&EPA
United States
Environmental Protectio.
Agency
Radiation
Office of
Radiation Programs
Washington DC 20460
EPA 520/1-82-025
December 1982
Draft
Environmental
Impact Statement
for40CFR191:
Environmental Standards
for Management and
Disposal of Spent Nuclear
Fuel, High-Level and
Transuranic Radioactive Wastes
-------
EPA 520/1-82-025
DRAFT
ENVIRONMENTAL IMPACT STATEMENT
40 CFR Part 191
ENVIRONMENTAL STANDARDS
FOR
MANAGEMENT AND DISPOSAL
OF
SPENT NUCLEAR FUEL, HIGH-LEVEL AND
TRANSURANIC RADIOACTIVE WASTES
DECEMBER 1982
U. S. ENVIRONMENTAL PROTECTION AGENCY
Office of Radiation Programs
-------
SUMMARY
DRAFT ENVIRONMENTAL IMPACT STATEMENT
40 CFR Part lyi Environmental Standards for Management and Disposal of
Spent Nuclear Fuel, High-Level and Transuranic
Radioactive Wastes.
Prepared By:
Office of Radiation Programs
U.S. Environmental Protection Agency
December 1982
1. The Environmental Protection Agency is proposing environmental
standards for the management and disposal of spent nuclear reactor fuel
and high-level and transuranic radioactive wastes. Subpart A of the
standards would limit the radiation exposure of members of the public from
management and storage of spent fuel and of waste prior to disposal.
Subpart B would establish both quantitative containment requirements for
disposal systems and qualitative requirements to assure that these
containment requirements will be met. The containment requirements would
limit the amount of radioactivity that may enter the environment for
10,000 years after disposal. The assurance requirements provide seven
principles necessary for developing confidence that these long-term
containment requirements will be complied with. These principles call for
well-designed, multiple-barrier disposal systems that would not rely upon
future generations for maintenance and would not be located near
potentially valuable resources. They also require that future generations
be provided information about the location and dangers of the wastes and
an option to recover the wastes if they need to. In addition, Subpart B
contains procedural requirements to ensure that the containment
requirements are properly applied. We think the proposed standards will
adequately protect the public health and the environment, and we believe
they can be satisfied without major economic consequences.
2. Copies of this Environmental Impact Statement and requests for comment
have been sent to the following Federal agencies:
Department of Commerce
Department of Defense
Department of Energy
Department of Health and Human Services
Department of the Interior
Department of Transportation
National Aeronautics and Space Administration
Nuclear Regulatory Commission
We have also sent copies to all State Clearinghouses and to other
individuals and organizations who have notified us of their interest.
-------
3. Comments on this Environmental Impact Statement should be received by
May 2, 1983. Please send comments (in duplicate if possible) to:
Central Docket Section (A-130)
Environmental Protection Agency
Attn: Docket No. R-82-3
Washington, D.C. 20460
4. For additional information, please contact Dan Egan at (703) 557-8610
or write to:
Director, Criteria and Standards Division
Office of Radiation Programs (ANR-460)
Environmental Protection Agency
Washington, D.C. 20460
11
-------
CONTENTS
Page
Chapter 1: Overview 1
Chapter 2: The Proposed Action 7
2.1 High Level Wastes 7
2.1.1 Commercial Reactor Wastes 8
2.1.2 Defense Wastes 11
2.1.3 Definition of High-Level Wastes in 40 CFR 191 13
2.2 Transuranic Wastes 16
2.3 Biological Effects of Radiation 17
2.4 Methods of Control 19
2.4.1 Storage of Spent Fuel and High Level Wastes 20
2.4.2 Disposal in Mined Repositories 21
2.4.3 Other Disposal Methods 21
2.5 Authorities 22
2.6 The Proposed Standards 24
2.6.1 Standards for Management and Storage (Subpart A) 24
2.6.2 Standards for Disposal (Subpart B) 27
2.7 Implementation 34
Chapter 3: Issues in Setting the Disposal Standards 37
3.1 Risks to Future Generations 37
3.1.1 Perspectives on Long Term Risks 39
3.1.2 Risks Associated with Natural Background Radiation 40
3.1.3 Risks from Uranium Ore Bodies 43
3.1.4 Risks from Nuclear Power Generation 44
3.1.5 Risks from Nuclear Weapons Fallout 45
3.2 Assessment and Reduction of Risks 48
3.2.1 Assessment of Risks from Waste Disposal 48
3.2.2 Reduction of Risks by Disposal Technology 49
ill
-------
Page
3.3 Dealing with Uncertainties 54
3.3.1 Technical Uncertainties 54
3.3.2 Uncertainties in Human Behavior 57
3.4 Choice of Format for Disposal Standard 62
3.5 Alternative Regulatory Time Periods 64
Chapter 4: Alternatives 69
4.1 Issue No Standard 70
4.2 Delay Action 71
4.3 Establish Only Qualitative Requirements 72
4.4 Delete or Deemphasize the Qualitative Assurance
Requirements 74
4.5 Select Containment Requirements on a Different Basis 76
4.6 Set Higher or Lower Release Limits 78
4.7 Set Disposal Standards in Terms of Limits on
Maximum Individual Exposure 81
4.8 Set Different Limits for Releases Due to Natural Causes
and for Those Caused by People 83
Chapter 5: Projected Health Effects From Disposal 87
5.1 Model Repositories 88
5.2 Release and Transport Mechanisms 94
5.3 Results of Risk Assessments 97
5.3.1 Population Risks 97
5.3.2 Risks to Individuals 105
5.4 Conclusions 107
IV
-------
Page
Chapter 6: Proposed Standards for Waste Disposal 111
6.1 Containment Requirements 112
6.1.1 The Accessible Environment 113
6.1.2 Radionuclide Release Limits 115
6.1.3 The Level of Protection 118
6.2 Assurance Requirements 120
6.2.1 Criterion 1 120
6.2.2 Criterion 2 121
6.2.3 Criterion 3 122
6.2.4 Criterion 4 123
6.2.5 Criterion 5 124
6.2.6 Criterion 6 125
6.2.7 Criterion 7 127
6.3 Procedural Requirements 128
Chapter 7: Proposed Standards for Waste Management Operations 131
7.1 Waste Management Operations for Various
Disposal Options 131
7.1.1 Operations for Disposal of Spent Fuel 131
7.1.2 Operations Prior to Disposal of Processed
High-Level Wastes 134
7.1.3 Collection and Disposal of Krypton-85 and
Iodine-129 Wastes 139
7.1.4 Extraterrestrial Disposal 141
7.1.5 Transmutation 142
7.1.6 Other Separations 142
7.2 Derivation of the Standards 143
7.2.1 Selection of Standards Format 144
7.2.2 Selection of Numerical Limits 145
7.2.3 Implications of the Operational Standards 148
Chapter 8: Environmental Impacts 151
Chapter 9: Regulatory Impact 157
-------
Page
Appendix A: The Proposed Standards 167
Appendix B: Risk Assessments of Geologic Repositories 183
B.I General Features of the Model 186
B.2 Risk Assessments for Populations 194
B.3 Dose Assessment for Individuals 212
References ' 223
VI
-------
TABLES
Page
2-1 Defense High-Level Waste Projections 12
2-2 Concentrations Identifying High-Level Radioactive Wastes
(Table 1 in 40 CFR Part 191) 14
3-1 Distribution of Natural Radiation Dose Equivalents 42
3-2 Dose Commitments from Fallout 46
5-1 Potential Health Effects (Fatal Cancers) Caused per Curie
Released to the Environment by Different Modes 96
7-1 Estimated Dose Equivalents to the Maximally Exposed
Individual from Solidification (Rural Site) 138
7-2 Dose Equivalents to Maximum Individual from Solidification
(Urban Site) 138
7-3 Population Doses from Waste Solidification 138
B-l Characteristics of Radioactive Waste (Spent Fuel) 188
B-2 Solubility Limits and Retardation Factors 189
B-3 Aquifer Characteristics 193
B-4 Reference Case Characteristics 199
B-5 Projected Population Risks Over 10,000 Years:
Reference Cases 205
B-6 Projected Population Risks Over 10,000 Years:
Various Canister Lives and Waste Form Release Rates 207
B-7 Projected Population Risks Over 10,000 Years:
Various Retardation Factors and Solubility Limits 209
B-8 Dose Equivalent Rates (rem/yr) from Drinking Grouadwater
Contaminated by Normal Groundwater Flow Through a
Basalt Repository 219
Vll
-------
FIGURES
Page
5-1 Reference Repository in Bedded Salt 89
5-2 Reference Repository in Granite 90
5-3 Consequences and Risks for Different Events
(Bedded Salt and Granite) 99
5-4 Projected Health Effects Over 10,000 Years
for Reference Repositories in Different Geologic Media 100
5-5 Projected Health Effects Over 10,000 Years
vs. Different Waste Form Release Rates 102
5-6 Projected Health Effects Over 10,000 Years
with Different Assumptions About Geochetnical Factors 103
5-7 Projected Health Effects Over 10,000 Years
vs. Different Canister Lifetimes 104
6-1 The Level of Protection 119
9-1 Variations in Waste Management Cost vs. Level of Protection
(Engineering Barrier Costs Only) 160
9-2 Variations in Waste Management Cost vs. Level of Protection
(Engineering Barrier Costs and Site Selection Costs) 162
B-l Reference Repository in Basalt 192
B-2 Probability of Population Risks Over 10,000 Years 197
vni
-------
Chapter 1
OVERVIEW
Although high-level radioactive wastes are produced in small
quantities, their proper management and disposal are important because of
the inherent hazards of the large amounts of radioactivity they contain.
This need for careful control has been recognized since the inception of
the nuclear age. The Federal Government has always assumed responsibility
for the ultimate care and disposal of high-level wastes, whether they are
produced by commercial or national defense activities. Over the last
several years, the Federal Government has intensified its program to
develop and demonstrate a permanent disposal method for high-level wastes.
President Reagan's April 28, 1982, message to Congress on nuclear waste
disposal reaffirmed this commitment and called for a Federally owned and
operated permanent repository to be available at the earliest practicable
date. The environmental protection standards described in this Draft
• Environmental. Impact Statement will provide the basic framework for the
long-term control of these wastes.
The Agency is proposing these generally applicable environmental
standards under authorities established by the Atomic Energy Act and
transferred to the EPA.by Reorganization Plan No. 3 of 1970. This Draft
Environmental Impact Statement includes detailed discussions of major
decisions and extensive summaries of our technical analyses. On
November 15, 1978, the Agency proposed Federal radiation protection
guidance for the disposal of all types of radioactive wastes (43 FR 53262),
-------
After reviewing the public comments on that proposal, we decided that the
characteristics of different kinds of radioactive waste are not
sufficiently similar for generally applicable criteria to be appropriate.
Therefore, we stopped developing this Federal radiation protection
guidance (46 FR 17567). However, several of the principles included in
this earlier proposal have been incorporated as integral parts of these
environmental standards.
About 500 million curies of radionuclides with half-lives greater
tnan 20 years exist in the wastes from reprocessing reactor fuel for
national defense activities. These wastes are stored in various liquid
and solid forms on three Federal reservations in Idaho, Washington, and
South Carolina, respectively. Relatively small additions are being made
from ongoing defense programs.
The total amount of spent fuel removed from commercial nuclear power
reactors contains about 800 million curies of radionuclides with
half-lives greater than 20 years. Over the next few years, this inventory
is expected to grow at a rate of about 200 million curies per year from
reactors currently licensed to operate. Virtually all of this spent fuel
is stored at reactor sites. At some reactor sites, spent-fuel storage
capacity is almost used up. Electrical utilities, the operators of
commercial reactors, are pursuing a variety of techniques to increase
storage capacities.
-------
These proposed standards apply to spent reactor fuel, high-level
wastes derived from reprocessing spent fuel, and wastes containing
long-lived radionuclides of elements heavier than uranium (transuranic
wastes). High-level wastes are covered by these standards if they contain
concentrations of radioactivity greater than the limits recently set by
the Nuclear Regulatory Commission (NRG) for acceptance at low-level waste
disposal sites (47 FR 57446). Transuranic wastes are covered if they
contain 100 nanocuries (one ten millionth of a curie) or more of
alpha-emitting transuranic radionuclides per gram of waste. The proposed
standards do not apply to wastes that have already been disposed of.
The objective of the proposed standards is to limit the risks to both
present and future generations and to adequately protect the public from
harm caused by management and disposal activities related to these
radioactive wastes. Separate standards were developed for those
activities related to waste management and storage operations preparatory
to disposal (Subpart A) and for the long-term performance of disposal
systems (Subpart B). The standards for management and storage are
intended to protect exposed individuals while these operations are in
progress. The standards for disposal are intended to assure long-term
isolation of the hazardous wastes from the biosphere and protection of the
public health.
