United States        Office of          EPA 520; 1-86-011
             Environmental Protection     Radiation Programs      August 1986
             Agency          Washington. D.C 20460


             Radiation
&EPA       Final  Rule for
             Radon - 222 Emissions from
             Licensed Uranium Mill
             Tailings

             Response to Comments

-------

-------
40 CFR Part 61                          EPA 520/1-86-011
National Emission Standards
for Hazardous Air Pollutants
                       RESPONSE TO COMMENTS

                  RULE FOR RADON-222 EMISSIONS
               FROM LICENSED URANIUM MILL TAILINGS
                         August 15, 1986
                   Office of Radiation Programs
               U.S. Environmental Protection Agency
                         Washington, D.C.

-------
Section
L
Table of Contents
INTRODUCTION
2.
3 .
LEGAL AND PROCEDURAL ISSUES
3.1
TECHNICAL ASPECTS
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
4.
Use of Interim Earth Cover and
of New Impoundments
Dewatering Tailings and Phased
Other Types of Cover
Groundwater Considerations
Timing of Standard
Evaporation Ponds
Source Term Estimates
Mill Descriptions
Alternative Standards
RISK ASSESSMENT
4.1
4.2
4.3
5.
Risk Modeling and Estimates
Significance of Risk
Risk Levels
ECONOMIC ANALYSIS
APPENDIX A
List of Commenters
iii
Design
Disposal
Page
1
3
7
7
9
10
10
11
12
12
13
13
15
15
18
19
22
25

-------
section 1:
INTRODUCTION
In this document EPA responds to comments received on the
proposed rulemaking "National Emission Standards for Hazardous
Air Pollutants; Standards for Radon-222 Emissions from Licensed
Uranium Mill Tailings," published in the Federal Reqister on
February 21, 1986 (51 FR 6382-6387).
Subsequent to
public hearing was
(51 FR 8205). The
1986.
the announcement of the proposed rule, a
held on March 25, 1986 in Denver, Colorado
comment period was held open until April 28,
In addition to requesting comments on the proposed
alternative work practices, the Agency specifically asked for
comments pertaining to the following questions:
L
Is it feasible to dewater tailings, as would be
required under a continuous disposal alternative?
2.
What is the maximum period for design licensing and
construction?
3.
Is the size limit of 20 acres for phased disposal
reasonable?
4.
Are current or potential ground water problems
severe enough to warrant termination of pumping
tailings into unlined impoundments?
5.
Are there any unidentified public health or
environmental problems associated with evaporation
ponds?
6.
Are interim controls for tailings piles a practical
alternative?
7.
Is it reasonable to assume for a reference case, a
40-year lag before compliance with UMTRCA?
8.
To what extent should the timing of UMTRCA
requirements be factored into the choice of a
control option?
1

-------
Copies of written documents and transcripts of the hearing
are available for inspec~ion and copying at EPA's Central
Docket Section, West Tower Lobby, Gallery One, waterside Mall,
401 M Street, S.W., Washington, D.C. 20460. The docket number
is A-79-ll. (A fee may be charged for copying).
Major concerns and issues arising from written and oral
comments on the proposed rulemaking are summarized. Each
commenter is identified by a letter and number after the
comment. EPA's 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
Appendix A.
2

-------
Section 2:
LEGAL AND PROCEDURAL ISSUES
Comment 2.1: There is no need for regulation under the
Clean Air Act because existing standards, regulations contained
in the Atomic Energy Act and Uranium Mill Tailings Radiation
Control Act and license conditions administered by NRC and
Agreement States adequately protect the public from risk due to
radon-222 (D-2, D-5, D-6, D-12, D-13, D-19. D-22).
Response: The Agency feels that there is a significent
risk to individuals living near active uranium mill tailings
piles. According to our model calculations, the risk may be as
high as 1 in a 100. The concentration limit of 3 pCi/1 at the
boundary of a mill established by NRC regulations may not
provide an ample margin of safety for the local population.
The number of committed fatal cancers per year to the local,
regional and national populations may be as high as 2 to 9 due
to the operating tailings piles depending on how dry the piles
become.
Comment 2.2: Confusion exists over the application of
this proposed regulation because the definition of "licensed
site" is unclear. sites which are currently licensed by
agreement states that are undergoing remedial action by DOE
under Uranium Mill Tailings Radiation Control Act should not be
included within the definition of licensed sites CD-9).
Response: This rule has no affect on existing licensed
piles that are undergoing remedial action leading to final
closure, or to sites being reclaimed by the DOE. However, it
does apply to all sites that are not reclaimed and hold a
license from the NRC or one of its agreement states.
Comment 2.3: EPA should not consider the cost and
technical feasibility of regulation under section 112 of the
Clean Air Act. The congressional mandate directs EPA to adopt
standards based exclusively on protection of public health
(D-20).
Response: section 112 of the Act requires EPA to
establish emission standards for hazardous air pollutants that
protect public health with an "ample margin of safety-" EPA
has interpreted that language to require standards that protect
against significant or unreasonable public health risks.
Similarly, Section 112 is construed to require standards that
do not necessarily eliminate all public health risks, but
minimize those risks without causing unreasonable social or
economic impacts.
3

-------
The merit of the Agency's interpretation is well
demonstrated in the context of regulating carcinogens (such as
radionuclides). EPA agrees with the current scientific
consensus that there is no threshold level below which a
carcinogen poses no risk to health. Consequently, to control a
carcinogen so that it poses absolutely no health risk would
require a standard permitting no emissions: yet such a standard
could, and often would, prove beyond the technological or
financial capacity of the affected industry to implement.

