., JNCI,
70:611-622, 1983) and Whittemore and McMillan, (A.S. Whittemore and A.
McMillan, Technical Report No. 68, prepared for SIMS, Stanford

University. California,  1983) have used the relative risk projection
model on estimating radon related lung cancer hazards.

     In conclusion, EPA  believes the record clearly justifies  the use of
relative risk for radon-222 exposures.
     Comment 5.2.1b:  The EPA assumption of a risk of  1 or 2  in  100  for
exposure to 0.007 to 0.014 working  level months seems  high.   R.D. Evans,
et al.'s risk estimates would lead  to 2 to 5 x 10~3; the NCRP
Subcommittee 57 report gives about  6 x 10~3.  The risk estimates used
by the EPA are consistently high.   For example, the radon risks are  cited
as 1-2 per 100 lifetime exposure to an equivalent concentration of 1-2
pCi/1 radon.  The same risks are given by R. D. Evans, et al. and by a
recent NCRP committee as 0.2 - 0.5  in 100.  (P-la, P-lb)

     Response;  The Agency does not feel its risk estimates for radon
exposure are too high compared to other estimates.
EPA (a)
WAS BEIR-3 (a)
AECB (b)

NCRP (c)
Evans, et al .
Fatalities per
106 person WLM

40 years
30 years

     (a)  Assumes increased exposure during childhood, due to high minute
          volume for lung size compared to adult.

     (b)  Adjusted for U.S. General Population.

     (c)  Assumes risk diminishes exponentially with a 20-year halftime.

     Specific aspects of the ICRP and NCRP (Harley and Pasternak) models are
mentioned in the response to Comment 5.2.1h.

     It is uncertain how the Evans et. al. estimate was made.  The paper goes
from selected annual risk coefficients in some miners to a lifetime estimate
without indicating latent periods, period of expression, etc.  Apparently
they assumed a 15 year expression period even though the documented risk is
over twice this long.  These problems, in addition to other questionable
assumptions, cast doubt on the validity of this estimate in comparison to
more fully documented estimates for which there is an adequate rationale.

     Comment 5.2.1c:  A uranium mine only has a working lifetime of 10 to
15 years.  (I-la,  I-3a)

     Response:   EPA's model mine was assigned a 20-year operational
lifetime.  The  choice of this parameter was based on information
presented in the Final Background Information Document (Table 5-8) which
shows that during 1978-1979 the age of underground uranium mines ranged
up to about 30  years, with the age of older mines averaging 21 years.
See response to Comment 5.2.1g.

     Comment 5.2.Id:  The EPA is bound by the requirements of the Clean
Air Act to evaluate the potential risks and establish standards based on
reasonable assumptions.  Moreover, the assumptions must be explained,
justified, and  presented with margins of error.  By failing to do this,
the EPA has misled the Administrator and the public as to the risk from
underground uranium mines.  (I-3b)

     Response:   EPA believes its risk estimates are based on reasonable
assumptions and valid methods.  The assumptions and methods used by the
Agency to estimate risks from radionuclide emissions to air, and the
uncertainties in these estimates are described in the Final Background
Information Document.

     Comment 5.2.2:  The EPA should better explain its rationale for the
differences between its estimates of risk for uranium mines and those for
uranium milling.  (I-lb)

     Response;   Radon produces the greatest risk from both of these
facilities.  There is little difference in the methodology used to
estimate risk from radon from uranium mines and from uranium mill
tailings.  The  most significant difference is that the sources of radon
have different  configurations - area source for tailings and point source
for uranium mines.  However, this difference is rather small since most
mines have several vents (several point sources) which in practice can be
considered an area source.  Therefore, in modeling environmental pathways
for radon risk  estimation, differences are minimal between mines and mill

     Probably the greatest difference between the two is that tailings
can be (and have been) used as a construction and/or fill material around
structures.  This leads to very high risk estimates from the buildup of
indoor radon decay products.  Two important considerations apply here.
First, this pathway of misusing tailings does not exist for uranium mine
radon releases.  Second, EPA does not believe there is a reasonable way
to model or estimate the impact of this pathway since it is highly
dependent on local mines, etc.  Therefore, this misuse problem must be
considered qualitatively, thereby greatly limiting comparative analyses.

     Comment 5.2.3;  Epidemiological studies  in Colorado and Pennsylvania
have failed to show any increased risk from radon.   (1-31)

     Response:  As noted by  the Colorado Department  of Health  (Comment
2.2.5c),  "It was stated earlier in this [Denver] hearing that  the
epidemiological studies done in Grand Junction by the Colorado Department
of Health have not indicated that there is a  bioeffect from exposure to
the radiation levels observed.  This is not a conclusion of the study.
[emphasis added]."

     The Pennsylvania study  (of Dr. Steven Lane) did not have  adequate
power to detect the presence of a radiation effect if it was present.
See response to Comment 2.2.5a.

     In neither case was a sufficiently large population followed for a
long enough time.  Neither study is capable of demonstrating the
increased risk of radiation  bioeffects if they are present.

     Comment 5.2.4:  The source term for radioactive airborne  emissions
used by the EPA is overestimated by the assumption that all particles
have a diameter of 1 micron.  Thus, the EPA overestimates the  source terra
that can be transported to individuals living near a uranium mine.  (1-17)

     Response;  EPA did not  assume a one micrometer  particle diameter in
estimating the source term for uranium mines.  The only radionuclide
considered for a mine was radon.  It is an inert gas (deposition velocity
of zero).  Its daughters were taken to be at  70 percent of their
equilibrium value.  This 70  percent factor has been  changed to a distance
dependent percent and is shown in the Final Background Information

     Comment 5.2.5;  Extrapolation of lung cancer risks from occupational
exposures of miners understates risks to children who may spend almost
100 percent of their time indoors, and ignores the synergistic effects of
simultaneous exposure to radiation and other  toxins.  (P-15a)

     Response;  The Agency disagrees since the lifetime risk of
continuous exposure has been calculated.  Infants, who spend most of
their time indoors, contribute little to the  lifetime risk of  continuous
exposure.  Even if children  remained indoors  100 percent of the time
until school age (age six),  their increased contribution to exposure
would be about 3 percent of  the total exposure, much less than the
uncertainty in the calculation.  Age-specific differences in physiology
are incorporated in the Agency model and, since a relative risk
projection model is used, all synergisms are  automatically included.

     Comment 5.2.6:  The EPA has presented the risk  estimates  in a manner
that is misleading.  As an example, in the case of underground uranium
mines,  the EPA states that a hypothetical maximum individual who lives
500 meters from an exhaust vent and is constantly exposed will have an

increased risk of lung cancer of 1 or 2 in 100.  This distorts the
potential risk to the public.  EPA also acknowledges that a large
underground uranium mine would require 30 years of operation to produce 1
health effect.  This is twice as long as the ordinary life of such a
mine.  (I-3b)

     Response;  The Agency did not intentionally overestimate the risk
from radon to nearby individuals.  There are 42 people who live within
500 meters of uranium mines (see Table 5-12 of the final BID).  There are
8 mines that have an average life of 21 years and range fron 17 to 29
years old (see Table 5-7 of the final BID).  [The risk estimates have
been modified by presenting a range of radon risks to reflect the
difference of opinion among the scientific community regarding the risk
from exposure to radon decay products.]

     Comment 5.2.7;  EPA's risk coefficient for exposure to radon decay
products is not supported by observations among underground uranium
miners.  (I-3a)

     Response;  The Agency disagrees.  The Agency risk coefficient for
radon is derived from the uranium miner data base.  It is numerically
almost identical to that in BEIR-3 (reference NAS806) and within the
environmental level risk coefficients derived by Archer for the same data
base (reference Ar79).  It is also quite similar to environmental level
risk coefficients in an AECB of Canada review of alpha radiation (D.C.
Thomas and K.G. McNeill, "Risk Estimates for the Health Effects of Alpha
Radiation," INFO-0081, Atomic Energy Control Board, Ottawa, Canada, 1982).

     Agency risk coefficients are not the lowest reported, but they are
not the highest either.  They are, hopefully, realistic, but prudent.

     Comment 5.2.8;  Measurements at Ambrosia Lake suggest that natural
background sources of radon overwhelm contributions of radon from uranium
mines.  (I-3a)

     Response;  While it is true that background sources will exceed
radon emitted from underground uranium mines with respect to population
exposures, this is not the case with respect to doses received by
individuals living near mine vents.  For individuals close to uranium
mines, the radon releases from the mine vents can cause a significant
exposure and risk.

     Comment 5.2.9;  EPA assumes an equilibrium of 70 percent for indoor
radon decay product concentrations.  Measurements at Ambrosia Lake
indicate that the equilibrium factor is far less than 70 percent.  (I-3a)

     Response;  See response to Comments 5.2.la. Part e, and Table 5-10
in the final Background Information Document (BID).

     5.3  Control Technology

     Comment 5.3.1:  The A.D. Little study done for the EPA identified a
number of options for the removal of radon from mine exhaust, the most
promising of which appears to be activated carbon filtration.  Given this
technology, the high level of risk from mine radon in the proposed
standards is unjustified.  (P-15a, P-15b)

     Response:  Studies performed for the Agency have identified a number
of control technologies to control radon releases for underground uranium
mines.  The effectiveness and costs of these technologies are discussed
in the BID.  Activated carbon filtration of radon in mine ventilation air
has not been demonstrated in actual mines and is not believed to be
practical.  In EPA's consideration of a work practice standard, the
Agency will consider the use of activated charcoal filters in conjunction
with bulkheading.

     Comment 5.3.2:  The three control options for mines are impractical.

     a.  Private companies have no power of eminent domain; therefore,
         purchase of surrounding land at a reasonable price will be

     b.  The methods suggested by the EPA to control radon emissions are
         not practical; if they were, they would already be in use.

     c.  The use of tall stacks is not practical for a number of
         reasons.  Even if they were, the proposed standard gives no
         credit for them.  (1-26, I-37b, I-43b)

     Response:  See response to Comment 5.1.2.

     Comment 5.3.3:  The EPA wrongly assumes that radon reductions from
sealant coatings, bulkheading, activated carbon, and backfilling are
additive.  (I-3b)

     Response:  Our analysis of radon control technologies in the BID
does not assume that radon reductions for various control technologies
are additive.  Only the reductions from bulkheading and adsorption on
activated carbon of the radon bled from the bulkheaded areas are additive
and treated as such in the BID.  The revised table in the final BID
reflects this more clearly.  EPA believes this approach is reasonable.

     Comment 5.3.4:  Curtailing operations to reduce emissions is
useless, as the ventilation system must operate continuously.  (1-22)

     Response;  Curtailing operations to reduce emissions is only
effective if the shutdown is for prolonged periods of time.  Short-term
shutdowns would not be effective in reducing radon emissions.

     Comment 5.3.5;  Emission limits under Section 112 of the Act are
intended to be technology forcing.  The EPA has ignored this and
underestimated the effectiveness of existing technology (e.g., sealants)
and the ingenuity and creativity of American industry.  (G-22, P-15a)

     Response:  See response to Comment 5.1.2.

     Comment 5.3.6;  The EPA has ignored several potential control
technologies for radon, including capture on activated charcoal,
cryogenic distillation, and scrubbers using dioxygenyl
hexafluorantimonate as an oxidant.  Our discussions with knowledgeable
scientists lead us to conclude that these are viable techniques and could
achieve 99 percent radon control.  (P-15a, P-15b)

     Response;  Numerous radon control techniques are discussed in the
BID, including adsorption on charcoal, cryogenic distillation, and
scrubbers.  None of the techniques mentioned by the commenter have been
demonstrated for controlling radon from mines; thus, information is very
sparse.  The available information suggests that such controls are not
viable for mines and/or entails very great costs.

     Comment 5.3.7;  We suggest that the land control standard be dropped
from further consideration and that standards be based on either emission
or dose rates.  (G-19)

     Response:  See response to Comments 5.1.2a through c.

     Comment 5.3.8;  We agree that the Agency has properly rejected
bulkheading, sealant coatings, activated'carbon adsorption, mine
pressurization, and backfilling as means of controlling radon emissions
from mines.  However, there are several inaccuracies in the assessment
that should be noted, as these inaccuracies cause the Agency to over-
estimate the effectiveness of these methods, and to underestimate their
costs.  Specifically:  Bulkheading is already used as extensively as
possible for worker protection; sealant coatings can only be used on very
limited areas of an underground mine; pressurization of deep mines will
not work; and activated carbon adsorption of radon will not work in the
moist atmosphere of a mine.  Also, the reference mine used by the Agency
fails to consider all sources of radon, and is not universally applicable
to all types of underground mines.  (I-3a, I-36b, 1-47)

     Response:  See response to Comment 5.1.2.

     Comment 5.3.9;  At the public hearing, the EPA suggested that
another means of control is to move the mine vents.  Mines follow the ore
body and mine vents are placed as necessary to bring in fresh air and
exhaust radon to protect the miners.  Moving the vents is therefore not
feasible.  (I-3b, 1-22, 1-23)

      Response;   EPA has  withdrawn the  proposed  standard  for  reasons
 described  in  response  to Comment  5.1.2.   Thus,  this  comment  is now moot.

      Comment  5.3.10:   The EPA  suggestion,  that  placing stacks  on mine
 vents is a practical way to  reduce ground level radon concentrations,  is
 neither practical  nor  effective.   (1-47)

      Response;   See response to Comment 5.3.9.

      Comment  5.3.11:   The proposed rule is  ambiguous with  respect  to
 government-owned land.   The  lands under the control of the Bureau  of  Land
 Management have  restrictions,  but not  prohibitions, on erecting
 residences.   Are such  lands  to be considered controlled  or uncontrolled
 areas?  (G-19)

      Response;   See response to Comment 5.3.9.

      5.4   Proposed  Limits

      Comment  5.4.la;   The 0.2  picocuries  per liter standard  is  a small
 percentage of the  fluctuation  in  natural  background and  is unreasonably
 low.  The  concentration  of radon  in the U.S.  Capitol ranges  from 0.7  to
 4.7 picocuries per  liter.  The radon monitoring program  in the  Ambrosia
 Lake  region of New  Mexico, before,  during,  and  after shutdown  and  peeling
 of the mines  showed no significant difference in radon concentration
 among the  three  periods.   (G-25,  I-3b, 1-31,  l-43b, P-lb)

      Comment  5.4.1b:   It  is  arbitrary  and inconsistent for the  EPA  to
 propose a  standard  for uranium mines that is much lower  than that  for
 uranium mills.   (1-31, l-37b,  1-47)

      Response (Comments  5.4.la and b):  Although the standard of 0.2
 pCi/1 represents a  small  increment of  the naturally occurring  indoor
 radon concentration, the  health risks  associated with these
 concentrations are  relatively  high,  and the  Agency believes  that any
 increases  in  these  concentrations from sources  of emissions  should  be
 kept  as small as possible.   However, for  the reasons described  in
 response to Comments 5.1.2a  through  c, EPA  has  withdrawn the proposed

      EPA does not agree  that monitoring data show no difference  in  the
 radon concentration in air when the  mines were  shut down.  On  the
 contrary,  when the  mines  were  shut  down the  radon concentration
 decreased.  EPA's interpretation  of  monitoring  data from the Ambrosia
 Lake  region is that radon emissions  from  the uranium mines significantly
 increase the  radon  concentration  in  air near the mines.

     EPA has not yet proposed  a radon  standard  for uranium mills under
 the Clean Air Act.   The need for  such  a standard is, however, now under
consideration by the Agency.   Because  the factors considered by EPA in

developing these standards may be different for uranium mills than for
uranium mines, it would not be inconsistent in the values if the standard
were not the same.