The rationale for the standards for management and storage (Subpart A)
is that these operations should not be permitted to substantially increase
the risk to people beyond that now accepted for the normal operations of
-------
the uranium fuel cycle. We conclude that these added risks can be very
small and that the exposure limits we have already established for the
fuel cycle (40 CFR 190) can be extended to cover these additional
operations. Therefore, the effect of Subpart A of the proposed standards
is to apply the same requirements to the normal operations of all parts of
the uranium fuel cycle, except transportation. Similar considerations
apply to the management and storage of high-level and transuranic wastes
from national defense activities, which are also covered by this Subpart.
The disposal standards (Subpart B) must deal with a yet unproven
technology and with the need to extend public health protection far into
the future. As our primary disposal standards, we are proposing
containment requirements that place numerical limits on possible releases
of radionuclides to the environment for 10,000 years after disposal.
Although these requirements are not specific to any particular method of
disposal, we focused on the use of mined geological repositories because
more information is available on this technique. We performed detailed
technical evaluations of the probabilities and consequences of possible
events that could disrupt such a repository and cause release of a part of
its contents. Our calculations show that the total adverse impact of
releases of radionuclides from such a facility over 10,000 years can be
kept very small and should be no greater than the risks from the unmined
ore from which the wastes were derived. The proposed containment
requirements—which would limit long-term risks to these low levels—apply
to any method of disposal. Thus, any other method would have to provide
at least as much protection as geologic disposal.
-------
Because of the uncertainties inherent in projecting disposal system
performance for 10,000 years, Subpart B also contains seven qualitative
criteria that are needed to develop appropriate confidence that our
containment requirements will be met. These assurance requirements set
forth a cautious approach to disposal of these wastes, and they provide
the context necessary for application of our containment requirements.
The assurance requirements call for well-designed, multiple-barrier
disposal systems that would not rely upon perpetual maintenance by future
generations and that would be located where it is unlikely that they would
be disturbed by natural forces or human activities. In addition, our
disposal standards contain procedural requirements to ensure that the
containment requirements are properly implemented.
We evaluated the effects of setting our containment requirements at
different levels of protection. We found that the increased costs of
setting these requirements at the proposed level could range from zero to
50 million (1981) dollars per year when compared to the costs of choosing
a level more than 10 times less stringent (release limits 10 times greater
than our proposed limits). This potential increase is much less than the
uncertainty in the total costs for waste management and disposal, since
these projected costs range from about 700 million to almost 1.5 billion
(1981) dollars per year. For comparison, electrical utility revenues were
about 100 billion dollars in 1980. We estimate that the potential
economic impact of choosing the more stringent level of protection could
-------
be about a 0.2-percent increase in the costs of generating electricity
from nuclear power plants and a much smaller increase (about 0.05 percent)
in average electricity rates.
The standards for waste management and storage (Subpart A) will be
implemented by the NRC for commercial nuclear power activities and by the
Department of Energy (DOE) for national defense facilities. Implementation
procedures for Subpart A will be similar to those for our Uranium Fuel
Cycle standards (40 CFR 190). Our standards for disposal (Subpart B) will
be implemented by NRC for all high-level wastes, whether the wastes come
from commercial or military activities. NRC will develop the necessary
regulations (primarily 10 CFR 60) and will issue appropriate construction
and operating licenses to DOE. DOE will select, design, and build all
disposal facilities for high-level wastes. Under current law, disposal of
transuranic wastes from military activities is not regulated by NRC;
therefore, DOE will apply our standards and guides to disposal of these
transuranic wastes.
-------
Chapter 2
THE PROPOSED ACTION
This Draft Environmental Impact Statement supports our proposed
environmental standards for the management and disposal of spent nuclear
fuel and high-level and transuranic radioactive wastes. It explains how
we developed the proposed standards, examines alternative actions we could
have taken, and discusses the related environmental impacts.
In this chapter we describe our proposed action. We identify the
radioactive materials covered by these standards, how they are produced,
their hazards, and how we can control them. We also explain the legal
authorities under which these standards were developed. Finally, we
summarize the various requirements we are proposing to protect public
health and the environment. The complete text of our proposed
environmental standards is contained in Appendix A.
2.1 HIGH-LEVEL WASTES
These proposed standards apply to the highly radioactive wastes
resulting from reprocessing irradiated (spent) nuclear reactor fuel—and
to the spent fuel itself if it is disposed of without reprocessing or if
it is stored pending a determination on disposal. These wastes are
products of the fission and activation processes associated with nuclear
reactors. The reactors are used for commercial power production,
research, and national defense activities. Other wastes containing large
quantities or high concentrations of radioactivity, which may represent a
-------
significant danger to the public health if not properly handled, are not
specifically included in these standards, but they may require similar
environmental standards for disposal of other categories of radioactive
waste.
High-level radioactive wastes from reprocessing have been commonly
designated as the material remaining after recovery of uranium and
plutonium from irradiated (spent) reactor fuel. These wastes include
residual fissionable materials, fission products, and activation
products. They may be in liquid or solid form, may be soluble or
insoluble in water, and may emit any of several types of radiation over a
wide range of energy.
Both spent fuel and high-level radioactive wastes from reprocessing
are intensely radioactive and generate substantial quantities of heat.
Radioactivity and heat production continue for long periods of time,
because the wastes contain a number of long-lived radionuclides. The
transuranium elements, especially, have long radiological half-lives and
present a possible hazard to people for tens of thousands of years.
2.1.1 Commercial Reactor Wastes
The fuel used in the present generation of commercial light-water
reactors consists of a mixture of uranium-238 and uranium-235 dioxides
encased in zircalloy or stainless steel tubes. During reactor operation,
fission of the uranium-235 produces energy, neutrons, and fission products.
8
-------
The neutrons produce further fission reactions and thus sustain the chain
reaction. The neutrons also convert some of the uranium-238 into
plutonium-239, which can fission like uranium-235. In time, the fissile
uranium-235, which originally constituted some 3 percent of the enriched
fuel, is depleted to such a low level that power production becomes
inefficient. Once this occurs, the fuel bundles are deemed "spent" and
are removed from the reactor (an annual removal rate of 30 tons per
reactor is typical).
Reprocessing of commercial spent fuel has been proposed with a view
towards recovering the unfissioned uranium-235 and the plutonium for reuse
as a fuel resource. However, spent fuel generated by commercial power
reactors is not currently being reprocessed. The radioactive materials
in spent fuel fall into two major categories: fission products and
actinide elements. Typically, fresh spent fuel contains more than
100 radioactive nuclides as fission products. Fission products of
particular importance, because of the quantities produced or their
biological hazard, are strontium-90, technetium-99, iodine-129 and -131,
the cesium isotopes 134, 135, and 137, tin-126, and krypton-85 and other
noble gas isotopes. The actinides consist of uranium isotopes,
transuranic elements (i.e., isotopes with atomic number greater than 92,
including plutonium-239, americium-241 and -243, and neptunium-237) formed
by neutron capture, and their decay products. Spent fuel also contains
tritium (hydrogen-3), carbon-14, and other radioactive isotopes created by
neutron activation. The exact composition of radionuclides in any given
-------
spent-fuel sample depends on the reactor type, the initial fuel
composition, the length of time the fuel was irradiated, and the elapsed
time since its removal from the reactor core.
If fuel is reprocessed, the resulting waste is an acidic aqueous
solution containing most of the actinides and the nonvolatile fission
products present in the fuel. About 0.5 percent of the original uranium
and plutonium would be left in the waste liquid. Current Federal
regulations require that commercial high-level liquid waste be converted
to a solid within 5 years. Calcination is the best available
solidification process. The calcined waste should be converted to a more
stable form for ultimate disposal.
The inventory of spent fuel through 1980 was about 7000 metric tons
of heavy metal (DOE 81). The activity of this spent fuel depends heavily
on its age, since fresh spent fuel contains large quantities of
short-lived fission products. By calculating waste activities as of
10 years after removal from the reactor, one can largely eliminate the
variability due to short-lived fission products. On this basis, the
activity of the 7000 metric tons of spent fuel corresponds to over
2 billion curies. A small amount of commercial fuel has been reprocessed,
producing about 600,000 gallons of liquid waste (EPA 80) containing about
30 million curies (BNWL 76).
10
-------
Predictions of high-level waste production from commercial generation
of electrical power depend on estimates of the growth of the nuclear power
industry. The Interagency Review Group (IRG 79) estimated a range of
cumulative totals by the year 2000 of about 70,000 to 100,000 metric tons
of spent fuel, corresponding to 148 and 380 gigawatt-electric (GWe) of
installed capacity, respectively. Arthur D. Little (ADL 79) developed an
upper estimate of 700 GWe of installed capacity by the year 2010, for
500,000 metric tons of spent fuel containing about 200 billion curies.
2.1.2 Defense Wastes
Weapons program reactors are operated to produce plutoniutn;
reprocessing to recover the plutonium is an integral part of the weapons
program operations. Naval propulsion reactor fuel elements are also
reprocessed to recover the highly enriched uranium they contain.
The present inventory of defense wastes in the U.S. is about 290,000
cubic meters (10 million cubic feet or 70 million gallons), stored at
three Federal reservations as liquids, sludges, and solids (IRG 79).
Operation of the weapons program reactors and other defense activities
will continue to produce defense wastes. Table 2-1 gives projections of
both the quantity and radioactivity of high-level defense wastes (ADL 79).
11
-------
Table 2-1
Defense High-Level Waste Projections
Volumes Radioactivity (Ci)
Total
Site Cubic Meters Fission Products Uranium Transuranics
Hanford(aJ 200,000 2.5 x 108 710 1.4 x 105
Savannah River(b) 83,000 3.2 x 108 48 7.4 x 105
Idaho^ ll.QQq l.Q x 1Q8 2 1.0 x 1Q3
TOTAL 294,000 6.7 x 108 760 8.8 x 105
(a) i99Q waste projection.
(b' 1985 waste projection.
12
-------
2.1.3 Definition of High-Level Wastes in 40 CFR 191
In developing its regulations for disposal of high-level wastes in
geologic repositories (10 CFR 60), the Nuclear Regulatory Commission (NRC)
defined high-level wastes as (NRC 81b):
". . . (1) irradiated reactor fuel, (2) liquid wastes
resulting from the operation of the first cycle solvent
extraction system, or equivalent, and the concentrated wastes
from subsequent extraction cycles, or equivalent, in a facility
for reprocessing irradiated reactor fuel, and (3) solids into
which such liquid wastes have been converted."
For the purposes of our environmental standards, we are proposing a
somewhat different definition of high-level wastes:
"... any of the following that contain radionuclides in
concentrations greater than those identified in Table 1:
(1) liquid wastes resulting from the operation of the first
cycle solvent extraction system, or equivalent, in a facility
for reprocessing spent nuclear fuels; (2) the concentrated
wastes from subsequent extraction cycles, or equivalent;
(3) solids into which such liquid wastes have been converted;
or (4) spent nuclear fuel if disposed of without reprocessing.
There are two substantive differences between our definition and the
one in 10 CFR 60. The first is that our definition does not identify
spent nuclear fuel as a waste unless it is determined that such fuel will
be disposed of without reprocessing. Thus, >--ovisions of our standards
that specifically apply to waste would not apply to spent fuel until such
a determination was made.
The other major difference is that our definiton of high-level waste
excludes materials with concentrations of radioactivity below those in
Table 2-2 (which reproduces Table 1 of the proposed standards).
13
-------
Table 2-2:
Concentrations Identifying High-Level Radioactive Wastes
(Table 1 in 40 CFR Part 191)
Radionuclide Concentration
(curies per gram of waste)
Carbon-14 8xl06
Cesium-135 8x 10~4
Cesium-137 5x 10~3
Plutonium-241 3xlO~
_3
Strontium-90 7x10
Technetium-99 3x 10~
Tin-126 7x 10~7
Any alpha-emitting transuranic
radionuclide with a half-life ----------- 1x10
greater than 20 years
Any other radionuclide with a half-life
_3
greater than 20 years --------------- 1x10
NOTE: In cases where a waste contains a mixture of radionuclides,
it shall be considered a high-level radioactive waste if the sum of the
ratios of the radionuclide concentration in the waste to the concentration
in Table 1 exceeds one.