The benefits of a particular industry or activity may make
some health risk reasonable to bear if eliminating the risk
would entail significant societal harm. EPA believes that
Congress intended that section 112 standards give primary
emphasis to protecting public health, but not at all costs:
instead, Congress sought standards which minimized risks to
public health posed by a pollutant without causing unreasonable
economic and societal consequences. Both the legislative
history of the Clean Air Act and the Courts support such a
common sense approach to setting standards for harzardous
pollutants. See Industrial Union Department, AFL-CIO v.
American Petroleum Institute, 448 U.S. 607, 642 (1980): Ethvl
Corp. v. EPA, 541 F. 2d 1 (D.C. Cir. 1976), cert. den. 426 U.S.
941 (1976): H.R. Rep. No. 95-294, 95th Congress, 1st Sess.
43-51, 127.
Comment 2.4: The Administrative Procedures Act was
violated by EPA by failing to provide adequate notice of its
proposed rule. The EPA has not set forth the terms or
substance of the proposed rule or a description of the subjects
or issues involved with sufficient specificity to allow
reasonable public comment (D-12).
Response: EPA maintains that the notice of proposed
rulemaking provided sufficent detail of the issues involved and
the possible regulatory actions to allow for reasonable public
comment. This judgment is supported by the extent and detail
of the public comment EPA has received on this rule.

Comment 2.5: Radionuclides should not be regulated under
section 112 of the Clean Air Act until a legal challenge to
EPA's listing of radionuclides as a hazardous air pollutant is
resolved (D-12).
Response: EPA has listed
pollutants under section 112.
an order of the U.S. District
with respect to uranium mills
deadline, foregoing action at
available option.
radionuclides as hazardous air
In addition, under the terms of
Court of California, EPA must act
by August 15, 1986. Given this
this time is not a legally
4

-------
Comment 2.6: EPA should consider alternative work
practices and regulatory programs that more fully address
localized or site specific circumstances (D-19).
ResDonse: EPA has considered other alternative work
practices but there are difficulties with all of them. For
example, water covers would prevent radon emission but this
would likely cause ground water contamination problems at most
sites. The Agency feels it is proposing a rule that provides
sufficient flexibility to protect the public health and
safety. EPA is legally unable to propose site-specific
criteria.
Comment 2.7: The proposed rule may impose regulations
that are duplicative of and yet inconsistent with existing EPA
and NRC rules (D-22).
ResDonse: It is true that there are inconsistencies with
several of the proposed alternatives as to which Federal agency
will implement the new rule. The language of the final rule
will correct this. The Agency is committed to work closely
with NRC to avoid duplication of effort.
Comment 2.S: Mill
standard within 90 days
CAA unless an extension
(D-20) .
operators should comply with the
as required by section 112 (c) of the
is granted on a site by site basis
ResDonse: Under Section 112(c) (1) (B) EPA may grant a
waiver permitting a period of up to two years if necessary for
the installation of controls. This waiver need not be site
specific. The purpose of this waiver is to provide existing
licensed operators the necessary time to meet the work practice
standard. EPA believes it is not practical to design, license
and build new mill tailings impoundments within two years.
Much of this delay is occasioned by the many Atomic Energy Act
requirements and NRC licensing procedures, and by the prac-
ticalities of constructing a new impoundment. Therefore, EPA
will require mill tailings pile owners to comply to a strict
schedule and build new impoundments as quickly as practical but
within a maximum of 6 years. As a result, EPA believes that
Section 112(c) (1) (B) is not applicable to these circumstances.
Comment 2.9: Ground water quality should not be
considered in regulating radon-222 under Section 112 of the
Clean Air Act (D-l, D-3, D-5, D-6, D-7, D-S, D-19, D-21).
ResDonse: This rule does not regulate ground water.
Ground water contamination is controlled by pre-existing
5

-------
regulations prepared under the Uranium Mill Tailings Radiation
Control Act. EPA has considered the effects of ground water on
the environment and industry under this rule. However, if
water cover is maintained or expanded in order to limit
radon-222 emissions to the atmosphere, the potential for
impacting ground water increases because of the greater
hydraulic head. The Agency is also concerned that extended use
of existing unlined piles could increase ground water impact
problems.

Comment 2.10: The provision of the rule which requires
owners to "begin negotiating a reclamation plan and an
agreement to implement the plan with the NRC within one
year..." would not be binding on NRC unless it is addressed in
the EPA-NRC Memorandum of Understanding. (D-22)
Response: It is anticipated that the EPA-NRC Memorandum
of Understanding will have to be updated as a result of this
new rule.
Comment 2.11: There is an inconsistency between the
EPA-NRC Memorandum of Understanding and the proposed rule as to
which agency has implementation authority (D-22).
Response: This inconsistency will be corrected in the
final rule. EPA will not usurp NRC implementation authority
under this rule.
Comment 2.12: The basis for granting exemption under
61.252(b) (2) should be extended to include facilities in
compliance with 40 CFR 192 secondary standard (D-21).