     Comment 5.4.1c;  The standard, to be credible, must be based on
health effects and techniques capable of demonstrating compliance.  There
appears to be no data demonstrating any public health impairment at radon
levels of 3 picocuries per liter, nor for occupational exposures to 30
picocuries per liter.  (G-16)

     Response:  The Agency disagrees.  The standard is based on expected
health effects for radiation exposure at low levels.  Our ability to
detect or demonstrate these effects does not impact on the probability
that the health effects estimates are correct.  With techiques available
to us at this time we cannot prove, or disprove, the effects of low
levels of radiation exposure.  The scientific consensus has always been
that it is provident for radiation protection purposes to treat the
estimated health effects as real ones.

     Comment 5.4.2:  By imposing such rigid standards, the EPA is
discouraging the energy conservative nuclear industry and encouraging the
tightening up of homes.  This will increase public exposure to
naturally-occurring radon.  (1-31)

     Response:  The proposed standards did neither.  Rather, they were
designed to protect the public with an ample margin of safety from the
effects of radionuclides emitted to the atmosphere from the facilities
covered by the standards.  See response to Comment 5.1.2.

     Comment 5.4.3;  The form of the standard should be that of a working
level, because the potential dose is not from radon, but from the radon
daughter products.

     Response:  See response to Comment 5.1.2.

     Comment 5.4.4a:  The standard should be 0.02 working levels which is
consistent with the EPA's standards for inactive sites and Canada's
standards for uranium mines and mills.  Alternatively, the standard
should not be less than 0.4 working level months.  These recommendations
are consistent with recommendations by the ICRP and NCRP and with the EPA
approach used for nuclear fuel cycle operations.  (G-24, I-3b, 1-23)

     Comment 5.4.4b:  The standard should not be lower than 3 pCi/1 for
unrestricted areas.  (1-17)

     Comment 5.4.4c:  If the standard were based on the traditional U.S.
formula of restricting the maximum dose to the member of the public to
one-tenth the dose permitted workers, the standard would be 0.4 WLM/year,
or approximately 2 pCi/1.  (1-17)

     Response (Comments 5.4.4a through c):  See response to Comments
5.1.2a through c.

     Comment 5.4.5;  The EPA should establish an annual emission  limit of
1 Ci/yr.  (P-15a)

     Response;  An emission standard of 1 Ci/yr would require large
underground mines to reduce their emission by factors of 10-^-10^.
Such reductions cannot be achieved.  See response to Comment 5.1.2.

     Comment 5.4.6:  The proposed limit, while quite restrictive, appears
reasonable based on ICRP-26 recommendations concerning acceptable risk.
Even though health effects cannot be accurately measured due to
inadequacies in epidemiological methods, such low limits are needed to
provide an acceptable level of risk.  (G-20)

     Response:  See response to Comment 5.1.2.

     Comment 5.4.7:  Background varies from below 0.1 pCi/1 to above
1.0 pCi/1.  The 0.2 pCi/1 limit above "background" will most often not be
distinguishable from background radon as measured by available
techniques.  (P-la, P-lb)

     Response;  EPA agrees that the 0.2 pci/l cannot be distinguished
from background.  This is why compliance with the proposed standard was
to be determined by calculating the radon-222 concentration using
appropriate dispersion models.

     Comment 5.4.8:  The 0.2 pCi/1 standard is too lax, and is
inconsistent with the other proposed standards.  (P-3a)

     Response:  See response to Comment 5.1.2.

     Comment 5.4.9;  EPA should establish work practice controls rather
than land use controls for uranium mines.  (P-3)

     Response:  See response to Comment 5.1.2.

     5.5  Implementation

     Comment 5.5.1:  The EPA has suggested curtailing operations to
reduce radon emissions.  This is not only impractical, but in the EPA's
proposed rule, par. 61.143(d). the source is to be treated as a
year-round source, so that no credit would be given for reduced
operation.  (I-3b, l-43b)

     Response:  Long-term curtailment of operations, with ventilation
systems shut down, will reduce radon emissions (see response to Comment
5.3.4) as much of the radon emanating from the surfaces of the mine will
decay before reaching the atmosphere.

     Comment 5.5.2;   Achieving compliance with the proposed standard by
purchasing additional land around the mines is not practical because:

     a.  The mine owners have no power of eminent domain and therefore
         cannot compel the owners to sell.

     b.  Even if an owner agreed to sell, experience indicates the price
         would be greatly inflated, and the total cost would be

     c.  The Federal regulations governing the sale of Indian land make
         it difficult and in some cases impossible to acquire such land.

     d.  Federal and state agencies controlling land near mines would be
         under no compulsion to sell or lease this land to the mine

     e.  The EPA offers no guidelines as to how neighboring mines should
         determine their 0.2 picocurie per liter boundary lines.

(G-24, 1-22, 1-23, 1-31, I-43b, 1-47)

     Response;  See response to Comment 5.1.2.

     Comment 5.5.3;   Demonstration of compliance with the proposed
standard is virtually impossible because the proposed limit is only a
small fraction of natural background.  Compliance is therefore to be
based on estimates of the boundary level concentration.  These estimates
are to be made using an unverified and questionable computer model.
(G-24, G-25, 1-22, 1-23, 1-26, P-la, P-lb)

     Response:  See response to Comment 5.1.2.

     Comment 5.5.4;   Once the EPA sets the required standard in terms of
emissions, compliance monitoring requirements should be set.  These
should include simultaneous measurements at all vents quarterly.  In
addition, records on total ventilation rates and maintenance practices
taken to minimize emissions should be required.  (P-15a)

     Response:  See response to Comments 5.1.2a through c.

     Comment 5.5.5;   The modeling method proposed by the EPA would allow
radon from multiple sources to accumulate and exceed the regulatory limit
which is based on a single source.  This would be avoided if the EPA
specified environmental monitoring (current field instruments can
determine concentrations below the proposed limits) to demonstrate
compliance.  (G-20)

     Response:  EPA does not believe that compliance with the proposed
standard can be determined by environmental monitoring.  See response to
Comment 5.1.2.

     5.6  Costs

     Comment 5.6.1:  Section 112 of the Clean Air Act neither expressly
requires nor precludes consideration of costs, and an interpretation that
precludes consideration of costs is neither reasonable nor warranted.

     Response;  See response to Comments 2.1.7a through 2.1.7i.

     Comment 5.6.2a:  The cost of installing activated charcoal beds as a
means of removing radon from mine vent exhaust would be on the order of
one billion dollars per mine, and the annual operating costs would be on
the order of seventy five million dollars per year.  Based only on annual
operating costs, the cost per health effect averted per year is about one
billion dollars.  (1-47, P-15a)

     Response:  The Agency agrees that adsorption of radon on activated
charcoal beds would be very costly.  Such technology has not been
demonstrated for mine applications, and the EPA has not considered it an
available control technology for uranium mines.

     Comment 5.6.2b;  The EPA erred in using county tax assessments for
its estimates of the cost of land in the Ambrosia Lake area.  Estimates
based on asking prices fall in the range of $1200 to $2000 per acre, not
the average assessed value of $3.00 per acre.  The actual fair market
value of the land is about $70.00 per acre.  (I-43b, 1-47)

     Response;  The EPA did not rely solely on county tax assessments for
its estimate of land costs.  Land values were determined by using several
data sources:  (1) information from the respective county and state tax
assessors'  offices; (2) detailed assessed valuations from these offices
and applying applicable assessment valuation to selling price ratios
obtained from the U.S. Department of Commerce, Bureau of Census (1978);
(3) estimates from local real estate agents; and (4) local newspapers.
The valuations were based on surface usage and rights only, since mineral
values would remain intact.

     Land value calculations were then made for each privately held
parcel within 5 km using two basic methods:  (1) using full assessment
data and applying appropriate ratios, or (2) using quoted local land
values and multiplying by the number of acres and/or dwelling units.  The
higher value was used in our estimates.  We believe our methodology gives
a fair and accurate estimate of land costs.

     Comment 5.6.2c:  The EPA's assumptions regarding the costs of
applying sealants are outdated and are much lower than estimates made by
the Bureau of Mines.  (I-3b)

     Response:   The cost estimates for applying sealant coatings are
based on 1980 data from the Bureau of Mines.  Recent data by PNL
indicates the cost may be higher ($5.80 per ton of ore mined).

     Comment 5.6.3;  The costs of installing stacks are unreasonable and
disproportionate to any public health benefit.  In addition, the safety
hazards to construction workers outweigh any potential hazards to some
hypothetical nearby resident.  (1-22)

     Response:   See response to Comment 5.3.10.

     Comment 5.6.4;  The EPA's proposed standards are not cost effective,
in that real costs would be expended to benefit hypothetical people.

     Response;   EPA is not considering standards to benefit hypothetical
people.  The proposed standard was based on protecting real individuals.
As outlined in Table 5-13 of the BID, a study was conducted in January
and February of 1983 to determine the population, type of ownership, and
cost of land around 30 large uranium mines.

     Comment 5.6.5;  The EPA, in avoiding discussion of relative risks,
fails to provide the perspective needed to see if the costs justify the
benefits.  (1-22)

     Response;   See response to Comment 2.6.7.

     Comment 5.6.6:  Imposition of this standard could drive domestic
suppliers of uranium out of business, thereby forcing the nation to
depend on foreign sources.  This is inconsistent with the Atomic Energy
Act and the EPA's regulatory impact analysis of environmental standards
for uranium mill tailings at active sites.  (1-47)

     Response:   It is not felt that this standard would have driven
domestic suppliers of uranium out of business.  On page 5-22 of the draft
BID, it was estimated that the cost of land control for this standard
would be only 1 percent of the cost of production.  However, this comment
has been rendered moot by EPA's withdrawal of the standard.

     Comment 5.6.7:  By ignoring combined techniques such as bulkheading
and backfilling on ventilation requirements, the EPA has overstated the
costs of controls.  It neglects to consider that a carbon system is
dependent on the bulkheading strategy used in connection with it.   (P-15)

     Response:   EPA did consider the combined techniques of bulkheading
and carbon absorption in our cost estimates.  The cost estimate for
carbon absorption by itself should not have been included.  The costs are
only accurate to order of magnitude.

     Comment 5.6.8;  Cost estimates for tall stacks at two existing mines
spread over existing reserves show annual costs would increase by 6
percent at one and 23 percent at the other.  (1-17)

     Response:  EPA agrees that the use of stacks on mine vents may
significantly add to the operating cost of mines.  See response to
Comment 5.3.9.

     Comment 5.6.9a:  Purchasing land to form a buffer is either not
feasible or prohibitively expensive.  At the Schwartzwalder mine, an
adequate buffer zone would cost 11 million dollars.  At Cotter's Colorado
Plateau mines, such a zone is not feasible since the surrounding land is
federally owned.  (1-17)

     Comment 5.6.9b:  EPA's proposed mandatory control requirement would
force uranium companies to either close operations or pay existing owners
windfall premiums to buy them out.  (1-3)

     Response (Comments 5.6.9a and b):  See response to Comments 5.3.2a
and 5.3.2b.

     Comment 5.6.10:  EPA's cost estimates are deficient because they
fail to take into account the loss of worker efficiency.   (1-3)

     Response:  It was not felt necessary to include a cost for the loss
of worker efficiency in our estimates because EPA's costs are only

     5.7  Other Comments

     Comment 5.7.1:  The EPA has illegally ignored the risks from surface
and in situ uranium mining.  The EPA's 1979 assessment of surface mining
indicates risks of 1.3 in 1000.  (P-15a)

     Response:  The Agency has not ignored the risks from surface and in
situ uranium mining; the discussion of uranium mining in the BID includes
both.  Standards were not proposed for either of these technologies as
the maximum ground level air concentrations of radon emitted from these
activities are significantly lower than those which result from
underground mining.  Given the level of risk, and the unavailability of
controls to reduce emissions from surface mining, the decision was made
that no standard is necessary.


     6.1 Basis for the Standard

     Comment 6.1.1:  The health of the population of southeastern Idaho
appears to be better than that of the country as a whole and has shown no
deterioration during the 30 years that the Pocatello phosphorus plant has
been in operation.  Therefore, there is no need to limit the amount of
polonium-210 in calciner off-gas.  (I-2c)

     Response:  The expected risk from the Pocatello phosphorus plant is
not great enough to be detected in the "spontaneous" cancer background in
any epidemiologic studies that could be done.  However, the fact that the
increase cannot be detected by the means available to us does not mean
there is no risk.  Also see response to Comment 2.2.59.

     Comment 6.1.2a: The EPA estimates that a person living one mile
downwind of the Pocatello plant for 70 years would have a 1 in 10,000
chance of developing cancer related to the plant operation.  This level
of risk is insignificant when applied to the population of southeastern
Idaho.  (l-2c)

     Comment 6.1.2b;  The EPA has failed to demonstrate a health-related
basis for regulating emissions of radionuclides from elemental phosphorus
facilities.  (I-20a, I-20b)

     Comment 6.1.2c;  The EPA has not shown that either maximally exposed
individuals or populations living near elemental phosphorus plants are
being exposed to a significant or unreasonable health risk from
radionuclides.  The modeling the EPA performed was too conservative and
uncertain and the actual doses that people are receiving are vanishingly
small.  (1-39)

     Response (Comments 6.1.2a through c);  EPA conducted extensive
additional emission testing at elemental phosphorus plants following
issuance of the proposed standard.  Based on information from these
tests, EPA estimates that radionuclide emissions from calciners at these
plants will cause about 0.06 fatal cancers in the population around these
plants for each year of plant operation.  About 80% of this population
risk is caused by emissions from two plants:  the FMC plant in Pocatello,
Idaho and the Monsanto plant in Soda Springs, Idaho.   The lifetime risk
of fatal cancer to the "most exposed individuals" is estimated to be
lx!0~3 (1 chance in 1000) at the Monsanto plant and 5xlO~4 (5 chances
in 10,000) at the FMC plant.

     After consideration of these risks, and other pertinent information
discussed below, it is the Administrator's judgment that the present
record does not support a conclusion that regulation of elemental
phosphorus plants is "necessary to protect the public health with an

ample margin of safety," within the meaning of the Clean Air Act.
Therefore, the proposed rule is withdrawn and the rulemaking is

     EPA considers the risks to the "most exposed individuals" to be
relatively high.  If risk to the most exposed individuals were the only
criterion for judgment, this relatively high risk might well have led to
a decision to regulate.  However, these individual risks must be weighed
both against the low population risk (aggregate risk) and against other

     Adding additional controls to these plants will be extremely
expensive measured against the limited public health benefits provided.
The two plants involved are both located in a single state which makes
the case for Federal action somewhat weaker than if there were a number
of sources located throughout the country.  Finally, a special
subcommittee of EPA's Science Advisory Board (SAB) in a report issued
last August stated that EPA's analysis in support of its proposed
standards was not presented in a format that provides a scientifically
adequate basis for regulatory decisions.  EPA disagrees with this last
comment; however, the Agency believes that the record as a whole does not
support issuing this standard.

     Comment 6.1.3;  Even if radionuclide emissions from elemental
phosphorus plants can be shown to be a significant health risk, they
should not be regulated under Section 112.  (I-20a, I-20b)

     Response:  See response to Comments 2.1.2a through 2.1.2J.

     Comment 6.1.4a:  The elemental phosphorus industry is operating
within standards set by the NRC and the EPA for normal operation of the
uranium fuel cycle (40 CFR Part 190), as acknowledged by the EPA in 48 FR
15087, April 6, 1983.  These standards provide the public with an ample
margin of safety.  (I-20a)

     Comment 6.1.4b:  The EPA has exempted from regulation some source
categories that emit higher levels of radionuclides than those proposed
for elemental phosphorus.  Uranium fuel cycle facilities and sources of
high level nuclear wastes are permitted emission levels that result in
annual doses of 25 mrem/yr to kidneys and lungs.  If this provides an
ample margin of safety for these sources, why is this not so for
elemental phosphorus plants?  (1-39)

     Response (Comments 6.1.4a and b);  Radionuclide emissions from
several elemental phosphorus plants result in radiation doses to the
lungs of individuals living near these plants which exceed EPA standards
for uranium fuel cycle facilities (40 CFR 190).  Nevertheless,  for the
reasons described in response to Comments 6.1.2a through c, the
Administrator has determined that the present record does not support a

conclusion that regulation of elemental phosphorus plants is "necessary
to protect the public health with an ample margin of safety" within the
meaning of the Clean Air Act.