For example, if a waste contains radionuclides A, B, and C in
concentrations Cfl, Cb, and Cc and if the concentration limits from
Table 1 are CLg, CLb, and CLC, then the waste shall be considered
high-level radioactive waste if the following relationship exists:
ca cb cc
CLa CLb CLC **
14
-------
We propose to make this exclusion because some wastes derived from
reprocessing spent fuel may not require the same isolation as the wastes
containing most of the radioactivity produced in the fuel. In particular,
processing of certain high-level wastes from national defense activities
may remove much of the radioactivity, leaving residual material that is
far less dangerous (for example, large volumes of salt cake containing
relative low concentrations of technetium-99). Our definition would allow
such material to be disposed of without having to meet the stringent
requirements we feel are appropriate for the most highly radioactive
wastes. The disposal requirements for materials excluded by our
definition will be addressed as we develop standards for low-level
radioactive wastes.
The levels in Table 2-2 are equivalent to the maximum concentrations
for acceptance at shallow-land burial sites that the NRG recently
promulgated as part of 10 CFR Part 61 (47 FR 57446). The NRG derived the
concentration limits in 10 CFR 61 so that a person intruding into a
shallow-land site—after institutional controls were no longer effective—
should not receive a radiation exposure greater than 500 millirem per year
(NRG 81a and WI 81). We converted the units of these concentrations from
curies per unit volume to curies per unit mass by assuming a density of
1 gram per cubic centimeter. For wastes with higher densities, the levels
in Table 2-2 would be somewhat more stringent than the levels in 10 CFR 61.
15
-------
2.2 TRANSURANIC WASTES
The proposed standards also apply to wastes containing alpha-emitting
transuranic nuclides with half-lives greater than 1 year at concentrations
greater than 100 nanocuries per gram (nCi/g). Alpha-emitting transuranic
nuclides represent a special type of hazard because of their long
half-lives and high radiotoxicity. Cohen and King (CO 78) considered the
levels at which these materials might require handling different than that
generally given to low-level radioactive wastes. They concluded that the
more hazardous transuranic elements could cause excessive radiation doses
if they are present at concentrations of about 1 microcurie per cubic
centimeter of waste. A task group (HE 79) recommended shallow-land burial
for Department of Energy (DOE) transuranic wastes with concentrations
below 1 to 10 nCi/g. The task group also suggested that somewhat deeper
burial was suitable for transuranic wastes with concentrations up to
100 nCi/g. In view of these studies, txansuranic wastes with
concentrations aoove 100 nCi/g were included under these standards.
There are about 1100 kilograms of transuranic elements at levels
above 10 nCi/g in defense wastes at DOE facilities (IRG 79). These
radionuclides are contained in about 400,000 cubic meters (14 million
cubic feet) of transuranic-contaminated wastes. Both the DOE and the EPA
have studies underway to define the amounts and characteristics of defense
transuranic wastes containing more than 100 nCi/g.
16
-------
2.3 BIOLOGICAL EFFECTS OF RADIATION
Radioactive materials may emit participate (alpha and beta) or
electromagnetic (gamma and X) radiation that can interact with matter.
In this interaction, energy is transferred from the radiations to the
matter. Energy transferred to matter is called absorbed dose. The unit
of dose is the rad (1CRU 71), which corresponds to the absorption of
100 ergs per gram of material (0.01 joules per kilogram).
Gamma radiation, X-rays, and beta particles are sparsely ionizing
radiations, with low linear energy transfer (low-LET). Alpha particles,
such as those emitted by transuranic radionuclides, are high-LET
radiations that exhibit a very dense ionization pattern in tissue. The
interaction of radiation with living tissue in man, animals, or plants may
damage the tissue. Since the absorption of the same quantity of energy
from different radiations can produce different biological effects in
living tissue, modifying factors are applied for the nature of the
radiation and other considerations. The product of the dose and the
modifying factors is called the dose equivalent (ICRU 73); the unit of
dose equivalent is the rem. The total impact on a population is the
population dose equivalent, the sum of the radiation dose equivalent
incurred by each person in the population.
Releases of high-level wastes to the environment may cause human
exposure from direct radiation or from breathing or ingesting radio-
nuclides. These modes of exposure may arise because of volatilization,
17
-------
particulate dispersion, and dissolution of the waste materials—the
predominant pathway(s) depend on the mode of release, the specific
environmental circumstances, and the isotopic composition of the released
material.
Exposure to large amounts of radiation, on the order of tens to
hundreds of rems, in a short time (acute exposure) can produce a number of
effects on humans ranging from barely detectable to severe. These effects
range from chromosome and blood cell changes through loss of appetite and
loss of hair, to diarrhea, vomiting, and even death. The severity of the
effect is directly related to the amount of radiation absorbed by the
individual.
Releases from a high-level waste repository, if they occur at all,
are expected to be much smaller than those that would cause acute effects.
These continuing (chronic) exposures to radiation, at levels well below
those noted in the preceding paragraph, have no detectable early effects.
However, in such cases, there may be delayed effects. The major delayed
effects are the development of cancer and the production of genetic
changes. In contrast to acute effects, for which the severity depends on
the dose equivalent incurred, delayed effects develop only in some of the
exposed individuals. The rate of incidence is a statistical function and
depends on the total population dose equivalent incurred.
18
-------
The relationship between the radiation dose and the number of delayed
health effects has been studied extensively. The National Academy of
Sciences (NAS 72 and 80) and the United Nations Scientific Committee on
the Effects of Atomic Radiation (UN 77) have published reviews of the
current state of knowledge on this subject. Because of the inherent
uncertainties in our scientific evidence, and as a prudent general policy
for public health protection, we use a linear, non-threshold relationship
between radiation dose and the risk of incurring cancer or genetic
abnormalities. This relationship assumes that any exposure can produce
some harm and that the harm is proportional to the absorbed dose. For the
low dose rates considered here, this assumption may overestimate the risk
from low-LET radiation. Because no method of disposing of hazardous
materials can be entirely free of risk, the development of standards
limiting the level of risk cannot be based solely on health
considerations. The extent to which risk can reasonably be reduced,
considering social, economic, and other factors, must also be evaluated.
2.4 METHODS OF CONTROL
Protection of humans from the hazards of radioactive wastes requires
shielding or permanent physical separation of wastes from people.
Specially designed storage facilities may provide short-term protection,
but long-term protection must rely on passive natural and engineered
barriers requiring no upkeep during their entire functional lifetime.
(Conceivably, the wastes could be removed entirely from the biosphere by
19
-------
transport away from the earth or by transmutation to less harmful
materials, but these alternative concepts do not currently appear
practical.)
Subpart A of the proposed standards applies to the management,
including storage, processing, and emplacement, of the wastes, but it does
not cover their offsite transportation. Subpart B applies to the wastes
after disposal. For a geologic repository, for example, Subpart B would
take effect when the mine is backfilled and sealed.
2.4.1 Storage of Spent Fuel and High-Level Wastes
There is considerable experience with storing spent fuel. Fuel
elements are kept in racks in water basins to remove the heat and provide
shielding. Dry, air-cooled storage of aged spent fuels, which produce
less heat than fuels recently removed from a reactor core, has been
proposed.
There is also considerable experience with storing liquid wastes from
the reprocessing of spent fuel, since most defense wastes are of this
nature. Most of these acidic liquids have been neutralized for storage in
large steel tanks. A small fraction of these existing liquid wastes has
leaked from the tanks, requiring transfer into new tanks. Liquid wastes
can be converted to solids, which are then stored like spent fuel.
20
-------
2.4.2 Disposal in Mined Repositories
Development work on methods for disposal of spent fuel and solidified
high-level and transuranic radioactive wastes has concentrated on mined
geological repositories. Such repositories would be constructed at depths
greater than 1000 feet by using conventional mining techniques in suitable
host media. Suggested media include granite, basalt, volcanic tuff, and
salt. Wastes in canisters would be placed in holes in the mine floor.
When the repository is full, it would be backfilled. After a validation
period, during which the wastes could be retrieved, the site would be
permanently sealed. Protection would be provided by a stable and
insoluble waste form, a durable canister, a stable host medium, and low
migration potential for radionuclides through the environment around the
host rock. Mined geological repositories are expected to be available for
use sooner than any other disposal method.
2.4.3 Other Disposal Methods
A number of other methods have been suggested for disposal of
high-level radioactive wastes. These methods include:
— Placement of fresh liquid or solid wastes directly into rocks by
melting. The heat of the fresh wastes would melt the rock, and
the wastes would become incorporated as an integral component of
that rock.
— Placement of containers of waste into holes 10,000 to 30,000 feet
deep.
21
-------
— Placement of containers of high-level wastes on or under the ocean
floor. Disposal of high-level wastes in the ocean is prohibited
under United States law by the Marine Protection, Research, and
Sanctuaries Act of 1972.
— Transport into space (extraterrestrial disposal) or transmutation
in fission or fusion reactors of the most long-lived or hazardous
radionuclides. Both of these methods would dispose of only a
portion of the wastes, leaving the rest for disposal on earth.
These alternative methods of disposal are discussed in more detail in
Chapter 3.
2.5 AUTHORITIES
Reorganization Plan No. 3 of 1970 transferred to EPA two functions
derived from the Atomic Energy Act of 1954: (1) the responsibilities of
the former Federal Radiation Council, and (2) the authority to set
generally applicable environmental standards. The authority to set
generally applicable environmental standards was transferred from the
Atomic Energy Commission:
". . . to the extent that such functions of the Commission
consist of establishing generally applicable environmental
standards for the protection of the general environment from
radioactive material. As used herein, standards mean limits on
radiation exposures or levels, or concentrations or quantities
of radioactive material, in the general environment outside the
boundaries of locations under the control of persons possessing
or using radioactive material."
22
-------
Our generally applicable environmental radiation standards are rules
that apply outside the controlled boundaries of facilities using
radioactive materials. They are designed to preserve the general
environment and protect the public health. The NRC—or in some cases the
DOE—implements and enforces these standards by issuing specific
regulations and by assuring that individual and interrelated components of
the fuel cycle are constructed and operated consistent with licensing
provisions and all applicable regulations.
The authorities inherited from the former Federal Radiation Council,
under Executive Order 10831 and the Atomic Energy Act (42 U.S.C. 2021(h)),
include the responsibility to:
". . . advise the President with respect to radiation
matters, directly or indirectly affecting health, including
guidance to Federal agencies in the formulation of radiation
standards ..."
We provide advice by preparing Federal radiation protection guidance
for approval by the President as executive direction to Federal agencies.
These guides establish overall direction for the radiation protection
programs of all Federal agencies. Executive Order 12088 makes the head of
each agency responsible for compliance with such guides, once the
President has approved them. In addition, the Order directs the EPA to
monitor compliance by Federal agencies. Conflicts on implementation may
be resolved by the Director of the Office of Management and Budget.
Exemptions from the guides may be granted by the President.
23
-------
The environmental protection requirements described in this Draft
Environmental Impact Statement are proposed solely under our generally
applicable environmental standards authority. However, we considered
alternatives to our proposed action that would involve our Federal
radiation protection guidance authority. These alternatives are discussed
in Cnapter 4.
2.6 THE PROPOSED STANDARDS
We are proposing generally applicable environmental standards for
both the management and the disposal of spent fuel, and high-level and
transuranic radioactive wastes. When promulgated, these standards will
become a new Part 191 to Title 40 of the Code of Federal Regulations
(40 CFR 191). Subpart A applies to all operations up to and including the
emplacement of the wastes in a disposal system. Subpart B applies to the
period after disposal, except that these standards would not apply to
disposal directly into the oceans or ocean sediments. Both Subparts are
summarized below; they are examined in more detail in Chapters 6 and 7 of
this document.
2.6.1 Standards for Management and Storage (Subpart A)
The preparation for disposal or storage of these materials and the
placement of the materials in a disposal system are included under these
standards. The transportation of the materials is not covered. These
standards for waste management and storage operations apply only to
anticipated normal events or releases.
24
-------
Most of the wastes subject to this regulation (40 CFR 191) were and
will be derived from the production of commercial nuclear power and from
activities in national defense programs. We have issued regulations
(40 CFR 190) covering one aspect of nuclear power production—the uranium
fuel cycle—from milling of ore through the generation of electricity at a
nuclear power plant and the reprocessing of spent uranium fuel. The
40 CFR 191 regulations proposed here ideally should include all operations
following those covered in 40 CFR 190 up to and including the final
abandonment of the waste materials. In this way, no portion of the
uranium fuel cycle subsequent to mining (which is not included under the
authority of the Atomic Energy Act of 1954, as amended) would be left
uncontrolled. Waste management problems in national defense programs, and
in other projected or contemplated fuel cycles, are similar enough to
those in the uranium fuel cycle that they can also be included under
40 CFR 191. The storage of spent fuel, whether or not it has been
designated as a waste material, is included in this regulation when such
storage is not already regulated under 40 CFR 190.