Response: The Agency has determined that existing
impoundment designs of 40 acres or less that have a liner
meeting the specifications of 40 CFR 246.221 will be provided
an exemption from the schedule requirements. This liner
requirement assures that the impoundment has the capability to
retain water, thereby keeping tailings wet and greatly reducing
radon-222 emissions.
6

-------
section 3:
TECHNICAL ASPECTS
3.1
Use of Interim Earth Cover and Desiqn of New Impoundments
Comment 3.1.1: Interim cover is not practical at
operating sites because tailings are discharged from the
perimeter of an impoundment. Thus fresh tailings may be
discharged onto interim cover~. Operation of earth-moving
equipment on wet areas is difficult. Fugitive dust from the
cover material may be a problem. It is impractical to place
interim cover on steep slopes of tailing sands. Interim cover
on dams would interfere with monitoring of their stability. In
addition, covering drains of these dams could impact their
stability (D-1, D-3, D-8, D-12, D-18, D-19, D-21, D-22, D-23).
Response: The option of applying an interim earth cover
to dry areas of tailings impoundments was reevaluated in the
final Background Information Document (BID) and Economic
Analysis (EA) to better assess its practicality, effectiveness,
and cost. A detailed discussion of these aspects associated
with interim cover has been incorporated into the BID both in
Chapter 7 and Appendix C. Because of the significant
uncertainties and difficulties associated with application of
interim cover to dams constructed of coarse tailings, a revised
explanation has been made. The revised evaluation assumes
slopes remain uncovered, areas that are currently dry would be
covered immediately, that a 5 year drying period is required
before wet or ponded areas could be covered and that evapo-
ration ponds would not be covered. The Agency concluded on the
basis of this revised evaluation that interim cover is
inappropriate as a generally applicable work practice.
Comment 3.1.2: There is no need for EPA to require an
immediate earth cover to decrease radon-222 emissions at
existing impoundments. The NRC and/or Agreement States,
through licensing procedures, will force use of interim covers
on a site-by-site basis when appropriate (D-3).
ResQonse: The NRC does require application of an interim
cover on impoundments on standby status to limit windblown
tailings. While a thin cover (i.e. 1 foot) of coarse material
may effectively limit fugitive emissions, its effectiveness in
controlling radon-222 emissions is minimal. In contrast, the
BID assumes that the cover material is a 1 meter thick
homogeneous, silty, clay' soil with an 8 percent moisture
content. The function of this interim cover is to control
radon-222 emissions whereas the basic function of the NRC's
interim cover is to limit the quantity of tailings particles
that are,blown by wind from the impoundment. The EPA is not
requiring an interim cover in the final rule.
7

-------
Comment 3.1.3: Any regulation of the design of future
tailings impoundments to achieve a reduction in radon-222
emissions should allow flexibility for site-specific
considerations and not dictate one type of design (D-3, D-7,
D-12, D-18, D-19, D-23).
Response: The standard will allow more than one type of
technology to be used. It also allows for site-specific
considerations in the design of future impoundments through
petitions to the Administrator, if the proposed designs are
sufficient to protect the public health and safety of nearby
populations.

Comment 3.1.4: The industry has minimal experience with
dewatering sands and no experience with dewatering slimes.
Dewatered slimes would not be stable enough to support the
cover. Therefore, continuous disposal is not practical from an
operational standpoint (D-12, D-19).
Response: Tailings dewatering systems have been used
successfully at nonferrous ore beneficiation mills. Although
continuous disposal has never been actually practiced on
uranium mill tailings in the united states, it has been
proposed by industry as the preferred method of tailings
disposal at three sites. A proposed mill on the East coast
submitted plans for above grade, continuous disposal of
dewatered tailings including a 40 foot thick cap. This
operation was licensed but never constructed. Additionally,
the Anaconda mining company proposed disposal of dewatered
tailings in trenches with continuous covering as an alternative
at the Bluewater mill. This option was never put into practice
because of the downturn in uranium production. pioneer Uravan
also submitted plans for continuous disposal using belt
filters. Umetco Minerals has proposed the use of dewatering
and continuous disposal at the Uravan mill. The EPA believes
that these proposals submitted by industry demonstrate that
continuous disposal can be a viable option.

Comment 3.1.5: Earth cover is preferred as an interim
cover, rather than water, for existing piles since it would
reduce the potential for ground water contamination (D-15).
Response: EPA agrees that earth is the preferred cover
material. The concern with potential ground water
contamination was stated both in the BID and in the proposed
standard. The use of water cover to control radon-222
emissions is most feasible at lined impoundments during
relatively short stand-by periods.
8

-------
Comment 3.1.6: Contrary to EPA's evaluation, the
single-cell approach is more advantageous than either phased or
continuous disposal impoundment design. It has less total
surface area and provides better long-term containment (D-12).