     Comment 6.1.5a;  The EPA does not explain why it is sufficient to
limit the emissions of elemental phosphorus plants to 1 Ci/y when
available technology will permit greater reductions.  The EPA also does
not explain why the standard should apply only to large plants and not to
small ones, when the only apparent significant difference among the
plants is their size.  (P-17)

     Comment 6.1.5b:  The EPA must set a standard based on curies per
metric ton of phosphate rock production basis.  This alternative was
rejected with only the explanation "...this type of standard may require
emission control retrofit by one or more additional plants even though
their emissions of polonium-210 would be significantly less than 1 Ci/y."

     Comment 6.1.5c:  An emission standard limiting emissions to a
certain number of picocuries per ton of material processed seems more
reasonable.  (G-24)

     Response  (Comments 6.1.5a through c);  EPA has withdrawn the
proposed standard for elemental phosphorus plants for the reasons
described in response to Comments 6.1.2a through c.  Therefore, the
issues raised  in these comments have been rendered moot.

     Comment 6.1.6a:  The EPA has proposed an emission rate standard for
polonium-210 of 1 curie per year.  This rate is related to dose through a
conservative, unrealistic modeling procedure.  The emission rate by
itself does not take into account the local population, meteorology.
topology, or land use, all of which affect dose.  (1-39)

     Comment 6.1.6b:  Concentrations of radionuclides in ambient air are
more closely related to doses to persons than are emission rates.  In
addition, airborne  radionuclide concentrations can be measured directly.
Therefore, any standard for elemental phosphorus plants should be based
on  radionuclide concentrations in air.  (1-39)

     Comment 6.1.6c:  There is an incongruity between the standards of
emissions for  radionuclides at phosphorus plants and the standards for
exposure at the border of Federal facilities.  The standard should be
based on exposure to an individual in the population and the
corresponding  concentration that would produce this exposure.  (G-14)

     Response  (Comments 6.1.6a through c);  EPA proposed a direct
emission limit for polonium-210 emissions from calciners at elemental
phosphorus plants because the Act indicates a preference for this type of
standard where practical.  The Agency considered indirect emission

standards in units of dose-equivalents or air concentrations for other
source categories because in those cases a direct emission standard would
have been extraordinarily complex.  However, because EPA has withdrawn
the proposed standard for elemental phosphorus plants  (see response to
Comments 6.1.2a through c), issues relating to forms of the standard are
no longer relevant.

     Comment 6.1.7a:  The risk estimates prepared by the EPA are smaller
than risks the average person accepts everyday.  This  includes such
things as the risk of cancer from cosmic rays caused by living in Denver
for two months or the risk of dying by electrocution.  This statement
takes into account the fact that health risks the EPA  estimated due to
emissions from phosphorus plants are upper limits.  (l-20a)

     Comment 6.1.7b:  The doses estimated by the EPA are smaller than
background levels and variations in background levels  in the United
States.  (I-20a, 1-39)

     Response (Comments 6.1.7a and b):  See response to Comments 2.1.11a
through 2.1.1If.

     Comment 6.1.8:  There is no evidence of a relationship between
cancer and employment at the FMC Pocatello plant in a  study that covers
30 years.  (I-20a)

     Response:   No attempt was made in the Background  Information
Document to assess occupational exposure at the FMC Pocatello plant.  No
exposure measurements were made for estimating occupational exposure.
The CAA dictated concern over airborne emissions to the environment, not
occupational conditions.

     Comment 6.1.9:  In the report of April 6, 1983, the EPA stated that
the emission of radionuclides greatly increases the risk of lung cancer
to residents living near producers of elemental phosphorus.  To the best
of the commenter's knowledge, no such health related problems have been
reported.  (I-2b, G-14)

     Response:   See response to Comment 6.1.1.

     6.2  Dose and Risk Calculations

     Comment 6.2.1:  Since only two plants appear to be affected by the
proposed regulations and since both of these plants are in southeastern
Idaho,  a more reasonable approach to assessing the need for emission
limitations would be to use site-specific information, including airborne
radioactivity measurements,  measurements of radionuclide content of
locally grown foods, and actual consumption rates of locally grown
foods.   Monsanto believes that this approach would produce realistic
results well below the EPA's estimated values.  (1-39)

     Response;   The Agency agrees that site-specific information is
desirable in developing standards for elemental phosphorus plants.
Radionuclide emission rates and particle size distribution measurements
were made at these facilities and are presented in the final BID.  Most
of the environmental measurements suggested are related to the ingestion
pathway which is a minor contributor to the risk from polonium-210 from
these facilities.  EPA agrees that measurement of environmental airborne
concentrations of radionuclides would be desirable to confirm predicted
concentrations based on models.  EPA will consider doing this in an
upcoming study of the area.

     Comment 6.2.2;  Even the conservative EPA estimated doses do not
represent a significant risk because they are well below the Radiation
Protection Guide for whole body dose of 500 mrem/yr.  (1-39)

     Response;   EPA does not believe that current FRC guidance and NEC
policy of limiting exposure to individuals to 500 mrem/y whole body and
1500 mrem/y to any organ protects public health with an ample margin of
safety, within the meaning of the CAA.  EPA estimates that a person
receiving 500 mrem/y to the whole body over a lifetime would have an
added potential fatal cancer risk of about 1 in 100 due to the radiation
exposure.  However, for the reasons described in response to Comments
6.2.la through c, the Agency has decided to withdraw the proposed

     Comment 6.2.3:  For the prime radionuclide (polonium-210) of
interest to EPA regarding emissions from elemental phosphorus plants, EPA
inexplicably ignores entirely the small body of scientific studies that
is the only known source of the health effects of polonium-210 in humans
and animals, as well as the sole recommended level for human exposure.

     Response:  The commenter is in error.  The Agency used ICRP 30
metabolic and dosimetry data (Background Information Document, Volume I,
Chapter 7).  ICRP-30, Part 1, (Ann. ICRP, Vol. 2, No.3/4, 1979) lists the
same references listed by the commenter as appropriate.  These were not
used by ICRP or EPA for risk estimates.

     Risk estimates based on BEIR-3 data were used in development of the
Background Information Document.  The ICRP limits are based on
occupational exposure and are not applicable to the general population.

     It may be noted in passing that the data base (references) listed in
ICRP 30, Part 1, is by no means exhaustive.  Other references include
"Metabolism and Biological Effects of an Alpha Particle Emitter,
Polonium-210," J.N. Stannard and G.W. Casarett, editors, Radiation
Research, Supplement 5, 1964; "Effects of Polonium-210 on the Organism"
by B.B. Moroz and Yu. D. Parfenov; Adomizdat, Moscow, 1971,
(AEC-translation-7300, 1972); L.M. Scott and C.M. West, Excretion of
210-Po Oxide Following Accidental Inhalation, Health Physics 28: 563-565,

 1975; J.F.  Stara,  et  al., Comparative Metabolism  of  Radionuclides  in
 Mammals:  A Review, Health Physics,  20:  113-137;  1971;  Sections  in
 UNSCEAR publications  of  1972,  1977,  and  1982.

     Unfortunately these studies,  reports,  and  reviews  provide data only
 on metabolism, dosimetry, and  acute  effects.  Long-term effects  (e.g.,
 cancer) at  low doses  must be estimated using models.

     6.3  Control  Technology

     Comment 6.3.la:  A preliminary  analysis shows that  it will  be
 technologically  impossible for  FMC to meet  the  proposed  standard.
 (I-20a, I-20b)

     Comment 6.3.1b:  The EPA  has  not justified its  determination  that
 high energy venturi scrubbers,  rather than  fabric filters, are the most
 effective available technology.  It  has  ignored the  statement in the BID
 that fabric filters are currently  in use  "... in  more difficult
 operations  such  as asphalt plants...."   It  is inappropriate to dismiss
 the application  of controls used successfully in  other  similar
 applications when  promulgating  regulations  under  Section  112.  (P-17)

     Comment 6.3.1c:  The EPA  has  failed  to provide  any data on
 particulate size in its discussion on control technology.  Without such
 data, it is impossible to comment  on control efficiency and cost.  (1-39,

     Comment 6.3.Id;  The EPA has  not demonstrated that at a commercial
 scale elemental  phosphorus plant there is control technology available
 that will enable nodulizing kilns  to meet the proposed emission  limit.
 The statement that a  "high energy  venturi scrubber is expected to  be at
 least 98 percent efficient for  polonium-210 removal  and reduce emissions
 of this radionuclide  for a large plant to less  than  1 Ci/y" is
 unsupported.  (1-39)

     Comment 6.3.1e:  Applying  generic or typical emission control
 performance efficiencies to nodulizing kiln/calciner emissions for the
 purpose of  determining polonium-210  removal efficiencies  is not valid.
 There is no indication that EPA took particle size distribution into
 account in  its analysis.  Since the  polonium-210  volatilizes and
 recondenses, it  is reasonable to assume that it is concentrated in the
 smaller particulates.  There are,  however,  no data that define the
 concentration of polonium-210 versus particle size.  These data are
 essential in evaluating emission control effectiveness.   (1-39)

     Comment 6.3.If;  In Unit V.C. (48 PR 15085), the EPA estimates that
 the existing spray towers control  particulate emissions to 0.5 to  1.0
pounds per  ton of  rock processed.  Based on the estimate  that venturi
scrubbers will reduce this to 0.1  pounds per ton, the EPA concludes that

polonium-210 emissions will be reduced accordingly and the 1 Ci/yr limit
attained.  However, emission data collected by Monsanto taken to show
compliance with state emission limits indicates that the spray towers
reduce emissions to 0.0095 and 0.0188 pounds per ton.  Based on this, the
use of venturi scrubbers will have little effect on polonium-210
emissions.  (1-39)

     Response (Comments 6.3.la through f);   To obtain additional
information, EPA conducted emission testing of calciner off-gases at
three elemental phosphorus plants.  The purpose of this testing was to
measure the radionuclide emission rates in the calciner off-gas streams
and to determine the particle size distribution of the radionuclides in
these streams.  Using this emissions data and plant specific off-gas
stream characteristics, EPA has reevaluated the costs and effectiveness
of control equipment for reducing polonium emission from calciners at
elemental phosphorus plants.  (See Section 6.3.6 of final BID).

     Comment 6.3.2; The EPA should require the use of fabric filters on
sources other than calciners in elemental phosphorus plants.  (P-17)

     Response:  Since radionuclide emissions from calciners will not be
regulated under the CAA (see response to Comments 6.1.2a through c),
regulation of these less significant sources is not justified.

     Comment 6.3.3:  Emission data collected by the EPA at three
elemental phosphorus plants is not valid.  Several statements from the
three sampling reports indicate that there were serious problems with the
validity of the data, due both to sampling and analytical problems.

     Response;  EPA has reviewed the emission data cited in this comment
and believes the data to be valid, reliable, and adequate for use in
determining the need for emission standards.  EPA has supplemented this
data with additional emission testing.  These additional tests have
confirmed the validity of the original data.

     Comment 6.3.4;  Polonium-210 is harmful when the particulates to
which it is attached are inhaled or ingested.  If the polonium-210 is
preferentially attached to the smallest particulates, the removal
efficiency for various types of pollution control equipment estimated by
the EPA may be high.  However, since a large percentage of inhaled
particulates in the 0.5 to 1.0 micron range are expelled when a person
exhales, it is possible that the smaller particulates are less of a
radiation hazard.  (I-2c)

     Response:  Recent emission test data has indicated that polonium-210
in calciner off-gas streams is mostly associated small particles.  More
than 90 percent of the polonium is associated with particles less than
1 micron.  Based on this new data, EPA has concluded that its original

estimates of the effectiveness of wet scrubbers for reducing polonium-210
emissions were high.  EPA has reevaluated the effectiveness of control
equipment based on this new information (see response to Comments 6.3.la
through 6.3.If).  However, contrary to the comment, the smaller particle
size distribution will result in a higher radiation dose to the lung not
a smaller dose.

     6.4  Proposed Limits

     Comment 6.4.1:  The EPA's proposed emission standard (limiting whole
body doses to about 10 mrem/yr) for elemental phosphorus plants, based on
available and reasonable control technologies, is laudable.  (G-21)

     Response:   This comment has been rendered moot by EPA's decision not
to set standards for elemental phosphorus plants.

     Comment 6.4.2:  The emission limits that the EPA is proposing would
produce a dose to the hypothetical maximum individual that is several
orders of magnitude smaller than limits set by such groups as NCRP and
ICRP as well as that recommended in several studies.  In addition, it is
less than one percent of the current standard set by the State of Idaho
for emissions from elemental phosphorus plants.  (I-20a)

     Response:   Response to this comment is addressed in responses to
Comments 6.1.2a through c and 6.2.2.

     6.5  Implementation

     Comment 6.5.1:  The EPA should review how the standard will be
enforced and what the impact on state agencies will be.   (G-14)

     Response:   This comment is no longer applicable since EPA has
withdrawn its proposed standard.

     Comment 6.5.2a:  The procedure given in 48 FR 15090 for
determination of polonium on air filters contains deficiencies.  The
procedure should be carried out using a polonium tracer along with alpha
pulse height analysis spectroscopy using a silicon surface barrier
detector.  (1-39)

     Comment 6.5.2b:  Determining the procedure efficiency as described
in step 4.3.3 (48 FR 15090) with a "clean filter" will not give the
efficiency of a "real world" dirty filter.  (1-39)

     Comment 6.5.2c:  In step 4.1.1, a statement should be added so that
the analyst will use care throughout the procedure to avoid polonium loss
through volatilization.   (1-39)

     Comment 6.5.2d:  Samples should be analyzed within a few weeks of
collection.  Since lead-210 decays to polonium-210, not knowing the
amount of lead-210 present leads to uncertainty in back calculating the
amount of polonium-210 originally present.  (1-39)

     Comment 6.5.2e;  In step 4.2.1, the statement "add 200 ml of
ascorbic acid" should probably read 200 mg of ascorbic acid.  (1-39)

     Response (Comments 6.5.2a through e):  EPA acknowledges the
advantages of using a polonium tracer and alpha spectroscopy for the
analysis of polonium-210 in filters.  EPA also agrees with the
suggestions in Comments 6.5.2b through e.

     Comment 6.5.3:  Problems that have been identified with procedures
for stack sampling for polonium-210 raise the question of whether or not
reliable measurements can be performed in order to show compliance.

     Response:  EPA has carried out additional emission testing at
elemental phosphorus plants for polonium-210.   This testing showed that
reliable sampling and measurement of polonium-210 can be made when proper
facilities for such sampling exist.

     Comment 6.5.4a;  It is not feasible to conduct emission tests as
specified using Test Methods 1, 2, and 5 of part 60 at Monsanto's
plants.  (1-39)

     Comment 6.5.4b:  Due to large stack diameters and short stack
height, it is physically impossible to meet the upstream and downstream
spacing requirements contained in the EPA methods for determination of
sampling location.  (1-39)

     Comment 6.5.4c:  Handling of fragile sampling equipment with the
required long probe at heights in excess of 100 feet where there is no
working area would be difficult if not impossible.  (1-39)

     Comment 6.5.4d:  The gas flow in the demister stacks is cyclonic,
thus making isokinetic sampling impossible.  In fact, it states in the
test methods that the method cannot be used if the flow is cyclonic.