We examined the radiation exposure to members of the public from
waste management and storage prior to geologic disposal and determined
that it is feasible, with available technology and proper siting of
facilities, to limit maximum individual doses from all normal operations
to those dose limits established for uranium fuel cycle operations in
40 CFR 190. Those limits are 25 raillirem to the whole body, 75 millirem
to the thyroid, and 25 millirem to any other organ. Therefore, we propose
25
-------
to extend the limitations of 40 CFR 190 to the operations covered by
40 CFR 191. Since some of these operations can be conducted with
individual doses significantly lower than these limits, Subpart A also
requires that exposures be reduced below these limits to the extent
reasonably achievable, taking into account technical, social, and economic
considerations.
These limitations for individual exposures from waste management
operations do not include transportation of spent fuel or high-level and
transuranic wastes. Normal exposures in transportation would be in the
direct radiation to persons near shipments; radioactive materials could be
released only due to accidents. Transportation of spent fuel was
considered in the development of the Uranium Fuel Cycle (40 CFR 190)
standards. We found that the average radiation dose to individuals and
the total dose to the general public were small for transportation
activities in the uranium fuel cycle. It is unlikely that an individual
would receive an annual exposure greater than 25 inillirem from exposure to
shipments. We concluded, however, that providing a regulatory guarantee
that no individual could receive annual doses greater than 25 milliretn
would be extremely difficult because of the problems involved with
identifying and monitoring individual exposures. The expected risks from
transportation of high-level wastes are similarly small. Thus, it is
inappropriate to include transportation under these standards.
26
-------
2.6.2 Standards for Disposal (Subpart B)
The intent of disposal is to isolate the wastes from the environment
for a long time without the need for human intervention. Therefore, the
standards for a disposal system can be implemented only through general
planning and selection criteria and by designing the system to meet
projected performance requirements.
The proposed standards for disposal apply to both normal and
accidental releases, and both types of events must be considered by those
who will design and license a repository. The most significant releases
may result from unplanned or accidental events or processes. These
accidental releases would occur long after the conclusion of normal
operations at the disposal site. A release may be undetected for long
periods of time, and there may be no way to correct the situation. There
may be no personnel on hand to take protective actions, and the very
concept of the site as a place under control of an organization may be
inappropriate. Addressing the problem of unplanned releases is,
therefore, a prime requirement of the proposed standards for disposal.
The proposed standards for disposal contain numerical containment
requirements for the first 10,000 years after disposal, accompanied by
qualitative requirements to assure that these containment requirements are
met. These two parts of our proposed action are complementary: the
containment requirements set limits on potential releases of radioactive
materials to the environment—limits that will serve as overall design
27
-------
objectives for disposal systems; the assurance requirements provide the
framework necessary to develop appropriate confidence in meeting these
projected release limits in spite of the inherent uncertainties of
predictions over 10,000 years. In addition, the proposed standards
contain procedural requirements to ensure that the containment
requirements are properly applied to specific disposal systems.
Containment Requirements
As our primary environmental protection standards, we are proposing
numerical limits on the projected maximum amount of radioactivity a
disposal system can release to the environment. These requirements would
apply to both expected and accidental releases that could occur within
10,000 years after waste disposal. They place a limit on the harm that a
disposal system can cause to future generations.
For high-level wastes, the release limits are expressed as the
maximum amounts of radioactivity that can be released from disposal of
tne wastes generated from 1000 metric tons of heavy metal (MTHM). For
transuranic wastes, the release limits are expressed as the maximum
amounts of radioactivity that can be released from disposal of transuranic
wastes containing 1 million curies of alpha-emitting transuranic
radionuclides. These units were chosen so that the standards would
require alpha-emitting radioactivity from either high-level or transuranic
wastes to be isolated with about the same degree of effectiveness.
28
-------
To determine the potential harm to future generations from waste
disposal, we studied the projected environmental impacts of high-level
waste disposal in mined geologic repositories. Our studies were based on
the potential releases from a generic model of a repository containing
100,000 MTHM as spent reactor fuel. This is about as much waste as would
be generated during the 40-year operating lifetimes of 100 large reactors
of current design. We examined the risks over the first 10,000 years
after disposal.
As a basis for comparison, we also evaluated other sources of
radiation exposures to present and future generations. We looked at the
radiation risks from natural background radiation, from untnined uranium
ore bodies (the initial source of these wastes), from commercial nuclear
power generation, and from fallout from previous atmospheric testing of
nuclear weapons.
Based on these analyses and evaluations, we decided we could choose
performance requirements that would satisfy two objectives:
First, the requirements would limit the harm to future generations to
no more than 1000 excess cancer deaths over 10,000 years (an average of
one extra death every 10 years) from disposal of the wastes from 100,000
MTHM. This risk is about the same as our smallest estimate of the harm
from an equivalent amount of unmined uranium ore, and it is much smaller
than the risks from the other sources of radiation exposure that we
studied.
29
-------
Second, the requirements appear to be achievable by well designed
geologic repositories at carefully selected sites. Any other method used
would have to provide at least as much protection as that projected for
geologic disposal.
Once we chose this level of protection as a basis for these
standards, we developed a set of numerical release limits for the various
radioisotopes in the wastes. We estimated how many curies of each
radioisotope would cause 1000 deaths over 10,000 years if it were the only
radionuclide released; this number of curies determined the release limit
for that radionuclide. For releases involving more than one radionuclide,
the allowed release for each radionuclide is reduced to a fraction of its
limit to assure the overall limit on harm is not exceeded. The procedure
for using the release limits is described in Table 2 of the proposed
standards (see Appendix A of this document).
Assurance Requirements
Closely associated with our numerical containment requirements are
seven qualitative requirements we believe are essential for developing the
needed confidence that these long-term release limits will be met. Our
assurance requirements address and compensate for the uncertainties that
necessarily accompany plans to isolate high-level and transuranic wastes
from the environment for a very long time. No matter how promising
analytical projections of disposal system performance appear to be, these
wastes should be disposed of in a cautious manner that reduces the
30
-------
likelihood of unanticipated releases. Our assurance requirements provide
the context necessary for application of our containment requirements, and
they should insure very good long-term protection of the environment.
Several of the concepts incorporated in these assurance requirements
were adapted from the Federal radiation protection guidance for all types
of radioactive waste disposal that we proposed for public comment on
November 15, 1978 (43 FR 53262). After reviewing the responses we
received, we decided that the characteristics of different kinds of
radioactive waste are not sufficiently similar for generally applicable
criteria to be appropriate. Therefore, we stopped developing this Federal
radiation guidance (46 FR 17567). However, we also determined that,
because of the uncertainties inherent in meeting our 10,000-year
containment requirements, several of the principles included in this
earlier proposal needed to be incorporated as integral parts of these
standards for disposal of high-level and transuranic wastes.
We expect that the specific steps taken by the NRG or the DOE to
comply with each of these seven assurance requirements will be described
in the Federal environmental impact statement—and other appropriate
decision documents—for each disposal system. These seven requirements
are:
1. Wastes shall be disposed of promptly once disposal systems are
available and the wastes have been suitably conditioned for
disposal.
31
-------
2. Disposal systems shall be selected and designed to keep releases
to the accessible environment as small as reasonably achievable,
taking into account technical, social, and economic
considerations.
3. Disposal systems shall use several different types of barriers to
isolate the wastes from the accessible environment. Both
engineered and natural barriers shall be included. Each such
barrier shall separately be designed to provide substantial
isolation.
4. Disposal systems shall not rely upon active institutional
controls to isolate the wastes beyond a reasonable period o£ time
(e.g., a few hundred years) after disposal of the wastes.
5. Disposal systems shall be identified by the most permanent
markers and records practicable to indicate the dangers of the
wastes and their location.
6. Disposal systems shall not be located where there has been raining
for resources or where there is a reasonable expectation of
exploration for scarce or easily accessible resources in the
future. Furthermore, disposal systems shall not be located where
there is a significant concentration of any material which is not
widely available from other sources.
7. Disposal systems shall be selected so that removal of most of the
wastes is not precluded for a reasonable period of time after
disposal.
32
-------
Procedural Requirements
The containment requirements were derived with the assistance of our
performance assessments of long-term repository performance (summarized in
Chapter 5). When these requirements are applied to a particular disposal
system, some of the procedures we used in our assessments must be retained
to ensure that the intent of our containment requirements is met. On the
other hand, some of the assumptions we made should be replaced with the
specific information developed for each particular system. Our three
procedural requirements provide instructions necessary for performance
assessments of specific disposal systems to properly demonstrate
compliance with our containment requirements:
1. The assessments shall consider realistic projections of the
protection provided by all of the engineered and natural barriers
of a disposal system.
2. The assessments shall not assume that active institutional
controls can prevent or reduce releases to the accessible
environment beyond a reasonable period (e.g., a few hundred
years) after disposal. However, it should be assumed that the
Federal Government is committed to retaining passive
institutional control of disposal sites in perpetuity. Such
passive controls should be effective in deterring systematic or
persistent exploitation of a disposal site, and it should be
assumed that they can keep the chance of inadvertent human
intrusion very small as long as the Federal Government retains
such passive control of disposal sites.
33
-------
3. The assessments shall use information regarding the likelihood of
human intrusion, and all other unplanned events that may cause
releases to the accessible environment, as determined by the
implementing agency for each particular disposal site.
2.7 IMPLEMENTATION
The standards for waste management and storage (Subpart A) will be
implemented by the NRC for commercial nuclear power activities and by the
DOE for national defense facilities. Implementation procedures for
Subpart A will be very similar to those for the Uranium Fuel Cycle
Standards (40 CFR 190).
The standards for disposal (Subpart B) will be implemented by the NRC
for all high-level wastes, whether the wastes come from commercial or
national defense activities. The NRC will do this by developing the
necessary regulations (primarily 10 CFR 60) and by issuing appropriate
licenses. Under current law, disposal of transuranic wastes from national
defense activities is not regulated by the NRC; therefore, the DOE will
implement our requirements for disposal of these transuranic wastes.
The containment requirements in Subpart B will be applied through
design specifications, and the implementing agencies will have to evaluate
long-term projections of the disposal system performance. As a result, a
vital part of implementation will be the use of adequate models, including
the probabilities of unplanned events, to relate appropriate site and
34
-------
engineering data to projected performance. The NRG has made substantial
progress in developing such analytical models to predict long-term
performance of actual geologic repositories. These models are quite
detailed, and they are capable of evaluating how important any
uncertainties in specific types of data are to the overall projections of
repository performance. Thus, they can provide information about any
needs for obtaining better data to determine if repositories meet the
containment requirements of these standards.
At our request, the National Academy of Sciences studied the
difficulties in verifying compliance with long-term environmental
protection requirements for geologic disposal (NAS 79). Our NAS panel
developed an approach that specifies the types of information needed and
outlines appropriate methods for obtaining this data at prospective
sites. Based on this NAS study, the NRC models, our own analytical
efforts, and the confidence that should be provided by our assurance
requirements, we have concluded that our containment requirements can be
effectively implemented.
35
-------
36
-------
Chapter 3
ISSUES IN SETTING THE DISPOSAL STANDARDS
Our standards for disposal of high-level and transuranic wastes are
intended to provide long-term protection over the very long periods that
these materials present a possible hazard to man. This approach
introduces questions not ordinarily considered in regulatory decisions
having only short-term impacts. For disposal of high-level wastes, most
of the potential harm would occur far into the future. This chapter,
therefore, describes how we addressed each of the following questions:
1. How should we treat risks to future generations?
2. How should we assess the capability of disposal systems to reduce
long-term risks?
3. How should we consider the uncertainties in predicting the future?
4. How should we express the standards to best reflect the
environmental protection objectives we desire?
5. How should we select the period of time over which we should
regulate?
3.L RISKS TO FUTURE GENERATIONS
The National Environmental Policy Act (NEPA) requires that we
consider risks to future generations. Since disposal of high-level and
transuranic radioactive wastes create some risks for present and future
generations, regulatory actions must attempt to reduce those risks to an
acceptable level.
37
-------
Some might argue that no risks to future generations from disposal of
these wastes should be allowed, since these generations receive no direct
benefits from present activities and since they cannot take part in
decisions to create such risks. Others would hold that the industrial,
medical, scientific, and military activities that produce the wastes may
contribute to the knowledge and security of future generations. Moreover,
some of these activities may be alternatives to activities that might
result in worse long-term pollution and in depletion of resources. These
indirect benefits may justify some risks to future generations from the
long-lived radioactive wastes generated by these activities.