ResDonse: The single cell tailings impoundment design
does have less total surface area than the other alternatives
as shown in the BID. However, during operations and
particularly during stand-by periods, the single cell
impoundment has more exposed surface area from which radon-222
would be emitted. Additionally, when operations cease, the
single cell impoundment would require dewatering and covering
of the total impoundment area at one time. The phased and
continuous impoundments have only a fraction of their total
surface area exposed during operations and at closure. Thus,
there is less potential for radon-222 emissions to escape if
final closure is delayed for any reason. Similarly, if an
extended stand-by period occurs during the life of the mill,
less surface area of tailings is exposed with the phased and
continuous alternatives. Each alternative is believed to
provide the same degree of long term containment.
3.2
Dewaterinq Tailinqs and Phased DisDosal
Comment 3.2.1: Technology to dewater tailings is believed
to exist, but increased energy and manpower to accomplish this
are probably too costly to be economically feasible (D-l, D-3,
D-18) .
ResDonse: EPA believes this option can be economically
feasible based on proposals for continuous disposal from the
industries and on the evaluation presented in the BID.
Comment 3.2.2: A 20 acre cell for phased disposal is
acceptable but the economic impact to operators due to this
option must be assessed (D-l).
ResDonse: The 20 acre cell size for the phased tailings
disposal alternative was determined to be an acceptable size as
described in the BID. The cost comparison of this option to
the others is presented in the BID. Additionally, the final
BID includes an evaluation of a three cell phased disposal
option, each cell having about 40 acres. In the final rule we
are requiring that tailings piles be no greater than 40 acres.
9

-------
3.3
other Types of Cover
Comment 3.3.1: Lead sheeting is being investigated as a
possible cover for tailings impoundments (D-4).

ReSDonse: EPA will follow the progress of this
investigation. Since the technology is not proved at this
time, the Agency has not considered this as an alternate cover
in their analysis.
Comment 3.3.2: A work practice standard related to
maximizing the wet area of a tailings impoundment to reduce
radon-222 would be acceptable at some sites (D-S, D-2l).

ReSDonse: Such a practice was incorporated into the EPA's
consideration of interim cover. sites with lined impoundments
and operational practices that keep tailings flooded during
normal operations effectively control radon-222 emissions
during operations. These practices are site specific and could
not be used at most existing impoundments.
3.4
Groundwater Considerations
Comment 3.4.1: Maintaining a water cover on a lined
impoundment has the following disadvantages: it would not allow
recl~mation in the shortest time, there may not be enough water
to keep the impoundment flooded, a full impoundment has
increased risk of failure, and the hydraulic head increases
potential for leaks (D-18,D-22).
ResDonse: The EPA agrees with the comment. For the
reasons cited in this comment and in the response to
comment 3.1.5, the preferred cover material in most cases is
earth. However, a lined impoundment does retain more water as
a result of normal operations. This extra water cover then
reduces the radon-222 emissions.
Comment 3.4.2: Present and potential ground water
contamination at uranium mills is not severe enough to justify
shutdown of all u~lined tailings impoundments (D-19).

ReSDonse: EPA is not considering the closure of tailings
impoundments based on groundwater contamination under this
rule. Ground water quality is already protected at licensed
uranium tailings sites by existing EPA regulations. The radon
emissions from existing tailings piles is of primary concern.
10

-------
Comment 3.4.3: The NRC stated that recent literature
indicates that a water cover may not be as effective in
reducing radon emissions as previously thought (D-22).

Response: Recent technical assissments of radon emission
rates from tailings indicate that radon emissions from tailing~
covered with less than one meter of water, or merely saturated
with water, are about 2% of emissions from dry tailings.
Tailings covered with more that-one meter of water are
estimated to have a zero emissions rate. The Agency believes
this calculated difference between 0% and 2% is negligible.
The Agency used an emission rate of zero for all tailings
covered with water or saturated with water in estimating radon
emissions.
3.5
Timinq of Standard
Comment 3.5.1: One commenter stated that the assumption
of a 40-year period between the end of an impoundment's useful
life and compliance with Uranium Mill Tailings Radiation
Control Act (UMTRCA) requirements is reasonable (D-1). Several
other commenters, however, stated that EPA's assumption of 40
years is excessive (D-6, D-S, D-12, D-15, D-1S, D-19).

Response: A 40-year time period between the end of an
impoundments useful life and compliance with UMTRCA is believed
by the EPA to represent a "worst-case" scenario. Reclamation
may be accomplished under NRC's or the Agreement State's
direction in a shorter time period. However, the Agency has
evaluate~ the cumulative impacts (i.e. health effects) over
this extended time to serve as a point of reference. The
Agency has evaluated in the final Economic Analysis a 20 year
period. This time period is being used in our calculations of
costs and benefits for the final rule.
Comment 3.5.2: Permits for new tailings management
processes can be obtained in 1 year (D-1, D-21). Implementing
and permitting new technologies may take substantially longer
(D-21). The entire process of acquisition, design, licensing
and construction is more like 6.5 years (D-S, D-1S, D-19). It
also could take more than 10 years (D-12).
Response: Based on the comments received from the NRC,
Agreement States, and individual companies, EPA realizes that
three years for constructing and operating a new tailings pile
is too short a time. The Agency now considers 6 years as the
time needed to design, permit, and construct a new tailings
impoundment.
11

-------
Comment 3.5.3: Setting a time limit on use of existing
tailings impoundments would preclude utilizing full capacity
and have adverse economic impacts on operators (D-3, D-6,
D-12) .
Response: The EPA is aware that setting a time at which
the use of existing impoundments would cease would have
economic impacts. The Economic Analysis includes this cost of
lost capacity in the evaluation of the different time frames
considered. However, it is difficult to evaluate these costs
since the future of the industry is uncertain.
3.6
Evaporation Ponds
Comment 3.6.1: Properly designed (i.e., lined) and
constructed evaporation ponds provide adequate protection for
public health and the environment (D-l, D-12, D-18, D-19).
Res~onse: Since phased and continuous tailings disposal
methods require evaporation ponds, any health or environmental
problem associated with the ponds is important. The Agency
agrees that adequate health and environmental protection is
provided with a properly designed pond.
Source Term Estimates
3.7
Comment 3.7.1: EPA assumed that radium-226 is evenly
distributed; however, coarser sands are much lower in
radium-226 content and yield lower radon-222 emissions (D-6).