     Comment 6.5.4e:  in order to comply with state regulations on
particulate emissions, Monsanto has developed modified methods that
provide representative results.  Section 61.154 of the proposed
regulations should be amended to allow state-approved methods when EPA
methods are not valid.  (1-39)

     Response (Comments 6.5.4a through e):  EPA recognizes that the
existing facilities at the Monsanto plant do not allow the use of

standard EPA sampling methods.  However, even  through EPA has withdrawn
the proposed standard for elemental phosphorus plants,  the Agency  still
believes that  the Monsanto, Soda Springs, Idaho, plant  should install  the
necessary facilities to allow for emission testing using standard  EPA
stack sampling methods.

     6.6  Costs

     Comment 6.6.1:  Implementation of  the proposed standard will
probably cause FMC to become non-competitive in the elemental phosphorus
market.  (I-2b)

     Response:  Our analysis indicates  that FMC has a relatively large
margin between selling prices and production costs.  Our analysis  also
indicates that although the use of additional control equipment would cut
into the economic rent that FMC currently enjoys, it would not endanger
the viability of its operations (see EPA-520/1-84-025).

     Comment 6.6.2;  Even if it is possible for FMC to  comply with the
proposed rule, the cost, because of FMC's unique process, may be much
greater than that predicted for the EPA's model plant.  (I-20a,  I-20b)

     Response;  EPA has evaluated the cost for various  alternative
control systems from reducing polonium-210 emissions from the FMC plant
based on plant specific information (see response to Comments 6.3.la
through f.)  These costs are shown on Section 6.3.6 of  the final BID.
Additional comments on costs for these  control systems  are presented in
Volume II of the Response to Comments.

     Comment 6.6.3a:  Even based on the conservative risk estimates, the
proposed rule is not cost effective.  No rules have been proposed  for
other industries due to cost effectiveness even though  the costs are in
the same range as those for the elemental phosphorus industry.  (1-39)

     Comment 6.6.3b:  Compliance with the proposed standard would  require
the expenditure of considerable capital and other resources without
producing a significant reduction in risk.  It is unreasonable to expend
the resources when estimated doses are  below natural background and 30
times less than doses considered safe (i.e. Radiation Protection Guide).

     Comment 6.6.3c:  The basis for the EPA's decisions regarding cost
effectiveness is not well defined and the results are inconsistent.

     Comment 6.6.3d;  The cost per fatal cancer avoided for the elemental
phosphorus industry should be higher, since the risk is overestimated by
a factor of 10 or more and the cost of  compliance has been underestimated.

Monsanto estimates the capital and annual operating costs for venturi
scrubbers to be $5 million and $2 million, respectively, while the EPA
estimates are $3 million and $1.5 million.  (1-39)

     Comment 6.6.3e;  The EPA has not performed an adequate cost/benefit
analysis for the control of polonium-210 and lead-210 emissions from
elemental phosphorus plants.  (1-39)

     Response (Comments 6.6.3a through e):  EPA did consider costs of
alternative standards in evaluating the need for standards for elemental
phosphorus plants.  This information is contained in the BID.

     Comment 6.6.4a;  The proposed rule is defined as a "Major Rule"
under Executive Order 12291, particularly because it would result in a
major increase in costs or prices and would have "significant adverse
effects on competition, employment, investment, productivity, innovation,
or the ability to compete with foreign-based enterprises..."  (1-39)

     Comment 6.6.4b:  The proposed rule would result in significant
economic dislocations to the elemental phosphorus industry which would
affect costs, prices, competitiveness, investment, productivity, and
innovation, and would seriously affect Idaho's ability to compete in
domestic markets with either domestic or foreign producers.  (I-2a)

     Comment 6.6.4c:  It is appropriate to perform a full regulatory
impact analysis of the proposed rule under Executive Order 12291, even if
the annual effect on the national economy is less than $100 million.
Such an analysis is also suggested under the Executive Order if a major
increase in costs or prices results for consumers, individual industries,
Federal, state, or local government agencies,  or geographical regions.
(I-2a, I-2d)

     Response (Comments 6.6.4a through c):  EPA has carried out a
regulatory impact analysis of various alternative standards for the FMC
and Monsanto elemental phosphorus plants (EPA-520/1-84-025).  These
plants were analyzed because they had the highest radionuclide emissions
and were the only plants affected by the proposed standard.  This
analysis showed that installation of control equipment to reduce polonium
emissions would not endanger the economic viability of the plants.

     6.7  Other Comments

     Comment 6.7.1;  it is clear from the discussion in Unit V (48 FR
15084) of the Supplementary Information that the EPA intends to regulate
only polonium-210 emissions from kilns/calciners.  However. Section
61.152 of the proposed regulations states "Emissions of polonium-210 to
air from sources subject to this subpart shall not exceed 1 curie in a
calendar year."  The phrase "sources subject to this subpart" is not
defined in subpart V.  it should be defined as nodulizlng kilns and
calciners.   (1-39)

     Response;  This comment is no longer applicable since EPA has
withdrawn the proposed standard.


     7.1  Basis for the Standard

     Comment 7.1.la:   Existing data show that coal-fired boilers do not
emit sufficient radionuclides to cause a high risk to the public.  A risk
to public health lower than comparable risks experienced in daily life
cannot be said to pose a high risk.  (G-15, l-4b)

     Comment 7.1.1b;   Section 112 authorizes regulation of emissions
causing high risk.  It was not intended to require regulation of
"insignificant" emissions of risks or possibilities of harm.  Emissions
from coal-fired boilers do not produce a discernible risk, when one
defines discernible risk as statistically different from background.
(I-4b, 1-44)

     Comment 7.1.1c;   The Administrator should consider factors such as
the degree to which the substance is already regulated, whether existing
regulations lead to the desired reductions, and whether further
regulations could result in other public health risks as great or greater
than posed by current emissions.  Regulating radionuclide emissions from
coal-fired boilers would not decrease (and might even increase) public
health risks.  (I-4b)

     Comment 7.1.Id;   Further reduction in particulate emission levels
from levels mandated by existing regulations is not justified either
economically or from the reduction in health risks.  Particulates are
among the most stringently regulated substances under the CAA.  (G-15,
I-4a, I-4b, 1-53)

     Comment 7.1.1e;   Based on information contained in a report entitled
"Behavior and Impacts of Radionuclides from Western Coal-Fired Power
Plants," we believe the Agency has properly concluded that emission
standards for coal-fired boilers are not needed.  (G-15, 1-9, 1-24,  1-51)

     Response:  (Comments 7.1.la through e):  The commenters agree with
the EPA determination that standards for coal-fired boilers are not
appropriate.  The Agency reached this conclusion by considering the
following criteria:
          1.  The radiation dose and risk to nearby individuals;
          2.  The cumulative radiation dose and risk to populations in
     the vicinity of  the source;
          3.  The potential for radiation emissions and risk to increase
     in the future;
          4.  The availability, practicality,  and cost of control
     technology to reduce emissions; and
          5.  The effect of current standards under the Act or other
     applicable legislative authorities.

      Applying these  criteria  to coal-fired  boilers,  it  is  EPA's finding
 that  the  existing  and  proposed  EPA  standards  for  particulate emissions
 also  limit  radionuclide  emissions and  provide an  ample  margin of safety
 from  radionuclides for the  following reasons:

      •  Individual and population risks  due to radionuclide
        emissions  from coal-fired boilers are not unreasonable
        compared to  the  cost  and feasibility  of reducing risks

      •  These  emissions  and risks are  likely  to decrease in
        the  future,  and  there is not a large  radionuclide
        inventory  that could  be rapidly  released  to  the

      •  For  industrial boilers,  planned  NSPS  revisions  would
        require new, modified,  and  reconstructed  industrial
        boilers capable  of  15 MW heat  input to employ control
        technology in  the form  of electrostatic precipitators
        or  fabric  filters.  This will  reduce  radionuclide
        emissions  along  with  particulate emissions.

      •  For  utility  boilers,  New Source  Performance  Standards
        require the  application of  Best  Available Technology
        for  particulate  emissions from new  sources;  no
        technologically  feasible controls are  available beyond
        this for new boilers.

      •  Existing particulate  emission  limits  for  utility and
        industrial coal-fired boilers  established by State
        Implementation Plans  require the application of
        various degrees  of  technology  for existing sources.
        Even the most  lenient of these standards  require the
        use  of particulate  control  technology.  It is not
        practical  to require  further control  technology beyond
        this in view of  the small increment in estimated
        health benefit in relationship to extremely  large

      Comment 7.1.2a:   Regulation of a  subcategory of coal-fired  boilers
 is unnecessary because over 99  percent of utility boilers have
particulate  control  technology  and  no  abnormally  high radionuclide coals
are being burned by  utilities.   The one  percent of boilers that  are
uncontrolled are very  small,  older  plants with low capacity  factors that
produce insignificant  radionuclide  emissions.   (I-4b)

     Comment 7.1.2b:   The EPA is required by  law  to  determine an adequate
margin of safety for radionuclide emissions from  coal-fired  boilers.  If

a pollutant causes human cancer, the EPA has a duty to eliminate that
risk.  If NSPS controls are enough to achieve a level of health
protection that provides an ample margin of safety for worst case (highly
contaminated) coals, the EPA should so specify.  (P-15a)

     Comment 7.1.2c:  The EPA is obliged to study and consider regulating
coal-fired boilers whose emissions create above average risks or burn
coals with a high radioactivity content.  (P-3b, P-17)

     Response (Comments 7.1.2a through c):  Boilers using coal having a
radionuclide content significantly above average or that operate in a
manner to cause elevated emission of radionuclides could constitute a
subcategory of boilers where a radionuclide emission standard might be
necessary.  This was the subject of a study by the Los Alamos National
Laboratory which was completed after the Federal Register Notice of
April 6, 1983.  A report of this study, filed in the Docket (A-79-11,
doc. # E-9), concludes that there is no commercially available coal which
has significantly elevated concentrations of radionuclides, and no boiler
has been identified which emits levels of radionuclides that are higher
than emission levels of the model plants upon which EPA has based its
determination that additional standards are not needed for
radionuclides.  Therefore, EPA has not proposed emission standards for
radionuclides for any subgroup of coal-fired boiler.

     Comment 1.1.2d:  The EPA should more closely study potential
exposure due to plants burning coals with higher than average
concentrations of radioactive elements.  (P-6)

     Comment 7.1.3a:  Coal contains numerous trace elements.  The EPA
should evaluate risks using a cumulative risk from all elements and
compounds, not just radionuclides.  The EPA approach ignores the total
risk to the population from burning coal.  (P-17)

     Comment 7.1.3b:  Controlling particulates would benefit reduction in
trace pollutant emissions, visibility, soiling index,  and meeting NAAQS
for particulate matter.  The EPA cannot declare the risks de minimus
without a comprehensive analysis of risks from all hazardous pollutants
in a coal-fired boiler's particulate emissions.  (p-15a, P-17)

     Comment 7.1.3c:  The EPA must consider controlling particulates
further because they contain four listed hazardous pollutants (i.e.,
radionuclides, arsenic, beryllium, and mercury).  (P-17)

     Comment 7.1.3d:  Any reduction in trace elements resulting from
standards to limit radionuclides would be insignificant in terms of the
overall emissions of such elements from all sources.  (l-4b)

     Response  (Comments  7.1.3a  through  d):   EPA  agrees  that  a
comprehensive  analysis of  all risks  due to  all hazardous  pollutants  from
coal-fired  boilers  should  be considered when particulate  emission
standards are  justified.   However, methodology for  performing  such a
calculation has not yet  been developed.   EPA did not  declare the risks
due  to  radionuclides de  minimus.  EPA only  determined that,  given the
high costs  and other emission control limits for particulates,  additional
particulate emission control limits  would not be proposed.

     Comment 7.1.4:  The EPA placed  undue emphasis  on cost of  control
technology  rather than health impacts.   Section  112 requires health-based
standards,  without  undue emphasis on cost.   (P-3a,  P-3b)

     Response:  EPA considered  a number of  factors  including:   (1) the
radiation dose and  risk  to nearby individuals, (2)  the  cumulative
radiation dose and  risk  to populations  in the vicinity  of the  source,
(3)  the potential for radiation emissions and risk  increase  in  the
future, (4) the availability, practicality,  and  cost  of control
technology  to  reduce emissions, and  (5)  the  effect  of current  standards
under the Clean Air Act  or other applicable  legislative authorities.  The
basic conclusion is that existing and proposed standards  to  control
particulates provide an  ample margin of safety to protect the public
health  from radionuclides  emitted by coal-fired  boilers and  that
additional  controls would  be extremely  expensive and  of small comparative
benefit.  EPA believes it  has not placed  undue emphasis on cost and  that
its conclusions are primarily health-based.

     Comment 7.1.5a;  We do not agree with  the EPA's  premise, as
presented in the FR notice and  the BID,  that  no  standards are needed for
coal-fired  boilers.  We  do not  agree with the premise that particulate
emissions are sufficiently controlled to  keep radionuclide emissions at
acceptable  levels and that increased control  would  not  be cost-effective.

     Comment 7.1.5b:  Although  the New  Source Performance Standards  for
utility boilers greater  than 250 MMBtu/hr do  require  best available
control technology, this limit  only  applies  to units  constructed after
September 19,  1979.  Thus, most large boilers are exempted,  and
industrial  boilers constructed  since 1971 are allowed emissions more than
three times this limit for utility boilers.   (G-21)

     Comment 7.1.5c:  Smaller industrial  boilers, approximately 67
percent of  total capacity, are  not covered by New Source  Performance
Standards.  Although reference  is made  to draft  proposed  limits that
reportedly  reduce emissions and corresponding doses to  less  than 1
mrem/yr, the draft  limits  are not given,  and  the 1 mrem/yr figure is not
supported by any calculations.  Thus, much appears  to depend on an
unknown standard,  not yet  adopted, that will  only apply to new sources.
Standards should be designed to control emissions from  existing as well
as new sources.  (G-21)

     Comment 7.1.5d:  The statement that, under current regulatory
programs, emissions should slowly decrease as old boilers are replaced
ignores the facts that the NSPS for most industrial boilers is unknown
(and may be set as high at 0.2 Ib/MMBtu), and that increases in overall
coal use could substantially cut into any reduction in emissions due to
retirements.  (G-21)

     Comment 7.1.5e:  We support the EPA's decision not to propose
additional standards for coal-fired boilers.  We concur in your
assessment that additional controls would not be cost-effective, and that
there is no potential for overall emissions to increase due both to
limited amounts of radionuclides in coal and the beneficial effects of
retiring older units.  (1-48, 1-50)

     Comment 7.1.5f;  EPA did not adequately justify its decision not to
impose standards on coal-fired boilers.  (P-6)

     Response (Comments 7.1.5a through f);  since the Background
Information Document was published, a report prepared for EPA by Radian
Corporation (See Docket A-79-1, Document #II-E-6-7) updates information
on controls used on coal-fired utility boilers.  According to the report,
all coal-fired utility boilers currently use some form of particulate
control device.  Boilers accounting for over 90 percent of the total
generating capacity are controlled by electrostatic precipitators.  A
large number of boilers are controlled with mechanical collectors only,
but these are generally small units with generating capacity of less than
25MW.  Industrial boilers not covered by New Source Performance Standards
are regulated by State plans.

     Total health effects due to radionuclide emissions from coal-fired
boilers are low as the result of these particulate emission controls.
These estimates are based on current information about total particulate
emissions from utility boilers and from industrial boilers.  This
information and the high cost to reduce emissions further, weighed
heavily in the decision not to regulate coal-fired boilers for
radionuclide emissions.

     Comment 7.1.6:  The EPA should examine emissions resulting from ash
collection and disposal.  (P-4)

     Response:  Solid waste management practices at plants burning coal
prevent airborne radionuclide emissions from ash collecton and disposal.
This source of emissions of radionuclides is insignificant compared to
emissions from the stack.