We have not tried to resolve this question of intergenerational
equity by attempting to make risk versus benefit judgments. We believe
that the risks to future generations should be small—at the very least,
we should not pass on to future generations any radiation risks greater
than risks we would be willing to assume ourselves. We do not know of
disposal technologies foreseeable within this century that can eliminate
all risks to future generations from disposal of these wastes. However,
our analysis of the risks associated with undisturbed uranium ore bodies
has reinforced our decision about the reasonableness of the residual risks
permitted under our proposed disposal standards. This analysis indicates
that leaving equivalent amounts of uranium ore unmined presents at least
as great a risk to future generations as disposal of the wastes covered by
these standards. Thus, there would appear to be no increase in future
risks caused by disposal of these wastes.
38
-------
We are not sure that this analysis provides an adequate means of
resolving the question of intergenerational risk. It has, however, helped
to influence our decision of what is an acceptable level of residual risk
given our current scientific, technological, and fiscal capabilities. The
following discussion provides more details on the perspectives we used to
evaluate risks to future generations.
3.1.1. Perspectives on Long-Term Risks
People have always been exposed to radiation. Life has evolved in a
natural background radiation field that gives most humans doses on the
order of 100 millirem per year. Man has added more radiation—from
medical, industrial, and national defense activities—to this natural
background. We believe our standards should limit the risks from
high-level and transuranic wastes after disposal to a small fraction of
today's radiation risks. As a basis for comparison, we present estimates
of the risks from four radiation sources. Two of these sources, natural
background radiation and uranium ore bodies, are relatively constant, and
the risks will continue far into the future. The risks from the other two
sources, generation of nuclear power and nuclear weapons fallout, are
largely short-term and will decrease with time. The estimates of risks
from natural background radiation, generation of nuclear power, and
nuclear weapons fallout are much higher than those we estimate from
high-level wastes after disposal. The estimates of risks from uranium ore
bodies vary over a large range. Our estimates of the risks from the
disposal of high-level wastes are slightly above the lower bound of this
range and are substantially less than the higher part of the range.
39
-------
3.1.2. Risks Associated with Natural Background Radiation
Humans are exposed to natural background radiation from cosmic rays,
from naturally occurring radionuclides in the earth, and from naturally
occurring radionuclides in the body, primarily potassium-40. Cosmic
radiation originates primarily in galactic and solar processes. Naturally
occurring radionuclides in the earth's crust comprise chiefly members of
the uraniunr-238 series, members of the thorium-232 series, and
potassium-40. There are approximately 2.7 grams of uranium-238, 9.6 grams
of thorium-232, and 20,900 grams of potassium per metric ton of rock in
the earth's crust (FO 71). Rock in the earth's crust contains about
-5 -5
1.4x10 curies of alpha activity per metric ton and about 2.7x10
curies of beta activity.
The United Nations Scientific Committee on the Effects of Atomic
Radiation (UNSCEAR) has estimated that the average annual radiation dose
from external exposure to natural sources in "normal" areas of the world
includes 28 millirads of cosmic radiation and 32 millirads of terrestrial
radiation (UN 77). The UNSCEAR committee stated that the dose from
external and internal radiation was 92 millirads per year to red marrow,
78 to gonads, and 110 to lungs. The corresponding dose equivalent rates,
with a quality factor of 10 for the alpha and neutron components, are 109,
86, and 425 millirem per year (mrem/yr), respectively. The difference
among organs is due to differing radionuclide contents in the organs.
40
-------
People receive from about 50 to about 195 mrem/yr natural background
radiation dose equivalent from cosmic, terrestrial, and internal radiation.
The highest dose equivalents received in large urban areas are about
130 mrem/yr in Denver, Colorado, and about 115 mrem/yr in Albuquerque,
New Mexico. The highest average for a state is about 115 mrem/yr in
Colorado. Table 3-1 gives the distribution of the dose received by the
population. It shows tnat about 60 percent of the United States
population receives from 65 to 90 rarem/yr; 20 percent more than
90 mrem/yr; and 5 percent more than 105 mrem/yr. Natural background
radiation levels in the United States vary by a factor of three.
Information on the variability of the dose equivalent received by
people in the United States has been published by Oakley (OA 72) and
revised by Bogen (BO 80). Cosmic radiation dose equivalent in the United
States varies from 29 mrem/yr at sea level to about 125 rarem/yr at
3200 meters (10,500 feet), the altitude of Leadville, Colorado (NCRP 75).
The terrestrial component dose equivalent from radionuclides in the earth
can be characterized as low (average of 23 mrem/yr) for the Atlantic and
Gulf coastal plain, moderate (average of 46 mrem/yr) for most of the rest
of the United States, and high (average of 90 mrem/yr) for a small region
in the Colorado Front Range.
-------
Table 3-1
Distribution of Natural Radiation Dose Equivalents
(Terrestrial Plus Cosmic Plus Internal)
Dose equivalent* Percentage of U.S. Population
(mrera/yr) receiving more than stated dose
50 100
65 80
75 50
90 20
105 5
120 1
*Dose equivalent includes 25 millirem per year from internal radiation,
Values have been rounded to the nearest 5 tnrem/yr.
42
-------
We estimate that an excess of about 200 fatal cancers per million
person-rem can be expected (NAS 72). This means that an increase of
1 mrem/yr in natural background radiation (about 1 percent) to everyone in
the present U.S. population would result in about 40 excess fatal cancers
per year. A 1 mrem/yr increase in radiation exposure to 5 million people
(a population that might be exposed to releases from a large high-level
waste repository) would result in about one additional fatal cancer per
year.
Risks associated with natural background radiation are similar to
those from radioactive wastes; both produce the same kinds of health
effects and both persist for long periods of time. The risk from natural
background radiation is much larger than that projected for releases from
a high-level waste repository. However, in general, the variability of
this risk does not seem to greatly influence the choice of people in
selecting a place to live.
3.1.3 Risks from Uranium Ore Bodies
A comparison of the risks from a high-level waste repository with
those from an undisturbed uranium ore body offers another useful
perspective. In making our assessment (WI 80), we consider that uranium
dissolved by the groundwater eventually reaches a stream. As the uranium
moves along the aquifer, radiunr-226 and other daughters are formed. We
estimated the risk from a model generic ore body containing 620,000 tons
of uranium oxide, the amount that would have to be mined to produce the
high-level wastes contained in our model repository. For this estimate we
43
-------
used the same generic environmental model (SMJ 82) that we used for
assessing the risk from high-level radioactive wastes in mined geological
repositories (SMC 82).
The risks from uranium ore bodies cover a wide range, depending on
the ability of uranium and its decay products (particularly radium-226)
to leave the ore body and reach people. To estimate a minimum impact, we
selected a uranium concentration in groundwater at the low end of the
reported range (FI 55). For this assumption, we calculate an occurrence
of two excess fatal cancers per 100 years for the model generic ore body.
Based on pre-operational data for three actual uranium mines, we calculate
the impact of releases of radionuclides to groundwater to cause between
100 to about 1000 excess cancers per year. These estimates from actual
ore bodies may be high, because the reported concentrations in the
groundwater include some measurements from the oxidizing uranium-rich
groundwater. Our estimate of the risks from the wastes after disposal is,
therefore, similar to the lowest estimates of risk from the unmined ore
bodies.
3.1.4 Risks from Nuclear Power Generation
Ellett and Richardson (EL 77) estimated that exposure to the radiation
incident to the generation of 1 gigawatt-year of electrical energy could
result in an average of about 1.2 fatal cancers* in the first 100 years.
*The estimated number of fatal cancers represents a probability of
incidence in the entire exposed population.
44
-------
More recent information reduces this estimate to 0.8 fatal cancers.
Operation of the 50 gigawatts of nuclear power generation available in 1977
(ADL 79) at 60 percent of capacity could cause about 24 fatal cancers per
year. To the extent that radionuclides such as carbon-14, thoriura-230,
and radium-226 and its daughters persist beyond 100 years, this estimate
may be low.
Since they are limited in time, the risks from generating nuclear
electric power are not directly comparable to risks that would be
associated with disposal of high-level radioactive wastes, but they do
offer a perspective. Our estimate of the risk from the wastes for 10,000
years after disposal is less than the current population risks from
generating the electricity.
3.1.5 Risks from Nuclear Weapons Fallout
Fallout from atmospheric testing of nuclear weapons has added to the
radioactivity of air, water, food, and residual radionuclides (primarily
tritium, strontium-90, cesium-137, carbon-14, and plutonium) and is a
source of continual radiation exposure. UNSCEAR has estimated the dose
commitments to various organs of an average individual in the northern
temperate zone from nuclear weapons testing through 1975 (UN 77). The
UNSCEAR values are shown in Table 3-2. When converted to millirems, the
dose commitments are 150 millirems to gonads, 260 to bone marrow, 300 to
bone linings, and 270 to lung. Almost all the dose commitment is due to
relatively short-lived (30 years or less) nuclides, because doses for
carbon-14 were calculated only to the year 2000.
45
-------
Table 3-2
Dose Commitments from Fallout (UN 77)
External
Short-lived
137CS
Internal
14C (a)
90Sr
106Ru
137Cs
144Ce
239Pu
TOTAL ^
(Northern Temperate Zone)
tnillirad
Bone Bone
Gonads Marrow Lining
48 48 48
62 62 62
7 32 29
84 120
27 27 27
1
150 260 290
Lun
48
62
41
27
65
1
260
accumulated up to the year 2000
Including nuclides not listed
46
-------
We estimate the number of projected fatal cancers in the United
States population from fallout by multiplying the average dose committnents
in Table 3-2 by 200 million and by the appropriate risk factors. We have
assumed that the soft tissue dose is equal to the gonadal dose and have
reduced ttie whole body risk factor of 200 per million person-rem to 110,
to avoid counting the fatal cancers from the listed organs twice. The
expected harm from nuclear weapons testing in the United States is then
about 3300 fatal cancers from soft tissue irradiation, 2080 from bone
marrow irradiation, 600 from bone lining irradiation, and 2160 from lung
irradiation, for a total of 8140 fatal cancers over the life of the
nuclides involved. If this number is averaged over the half-lives of
cesiutn-137 and strontium-90, the largest contributors to the dose
commitment, the result is an average of about 190 fatal cancers per year
over about 43 years.
The risks from nuclear weapons fallout are primarily near-term risks
and are relatively uniformly distributed over the northern hemisphere.
Therefore, they are not directly comparable to the risks from the disposal
of waste. The present annual risk from fallout is about 2000 times
greater than the projected future annual risk from the wastes after
disposal.
47
-------
3.2 ASSESSMENT AND REDUCTION OF RISKS
An important element in the development of environmental standards is
an assessment of technology to control risks. In this discussion, we deal
with the quantitative assessment of risks, including the effect of
uncertainty, and consider some methods suggested for the disposal of
high-level and transuranic radioactive wastes.
3.2.1 Assessment of Risks from Waste Disposal
No high-level or transuranic radioactive wastes have ever been
disposed of permanently. Because there is no background of experience
that can be used as a basis for analysis, we must rely on modeling
procedures for an assessment of risk. To model the characteristics and
performance of a disposal system, mathematical equations must be developed
that predict potential releases of waste and describe their subsequent
transport through the environment. These models predict the consequences
of releases over long periods of time. The models can be no better than
our understanding of all the processes involved and, in fact, are often
simplified to facilitate the complex computations.
The mathematical equations describing the behavior of the system and
the movement of radionuclides through the environment to man are designed
to be general, and they require appropriate numerical data to produce
useful results. Therefore, there are uncertainties both in how well the
models describe the actual physical and chemical processes that may occur
and also in how good the numerical input data are in terms of producing
realistic results.
48
-------
When attempting to model the behavior of a generic system, we must
select numbers descriptive of system characteristics. It is customary to
make several model assessments, using several different sets of values for
the characteristics we consider important. This practice, which is
generally called sensitivity analysis, gives us information on the extent
to which each system characteristic influences the overall estimate of
risks.
3.2.2 Reduction of Risks by Disposal Technology
Although several methods of isolating high-level and transuranic
wastes are being developed, we have chosen to assess disposal in mined
geologic repositories. We believe that this is the only method for which
sufficient information is available to provide the basis for judgments
needed to develop disposal standards. Any other method for disposal of
high-level and transuranic wastes must result in risks no greater than
those we have judged would result from disposal in carefully selected and
well-engineered geologic repositories.