Response: The Agency is aware that radium-226 is not
uniformly distributed in the tailings. Th~ fines contain much
higher radium-226 and the sands contain less radium-226. Data
on the variations of flux and radium over the surface of
individual piles is not known at this time. Reported
radium-226 concentrations represent average values reported by
milling companies and are based on their measurements and/or
estimates. This is considered to be the best available data.
Comment 3.7.2: The area of tailings and/or radium-226
contents used in BID are not correct (D-6, D-8).

Response: Tailings areas were based on aerial photographs
taken in the late summer of 1985. Radium-226 concentrations
were based on the milling company's measurements or previous
EPA estimates and are considered to be the best available
data. Corrections to this data does not affect the overall
results by more that 10%.
12

-------
Comment 3.7.3: More accurate site-specific emanation
factors2should be used as opposed to using the relationship of
1 pCi/m s per pCi Ra-226/g tailings (D-12, D-13, D-14).
ResDonse: Site-specific emanation factors were not used
because most of the information needed to estimate these
factors, such as moisture content, porosity, density, and
emanating power, are not known for ea~h site. The Agency thus
used a conservative factor of 1 pCi/m s per pCi Ra-226/g of
tailings for all dry areas and a factor of zero for wet areas.

Comment 3.7.4: Water may not be as effective in reducing
radon-222 as previously thought (D-21).
ResDonse: As stated in the response to Comment 3.7.3, the
Agency believes that the use of a conservative emanation factor
adequately approximates the overall rate. We acknowledge that
water may not be 100 percent effective as a radon-222 control.
3.8
Mill DescriDtions
Comment 3.8.1: Site-specific corrections im mill and
tailings impoundment descriptions should be made (D-ll, D-19).
ResDonse: Corrections in mill and impoundment descrip-
tions will be made in the final BID where appropriate based on
comments from the industry.
3.9
Alternative Standards
Comment 3.9.1: A site-specific alternative standard based
on radium-226 content and a risk of 1 in a million should be
set for each mill to determine the allowable exposed surface
area (D-20).
Re~Donse: Setting a site-specific standard for each mill
requires detailed knowledge about its size, radium-226 content
of the tailings, and information on local meteorology and
topography. The radium-226 content of the tailings also varies
widely, especially between the sands and fines fractions. The
Agency has not accepted the proposition that the standard must
reduce risk to a level of 1 in a million. EPA believes that it
must protect the public with an ample margin of safety which is
met by the final rule.
13

-------
Comment 3.9.2: An emission standard should also be
implemented to monitor emissions (D-15, D-22).

ResDonse: The Agency did consider an emission standard.
It was felt that boundaries could be changed to comply with an
emission standard which is not an acceptable practice under the
Clean Air Act. Also, methods to determine emissions from
tailings piles also have not been sufficiently developed to
provide accurate and consistent measurements of radon
emissions.
14

-------
section 4:
RISK ASSESSMENT
4.1
Risk Modelinq and Estimates
Comment 4.1.1: Risks from radon-222 are very approximate
and overestimated. Lower relative risk coefficients should be
used. EPA should stress the low side of the risk which is zero
(D-2, D-5, D-6, D-8, D-12, D-19, D-21).
ResDonse: Scientific evidence does not support the idea
of a threshold at low doses below which the risk coefficient
for cancer induction by radon daughters approaches zero. On
the contrary, laboratory studies of in vitro cell transfor-
mation and of cancer induction in animals indicate that the
risk per unit dose of alpha radiation is maximal at low doses
and dose rates. Epidemiological studies of miners exposed to
radon also show a decreasing risk coefficient at higher doses
and dose rates.
For any carcinogen, however potent, there will always be a
dose level below which it becomes practically impossible to
directly demonstrate harm in a human population. In
particular, it may never be possible to observe harmful effects
of radiation at dose levels comparable to natural background or
below. Risk estimates in this dose region are derived through
empirical extrapolations and theoretical calculations based on
the entire body of relevant scientific evidence.
It remains the view of EPA that a risk coefficient of
1%-4%/WLM, is reasonable in light of current evidence from
epidemiological studies of miners. This view has been endorsed
by an independent group of experts, the Radiation Advisory
Committee of the Agency's Science Advisory Board.
The substantially lower risk estimates which have been
suggested are in general either based (1) solely on analysis of
the u.S. miner data set, failing to take into account the
evidence for decreasing risk per unit dose at high doses and
the probable upward bias in dose estimates or (2) on model
assumptions which do not seem prudent in light of current
evidence (e.g., an absolute risk coefficient which decreases
over time) .

It should also be noted that in extrapolating its risk
estimates from miners to the general public, EPA employs an
"exposure equivalent", which corrects for the lower average
breathing rate in the latter. This correction effectively
implies about a 40% reduction in the risk estimate for average
15

-------
members of the general public. Some recent dosimetric
calculations, however, suggest that radon daughter deposition
in the lung may vary only slightly with breathing rate. As a
result, there can be little support for the contention that the
range of risk coefficients used by EPA represent a substantial
overestimate in light of current evidence.