     Comment 7.1.7:  The EPA should note that some fly ash is more
radioactive than normal building material and should be tested before
being sold as a concrete additive.   (G-4,  P-5)

      Response:   Since this comment  refers to fly ash that is collected by
 control  devices  and  does  not  constitute  an air  pollutant,  it would be
 more  appropriately considered in  other proceedings.

      7.2  Dose and Risk Calculations

      Comment  7.2.1;   EPA's data bases on particulate emissions and
 control  technology are out-of-date.  Less than  0.1 percent of total
 generating capacity  (representing 1.1 percent of boilers)  have either no
 controls or have not  reported controls.   National particulate emissions
 are 20 percent lower  than the value reported by EPA.   (I-4b)

      Response:  Some  of the data  used in the Background  Information
 Document have proved  to be out-of-date.   A report recently prepared for
 EPA by Radian Corporation (See Docket #A-78-ll,  Document  ttll-E-6-7)
 updates  these data.   In the BID,  EPA assumed some utility  boilers  have  no
 controls;  however, the recent data  indicate,  in fact,  all  coal-fired
 utility  boilers have  some particulate controls.   This  change does  not
 affect EPA's determination not to propose standards  for utility boilers.

      The total particulate emissions from all utility  boilers of 0.9
 million  metric tons per year  used in the Background  Information Document
 is a  recent estimate  based on current data by EPA's  National Air Data
 Branch,  Monitoring and Data Analysis Division,  in November,  1982.   This
 remains  EPA's best estimate.

      Comment 7.2.2a:   The radionuclide content  of coal will  tend toward
 the average value  for  coals because of the large  tonnages  burned.   The
 radionuclide content  of coal  generally is not region nor coal-type
 dependent.  Claims that high  radionuclide content coal exists  are
 unfounded.  (I-4b)

      Comment 7.2.2b:   The radioactive content of  coal  is highly
 variable.  This variability is well established and should be  explicitly
 addressed  in the standard-setting methodology.  A more realistic (lower)
value for  fly ash  radioactivity should be used.   (G-15)

     Comment 7.2.2c:   The  concentrations  and  ratios of radionuclides in
 fly ash were found to  be  similar  to those used  by the EPA  and  are highly
variable.  The EPA assumption  of  9 pCi/g  of  U-238 in fly ash  and a  100
mCi/yr source term are conservative, upper bound  estimates of
radionuclide emissions from existing coal-fired power plants.  The  total
U-238 emission rate for all coal-fired power  plants  is probably half or
 less than half that value.  (G-15, I-4b,  1-48,  1-51)

     Response (Comments 7.2.2a through c):   A draft of a report prepared
for EPA by the Los Alamos National Laboratory (See Docket tA-79-11,

Document ttIII-E-9) confirms the comment on the variability of
radionuclide content of coal and that the radionuclide concentration in
the coal of a specific plant will tend toward the national average value
because of the large amounts of coal that are burned.  EPA asked
specifically when the determination was proposed if anyone knew of
coal-fired plants burning coal with elevated radionuclide content. No
examples were provided.  EPA contractors have looked for coal-fired
plants burning coal high in radionuclide content above the average and
have not found any.

     However, the concentration of uranium-238 in fly ash is highly
dependent on particle size which is, in turn, dependent on the operating
characteristics of a specific plant; the smaller the particle size the
higher the concentration will be.  EPA's choice of 9pCi/g is an average
of values for various particle sizes given in the open literature.  It is
not considered to be upper bound.

     The uranium-238 source term for the reference utility boiler of 100
mCi/y is not meant to be representative of emissions from a typical
boiler, but rather an estimate of the largest emission levels likely to
be encountered.  It is used to establish a relationship between
radionuclide emissions and dose  and risk of health effects.  This
relationship is also used with the industry's yearly particulate emission
rate to determine total health effects.  Therefore, EPA sees no reason to
lower its estimate of total uranium-238 emission rates or its estimate of
health effects.

     Comment 7.2.3;  There is no consideration of natural sources,
including radioactive dust, contributing to cancers.  Natural dust bears
radioactivity very close to the activity of the coal burned.  (I-4b)

     Response;  There is no need to consider natural sources including
dust.  The risk estimates for cancer are attributable risks.  That is,
the models and calculations consider only a single source of
radioactivity and carry through to estimation of the cancer risk
attributable to that single source.  While risks from various single
sources can be summed to get an estimate of total risk, each risk
estimate is independent and can be considered or discussed that way.

     The risks addressed in the Background Information Document are for
the sources as described.  So the risk for coal-fired boilers addresses
only that source of radioisotopes.

     Comment 7.2.4a:  Both the AIRDOS EPA and DARTAB/RADRISK models
predict maximum dose commitments using built-in conservative
assumptions.  Predicted doses and risks to the maximum individual are far
above what any individual will ever experience.  The analysis should
emphasize doses and risks to average individuals and to the population at
risk.  (G-15, I-4b, 1-48)

     Comment 7.2.4b:  The EPA's dose estimates are overly conservative,
upper-bound values.  The EPA used multiple conservative assumptions  in
their analysis, typically using the maximum value found as the average
value for modeling.  As a result, the health effects from coal-fired
plants are actually much lower. A more realistic risk value from
coal-fired boilers is an upper-bound estimate of 0.2 to 0.6 effects per
year.  (I-4b, 1-48, 1-50)

     Response (Comments 7.2.4a and b):  EPA has developed its dose and
risk models using reasonable parameters and assumptions.  It did not use
overly conservative upper-bound values.  Use of these models to predict
doses and risks to nearby individuals and to the regional population is
the usual procedure.  Calculating the dose and risk to the average
individual living within an 80 kilometer radius of a facility does not
provide the most useful information for determining if standards are
need.  While recognizing that there is a large uncertainty in estimating
risks, EPA considers 1 to 2 health effects per year to be a reasonable
estimate of population risk.

     Comment 7.2.5;  Risk assessments by the EPA could be improved by
using: 1) better information on dose transfer factors; 2) a more
realistic period for surface build-up of radionuclides reflecting the
actual useful life of power plants; and 3) more realistic plume rise and
dispersion characteristics.  (I-4b, G-15)

     Response:  EPA considers the present models adequate to predict dose
and risks for the purpose of determining that standards are not
necessary.  The refinements suggested by the commenters are not judged to
change our estimates of dose and risk by amounts large enough to alter
our conclusions.

     Comment 7.2.6:  In urban areas, the exposure due to inhalation may
be greater than the dose due to ingestion.  Population doses are similar
to those for persons exposed to routine radiation releases from nuclear
power plants.  (P-la, P-lb)

     Response:  EPA's studies of the reference utility boiler as the
source term situated in urban, suburban, rural, and remote locations show
that inhalation was the most significant pathway for population dose.

     Population doses from coal-fired boilers are not similar to those
from routine operation of nuclear reactors.  Coal-fired boilers cause the
irradiation of lung tissues with alpha particles, while nuclear reactors
cause irradiation in other ways.  Nuclear reactors are regulated for
radionuclide emissions because they have the potential for releasing
large quantities of radioactivity, far larger than coal-fired boilers.

     Comment 7.2.7;  Models used in performing risk assessments are just
state-of-the-art type models and have inherent sources of error.  The
models used have not been verified for real-world situations.  (G-15)

     Response;  The commenter is partially correct.  Models used for risk
assessments of radionuclide emissions from coal-fired boilers have not
been specifically verified for coal-fired boilers.  These models have
been developed for releases of specified radionuclides from any point
source.  On this basis, the models have been verified.  In EPA's opinion
the use of these models is appropriate in determining the need for
regulating the release of radionuclides from coal-fired boilers.

     Comment 7.2.8a:  The Teknekron report (EPA) did not use the
conservative assumptions required for regulating with an ample margin of
safety.  It did not address genetic defects or birth defects.  The
reference boiler assumed some features characteristic of new boilers
rather than older boilers.  Particulates were not characterized correctly
either for particle size distribution or collection efficiency.  (P-3b)

     Comment 7.2.8b:  The EPA uses average boiler emission rates in its
analysis.  This is insufficient under the Clean Air Act mandate.  The EPA
must evaluate emissions that create above-average risks.  (P-17)

     Comment 7.2.8c:  The assessment of emissions from coal-fired boilers
is not based on worst-case scenarios.  Indeed, the stack heights used for
the reference boilers may represent a better than the average case.
NESHAPS should be set for worst case, not the average case.  (G-21)

     Response (Comments 7.2.8a and c):  EPA uses procedures that will
result in dose estimates that are realistic, then, based on these
estimates, determines whether there is an ample margin of safety.  The
calculations presented in the Background Information Document are new
calculations made by EPA and not by Teknekron, Inc.

     Genetic effects were not explicitly discussed in the Background
Information Document for coal-fired boilers because compared to the
somatic effects they are smaller in number.

     Particle size ranges from boiler stacks are not known sufficiently
to determine average size ranges for the reference boiler.  Therefore,  a
default value was assumed, as recommended by the International Commission
on Radiological Protection, whose lung model we are using.

     Comment 7.2.9;  Why is there no discussion in the Background
Information Document of why the predicted dose rates given in the
December 27, 1979 announcement of the addition of radionuclides to the
NESHAPS list (44 PR 76738) are no longer valid?  (G-21)

     Response:  The intent of the Background Information Document is to
present current and best available information.  Since December 1979,
several studies were completed which update and supersede some of the
information previously published.

     Comment 7.2.10:  Contrary to assertions made, the EPA model does
account for the effect of scrubbers on radionuclide emissions, and is
extremely conservative with respect to source term and other parameters
such as particle size, plume rise, environmental accumulation, shielding,
terrain, lung classification, assessment area, and population.  (I-4b)

     Response:  The source term used in the EPA model considers the
amount of radioactivity emitted to the environment from the stack of the
reference boiler.  Emission controls are thus accounted for.
     EPA has not intentionally used conservative parameters in its model,
but has, in its opinion, used representative values to provide a
realistic estimate of dose and risk.

     Comment 7.2.11:  A realistic assessment of dose shows the EPA
estimate is high by an order of magnitude. (l-4b)

     Response:  Since it is the commenter's opinion that EPA used high,
upper-bound values for particulate emissions and radionuclide content of
coal ash, and then used these values in a model having built-in extremely
conservative parameters, it is not surprising that EPA's dose estimates
would be considered high by an order of magnitude.  However, in EPA's
judgment, the values used to estimate source terms and doses are
realistic based on the information available.

     7.3  Control Technology

     No comments.

     7.4  Proposed Limits

     Comment 7.4.1;  The EPA should adopt the following recommendations
to reduce radionuclide emissions from coal burning:

     1. Change the NSPS for large industrial boilers to the NSPS for
        utility boilers;

     2. Establish a NSPS limit of 0.1 Ib/MMBtu or less for small
        industrial boilers;

     3. Require existing large utility and industrial boilers to limit
        emissions to 0.1 Ib/MMBtu; and

     4. Require that replacement particulate controls be limited to
        0.03 Ib/MMBtu for large industrial and utility boilers, and
        0.1 Ib/MMBtu for units less than 250 MMBtu/NR. (G-21)

     Response:   These recommendations, with the exception of the second,
require added control technology to reduce particulate emissions.  The
second recommendation refers to New Source Performance Standards (NSPS)
for small industrial boilers for which a limit has not yet been set.  EPA
considered reducing radionuclide emissions by reducing particulate
emissions, but  concluded that the cost of additional control technology
for new boilers and retrofitting existing boilers with best available
control technology is not warranted by the small increase in health
benefits estimated to accrue from the added control.  (See response to
Comments 7.1.la through e.)
     7.5  Implementation

     No comments.

     7.6  Costs

     Comment 7.6.1:  Attempts to further control particulate emissions
will cause control costs to rise significantly with very little
additional particulate reduction benefit. Radioactive elements cannot be
further beneficially controlled with existing particulate control systems
at about half the plants (i.e., those plants 10 or less years old).

     Response:   EPA agrees.

     Comment 7.6.2a:  The absolute magnitude of costs imposed by a given
level of control is relevant in judging how much control of emissions is
warranted.  The EPA has underestimated the capital, operating, and
maintenance costs for further particulate control.  The cost-per-life
saved is such a high value that an extremely costly standard could create
its own health  risks due to associated activities or economic dislocation
(i.e., higher fuel costs lowering household disposable income,
construction-related accidents, or replacement energy production health
impacts).  (I-4b, 1-48)

     Comment 7.6.2b:  Control costs presented are grossly exaggerated.
No supporting cost data were presented in the BID.  (P-17)

     Response (Comments 7.6.2a and b):  While one commenter asserts that
EPA has underestimated costs of controls to reduce particulates, another
states that control cost estimates are exaggerated.  In EPA's

opinion, the costs are not unrealistic.  Information to support the cost
data may be found in a report prepared by Radian Corporation for EPA (see
Docket #A79-11, Document ttII-E-6-7).

     Comment 7.6.3:  The EPA estimated the retrofit costs of the entire
industry, not those facilities most in need of regulation.  Yet the
cost-benefit analysis did not include the benefit from the significant
reduction of other pollutants as well.  That is, the entire cost of
controls was measured solely against the incremental reduction of
radionuclide emissions.  (P-3b)

     Response:  When estimating retrofit costs of the entire industry,
EPA did take into consideration that some facilities had better emission
controls than others.

     EPA agrees that it would be best if a cost-benefit analysis
considered all hazardous pollutants, not just radionuclides.  However,
this calculation cannot be done based on information that is available
now.  Furthermore it is not likely, should such a calculation be made and
additional controls proposed as the result, that radionuclide emissions
would play a significant part in the decision.

     7.7  Other Comments

     No comments.


     8.1  Basis for the Standard

     Comment 8.1.la:  Radon-222 emissions from phosphate mining and
gypsum piles have been ignored by EPA.  The EPA should propose standards
for radionuclide emissions from phosphate mining and phosphoric acid
plants because the phosphate industry mines more total uranium than the
uranium industry, and counties with phosphate mining and processing
facilities frequently also have significantly elevated rates of mortality
from lung cancer.  (P-2a, P-2b)

     Comment S.l.lb:  The EPA's analysis of the phosphate industry is
totally inadequate since it ignores emissions from mining and
beneficiation.  The EPA's 1979 analysis shows significant emissions from
these activities.  The current assessment also ignores radon from drying
and grinding.  (P-15a)

     Comment S.l.lc:  The EPA has no basis for regulating radon emissions
from phosphogypsum and clay wastes.  The actual radon flux from clay
wastes is almost non-existent because the material is <10 percent solids
and radon is not very soluble.  The radon flux from phosphogypsum
approximates the level proposed by the EPA as the ("overly restrictive")
standard for uranium mill tailings.  (1-45)

     Response (Comments 8.1.la through c):  EPA did not assess radon-222
emissions from phosphate raining and gypsum piles in determining the need
for standards for the phosphate industry because the sources of these
emissions are waste piles which the Agency believed, at the time of
proposal of the standards,  would better be considered under the Resource
Conservation and Recovery Act (RCRA).  In response to these comments, EPA
is now carrying out an assessment of the radionuclide emissions to air
from phosphate gypsum piles.  No assessment is planned for evaluation of
releases from mining activities since, based on available information of
mining practices, it is not likely this constitutes a significant source
of emissions.

     EPA did not ignore the radon-222 emissions in its assessments of
phosphate rock drying and grinding plants; rather it concluded, based on
previous assessments, that  these emissions were small and further
consideration of the impact of the emissions was not necessary.

     Comment 8.1.2a:  EPA should use the same reasonably available
control technology basis it applied to the source categories it proposed
to regulate when considering whether to regulate the radionuclide
emissions from the other phases of the phosphate industry.  (G-21)

     Comment 8.1.2b:  We support the EPA's model facilities approach and
agree with its conclusions  that the risks are acceptable and no further
regulations are required.  (1-24,  1-35)

     Response  (Comments  8.1.2a and b):  EPA did use  the  same  factors  in
determining  the need  for standards for the phosphate  industry facilities
as it used for source categories  for which standards  were proposed.
These factors were described  in the Federal Register  Notice on the
proposed standards (PR 48  15076).