As part of the development of these proposed standards, EPA asked the
MITRE Corporation (AL 79a) to review the scientific and developmental
status of alternative disposal concepts. The Interagency Review Group on
Nuclear Waste Management, created by the President, and the Department of
Energy have also examined alternative disposal technologies (IRG 79),
(DOE 80). The Ford Foundation (NU 77) and the American Physical Society
(APS 78) have previously reviewed high-level waste disposal concepts, and
49
-------
several other groups, including an Ad Hoc Panel of Earth Scientists
commissioned by EPA (EPA 78a), nave analyzed aspects of the status of
basic science and technology relevant to deep geologic disposal. The
following paragraphs describe the status of waste disposal concepts, based
on information that is currently available.
Emplacement in Deep Mined Repositories
Radioactive wastes could be placed in underground repositories built
with conventional raining techniques. The primary natural barriers to
migration of the radionuclides would be the emplacement medium and the
surrounding geologic and hydrologic environment. Engineered barriers
would include a stable and insoluble waste form, corrosion-resistant
containers, and absorbent packing materials. Removal of the material from
the immediate proximity of people and other living things would reduce the
possibility of accidents during the several-hundred-year period during
which fission products dominate the radioactivity. Careful attention to
site selection and deep emplacement would make catastrophic release of the
waste by severe surface disturbances extremely unlikely.
Our assessments indicate that exploration for resources, in which
holes would be drilled into the repository, would be the most common mode
of inadvertent release of radionuclides to the biosphere. Drilling could
release radionuclides to both the land surface and groundwater. While
groundwater would be a pathway by which radionuclides from the waste can
enter the biosphere, it would also be a protective barrier. Groundwater
50
-------
velocities are slow enough, in general, that many of the radionuclides in
the waste would decay before reaching the surface. Furthermore,
adsorption of many radionuclides by the media through which the flow
occurs can slow their movement even more and can provide considerable
added protection against exposure to radionuclides with moderately long
half-lives.
Other Methods of Deep Continental Disposal
Rock Melt. Fresh high-level waste might be emplaced into continental
rock formations. The heat generated by the radioactivity of the waste
would melt the surrounding rock. When the rock cooled, the wastes would
be incorporated in a rock mass that would be impermeable and in good
chemical equilibrium with its environment.
Rock melting disposal methods may offer the same protection against
catastrophic releases by natural forces and human intrusion as deep-mined
geologic repositories. There would, however, be considerable
volatilization of radionuclides during the melting period, after which
retention of the radionuclides would depend on the extent that they were
bonded to the rock during melting. There has been no evaluation of the
extent to which radionuclides would be immobilized in this disposal
concept, and, therefore, we cannot make an assessment of risk.
51
-------
Deep Holes. Containers of waste could be placed in the bottoms of
holes drilled as deep as 50,000 feet. At such depths, human intrusion
would be unlikely. The emplacement of wastes to this depth is not
feasible now. Present drilling methods are believed to be capable of
drilling a hole 20 centimeters (about 8 inches) in diameter to a depth of
about 11,000 meters (35,000 feet) or a hole 31 centimeters (about 12-inch)
to a depth of 9.1 kilometers (about 30,000 feet). The ability to seal
such a hole has not yet been demonstrated.
Risks from disposal in deep holes would depend on the ability of the
rocks around the hole to dissipate the heat and on the absence of
groundwater movement in any specific location. Information for such risk
assessment is not now available.
Emplacement in Deep Ocean Sediments
Containers of waste or spent fuel could be placed tens of meters or
more below the ocean floor in stable regions where the sediments are thick
and uniform. The wastes would be far from human activity, and the
sediments would be expected to prevent the waste nuclides from reaching
the ocean water.
Lack of knowledge of the stability and adsorptive properties of the
sediment barrier when wastes are emplaced in it makes it impossible to
assess the risks of this method of disposal. Disposal in the ocean is
52
-------
prohibited under United States law by the Marine Protection, Research and
Sanctuaries Act of 1972. Legislative action would be required before
seabed disposal could be implemented.
Extraterrestrial Disposal
High-level and transuranic radioactive wastes could be launched into
space. This method of disposal would potentially eliminate future risk to
living things. A solar orbit appears to be the most technically and
economically achievable option and would provide long-term isolation.
This method would not provide for injection of all the waste into space,
because costs would be too high. Present concepts would require chemical
separation of the wastes and would send only the long-lived radionuclides
into space .
Disposal in space would permanently isolate radioactive nuclides from
humans. However, methods for assuring the protection of public health in
the event of launch failures have not yet been developed, and separation
of wastes introduces added risks. The overall costs of this method of
disposal are high.
Transmutation of Waste Radionuclides
Radionuclides in waste could be transformed to less hazardous
nuclides by irradiation in nuclear reactors or in particle accelerators.
Long-lived and highly radiotoxic actinide radionuclides, including
neptunium-237, plutonium-239 and -240, and araericium-241 and -243, are the
53
-------
most important candidates for transmutation. However, other long-lived
nuclides, such as iodine-129 and technetiura-99, have also been
considered. Materials to be transmuted would have to be chemically
separated from the irradiated fuel. High-efficiency actinide separation
has not yet been developed.
Transmutation of nuclides would eliminate any possibility that they
would ever affect people. However, other risks of transmutation have not
been fully evaluated and may be high. Separation and recycling of the
actinides and other radionuclides could increase occupational,
transportation, and routine release risks.
3.3 DEALING WITH UNCERTAINTIES
In considering the possible harm that might be incurred in the future
by people as a result of releases from a disposal system, we must allow
for uncertainties in our expectations of the future. In our development
of this proposed action, we did this in two ways: by using conservative
values in our assessments and by including the qualitative assurance
requirements in our proposed standards. In the following paragraphs, we
will discuss the issues of uncertainty in disposal system performance and
of future human behavior.
3.3.1 Technical Uncertainties
Disposal systems limit potential harm to people from the disposal of
high-level and transuranic radioactive wastes by placing barriers between
the wastes and the human environment. Judgments of the acceptability of
54
-------
the system depend on estimates of the effectiveness of these barriers. We
can make such estimates by considering the performance expected from each
barrier, using conservative values, and combining these into an assessment
of the performance of the entire system. We allowed for uncertainties in
these estimates by selecting values for the effectiveness of each barrier
that we are confident are well within the capabilities of engineering and
site characterization. The actual effectiveness of each barrier will
probably be considerably better than our selected value. The entire
system should perform much better than our assessment of its performance,
which we obtain by combining our values for the effectiveness of all the
barriers.
We also allowed for uncertainty qualitatively by providing for
redundant independent barriers. Each barrier would function whether or
not other barriers perform in the manner we expect. We believe such
redundancy and independence is particularly appropriate for a system that
must operate over long periods of time without human control or
intervention. Accordingly, we have adopted both the conservative system
analysis, which we used to select limits on the quantities of
radionuclides projected to be released into the accessible environment,
and the qualitative concept of redundant independent barriers included in
our assurance requirements.
We could also have required that the disposal system meet our
containment requirements even if some of their barriers are assumed to
fail. Such a requirement might be stated as: (L) meet the numerical
55
-------
requirements even if any one of the barriers fails, or (2) meet the
requirements if only one of the barriers works. A repository built to
comply with such requirements probably would be much more expensive than
other designs we have studied, and much of the necessary technology does
not appear to exist yet. Although some additional protection might be
provided by such an approach, we do not believe this would justify the
extra costs, difficult implementation, and long delays that would result.
We must also consider the possibility of presently unknown processes
and phenomena that could produce much greater risks than we project from
the system. Development of technical and scientific knowledge has often
revealed such unexpected processes or phenomena. Our knowledge of
ionizing radiation itself is less than a century old. In geology, the
theory of tectonic plate movement is quite recent. Therefore, to allow
for such possible major technical uncertainties, we believe that we have
an obligation to leave future generations the option of removing the
wastes from a disposal system found to be excessively hazardous.
Accordingly, we are proposing an assurance requirement that requires
disposal systems to be designed so that recovery of most of the wastes is
not precluded for a reasonable period after disposal.
Some disposal methods, such as deep-hole disposal and rock-melting
concepts, may not be able to comply with this assurance requirement.
If we did not include this provision, it might be possible to develop
disposal methods that would have smaller projected risks than those we
56
-------
project for mined geologic repositories. However, if we were to dispose
of wastes in a way that could not be reversed, then future generations
would not have the option to take advantage of new information or better
technologies. We believe that mined geologic disposal methods, from which
wastes can be recovered, can keep the projected risks to future
generations small, and we judge that any additional protection is less
important than retaining future options to alter the disposal method, if
necessary.
3.3.2 Uncertainties in Human Behavior
Human behavior in the future may affect the risks from disposal
facilities in two ways. First, human actions may change the ability of
the system to isolate the wastes (for example, if a person drills through
a geologic repository while exploring for some mineral resource, he may
provide a pathway through which radionuclides can reach people). Second,
human actions may affect the likelihood of human exposure to radionuclides
that have escaped from isolation (for example, exposure to radionuclides
in contaminated groundwater occurs only as a result of using the water).
Institutional Controls
Institutional controls are intended to assure the integrity of a
waste isolation repository and to eliminate the possibility of human
damage to the isolation capacity. The pertinent issue regarding
institutional controls is how long we should rely on them in our planning
and design of disposal facilities. Different attitudes toward
57
-------
institutional controls can lead to very different strategies for
protecting the environment. This issue was discussed extensively during
the development and review of the Federal radiation protection guidance
for disposal of all types of radioactive wastes that we proposed on
November 15, 1978. Institutional control was defined as "activities,
devices, and combinations thereof which involve the performance of
functions by human beings to limit contact between the waste and the
humans or the environment." The proposed criteria stated ". . . Controls
which are based on institutional functions should not be relied upon for
longer than 100 years . . .".
Public comments with specific recommendations about how long we
should rely on institutional controls were divided approximately evenly
among four positions:
1. That institutional controls should be relied upon for only about
20 to 30 years.
2. That the 100-year period was about right.
3. That the 100-year period should be extended to 500 to 1000 years,
or even longer.
4. That we should limit reliance on controls, but let the
regulatory agencies select the appropriate time period.
On balance, a few more commenters felt that the 100-year period was
too short compared to those who felt it was too long. Several other
comments made a distinction between "active" controls, such as restricting
58
-------
access to disposal sites, and "passive" controls, such as warning future
generations about what we have done by creating extensive records and
markers.
In the proposed assurance requirements that are part of these
standards, we have decided to limit reliance on "active" controls—such
as guarding a disposal site, performing maintenance operations, or
controlling or cleaning up any releases from a site—to a "reasonable"
period of time after disposal, which we believe should be no more than a
few hundred years. However, because the Federal Government is committed
to retaining control over these disposal sites in perpetuity, we expect
that "passive" institutional measures should substantially reduce the
chance of inadvertent human intrusion well beyond this period. Such
passive controls will include permanent markers placed at a disposal site,
public records or archives, Federal ownership or control of land use,
and other methods of preserving knowledge about the disposal system.
These passive controls should not be assumed to prevent all possibilities
of inadvertent intrusion, because there is always a chance that the
controls will oe overlooked or misunderstood. However, such measures
should be effective in deterring systematic or persistent exploitation of
a disposal site.
As one alternative to this proposal, we could have assumed that
institutional controls such as those we now use would be effective for a
very long time (thousands of years). Inadvertant human intrusion would
59
-------
not be a concern, and few areas would have to be avoided because of
resource potential. This would permit use of monitored, engineered
storage systems—carefully designed to reduce the chances of accidental
releases—which would have a number of advantages over disposal systems.
In particular, failures of the waste containment could be quickly detected
and corrected before significant contamination of the accessible
environment occurred.
On the other hand, we could have assumed that society could lose all
knowledge of the disposal system and the dangers of its contents in a
relatively short period of time. Thus, not only could people intrude into
the system, they might not respond to any unusual conditions they
encountered and might continue to spread the wastes throughout the
environment for a long time. In this case, the first priority for a
disposal method would be to reduce the chance of inadvertant intrusion as
much as possible, even if this resulted in higher risks from other
events. This would strongly affect the relative attractiveness of various
disposal methods. Techniques such as deep-hole placement, which removed
the wastes far from man's typical activities, would be favored. We also
might have to delete our assurance requirement that the wastes be
recoverable, to avoid ruling out many of these disposal techniques.
When there are no active institutional controls, the likelihood of
inadvertant human intrusion into the disposal area may be reduced if the
existence of the disposal site is known. The waste disposal system can
60
-------
provide such information in the form of archival records and permanent
markers at the site. In the public workshops leading up to our
November 15, 1978, proposal of Federal radiation protection guidance for
radioactive waste disposal, there was general acceptance of the concept of
passive communication of the disposal system (EPA 78b). Over long periods
of time, any particular marker or record may be destroyed or lost, but
wide distribution of duplicate records would greatly reduce the likelihood
of all knowledge of the disposal location being lost. Accordingly, we
have included an assurance requirement for markers and records.