Comment 4.I.la: The linear nonthreshold hypothesis and
relative risk model have not been affirmed (D-19).
Response: The statement referred in the Draft BID was
meant to convey the point that recent evidence has lent support
to these assumptions. The word "affirm" was too strong, and
the statement has been modified accordingly.

Comment 4.1.2: Radon-222 exposure from mill tailings on a
regional and national level is overshadowed by background
radon-222 sources. Therefore, making regional and national
risk estimates is meaningless because it is indistinguishable
from other sources (D-2, D-5, D-6, D-7, D-S, D-2l).
Response: It is acknowledged that radon exposures from
mill tailings, at locations distant from mill tailings sites,
are small compared to exposures from other sources. It does
not, however, follow that it is meaningless to calculate
exposure and risk due to emissions from such sites. These
calculations are based on procedures generally regarded as
sufficiently accurate to support a rulemaking. The
significance of the risk is based on the absolute value of the
risk and the practicality of control measures.

Comment 4.1.3: The calculated indivIdual lifetime
exposure risks are not realistic because: (1) people do not
live at the point of maximum exposure for a lifetime; (2)
tailings are covered in less than a lifetime; and (3) risk
estimates are questionable because they are based on
assumptions for radon-222 emissions, lifetime exposures,
occupancy figures, nearby populations, and WLM-to-risk factors
that are all high (D-2, D-5, D-6, D-12, D-19, D-2l).
Response: The Agency has noted that "Estimates of risks
to nearby individuals must be interpreted cautiously, as few
people generally spend their whole lives at such locations."
(draft BID, p. 6-1). The assumptions used to estimate
occupancy and risk factors are documented in Chapter 2, those
for radon-222 emissions in Chapter 5, and those for population
exposed in Chapter 6 of the draft BID. EPA believes these
assumptions are a reasonable basis for a rulemaking.
16

-------
Shorter periods of exposure, as might be ass~ciated with
an active pile, do not necessarily chang7 conclus~ons about the
associated risk in a direct proportional~ty, although.
shortening the time of exposure does diminish the risk. R~sk
is related to age at first exposure and duration of exposure.
Exposures of 5 years or more generally constitute an
appreciable proportion of the risk of lifetime exposure (see
table below).
Percent of Lifetime Risk for Lifetime
Exposures Accumulated in Shorter
Time Periods
Age at first  Duration of Exposure  
Exposure     Lifetime
 5 yrs 10 yrs 20 yrs 30 yr
Birth 11% 25% 47% 61% 100%
25 years 8% 16% 29% 41% 50%
40 years 8% 14% 23% 28% 29%
If tailings are uncovered for at least 20 years, the difference
between the risk associated with the pile for that time period or
for 70 years or more (lifetime) is less than a factor of two.

Comment 4.1.5: EPA should base any radon-222 standard on
actual measured exposure near tailings impoundments (D-5, D-6).
Response: Due to time constraints we have not been able to
gather sufficient data on actual exposure measurements to
individuals near tailings piles. Exposures based on models have
been used for developing the standard. A population study was
conducted by EPA in 1983 to determine actual populations near
active uranium mills.
Comment 4.1.6: Effective radon-222 decay product equilibrium
factors are incorrect because they are based on a simple model that
ignores atmospheric removal processes such as plate-out (D-12,
D-19, D-21).
Response: The outdoor equilibrium fraction used by EPA is
limited to 0.85 in order to take into consideration atmospheric
removal processes. The outdoor equilibrium value used to calculate
the effective equilibrium fraction is for the released radon-222
which has been transported to the location under consideration.
This value can be substantially greater than a measured equilibrium
fraction since that value is influenced by the equilibrium
fractions for all sources of radon-222.
17

-------
Comment 4.1.7: EPA has not, but should, include the risk from
radon-222 exposure due to windblown tailings around the mill sites
(D-18) .
Response: The radon-222 exposure from windblown tailings was
not considered as a source because it is believed to be small
compared to exposures due to radon-222 from the pile. Also, an
extensive field study would be necessary to obtain data necessary
to make this correction.
Comment 4.1.8: Exposure and dose should be evaluated on a
site-specific basis over the estimated life of each tailings
impoundment. The model used disregards concentration differences
resulting from average wind rose data at each site (D-21).

Response: Metero1ogica1 data (wind rose and stability arrays)
were not available for each tailings impoundment. Metero1ogica1
data from representative nearby reporting areas was used for each
site. Obtaining site special metero1ogica1 data for each
individual site would have imposed unacceptable time delays and
financial burdens on performing a timely risk assessment.
4.2
sianificance of Risk
Comment 4.2.1: Significance of the radon effects from mill
tailings on total population is negligible. There are no proven
adverse health effects (D-2, D-3, D-6, D-8, D-12, D-13, D-19).
Response: Adverse health effects due to radon-222 emissions
from mill tailings piles cannot be directly measured because of the
very high incidence of lung cancer attributable to other causes.
However, it would be imprudent to regulate exposure to carcinogens
on that basis. It is the position of EPA that, based on current
scientific evidence, excess lung cancers can result from radon-222
emitted by tailings piles and that the numbers of cancers
calculated in the BID are a reasonable estimate.
Comment 4.2.2: Under Section 112 of the Clean Air Act, EPA
must prove a significant risk of harm to public health before a
standard is implemented. EPA has failed to prove that radon-222
emissions pose risk sufficient to justify regulation under CAA
(D-6, D-12, D-19).
Response: EPA maintains that the record supports its
contention that radon-222 emissions pose a sufficent public health
risk. Our model indicates that persons living near an active
uranium mill tailings pile may have a risk of 1 in 100 of
deve1oping'lung cancer over their lifetime. It is estimated that
the number of fatal cancers to the local, regional and national
populations may be as high as 2 to 9 deaths due to the operating
piles.
18

-------
Risk Levels
4.3
Comment 4.3.1: Nearby individuals are to be protected from
excess risk as well as large populations. The standard must be,
but is not, sufficient to reduce each individual's risk
attributable to emission from the mill and tailings to a de minimis
level of 1 in a million (D-20).