     8.2  Dose and Risk Calculations

     Comment 8.2.la:  The  EPA overestimated radionuclide emissions from
the phosphate industry.  EPA's analysis of the phosphate industry was
overly conservative.  In addition, the EPA's analysis did not  take into
account future reductions  in  emission levels that will occur  as a result
of emission  regulations established under Sections 108,  111,  165, 172,
and 173 of the Clean Air Act.  Thus, the EPA has more than adequate
justification for its determination that standards for radionuclide
emissions in the phosphate  industry are unnecessary.  (1-45,  1-46, 1-53)

     Comment 8.2.1b:  The  EPA assumed that all emissions from  processes
within the phosphate fertilizer industry have radionuclide concentrations
equal to that of the product.  In fact, much of the particulate emissions
consist of clays, sand, or  fluorosilicates.  The percentage of P2°5
in the emissions should be  used as a means of estimating the  radionuclide
concentrations.  (1-45)

     Comment 8.2.1c:  The  particulate emission factors used by EPA for
rock dryers, granular triple  superphosphate production, and diammonium
phosphate production exceeded the highest average emission rates
submitted during public hearings by factors of 4.2, 2.2, and 2.0,
respectively.  (1-45)

     Comment 8.2.Id:  The  EPA should have used actual or average
particulate emissions in their calculations rather than maximum allowable
emission levels.  Emissions have thus been overestimated by a  factor of
2 to 4.  (I-36b, 1-45)

     Comment 8.2.1e:  The  distances to the maximum individuals used by
EPA were much less than the distances that actually occur in  the
industry.  (I-36b, 1-45)

     Comment 8.2.If:  The  EPA used a regional population of 1,400,000,
which may be typical of central Florida, but does not represent
Pocatello,  Idaho, where the regional population is 138,000.   (I-36b)

     Response (Comments 8.2.la through f):  EPA estimated the
radionuclide emissions from phosphate rock processing facilities based on
information on the radionuclide concentration of phosphate rock and the

amount of particulate matter which could be emitted under existing
regulations for control of particulates.  The specific activity of the
particulate matter emitted was assumed to be equal to the specific
activity of the phosphate rock.  EPA agrees that this procedure is
conservative when applied to the entire phosphate industry.  The emission
estimates used are most likely representative of upper limits
representative of a few facilities with the highest emissions.  The
likelihood that the emission estimates are conservative or upper limit
estimates only reenforces EPA's decision that standards are not needed
for these types of facilities.

     EPA believes that the locations used to estimate risk to the most
exposed individuals are realistic locations, representative of places
where people actually live.  The use of a Central Florida site to
estimate population risks from phosphate rock processing is appropriate
because more than 75 percent of the phosphate rock processing takes place
in Florida.  Use of a different site to represent that segment of the
industry located in less populated areas would not significantly change
the population impact assessment.

     Comment 8.2.2a:  Many of the calculations and statements in a
Florida Department of Environmental Regulation report relating to
emissions from gypsum stacks are inaccurate and incorrect.  (1-46)

     Comment 8.2.2b:  The EPA's Horton Report regarding radon exhalation
from phosphate gypsum piles is inadequate. (P-2a, P-2b)

     Response (Comments 8.2.2a and b);  EPA did not assess radon-222
emissions from gypsum piles in determining the need for standards for the
phosphate industry (see response to Comment 8.1.1).  The usefulness of
the information in the cited reports will be considered by EPA in its
future assessments of gypsum piles.

     Comment 8.2.3:  The predicted doses from the phosphate industry in
the BID are significantly lower than those given in the 1979 EPA
document, "Radiological Impact Caused by Emissions of Radionucides into
Air."  (P-2a, P-2b)

     Response:  The assessments presented in the 1979 report
"Radiological impact Caused by Emissions of Radionuclides into Air" were
preliminary assessments based on information (and assessment methodology)
available at that time.  The assessments presented in the Background
Information Document supporting the proposed standards were based on the
latest available information and assessment methodology.

     Comment 8.2.4;  The regional population figures based on the 1970
census are grossly inadequate for current risk projections in Florida.
(P-2a, P-2b)

      Response;   The  population  of  Florida  as  a whole  increased  by 43.4
percent  between  1970 and  1980.  Had  the  1980  census data  been available
for our  assessment,  the estimated  doses  to the regional populations  in
our assessments  of rock processing and wet process fertilizer plants
would have  been  about  50  percent higher.   The estimated exposures of the
nearby individuals would  not  be affected.   The greater impacts  on the
regional population  would not change our overall  assessment  of  the need
for standards for these segments of  the  phosphate industry.

      8.3  Control Technology

      Comment 8.3.1:   It is unlikely  that the  control  systems on phosphate
rock  dryers work as  they  are  proposed to work because the whole machinery
has to be shut down  for cleaning,  and this is unlikely to happen.  (P-2a,

      Response;   The  efficiency  achieved by control systems does,  in  part,
depend on proper cleaning and maintenance.  However,  the Agency does not
agree  that  such  maintenance is  unlikely  to occur, even in those systems
where  the machinery  must  be shut down for  cleaning.   Process equipment is
not operated on  a continuous  duty  cycle; thus, cleaning can  be  scheduled
for down times.  Further, the emission limits imposed by state
implementation plans assure that controls  will be maintained to achieve
the requisite level  of efficiency.

      Comment 8.3.2;   Even if  polonium-210  is  volatilized in  the
calciners,  it will probably recondense on  the particulates in the  gas
stream before it reaches  the  pollution control system and be removed
along  with  the particulates.  Tests  of a pollution control system  (with a
replacement scrubber installed  in  1975) on a  calciner owned  by  the
commenter indicated  the control efficiency varied from 99.3  to  99.8
percent.  (I-36b)

     Response:   EPA  has conducted  radionuclide emission testing of
phosphate rock calciners  at two wet  process fertilizer plants.  Data from
one of these plants  show  only relatively small quantities of lead-210 and
polonium-210 are released to  air.  However, data  from the second plant
show substantial quantities of  these  radionuclides are released.  EPA has
not yet completed its analysis  of  these data  or assessed the risks
resulting from these emissions.

     8.4  Proposed Limits

     Comment 8.4.1:  The  10 mrem per  year  standard is unrealistically
low.  A person living in  Colorado  is  exposed  to 40 mrem per  year more
than someone living  near  the EPA reference phosphate  fertilizer facility
in central Florida.  A dose no  lower  than  50-60 mrem  per year should  be
considered.   The EPA should also consider  the maximum annual radiation
dose of 500 mrem per year and 100  mrem per year established  by  the
National Council on  Radiation Protection and  Measurements and the
International Council on  Radiation Protection and Measurements.    (I-36b)

     Response:  EPA did not propose any radionuclide standards for
phosphate fertilizer facilities.  The 10 mem/yr standard referred to in
the comment was the level of the standard proposed for NRC licensed

     8.5  Implementation

     No comments.

     8.6  Costs

     Comment 8.6.1:  The EPA's estimates of the pollution control costs
for new equipment  or plant modifications are within 25 percent of
industry estimates.  Retrofit costs for DAD and GTSP plants are low by a
factor of 2 to 2 1/2,  based  on recent industry experience.  (1-45)

     Response:  EPA recognizes that some uncertainty exists in its cost
estimates for pollution control equipment for phosphate rock processsing
facilities and that greater uncertainty exists in retrofit costs compared
to new plants.  The cost estimates presented in the final BID are
presented only as  approximate costs.

     8.7  Other Comments

     No comments.


      9.1   Basis Cor  the Standard

      Comment  9.1.la:   The  EPA has  supplied  no basis  for  treating annual
 emissions  from mineral extraction  activities  as  an extreme hazard for
 regulation under  Section 112  of  the  CAA.   (1-47)

      Comment  9.1.1b:   Emissions  and  exposures are  so insignificant as  to
 preclude regulation  under  any program under the  CAA.   (I-3b,  1-12, 1-47)

      Comment  9.1.1c:   Small emissions from  the mineral extraction
 industries must be considered de minimis  and  do  not  present a significant
 risk  to public health.   (I-3b, 1-12,  1-47)

      Comment  9.1.Id:   Exposures  attributable  to  radon and  its decay
 products are  acceptable as defined by 40  CFR  190 even with background
 included.   (1-47)

      Comment  9.1.le:   When compared  to natural sources and other  risks,
 potential  health  effects from the  minerals' extraction industries  are
 insignificant.   (I-3b,  1-47)   The  risk due  to the  mineral  extraction
 industries conservatively  computes to 1 to  6  x 10^, which  is  extremely
 small compared to risks involved in  daily life.  (1-47)

      Response  (Comments 9.1.la through e):  EPA did not propose air
 emissions  standards  for mineral  extraction  industries facilities.   The
 available  data at time of  proposal showed that the radiation  doses to
 individuals and populations from radionuclide emissions from  these types
 of facilities  were relatively small  and could not  be  reduced  at
 reasonable cost.  EPA  has  continued  to analyze information from its
 studies of radionuclide emissions  from the  mineral extraction
 industries.  These further evaluations confirmed our  previous conclusion
 about the  need for standards.

     When  EPA  promulgated  the 40 CFR  190  standards, it excluded radon-222
 and its daughter products  from the standards.  This exclusion did  not
mean  that  the  exposures from  radon-222 were acceptable but  rather  that
 the Agency believed  that additional  time  was  required to develop an
 appropriate standard for radon-222 and its  daughter products.

     Comment 9.1.2;  The minerals' extraction industries are  already
 subject to controls under  Sections 108 and  111 of  the Clean Air Act that
would reduce particulate emissions.   (I-3b)

     Response:  The effect of current  standards under the  CAA or other
applicable legislative  authority was  taken  into consideration by EPA in
determining the need for a radionuclide emission standard  under Section
112 (see PR 48  15076).

     Comment 9.1.3:  The EPA incorrectly listed radon or other
radionuclides from mineral extraction activities for purposes of section
122.  The major radioactive releases are ubiquitous in the natural
environment.  The mineral extraction industries contribute 0.4 percent of
natural background emissions, well within the range of naturally
occurring variations.  (1-47)

     Response;  Section 122 of the Act required the Administrator to
determine whether emissions of radioactive pollutants cause or contribute
to air pollution that may reasonably be anticipated to endanger public
health.  Section 122 did not exclude naturally occurring radionuclides.

     Comment 9.1.4:  It is unfortunate that the EPA does not propose to
use the same reasonably available control technology basis it applied to
the source categories it proposes to regulate to the radionuclide
emissions from metal mining and milling facilities.  (G-21)

     Response:  EPA did use the same basis in determining the need for
standards for the mineral extraction industry as it used for source
categories for which standards were proposed.  These factors were
described in the Federal Register notice on the proposed standards
(48 FR 15076).

     9.2  Dose and Risk Calculation

     Comment 9.2.1;  The EPA's estimates of emissions and doses are not
best estimates.  The EPA has significantly overestimated the risk of
mining-related categories to both regional populations and maximally-
exposed individuals.  The magnitude of the estimated maximum individual
risks and population doses would be reduced if more reasonable
assumptions and parameters were considered.  (I-3b)

     Response;  The Agency disagrees that radiation risks to both nearby
individuals and populations were significantly overestimated for the
extraction industries.  These risks are reduced in the final BID by a
factor of 2.5 as discussed in Chapter 8, Vol. 1, since almost all risk
was from high-LET radiation.

     The radiation risks presented in the draft BID were from preliminary
assessments of the extraction industries.  A more detailed analysis of
the measurements made at extraction industry facilities has been
completed and the results presented in the final BID.   These results are
both higher and lower than those presented in the draft BID.  The Agency
believes the risk values listed in the final BID are reasonable estimates
based on measured emissions and on widely accepted models for pathways
and receptor risk as discussed in Volume 1, Chapters 7, 8, and 9 of the
final BID.

      Comment  9.2.2:   The risk associated  with radon is  slight  because it
 is  dispersible and has no daughter  products.   Radon concentrations are
 principally affected by local geography and raicrometeorology.   (1-47)

      Response:   These factors are taken into  consideration  in  estimating
 risk  from  radon.   Dispersion  of  radon  in  air  as  it  travels  is  included in
 the calculation of radon concentration at various distances from the
 source.  Growth of radon daughters  during travel is also  calculated.   So
 the radon-radon daughter concentration at any point as  calculated
 includes dispersion  and in-growth of daughters.  As shown in response to
 Comment 5.2.1e,  growth of daughters for radon entering  a  structure was
 also  calculated.

      Any data on  local geography or micrometeorology could  also  be
 considered  and  perhaps incorporated into  the  calculations.   However,  the
 calculations  in the  Background Information Document were  of a  generic
 nature, not site  specific.

      9.3  Control  Technology

      No comments.

      9.4  Proposed Limits

      No comments.

      9.5  Implementation

      No comments.

      9.6  Costs

      Comment  9.6.la:   The application  of  additional  control  technology to
 reduce small  risks due to radon  would  be  neither reasonable  nor
 practicable.   (I-3b,  1-47)

      Comment  9.6.1b:   There are  no  technological means available at
 reasonable cost to further reduce fugitive and controlled emissions of
 particulate matter from extractive  and processing operations.  (1-12)

      Comment  9.6.1c:   Stringent  controls  on the extractive mineral
 industries for radon would impose an economic penalty that would insure
 reliance on undependable  foreign sources  and,  through conservation
measures,  expose the public to higher  risks.  (1-47)

     Response (Comments  9.6.la through c):  See response to
Comments 9.1.1.a through  9.1.I.e.

     9.7  Other Comments

     No comments.


     10.1  Basis for the Standard

     Comment 10.1.la;  The EPA's proposed decision not to issue
radionuclide standards for uranium fuel cycle facilities covered under
4O CFR 190 is appropriate.  The 40 CFR 190 standards, in conjunction with
the NRC's regulation of these facilities, protect public health with an
ample margin of safety.  Additional regulation is not needed, and the
decision not to propose standards for uranium fuel cycle facilities is in
keeping with the Congressional mandate to avoid duplicative regulation.
(G-2a, G-2b, l-lb. I~3b, I-4a, I-4b, 1-44, 1-48, 1-51)

     Comment IQ.l.lb:  We concur with the EPA's decision not to issue new
radionuclide emission standards for uranium-fueled light-water reactors.
Current standards are adequate to protect the public health with an ample
margin of safety.  (1-5, 1-48, 1-50)

     Comment IQ.l.lc;  The Act requires the EPA to set emission standards
for the uranium fuel cycle, even though standards already exist under
other authorities.  The legislative history of Section 112 of the 1977
amendments makes it clear that this was Congress's intent.  (P-15a, P-15b)

     Response (Comments 10.1.la through c);  EPA carefully considered its
decision not to propose radionuclide emission standards under the Clean
Air Act for uranium fuel cycle facilities.  EPA fully considered the
position that there is no discretionary choice and that the Act requires
such a standard.  EPA does not believe it was the intent of Congress that
duplicative standards be issued if standards under other authorities

     For uranium fuel cycle facilities, EPA concludes that Clean Air Act
standards are not justified or needed at this time for the following

     a.  EPA has recently promulgated emission standards for the Uranium
         Fuel Cycle (40 CFR 190) under the authority of the Clean Air Act.

     b.  These standards limit emissions to low levels,  thus limiting
         risks to low levels also.

     c.  40 CFR  190  is protective of public health with ample margins of

     d.  Part 190 standards apply to all pathways, including direct
         radiation,  and for all sources that affect a single individual.
         For this reason, direct comparison with a Clean Air Act standard
         is not possible; but for a given limit the Part 190 standard is
         more conservative because it limits emissions from all pathways,
         not just emission to air.