Resource Potential
We must consider the possibility of human activities that would
degrade the isolation capability of a waste disposal system in the absence
of active institutional controls or of knowledge of the disposal system's
existence. Human intrusion is more likely if the disposal site can be
expected to yield valuable mineral, energy, or other resources. Locations
that are now known to contain actual or potential resources, such as coal
or oil shale, must be avoided.
We must be cautious in our judgments as to what would be attractive
to future prospectors, because we cannot predict what materials will be
valuable to people in the future. A few centuries ago oil, uranium ores,
and aluminum ores were not significant resources. In our assurance
requirements, therefore, we state that wastes should not be placed in an
area where there is a reasonable expectation of future mining or
61
-------
exploration. To allow for our uncertainty about what materials may be
considered valuable in the future, we extend this requirement to sites
containing a substantial concentration of any material that is not widely
available.
3.4 CHOICE OF FORMAT FOR DISPOSAL STANDARD
In developing the proposed standards, we considered a number of
possible ways to limit risks to individuals or population groups. We also
considered a number of possible time periods over which we should
regulate. This section and the following one evaluate these alternative
approaches to the disposal standards.
Alternative 1; Standards expressed in numbers of health effects
projected over some period of time. This form most directly addresses the
population risks and would be preferred, all other things being equal.
However, regulations in this form require assumptions on the exact
relationship between dose and health effects and would also require
knowledge of future population and demography at a site, an obvious
impossibility. We therefore rejected Alternative 1.
Alternative 2; Standards expressed in terms of population dose
equivalents (person-rems) over some period of time. This form requires no
assumptions as to the nature of the dose response (health effects per
person-rem) but is otherwise similar to Alternative 1. It would require
the same knowledge of future population and demography as Alternative 1,
and therefore we rejected it also.
62
-------
Alternative 3: Standards expressed in dose-equivalent to individuals
over a lifetime. This form would provide an estimate of the total risk to
an individual. However, this alternative might not assure adequate
protection to the total population, and it appears very difficult to
implement. We therefore rejected this alternative.
Alternative 4; Standards expressed in terms of dose equivalent rates
(rem per year) to individuals. This form measures risk in terms of annual
increments of exposure and is more useful in regulating short-term hazards
than for controlling long-lived radionuclides. However, this alternative
(or equivalent standards which limit radionuclide concentrations in air or
water) is a traditional form of radiation protection standard. Therefore,
although we rejected this option, it is further examined in Chapter 4 as
one of the major alternatives to our proposed action.
Alternative 5; Standards expressed in terms of total integrated
radionuclide releases (curies) to the accessible environment over a long
period of time. This choice would serve as a surrogate for Alternatives 1
and 2 in estimating population risk. Standards in this form would be
easier to implement, since they only require estimation of the probable
quantities ot radionuclides leaving the disposal site and their physical
transport to a location accessible to man. For these reasons, we selected
this form for the standards.
63
-------
Alternative 6; Standards expressed in terms of radionuclide release
rates to the accessible environment (curies per year). Our investigations
(SMC 82) snowed that radionuclide release rates were less closely related
to population risk than total radionuclide releases. We rejected
Alternative 6 for this reason.
Alternative 7; Standards expressed in concentration of radionuclides
(curies per cubic meter) in environmental media. Concentrations of
radionuclides in specific environmental media are less closely related to
population risk than total curie releases, and standards in this form
could encourage siting where there are large volumes of diluting water.
Although this could reduce doses to exposed individuals, the ultimate
population dose would be the same. The calculations required for
implementation of this standard would be much more difficult, because they
would require detailed analysis of environmental pathways. We therefore
rejected this alternative, but—as discussed for Alternative 4 above—this
overall approach is considered as one of the major alternatives to our
proposal in Chapter 4.
3.5 ALTERNATIVE REGULATORY TIME PERIODS
The goal of our disposal standards is to provide protection as long
as the wastes present an unacceptable risk. The standards require a
reasonable expectation that releases over a specified time will be less
than the specified limits. Stating a period for regulation in the
standard does not imply that the radioactive wastes will be released after
64
-------
that time; it establishes the period over which the releases must be
assessed. We considered several alternative time periods for assessment:
100 years; 1000 years; 10,000 years; 1,000,000 years; and forever. We
could also have based our assessment on the risk to any future generation.
Alternative 1; 100 years. The Uranium Fuel Cycle (UFC) regulations
(40 CFR 190) are based partially on 100-year projections of risk. In
those regulations, we limited the projections to 100 years because of
uncertainties in population, demography, and use of the environment. We
do not need precise estimates of risk for development of these proposed
high-level and transuranic radioactive waste standards, because balancing
these risks and costs is not a major basis for the standards. Our risk
assessments showed that few health effects are predicted in the first
hundred years, mainly because of long groundwater transport times and the
low probability of disrupting events in such a short time period.
Therefore, an analysis of consequences limited to 100 years after disposal
would give an unrealistically low estimate of the total potential
impacts. We rejected this alternative for these reasons.
Alternative 2: 1000 years. Some have suggested that most risks
occur in the first 1000 years, because strontium-90 and cesium-137 decay
in that time. The same considerations that argue against the 100-year
period apply to the 1000-year period. Few health effects occur during
this period; many nuclides would still be in transit towards the
65
-------
accessible environment; disruptive event probability is still low; and a
considerable number of the important nuclides would not have undergone
significant radioactive decay. We therefore rejected Alternative 2.
Alternative 3: 1,000,000 years. A period of 1,000,000 years has
been advocated on the grounds that the hazard from high-level radioactive
wastes is not adequately reduced by radioactive decay until then.
However, it would be difficult to make quantitative assessments of the
potential environmental impacts of waste disposal over such a long time,
and prediction of geological changes over such a long period is not
reliable. These reasons make Alternative 3 unacceptable.
Alternative 4; Forever. The arguments against selection of the
1,000,000-year period apply even more strongly to regulation over all
time. This option, too, was rejected.
Alternative 5; 10,000 years. We have selected 10,000 years as a
regulatory period. In 10,000 years, many of the radionuclides that can
pass through groundwater transport paths will have reached surface water
and the accessible environment. At 10,000 years, the radiological hazard
of the wastes would be substantially reduced through the decay of most of
the significant fission products and of many of the actinides. The only
remaining nuclides presenting a potentially significant health risk would
be technetiura-99 and iodine-129 and the actinides plutonium-239 (about
75 percent of the original inventory), plutonium-240, and neptunium-237.
66
-------
This inventory will decrease fairly rapidly after 10,000 years by decay of
plutoniura-239 and -240; it will then decrease slowly because of the long
half-lives of technetiutn-99, iodine-129, and neptunium-237. The
radionuclide hazard potential at 10,000 years is, therefore, fairly
descriptive of the situation for a long time thereafter. Population and
demographic conditions 10,000 years from now are unknown and unknowable,
but they do not have a controlling effect on setting the standards. Since
10,000 years is a short time geologically, changes in geological
conditions are expected to be small. Only massive changes in climate,
such as could conceivably be induced by man-made additions to the
atmosphere (e.g., the carbon dioxide "greenhouse" effect ), could produce
large changes in the transport of radionuclides. The adverse effect of
these climate changes would be so great that any additional impact of the
radioactive wastes would be relatively insignificant.
Alternative 6; A reasonable objective for waste disposal is that no
future generation should incur more than a fraction of the risks that
would be acceptable to the current generation. We have considered this
principle in formulating these standards., However, we also recognize that
it could be difficult to implement as stated. The risks to each
generation for a long period of time would have to be examined to find out
if they were below the appropriate level. Our evaluations of the
environmental effects of disposal, described in Chapter 5, indicate that
total risks over 10,000 years can be made small. Limiting total risks to
this level ensures that no single generation within this period can incur
67
-------
greater risks. These risks do not appear unreasonable even if incurred by
only one generation. Therefore, we believe it is simpler and adequate to
restrict risks over the total period of 10,000 years rather than provide a
limit on risks to a single generation.
68
-------
Chapter 4
ALTERNATIVES
In developing our proposed environmental standards, we considered a
number of alternatives. This chapter examines those alternatives that
would involve major changes to our proposed action. Additional
alternatives that would affect specific details of the standards are
discussed in other chapters of this Draft Environmental Impact Statement:
Chapter 3 examines some different approaches we could take to compensate
for the uncertainties of the future and also considers different time
periods and alternative units for the disposal standard, and Chapter 7
considers alternative dose limits for the waste management standards.
Each of our discussions of alternatives indicates why we think our
proposed action is preferable to the other options we considered.
However, in each case we would like to receive public comments that
evaluate whether we made appropriate choices.
Our consideration of alternatives is influenced by the limits on our
information. High-level radioactive wastes have never, to our knowledge,
been permanently disposed of. Furthermore, since we believe these
disposal systems should isolate the wastes for many thousands of years,
the relative benefits of different systems can be judged only by
predictions of their long-term performance. With this pervasive
limitation in mind, the following are the major alternatives to our
proposed action.
69
-------
4.1 ISSUE NO STANDARD
The decision to prepare these standards was an administrative action
taken by EPA and was not mandated by law. We were directed to prepare
standards as part of President Ford's Nuclear Waste Management Plan on
October 27, 1976. President Carter established an Interagency Review
Group (IRG) on Waste Management in March 1978 to review existing programs
and recommend new policies where necessary; the IRG recommended that EPA
set standards for nuclear waste management and disposal activities and
accelerate its programs to do so. President Carter approved this
recommendation as part of his Program on Radioactive Waste Management
announced on February 12, 1980. The NRC best described (NRC 80) the
expected goal of these standards:
". . . (EPA) standards represent a broad social consensus
concerning the amount of radioactive materials and levels of
radioactivity in the general environment that are compatible
with protection of the health and safety of the public."
Although satisfactory management and disposal of high-level
radioactive wastes could be attained by other regulatory means (e.g., NRC
licensing requirements), such means would not likely receive broad public
acceptance. By promulgating standards, EPA directly focuses on
environmental protection requirements without the constraints of many of
the complex engineering requirements necessary for consideration in
licensing. Also, because of its broad environmental charter, EPA is in
the best position to assure broader input into the standards by groups
outside nuclear interests—e.g., those charged with protecting
groundwater. Thus, not issuing standards is an unacceptable option.
70
-------
4.2 DELAY ACTION
We could have chosen to delay issuing standards until we developed
more information about disposal systems. This delay could allow us to
better evaluate how well disposal systems can protect the environment and
might lessen the chance of poor judgments of how much protection is
necessary and reasonably achievable. We might then be able to develop
information on disposal systems other than geologic repositories and to
develop quantitative comparisons of the costs and benefits of these
alternatives. However, we do not expect enough information to become
available during the next decade to allow more comprehensive evaluations.
We choose to propose these standards now, because we believe that
environmental standards must be developed to guide the progress of the
national waste disposal program. DOE has decided to focus the national
program on mined geologic repositories (46 FR 26677), after evaluating
alternatives through a comprehensive environmental statement process
(DOE 80). Ttie DOE program anticipates identification and characterization
of three sites at which to begin exploratory shaft construction in 1983.
The overall goal is to provide the first licensed, fully operational
repository within the period 1997 to 2006 (NE 81).
Delay in proposing these standards could diminish public confidence
in the national waste management program. Loss of confidence might lead
to action by more state governments to restrict exploratory and research
activities. This could reduce the options available to the Federal
71
-------
government in exploring for repository sites. We believe that proposing
standards now will enable the national radioactive waste disposal program
to proceed on an orderly schedule. This is consistent with the synopsis
of the public comment received by the IRG (IRG 79):
"Comment from both the industrial sector and the
environmental community urged the acceleration of EPA standards
particularly to instill confidence that proper protection of
the public's health and safety is being provided. They
expressed the concern that early standards are essential to
permit the waste management program to proceed expeditiously."
In addition, proposing these standards now will reduce the chances of
harm from long-term storage of existing wastes in surface facilities.
Although such storage offers no significant danger under normal
conditions, the wastes are more vulnerable to accidental release. For
example, several leaks have occurred from the high-level waste storage
tanks at Hanford (ERDA 75). These leaks have not caused any exposures of
the general population, nor have they contaminated any areas beyond the
site boundary. However, they do represent unplanned releases that could
have had serious consequences. We believe that our standards will
adequately protect public health and expect that disposal technologies
satisfying these standards will be available within the next decade.