Response: The Agency feels that reducing the risk to 1 in a
million for radon-222 is too Iowa level to be practical and that
the Clean Air Act does not require EPA to protect all individuals
at a 1 in a million risk level. The Agency's position in this
matter is discussed in the response to
Comment 2.3.
Comment 4.3.2: Risks should not be compared with general
population cancer risks but instead with lung cancer risk for a
prudent person who has avoided exposure to high risk contaminants
(e.g., cigarette smoke) in the course of his or her daily life
(D-20) .
Response: There is some evidence that non-smokers are at
lower risk from radon exposure than smokers. This has been
implicitly assumed by EPA in adopting a relative risk model in
which exposures to radon daughters act multiplicatively with other
carcinogens in inducing lung cancer. Thus, as the baseline rate of
lung cancer in the U.S. has risen, primarily due to the effects of
cigarette smoking, EPA has revised its estimates of the number of
radon daughter induced lung cancers upward proportionally.
Based on the EPA model, individuals at low risk for lung
cancer (e.g., non-smokers) would have the same relative increase in
their risk from a given exposure to radon as would individuals at
high risk for the disease (e.g., heavy cigarette smokers).
Therefore, the proposed standard is, from the standpoint of
absolute risk, more protective of individuals in the former group.
Comment 4.3.3: The NRC stated it continues to believe
existing release limits and interim stabilization practices
adequate to protect public health and safety under their
legislative mandates (D-22).
the
to be
Response: The existing release limits are the concentration
limits listed in the Commission's rule 10 CFR 20. The
concentration limit for radon-222 is 3 pCi/l in this rule for any
uncontrolled area. A person exposed to this concentration for
their entire lifetime would incur a risk of about one in one
hundred. The Agency judges that the regulatory standards
established by this regulation are necessary to meet the
requirements of section 112 of the Clean Air Act. EPA believes
19

-------
that the phased management scheme of this standard provides
significant additional health benefits in keeping with the "ample
margin of safety" provision of section 112. This is shown in the
following table on cost-effectiveness estimates:
20

-------
Alternative
Alternative Work Practice St:arxJards for CantrollinJ Radan-222 Emissicms iran Ex:istin:1 piles
(Assumes IDi1 Yellowcake Deman:l scenario am. 20 year Baseline)

Cost Effectiveness Estjmates (a)
(d)
Costsct~arn B(f) Costs~am. B(f)
Discamt:ed Discamt:ed Discamt:ed Discamt:ed
Benefits (b)
(~ cancers
Averted)
Costs (c)
($ M)
 1. status 0
 Quo (0%) (g)
 2. New Tech. 31
 in 15 years (25%)
 3. New Tech. 50
 in 10 years (40%)
 4. New Tech. 65
 in 5 years (52%)
N  
.....  
 1. status 0
 Quo (0%) (g)
 2. Interiln (h) 36
 Cover (29%)
o (0)    
64 (0) 2 7 2 7
166(33) 3 10 5 13
279(78) 3 11 8 13
o (0)
131
4
8
4
8
(a) Expressed as $M per fatal 1un;J cancers averted: benefits expressed as mi~e of estimates:
(b) benefits at high confidence level would be twice shown am. C/B would be half.
124 Lung cancers in absence of controls: benefits expressed in fatal 1un;J cancers
avoided in 100 years.
(c) Present value costs for early disposal am replacement capacity, in millions of 1985 dollars,
d discounted at 5%, ( ) are replacement capacity costs only.
( ) Incremental to above alternative.
(e) Costs discounted at 5%, benefits not diSCOlUlted.
(f) Costs and benefits diSCOlUlted at 5%.
(g) Percent of the 124 cancers in absence of controls.
(h) Incremented to status Quo (option # 1).

-------
Section 5:
ECONOMIC ANALYSIS
Comment 5.1: Proposed rules will have significant adverse
effects on industry's ability to contain costs and threaten the
industry's future. The costs of complying with the proposed
regulation is unreasonable. Therefore the regulation should be
rejected (D-3, D-5, D-6, D-7, D-12, D-19, D-2l).