     Therefore, EPA  believes a Clean Air Act standard for uranium fuel
cycle facilities would represent an unreasonably large expenditure of
effort to achieve a  result that would not be significantly different from
the regulations now  enforced by EPA under the Atomic Energy Act.  This
would be unreasonable.

     Comment 10.1.2;  The EPA's proposed decision not to issue
radionuclide standards for uranium milling activities under Section 112
is appropriate.  The current standards are more restrictive than required
under either the Atomic Energy Act, as amended, or the Clean Air Act.
Radon from uranium mill tailings does not pose a significant threat to
public health within the meaning of either the Clean Air Act or the
Uranium Mill Tailings Radiation Control Act.  (I-3b, 1-50)

     Response;  The  arguments for not regulating uranium milling
activities under the Clean Air Act are much the same as for uranium fuel
cycle facilities (see response to Comment 10.1).  EPA has promulgated
standards for inactive tailings piles for both active and inactive
uranium mills under  the Uranium Mill Tailings Radiation Control Act of
1978.  This was a major long-term effort that limits the risks from
tailings piles to low levels and thus protects the public with an ample
margin of safety.

     The tailings piles are the source of radon-222 emissions which is
the most serious source of risk due to air emissions from milling.   Other
kinds of particulate emissions to air from mills are limited by the Part
190 standard.

     Comment 10.1.3a;  The EPA should set radionuclide standards for
operating uranium mill sites.  The language of the UMTRCA, Section
2022(e), clearly preserves the Agency's duty to regulate radioactive
pollutants at uranium mill sites under the Clean Air Act.  (G-20, P-3b,

     Comment 10.1.3b:  At present, there are no radiological performance
standards for uranium mills and associated tailings piles under  the
UMTRCA.  Thus, the EPA should amend its proposed Clean Air Act standards
to cover active uranium mills or set a new NESHAP standard.  (G-5, G-20,

     Response (Comments 10.1.3a and b);  Air particulate emissions from
licensed mills are limited by the Part 190 standard for the uranium  fuel
cycle as noted above.  What is not covered by that standard or by EPA's
standards under the Uranium Mill Tailings Radiation Control Act  are
radon-222 emissions from licensed uranium mill tailings piles.

     EPA agrees that radon emissions from active piles are not limited by
EPA standards, but only by NRC regulations to that amount causing a
concentration of 4 pCi/1 in unrestricted areas.  This is considerably
higher than EPA's final rule for uranium mines.  EPA agrees this
inconsistency should be investigated and is publishing an Advance Notice
of Proposed Rulemaking to this effect.

     Comment 10.1.4a;  The EPA's decision not to regulate waste disposal
sites is illegal and must be reversed.  (P-15b)

     Comment 10.1.4b:  The decision not to issue standards for several
source categories, particularly nuclear power plants,  is a serious
oversight.  (P-13b)

     Response (Comments 10.1.4a and b):   The arguments for not regulating
the management of high level radioactive waste under the Clean Air Act
are much the same as for uranium fuel cycle facilities (see response to
Comment 10.1).  EPA has proposed and intends to promulgate emission
standards for this source.   The proposed standard is identical to the
standard for the uranium fuel cycle and is issued under the same
authority, the Atomic Energy Act.   It limits emissions to low levels,
thus also limiting risks and protecting the public health with an ample
margin of safety.

     A second standard issued under the Clean Air Act  would not serve a
useful purpose and would be unreasonable.   EPA has chosen not to propose
such a standard.

     Comment 10.1.5:   The proposed 10 mrem/yr limit could impact the
required review of 10 CFR 190 limits.   Since the limits of 10 CFR 190 are
already restrictive enough,  we suggest the EPA reexamine the 10 mrem/yr
proposed standard from a cost/benefit viewpoint to assure an adequate
regulatory basis for  its decision.   (1-50)

     Response;  The proposed 10 mrem/y limit has been withdrawn.  The
cost-benefit viewpoint has been previously discussed.  (See response to
Comment 2.1.7.)  Also, the reasons for not proposing Clean Air Act
standards for uranium fuel cycle standards have been previously discussed
(See response to Comment 10.1.1).

     10.2  Dose and Risk Calculations

      No comments.

     10.3  Control Technology

      No comments.

     10.4  Proposed Limits

     Comment 10.4.1;  The existing 10 CFR 190 limits cannot be said to
provide an "ample margin of safety" when 10 mrem/yr is required to
provide an ample margin of safety from DOE and other NRC-licensed
activities.   (P-15a)

     Response;  "An ample margin of safety" is not a single value but may
vary within limits for different sources,  in this particular case,  the
10 CFR 190 limits are not greatly different from the Clean Air Act
proposed standard for DOE facilities but they cannot be directly compared
because of differences (see response to Comment 10.1.1).   This argument
cannot be reasonably used to show that 10 CFR part 190 does not provide
an ample margin of safety.

     10.5  Implementation

      No comments.

     10.6  Costs

      No comments.

     10.7  Other Comments

      No comments.


     11.1  Basis for the Standard

     Comment 11.1.1:  We agree with the EPA's decision not to propose
standards for low-energy accelerators.  (1-19)

     Response:   Our analysis of low-energy accelerators found the risks
from such facilities to be negligible.

     11.2  Dose and Risk Calculations

     No comments.

     11.3  Control Technology

     No comments.

     11.4  Proposed Limits

     No comments.

     11.5  Implementation

     No comments.

     11.6  Cost

     No comments.

     11.7  Other Comments

     No comments.


 Appendix A Government Commenters

 I.D.    DOCKET                                                DATE OF        DATE
 CODE    NUMBER        COMHENTER                             SUBMISSION     DOCKETED

 G-l     111-53        Alvin W.  Trivelpiece                  07-14-83       07-14-83
                      Director  of Energy Research
                      Department of Energy
                      Washington.  D.C.   20585

 G-l     III-J-01       Dr.  Alvin Trivelpiece                  04-28-83       05-03-83
                      Department of Energy
                      Washington,  D.C.   20545

 G-2     III-B-07       William J.  Dircks                      06-21-83       06-24-83
                      Executive Director for  Operations
                      U.S.  Nuclear Regulatory Commsission
                      Washington,  D.C.   20555

 G-2     III-J01        Dr.  William Mills                      04-28-83       05-03-83
                      U.S.  Nuclear Regulatory Commission
                      Office of Regulatory Research
                      Washington,  D.C.   20555

 G-3     III-B-06       William H.  Spell                       06-06-83       06-10-83
                      State of  Louisiana
                      Department  of Natural Resources
                      Nuclear Energy Division
                      P.O.  Box  14690
                      Baton Rouge,  Louisiana  70898

 G-4     III-E-06       John  F. Kowalczyk,  Supervisor          05-24-83        06-10-83
                      State of  Oregon
                      Department of Environmental Quality
                      Air Quality  Division
                      522 S.W.  Fifth Avenue
                      Box  1760
                      Portland, Ogegon   97207

 G-5     III-E-07       Robert R. Mooney,  Supervisor           06-14-83       06-20-83
                      State  of Washington
                      Department of  Social Health Services
                      Environmental  Radiation and Emergency
                      Response Unit
                      1409  Smith Tower B17-9
                      Seattle, Washington  98104

G-6    III-C-01      Randy Brich                           05-10-83       05-17-83
                     Office of Air Quality and Solid Wste
                     Dakota Department of Water and
                       and Natural Resources
                     Joe Foss Building
                     523 East Capitol
                     Pierre, South Dakota  57501

I.D.   DOCKET                                                DATE OF        DATE
CODE   NUMBER        COMMEMTER                             SUBMISSION     DOCKETED

G-7    111-58        James E. Watson. Jr. (Chairman)       07-12-83       07-15-83
                     The Northe Carolina Radiation
                       Protection Commission
                     Department of Natural Resources
                     P.O. Box 12200
                     Raleigh, North Carolina  27605-220

G-8    III-19        Seymour Abrahamson (Chairman)         05-20-83       05-25-83
                     Wisconsin Radiation Protection Council
                     Zoology Research Building
                     1117 West Johnson Street
                     Madison, Wisconsin  53706

G-9    111-26        L. D. Lukin, P.E.                     05-26-83       06-02-83
                     Director, Division of Environmental
                     Florida Department of Environmental
                     Twin Towers Office Building
                     2600 Blair Stone Road
                     Tallahassee, Florida  32301-8241

G-10   111-29        Lillian Roberts                       05-26-83       06-20-83
                     Commissioner of Labor
                     New York Department of Labor
                     Two World Trade Center
                     New York, New York  10047

G-ll   111-30        James C. Hardeman, Manager            06-01-83       06-21-83
                     Environmental Radiation Program
                     Georgia Department of Natural Resources
                     290 Washington Street,  S.W.
                     Atlanta, Georgia  30334

G-12   111-31        Audrey V. Goodwin, Director           06-21-83       06-24-83
                     Bureau of Radiological Health
                     State Office Building
                     Montgomery,  Alabama  36130

G-13   111-32        Mohamed T.  El-Ashry,  Ph.  D.           06-21-83       06-23-83
                     Director of Environmental Quality
                     Tennessee Valley Authority
                     Knoxville,  Tennessee  37902

G-14   111-37        Lee W. Stokes,  Ph.D.  (Administrator)  07-5-83        07-12-83
                     Division of Environment
                     Idaho Dept.  of Health and Welfare
                     Boise, Iowa  83720

 I.D.    DOCKET                                                DATE OF        DATE
 CODE    NUMBER        COMMEMTER                             SUBMISSION     DOCKETED

 G-15    111-41        Prem S.  Bhardwaja                     07-11-83        07-14-83
                      Environmental  Services
                      Salt River  Project
                      Box  1980
                      Phoenix,  Arizona  85001

 G-16    111-50        Jerry L.  Calhoun                       07-14-83        07-14-83
                      Acting Principal  Deputy  Assistant
                      Secretary of Defense,
                      Department  of  Defense
                      Washington, D.C   20301

 G-17    111-64        Henry G.  Williams                     07-13-83        07-18-83
                      State of  New York
                      Department  of  Environmental
                      Albany. New York  12233-0001

 G-18    111-69        Lawerence R. Jacobi, P.E.              07-13-83       07-20-83
                      General Manager
                      Texas Low-Level Radioactive
                       Waste Disposal  Authority
                      1300-C East Anderson Lane, Suite 175
                      Austin, Texas  78752

 G-19    111-74        Bruce Blanchard,  Director              07-15-83       07-20-83
                      Environmental  Project Review
                      Department  of  the Interior
                      Office of the  Secretary
                      Washington, D.C.  20240

 G-20    III-C-05       Toney Anaya, Governor                  07-18-83       08-12-83
                      State of  New Mexico
                      Santa Fe, New  Mexico  87503

 G-21    III-E-10       Steven G. Kuhrtz, Director            07-21-83       08-01-83
                      State of  New Jersey
                      Department of  Environmental Protection
                      Division  of Environmental Quality
                      John  Fitch Plaza, CN027
                      Trenton,   New Jersey  08625

G-22   III-J-09       The Honorable Gary Hart               06-14-83       07-01-83
                     U.S.   Senator
                     State of Colorado
                     221 Russell Senate Office Building
                     Washington,  D.C.   20510

I.D.   DOCKET                                                DATE OF        DATE
CODE   NUMBER        COMMEMTER                             SUBMISSION     DOCKETED

G-23   111-79        David Axelrod, M.D.                   06-29-83       08-02-83
                     State of New York
                     Department of Health
                     Albany, New York
                     (forwarded by Paul Giardina, Reg. II)

G-24   III-J-09      James Greir                           06-14-83       07-01-83
                     James Lents
                     Colorado Department of Health
                     4210 East llth Avenue
                     Denver, Colorado

G-25   III-J-09      Gerald Stewart                        06-14-83       07-01-83
                     State of New Mexico
                     Uranium Licensing Section
                     Albuquerque,  New Mexico

G-26   III-J-09      Ronald L. Ostop                       06-14-83       07-01-83
                     Superintendent of Environmental
                     Colorado Springs Department
                       of Utilities
                     Colorado Springs,  Colorado


 Appendix B Industry Conunenters

 I.D.    DOCKET                                                DATE  OF         DATE
 CODE    NUMBER        COHMENTER                              SUBMISSION      DOCKETED

 1-1     III-j-09       Dr.  Calvin Brantley                    06-14-83        07-01-83
                      Atomic  Industrial  Forum
                      549  Albany Street
                      Boston, MA  02118

 I-la    III-J-01       Dr.  Shepard Bartnoff                   04-29-83        05-03-83
                      Atomic  Industrial  Forum, Inc.
                      7101 Wisconsin Avenue
                      Bethesda, Maryland  20814

 I-lb    111-52         Carl Walske (President)                07-14-83        07-14-83
                      Atomic  Industrial  Forum, Inc.
                      7101 Wisconsin Avenue
                      Bethesda, Maryland  20814

 1-2     111-67         Jack G. Peterson                       07-11-83        07-19-83
                      Executive Director and
                       Chief Economist
                      Idaho Mining Association
                      Post Office box 1660
                      Boise,  Idaho  83701

 1-2     III-J-01       Jack G. Peterson                      04-29-83       05-03-83
                      Executive Director and
                       Chief Economist
                      Idaho Mining Association
                      Post Office box 1660
                      Boise,  Idaho  83701

 1-2     III-J-09       Jerry Reeve                           06-14-83       07-01-83
                     Greater Pocatello, Idaho
                       Chamber of Commerce
                     Pocatello,  Idaho
                      (With Idaho Mining Association)

 1-2    J-09          Dr. Lynn Anderson                     06-14-83       07-01-83
                      1448 East Center
                     Pocatello,  Idaho  83201
                      (With Idaho Mining Association)

1-3    111-56        J. Allen Overton,  Jr.  (President)     07-14-83       07-14-83
                     American Mining Congress
                      1920 N Street,  N.W.
                     Suite 300
                     Washington,  D.C.   20036

I.D.   DOCKET                                                DATE OF        DATE
CODE   NUMBER        COMMENTER                             SUBMISSION     DOCKETED

1-3    III-J-01      John Zimmerman                        04-28-83       05-03-83
                     American Mining Congress
                     1920 North Street, N.W.
                     Suite 300
                     Washington, D.C.  20036

1-3    III-J-09      Dr. Marvin Goldman                    04-28-83       07-01-83
                     American Mining Congress
                     1920 North Street, N.W.
                     Suite 300
                     Washington, D.C.  20036

1-4    111-55        Henry V. Nickel, et. al               07-14-83       07-14-83
                     Hunton and Williams
                     1919 Pennsylvania Avenue, N.W.
                     P.O. Box 19230
                     Washington, D.C.  20036
                     (On behalf of the Utility Air
                       Regulatory Group)

1-4    111-80        P. William Brownell                   08-11-83       08-12-83
                     Hunton and Williams
                     1919 Pennsylvania Avenue, N.W.
                     P.O. Box 19230
                     Washington, D.C.  20036
                     (On behalf of the Utility Air
                       Regulatory Group)

1-4    III-J-01      F. William Brownell                   04-28-83       05-03-83
                     Hunton and Williams
                     1919 Pennsylvania Avenue, N.W.
                     P.O. Box 19230
                     Washington, D.C.  20036
                     (On behalf of the Utility Air
                       Regulatory Group)

1-5    III-H-01      Yankee Atomic Electric Company        07-14-83       07-18-83
                     1671 Worchester Road
                     Framingham, MA  01701

1-6    III-B-03      Henry C. Briggs (Radiation Safety     05-06-83       05-12-83
                     Indiana University
                     Department of Environmental Health
                       and Safety
                     625 North Jordan Avenue
                     Bloomington,  Indiana  47405

1-7    III-B-04      Nathan J. Treinish (Attorney)         05-27-83       05-31-83
                     Abbott Laboratories
                     Abbott Park
                     North Chicago, Illinois  60064