Response: Control measures to meet the provisions of the
final rule do have significant impact on the conventional
uranium mining and milling industry. As stated in the Economic
Analysis for this rulemaking, EPA projects that this impact,
although significant, will not threaten the viability of this
industry. The EPA also feels that the costs are reasonable in
relation to the benefits derived and consistent with previous
Agency actions.
Comment 5.2: EPA should calculate cost benefit ratio to
justify a standard (D-3, D-5).
Response: EPA's policy is not to establish a dollar value
on life, or on the morbidity effects of cancer. Therefore, in
the case of radiation risks, costs and benefits are not
expressable in commensurate units, so a cost benefit ratio is
not calculated.
Comment 5.3: EPA's uranium production projections are
fundamentally flawed. The number of future mills and projected
volume of tailings are vastly overestimated. As a result, EPA
has not established an appropriate model for evaluating any of
its regulatory proposals (D-12).
Response: It is agreed that it is difficult
the future of the conventional uranium mining and
industry. In addition to all unknowns associated
projected future of the nuclear power industry is
to foresee how Congress will react to the present
this industry.
to predict
milling
with the
the inability
problems of
As stated in the Economic Analysis, EPA has assumed that
imports will continue to be a significant factor in the market
for uranium fuel, but that they will not take over the entire
market. Thus the domestic fuel production remains viable.
This implicitly assumes that Congress does act to restrict
imports sufficiently to maintain a viable, conventional,
22

-------
domestic mining and milling industry. In the model used for
the Economic Analysis, EPA assumes that the domestic
conventional mining and milling industry attains viability
within a short time (1 to 3 years) and maintains this viability
well into the future. We do not, however, assume that the
industry ever becomes large, or flourishes in a manner similar
to its structure in 1979-1980. The growth rates assumed are
low, somewhat below the Department of Energy's low growth
projection through the year 2000. EPA than assumes only modest
growth beyond the year 2000.
Comment 5.4: EPA's standard threatens the economic value
of the uranium ore in the ground because it threatens the
viability of the conventional uranium mining and milling
industry (D-6).
Response: It is true that the economic value of much of
the uranium ore in the ground is dependent upon the viability
of the conventional uranium mining and milling industry, and if
this industry does not maintain viability, the value of this
ore will disappear. This is true in the presence or absence of
EPA's standard. As stated previously, EPA has projected that
the conventional mining and milling industry will attain
viability and be sUfficiently strong to absorb the costs of the
regulations proposed under this rulemaking.

Comment 5.5: If EPA requires interim cover on existing
piles, the volume of this cover threatens the economic value of
the remaining capacity of these piles (D-6).
ResDonse: The Nuclear Regulatory Commission has estimated
that'a one meter interim cover on all piles will use up
approximately fifty percent of the total remaining capacity of
all existing mill tailing plies (EPA BID). However, the total
remaining capacity on these piles is much larger than needed to
sustain the industry at the growth rates projected by EPA in
its economic analysis. Therefore, lost capacity on existing
piles does not threaten the overall viability of the milling
industry. However, the viability of some individual mills with
little remaining capacity on their piles are threatened. EPA
is not requiring interim cover in its final rulemaking.
23

-------
Comment 5.6: The cost of EPA's standard is not justified
on the basis of the cancers prevented (D-2).

Response: Uncontrolled emissions from uranium mill
tailings piles impose significant risk of concer to those
persons residing in close proximity to the piles and to the
U.s. population. The benefit to be derived from this standard
is the reduction in this risk. The costs and benefits of
reducing the risk to the U.s. population for alternative
control levels were compared in establishing this standard.
The costs per health effect averted for this standard are in
accord with EPA practice and with other standards promulgated
under the Clean Air Act. An additional benefit that cannot be
quantified, therefore cannot be considered in a benefit-cost
evaluation, is that controls significantly reduce the risk to
maximally exposed individuals (those residing near the piles),
thereby reducing the disparity in the range of risks imposed on
the population.
The Agency does expect that extensions would be granted
for mills having piles smaller than 200 acres and having no
people living within 5 kilometers of the pile. Such mills are
remote enough to present small risks to maximally exposed
individuals and are small enough to present very small risks to
regional and national populations.
24

-------
APPENDIX A
LIST OF COMMENTERS
A.l
Introduction
The following is a list of the commenters whose comments
were responded to in this document. The identification number
used in this document is the same number used by the EPA Docket
Section (except that the Docket number A-79-ll, Category: VIII
is deleted from each number). Where there are missing
sequential numbers, it means that the document material was not
a letter of comment but some other reference material.
A.2
Commenter Listinq
Commant No
D-1
D-2
D-3
D-4
D-5
D-6
D-7
D-8
D-9
D-11
D-12
Cateqorv
Date Docketed
Wyoming Department of
Environmental Quality
3/13/86
State of Utah Department
of Health
4/01/86
Pathfinder Mines Corp.
4/04/86
4/02/86
st. Joe Minerals Corp.
The Colorado Mining Association
4/22/86
4/22/86
Umetco Minerals Corp.
State of Wyoming Office of the
Governor
4/24/86
Chevron Resources Company
4/24/86
4/24/8t::
Hecla Mining Company
Atlas Minerals
4/28/86
4/28/86
Kerr-McGee Corp. and Quivira
Mining Company
25

-------
Comment No
Cateaorv
Date Docketed
D-14
Rocky Mountain Energy
Hamel and Park for AMC
4/28/86
4/30/86
D-13
D-15
The Navajo Nation
4/30/86
4/30/86
D-18
New Mexico Environmental
Improvement Division
D-19
American Mining Congress (AMC)
D-20
Environmental Defense Fund
4/30/86
5/02/86
D-21
Texas Dept. of Health

u.s. Nuclear Regulatory
Commission (NRC)
5/16/86
5/28/86
D-22
D-23
u. S. Department of the Interior
7/10/86
26
"u.s. GOVERNMENT PRINTING OFFICE: 1986-621-735/60534

-------