 1-8    III-B-11
1-9    III-E-08
1-10   III-F-01
1-11   III-F-02
1-12   III-G-01
1-13   III-E-04
1-14   III-E-05

 Lincoln Clark,  Jr.  (Chairman)
 National Organization of Test
   Research and  Training
     Reactors  (TRTR)
 c/o Nuclear Reactor  Laboratory, MIT
 138 Albany Street
 Cambridge,  Massachusetts  02139

 Lincoln Clark,  Jr.  (Chairman)
 National Organization of Test
   Research and  Training
     Reactors  (TRTR)
 c/o Nuclear Reactor  Laboratory, MIT
 138 Albany Street
 Cambridge,  Massachusetts  02139

 John R.  McNamara, Chairman
 West Associates Management Committee
 P.O. Box 1980
 Phoenix,  Arizona  85001

 Frank B.  Silvestro,  Vice President
 Ecology and Environment,  Inc.
 195 Segg Road
 P.O. Box D
 Buffalo,  New York  14225

 Thomas  J.  Sayers, (Administrator)
 Environmental Control
 Stauffer  Chemical Company
 Westport, Connecticut  06881

 D.S. Cahn,  (Vice President)
 Regulatory  Matters
 California  Portland  Cement Company
 P.O. Box  17964
 Los  Angeles, California  90017-0964

 Timothy R.  Gablehouse
 Manager Regulatory Affairs
 Adolf Coors Company
Golden, Colorado  80401

David T. Modi
E.I. duPont de Nemours & Co.
Legal Department
Wilmington, Deleware   19898




1-15   III-C-02
1-16   III-C-03
1-17   III-C-04
1-18   111-02
1-19   III-B-12
1-20   111-66
1-20   III-J-09
1-21   111-77
                                         DATE OF
COMMENTER                             SUBMISSION

Lyda W. Hersloff, Ph.D.               05-18-83
Rocky Mountain Energy
10 Longs Peak Drive
Box 2000
Broomfield, Colorado  80020

C.E. Wolff, Resident Manager          06-29-83
Mortin Ranch Project
Silver King Mines, Inc.
P.O. Box 560
Casper, Wyoming  82602-0560

J.P. McCluskey                        07-14-83
Executive Vice President
Cotter Corporation
Suite 201
9305 West Alameda Parkway
Lakewood, Colorado  80226

Janet Trunzo (Radiation Safety
  Officer)                            04-27-83
Scripps Clinic and Research Foundation
10666 North Torrey Pines Road
La Jolla, California  92037

John Jennings                         07-14-83
Pharmaceutical Manufacturers Assoc.
1100 Fifteenth St., N.W.
Washington, D.C.  20005

Gary H. Baise                         07-14-83
Beveridge and Diamond, P.C.
1333 New Hampshire Avenue, N.W.
Washington, D.C.  20036
(On behalf of FMC Industrial Chemical
  Group, Philadelphia)

Gary H. Baise, Dan Scroggins          06-14-83
Beveridge and Diamond, P.C.
1333 New Hampshire Avenue, N.W.
Washington, D.C.  20036
(On behalf of FMC Industrial Chemical
  Group, Philadelphia)

A.E. Scherer, Director
Nuclear Licensing
C-E Poser Systems
Combustion Engineering, Inc.
1000 Prospect Hill Road
Windsor, Connecticut

 I.D.    DOCKET                                                DATE OF        DATE
 CODE    NUMBER        COMMENTER                              SUBMISSION     DOCKETED

 1-22    111-38        Grey Bogden  (Director)                 07-12-83        07-13-83
                      Environmental  &  Industrial  Safety
                      Western Nuclear,  Inc.
                      Executive Office
                      134  Union Boulevard
                      Lakewood, Colorado   80228

 1-23    111-76         Pathfinder Mines  Corporation           07-11-83        07-20-83
                      550  California Street
                      San  Francisco, California   94104

 1-24    111-75         J Jeffrey Zimmerman                    07-13-83        07-20-83
                      Occidental Petroleum Corporation
                      10889 Wilshire boulevard
                      Suite 1500
                      Los  Angeles, California  90024

 1-25    111-21         Jacob C. Stucki                        05-26-83        05-31-83
                      The  UpJohn Company
                      Kalamazoo, Michigan  49001

 1-26    111-73         E.W. Shortridge,  Chairman              07-14-83        07-20-83
                      Uranium Committee
                      The  Colorado Mining  Association
                      410  Denver Hilton Office Building
                      1515 Cleveland Place
                      Denver, Colorado  80202-5192

 1-27    111-27         W. G. Council  (Sr. Vice President)    05-26-83       06-07-83
                      Northeast Utilities  Service
                      P.O. Box 270
                      Hartford, Connecticut  06141-0270

 1-28    111-28         Adrienne E. Shirk (Attorney)          06-01-83       06-09-83
                      Medi-Physics,  Inc.
                      340 Kingsland Street
                      Nutley, New Jersey   07110

1-29   111-33        Marvin W. Marsh, Vice President       07-01-83       07-01-83
                     Mallinckrodt, Inc.
                      675 Brown Road
                      St. Louis,  Missouri  63134

1-30   111-72        H.N Wellhouser, Manager               07-14-83       07-20-83
                     Compliance Administration
                     GA Technologies, Inc.
                     P.O. Box  81608


1-31   III-J-09
1-32   111-43
1-33   111-36
1-33   IH-J-09
1-33   III-J-01
1-34   III-J-09
1-35   111-68

Geroge Rice, Charles H. Montange
Kerr McGee
P.O. Box 25861
Oklahoma City, Oklahoma  73125

Jerome M. Smith
Vice President and Director,
  Medical Products
New England Nuclear
549 Albany Street
Boston massachusetts  02118

Otha W. Linton
Director of Governmental Relations
American College or Radiology
6900 Wisconsin Avenue
Chevy Chase, Maryland  20815

Russell Ritenour
American College of Radiology
Health Science Center
4200 East 9th Avenue
Denver, Colorado  80262

Thomas Bruderle
American College of Radiology
Health Science Center
4200 East 9th Avenue
Denver, Colorado  80262

Al Hazle
Conference of Radiatin Control
  Program Directors,  Inc.
4210 Bash llth Avenue
Denver, Colorado

Joseph M. Baretincic, Manager
Environmental Services &
  Quality Control
International Minerals &
  Chemicals Corporation
New Wales Operations
P.O. Box 1035
Mulberry, Florida  33860



 1-36    111-59

 J.F.  Cochrane  (Director)
 Environmental  Affairs  Department
 J.R.  Simplot Company
 P.O.  Box 912
 Pocatello,  Idaho   83201


 1-36   III-J-09
1-37   III-J-71
1-37   III-J-09
1-38   111-63
1-39   III-D-01
1-41   111-35
J. R. Cochrane  (Director)             06-14-83       07-07-83
Environmental Affairs  Department
J.R. Simplot Company
P.O. Box  912
Pocatello,  Idaho   83201

C.J. Konnerth,  Manager               07-11-83       07-20-83
Health, Safety  & Environmental
Union Carbide Corporation
Medical Products Division
P.O. Box  324
Tuxedo, New York   10987

Dr. Richard Beverly                   06-14-83       07-01-83
Director of Environmental Affairs
Metal Division
Union Carbide Corporation
Medical Products Division
Grand Junction, Colorado

Joyce P. Davis                        07-12-83       07-18-83
Chief Licensing Engineer
General Physics Corporation
10650 Hielcory Ridge Road
Columbia, Maryland 21044

J. H. Waldebeser (Manager)            07-07-83       07-13-83
Environmental Operations
Monsanto Industrial Chemicals Co.
800 N. Lindbergh Boulevard
St. Louis,  Missouri 63167

Linda A. Bagby  (Manager)              07-06-83       07-13-83
Environmental and Safety Regulatory
Amersham Corporation
2636 South Clearbrook Drive
Arlington Heights,  Illinois  60005

Edmond E. Griffin,  Ph.D.              06-29-83       07-05-83
Medical Program Science Administrator
American Heart Association
National Center
7320 Greenville Avenue
Dallas,  Texas  75231

I.D.   DOCKET                                                DATE OF        DATE
CODE   NUMBER        COMMENTER                             SUBMISSION     DOCKETED

1-42   111-40        David F. Zoll                         07-12-83       07-13-83
                     Vice President, General Counsel
                     Chemical Manufacturers Association
                     2501 M street, N.W.
                     Washington, D.C.  20037

1-43   111-44        G. Stanley Grout (Attorneys)          07-13-83       07-14-83
                     Stephenson, Carpenter, Grout,
                       and olrastead
                     Bokum Building
                     142 W. Palace Avenue
                     P.O. Box 669
                     Santa Fe, New Mexico  87504-0669
                     (on behalf of Homestake Mining Corp.
                       and United Nuclear Corporation)

1-44   111-46        George P. Green (Manager)             07-08-83       07-14-83
                     Governmental Licensing & Planning
                     Public Service Company of Colorado
                     P.O. Box 840
                     Denver,  Colorado  80201

1-45   111-48        Gary D.  Meyers (President)            07-14-83       07-14-83
                     The Fertilizer Institute
                     1015 18th Street,  N.W.
                     Washington, D.C.  20036

1-46   111-57        G.E. Wilkinson                        07-11-83       07-15-83
                     Gardiner Inc.
                     P.O. Box 3269
                     Tampa, Florida  33601

1-47   111-49        Peter J. Nickles,                      07-14-83       07-14-83
                       Charles H. Montagne
                     Covington and Burling
                     1201 Pennsylvania  Avenue,  N.W.
                     Washington, D.C.
                     (on behalf of Kerr-McGee Corp.
                       Homestake Mining Co. & United Nuclear  Corp.)

1-48   111-62        Joel D.  Patterson                      08-14-83       07-14-83
                     Middle South Services
                     Box 6100
                     New Orleans, Louisianna  70161

1-49   111-51        David W. Delcour (Vice President)     07-12-83       07-14-83
                     External Affairs
                     Climax Molybdenum  Company
                     Amax Inc.
                     1707 Cole Boulevard
                     Golden,  Colorado  80401


1-50   111-60
1-51   111-61
1-52   III-J-09
1-53   111-90

W.J. Hurford
Carolina Power & Light Company
P.O. Box 1551
411 Fayetteville Street
Raleigh, North Carolina  27602

Ronald V. Shearin
Duke Power Company
Legal Department
P.O. Box 33189
Charlotte, North Carolina  28242

Chris Shuey
Southwest Research & Infor Center
P.O. Box 4524
Albuquerque, New Mexico

American Petroleum Institute
2101 L. Street, Northwest
Washington, D.C.  20037


No date


Appendix c Public Commenters
P-lb   111-45

              Dr. Warren Sinclair
              National Council on Radiation
                Protection & Measurements
              7910 Woodmont Ave.
              Suite 1016
              Bethesda, Md. 20814

              Warren W. Sinclair (President)
              National Coucil on Radiation
                Protection & Measurements
              7910 Woodmont Ave.
              Suite 1016
              Bethesda, Md.  20814
                                        DATE OF


Thomas W. Reese, Attorney at Law
123 Eighth Street North
St. Petersburg, Florida  33701
(on behalf of Manasota-88, Inc.,
 Booker Creek Preservation, Inc.,  and Manatee
 County Save our Bays Assoication,  Inc.)
Gloria C. Rains
5314 Bay State Road
Palmetto, Florida  33701

I.D.   DOCKET                                                DATE OF        DATE
CODE   NUMBER        COMMENTER                             SUBMISSION     DOCKETED

P-3    111-22        Walter G. Wells                       05-23-83       06-01-83
                     Conservation Chairman
                     Potomac Chapter
                     Sierra Club
                     3606 Veazey Street, N.W.
                     Washington, D.C.  20008

P-3    111-42        Beers and Dickson                     07-13-83       07-14-83
                     380 Hays Street, Suite One
                     san Francisco, California  94102
                     (on behalf of the Sierra Club)

P-3    III-J-01      Kathryn Burkett Dickson               04-28-83       05-03-83
                     Beers and Dickson
                     380 Hays Street, Suite One
                     Civic Center
                     San Francisco, California  94102
                     (on behalf of the Sierra Club)

P-4    III-E-01      Comment from "a curious citizen"      04-14-83       04-26-83
                     no return address

P-5    III-E-02      Peter B. Bossman, P.E.                04-29-83       05-03-83
                     14817 S.E. Raintree Court
                     Milwaukie, Oregon  97222

P-6    III-E-03      N. Roy Greiner                        05-18-83       05-24-83
                     9 Loma Vista
                     Los Alamos, New Mexico  87544

P-7    111-03        C.G. Bacon                            No Date        05-02-83
                     2960 Hannah Avenue
                     Norristown, Pennsylvania  19401

P-8    III-B-02      W. J.  Richards (Chairman)             05-04-83       05-11-83
                     ANS-15 Research Reactor Standards
                     American Nuclear Society
                     C/o Argonne National Laboratory
                     P.O. Box 2528
                     Idaho Falls,  Idaho  83401

P-9    III-B-09      Ralph G. Robinson,  M.D. (President,    07-12-83       07-14-83
                       American College of Nuclear
                       Physicians),  and
                     Merle K. Loken,  M.D.,  (President,
                       Society of Nuclear Medicine)
                     1101 Connecticut Avenue,  N.W.
                     Suite 700
                     Washington, D.C.  20036

 I.D.   DOCKET                                                DATE OF        DATE
 CODE   NUMBER        CQMMENTER                             SUBMISSION     DOCKETED

 P-10   iri-20        Mark P.  Oncavage (President)          05-28-83       05-31-83
                      Floridans United for Safe Energy,  Inc.
                      87 Herrick Way
                      Coral Gables,  Florida  33134

 P-H   111-23        William  Boek,  Ph.D.                    05-25-83       06-02-83
                      C/o Department of Physics
                      Niagara  University
                      Niagara,  New York  14109

 P-12   111-24        Tim Johnson                           05-25-83       06-02-83
                      No address given

 P-13a   111-34        Carl J.  Johnson,  M.D.                  06-14-83       07-01-83
                      42 Hillside Drive
                      Denver,  Colorado  80215

 P-13b   III-J-09       Carl J.  Johnson,  M.D.                  06-14-83       07-01-83
                      Medical  Care and Research Center
                      42 Hillside Drive
                      Denver,  Colorado  80215

 P-14   111-18        Gene J.  Triano,  M.D.  (President)       05-20-83       05-24-83
                      Pennsylvania Radiological Society

 P-15   111-65        Robert E.  Yuhnke                       07-14-83        07-18-83
                      Regional  Counsel
                      Environmental  Defense Fund
                      1405 Arapahoe  Avenue
                      Boulder,  Colorado 80302

 P-15    III-J-09       Robert E.  Yuhnke                       06-14-83        07-01-83
                      Regional  Counsel
                      Environmental  Defense Fund
                      1405 Arapahoe  Avenue
                      Boulder,  Colorado 80302

 P-16    111-25         Roger J.  Cloutier  (President)          05-27-83       06-02-83
                      Health Physics Society
                      C/o  Oak Ridge Associated  University
                      P.O. Box  117
                      Oak  Ridge, Tennessee  37830

P-17   111-54         David D.  Doniger, et. al.             07-14-83       07-14-83
                     Natural Resources Defense Council,  Inc.
                      1725 I Street, N.W.
                     Suite 600
                     Washington, D.C.  20006

P-18   III-39
P-18   III-B-10
P-19   111-47
P-20   III-I-01

Eric J. Boeldt
210 Ohio Avenue
Madison, Wisconsin  53704

Eric J. Boeldt
210 Ohio Avenue
Madison, Wisconsin  53704

Carl G. Benson
Solado, Texas  76571

J. Clark Pontius
324 W. Chestnut Street
Oxford, Ohio  45056
  DATE OF        DATE