4.3 ESTABLISH ONLY QUALITATIVE REQUIREMENTS
The large uncertainties inherent in predicting the performance of
disposal systems over thousands of years present an argument for issuing
environmental protection requirements that contain only qualitative
criteria—such as our proposed assurance requirements. Application of
72
-------
such criteria would be easier than implementing our proposed standards:
regulation on the basis of qualitative criteria would not require any
projection of releases over time, and the validity of models for
predicting those releases would not come into question. Instead,
selection and licensing of disposal methods could rely more upon the
informed judgments of technical experts (subject to appropriate review)
than upon uncertain numerical estimates. Issuing only qualitative
requirements would still provide instructions that could effectively
reduce the dangers of disposal, and such criteria could be used to compare
alternative disposal systems and to select among them. Since such
qualitative requirements—by themselves—would not be appropriate
generally applicable environmental standards, we would issue them under
our Federal radiation protection guidance authority.
Our quantitative environmental protection requirements provide at
least two important benefits, however:
1. By selecting the form of these protection requirements as
numerical release limits over a particular time, we identify those
overall objectives that should be considered in disposal system
design. (One example: setting stringent release limits rather than
individual exposure limits encourages very good containment rather
than suggesting dilution as part of a disposal strategy. Another
example: a disposal system designed to limit releases for 1000 years
could rely primarily on engineered barriers, whereas a system
73
-------
designed to retain wastes for 10,000 years also would require both
good geological and hydrological characteristics at the disposal
site.)
2. Setting quantitative standards requires a full assessment of
system performance to assure that these standards will not be
exceeded—and this requirement for a comprehensive examination of the
potential harm can provide confidence that disposal systems will
behave as expected and will not create more than a small risk to
public health.
4.4 DELETE OR DEEMPHASIZE THE QUALITATIVE ASSURANCE REQUIREMENTS
In contrast to the previous alternative, this approach would increase
the reliance on our quantitative containment requirements. It would
provide greater flexibility for selection and design of disposal systems—
since the emphasis of our action would be on achieving specific
environmental protection objectives without restrictions on how those
objectives should be met. (For example: system designers might choose to
use a single, extremely effective barrier, or they might decide to depend
on indefinite control of the site to meet our requirements.) As part of
this alternative approach, the NRC could be expected to provide any
appropriate procedural, site selection, or system design requirements that
might be needed beyond our quantitative standards.
74
-------
There are two variations of this approach that we considered:
— to simply delete the qualitative assurance requirements, leaving only
the containment requirements and the procedural requirements needed
to assess compliance, or
— to issue the assurance requirements as Federal radiation protection
guidance, rather than as part of the generally applicable
environmental standards.
The second variation would make the qualitative criteria less binding
since—as executive directives rather than formal rules—they could be
waived if specific situations warranted. Such waivers would be considered
in accordance with the procedures established under Executive Order 12088.
The concept of allowing maximum flexibility to achieve specific
environmental protection goals (saying "what to do", not "how to do it")
can be very effective in many situations. However, we do not believe we
should rely upon it in this case because of the unusually large
uncertainties inherent in designing disposal systems to isolate wastes for
at least 10,000 years. Instead, we think the cautious steps called for by
our assurance requirements should be an integral part of our standards in
order to develop appropriate confidence that our containment requirements
will be complied with.
Altnough the specific design requirements that will be developed by
the NRG could meet the objectives of our assurance requirements, we
believe that our standards should be expected to establish the complete
75
-------
framework of quantitative and qualitative environmental protection
requirements that are needed. Furthermore, one category of wastes covered
by these standards—transuranic wastes from national defense
activities—is not subject to NRG regulation.
4.5 SELECT CONTAINMENT REQUIREMENTS ON A DIFFERENT BASIS
We have proposed containment requirements that we believe are
achievable by well chosen and carefully designed geologic repositories and
that would limit harm to future generations to low levels. We could have
used different bases for selecting these release limits. In the following
paragraphs we examine some alternative approaches for picking the
containment requirements.
— Develop Projected Release Limits Considering Different Disposal
Methods
An analysis of the potential performance of several disposal methods
besides mined geologic repositories might allow us to set more restrictive
limits. These alternative disposal methods are described in Chapter 3.
However, we cannot do the necessary analyses now or in the near future,
because the needed information for any other disposal method will not be
available for a long time. Therefore, since this alternative would
significantly delay these standards, it is essentially the same as the
"Delay Action" alternative discussed previously.
76
-------
— Limit Risks to Future Generations from the Entire Nuclear Fuel Cycle,
including Waste Disposal, to Those That Would Exist from the Original
Unmined Uranium Ore
The Natural Resources Defense Council (NRDC) (CO 79) has proposed
this criterion for radioactive waste disposal. The underlying principle is
that nuclear power should cause no increase in risk to future
generations. This alternative would involve consideration of all
long-lived wastes (not limited to high-level wastes) and effluents
associated with the nuclear fuel cycle. It would also raise the question
of comparing risks to future generations from nuclear power with those
from any other sources of energy from which some residual risk is passed
on to future generations. We believe this is beyond the scope of this
standards-setting effort and, therefore, have not directly used this
criterion in developing these standards.
— Limit Risks to Future Generations from High-Level Waste Disposal to
Thoae That Would Exist from the Original Unmined Uranium Ore
This alternative is related to the criterion proposed by the NRDC.
However, it is less restrictive since it considers only one part of the
future risks from the nuclear fuel cycle. The risks from ore bodies vary
over a wide range, depending on local conditions such as solubility of
uranium in the local water, hydrological factors, and transport of radon
through environmental media. There are also many uncertainties in our
ability to assess the risks from a particular ore body, including
uncertainties in sampling, in measuring the characteristics of
77
-------
environmental pathways, and in demography. We estimated risks from
unmined uranium ore ranging from somewhat less than, to one-thousand times
greater than, the risks from disposal of the high-level wastes derived
from the ore (WI 80).
If we used this concept to select containment requirements, the large
uncertainties would make it difficult to decide where to set the limits.
We believe the risks from disposal in systems complying with our proposed
release limits will be no greater than the risks from most equivalent
amounts of unmined uranium ore, and most likely will be less. Therefore,
we did not choose this alternative, since it would not provide more
protection and would be a very uncertain basis for regulation.
4.6 SET HIGHER OR LOWER RELEASE LIMITS
For the release limits we have proposed as our containment
requirements, the residual risks projected by our generalized
environmental pathway models would be less than 1,000 premature deaths
from cancer over the 10,000 year period, an average of one premature death
every 10 years. To judge the effects on disposal costs of changing this
level of protection, we also compared release limits corresponding to
residual risk values of: 100, 1000, 5000, and 10,000 premature deaths over
the 10,000 year period. We chose this range of stringency levels because
it appears to represent the range of performance that may be expected of
mined geologic repositories.
78
-------
To do this assessment, we evaluated the long-term performance of
generic models of geologic repositories in three different geologic media:
bedded salt, granite, and basalt. We did the assessment in two steps.
First, we used the performance projections decribed in Chapter 5 to assess
the quality of the engineering controls that would be needed in each of
the three model repositories to meet each of the four different levels of
protection. Second, we tried to allow for the possible effect of
alternative stringency levels on site selection. This is particularly
relevant because our analyses indicate that the most important part of the
protection offered by a mined geologic repository comes from the
hydrological and geochemical characteristics of the site itself. These
calculations are further described in Chapter 9 and in our Draft
Regulatory Impact Analysis (EPA 82).
The results of these assessments of disposal costs and alternative
stringency levels indicate that the costs are not very sensitive to
different levels of protection, particularly for the geologic media
(bedded salt and granite) that appear to be better at reducing long-term
risks. The differences in costs for different levels are much smaller
than the overall uncertainties in waste management costs. For example,
consider the increased costs of complying with the release limits we have
proposed, rather than release limits 10 times less stringent. The
potential increase ranges from zero to 50 million (1981) dollars per
year. For comparison, the total costs of high-level waste management and
disposal (independent of our action) have been estimated as between
79
-------
700 million and almost 1.5 billion (1981) dollars per year. Electrical
utility revenues were about 100 billion dollars in 1980. The possible
impacts of this potential increase in disposal costs are estimated to be
less than a 0.2-percent increase in the costs of nuclear power and less
than a 0.06-percent increase in average electricity rates (EPA 82).
These analyses, while indicating that disposal costs appear to be
relatively insensitive to differences in the level of protection, do not
provide a way to determine the acceptability of the residual risks from a
societal perspective, nor do they indicate a level of protection that is
preferable from a balancing o£ costs and benefits. One possible approach
to balancing costs and benefits would be to judge the cost per life saved
by different levels of protection, perhaps taking into account some method
of discounting costs and benefits. However, our calculations of residual
risks are not reliable as absolute values. Thus, we have no meaningful
way to calculate an absolute value of the cost per life saved by different
levels of protection.
In the absence of the ability to make meaningful cost and benefit
comparisons, we have used other tests of economic feasibility and
acceptability of risk to judge the appropriateness of the level of
protection we have proposed. As discussed above, setting the release
limits at the level we chose—as opposed to a level 10 times less or
10 time's more stringent—appears to cause only very minor effects on the
costs of high-level waste disposal. This is why we did not choose higher
(less protective) release limits.
80
-------
To judge the acceptability of the remaining long-term risk, we
considered the risks that would otherwise be caused if the uranium ore
used to produce the wastes had not been mined. As described in the
previous section, the magnitude of the risks from these unmined ore bodies
is very uncertain due, in part, to the wide variety of settings in which
uranium ore is found—many of which are closer to the surface than a
geologic repository would be. Using the same generalized environmental
pathway models that were used to assess the risks from our models of
geologic repositories, the risks from a comparable amount of unmined
uranium ore are estimated to range from a few hundred to more than
1 million health effects over 10,000 years. The lower end of this range
is roughly equal to the residual risk associated with our proposed release
limits. Thus, the upper limit of the risk that our standards would allow
from the disposal of high-level wastes appears to pose a threat very close
to the minimal risk posed by nature had the uranium ore never been rained
and the high-level wastes never been generated. This is why we did not
choose lower (more protective) release limits.
4.7 SET DISPOSAL STANDARDS IN TERMS OF LIMITS ON MAXIMUM INDIVIDUAL
EXPOSURE
Individual exposure limits, or equivalent standards that limit
radionuclide concentrations in air or water, are a traditional form of
radiation protection standard. Particularly when the limits are
comparable to or less than natural background levels, they may be more
effective than our proposed standards at communicating how small the
81
-------
chance of harm from disposal of these wastes should be. However, we chose
not to use individual exposure limits in Subpart B because of two unique
aspects of the situation regarding our disposal standards:
First, these disposal systems have to protect the environment from
these highly concentrated radioactive wastes for much longer than
institutional controls can be guaranteed to be effective. Any individual
exposure limit we set could apply only at some distance from a repository,
or it would have to ignore the risks from unplanned events such as
inadvertent intrusion—because individuals who fail to understand passive
warnings and penetrate directly into or close to a disposal system
(through exploratory drilling for water or mineral resources, for example)
could receive very large exposures. These exposures would probably exceed
any reasonable individual exposure standard.
Second, the disposal standards have to be applied through analytical
performance projections—implementing such standards through environmental
monitoring and potential remedial actions over thousands of years is not a
credible approach. When we compared the analyses needed for compliance
with exposure limits to the analyses needed for compliance with release
limits, we found that our proposed disposal standards would be much easier
to implement than exposure limits. The NRC, which is responsible for
applying our standards for high-level waste disposal, made a similar
evaluation and also found that standards based on radioactivity release
limits could be implemented more readily than standards based on exposures
to individuals.
82
-------
The proposed standards, although they do not directly limit doses to
individuals, place requirements on disposal systems that make it unlikely
that many individuals will be unduly exposed. The assurance requirements
are intended to reduce the chance that a person will intrude into the
disposal system. Furthermore, our analyses show that the containment
requirements will limit the area of the environment that can become
contaminated and, thus, reduce the probability of significantly exposing
people who might inadvertantly disturb the lithosphere near the disposal
site.
From these considerations, we believe that our proposed action will
better facilitate licensing of good disposal systems while providing
appropriate environmental protection from the long-term "risks presented by
these wastes. Therefore, we did not choose this alternative.
4.8 SET DIFFERENT LIMITS FOR RELEASES DUE TO NATURAL CAUSES AND FOR THOSE
CAUSED BY PEOPLE
We could have set standards containing two sets of containment
requirements: cumulative limits on releases due to natural processes and
events and limits on releases from each individual event caused by human
actions. The separate containment requirements would then better reflect
the greater uncertainties involved in predicting human behavior.
For releases caused by natural processes and events, we would use
numerical limits similar to those in the proposed standards. The
probability of these releases would still have to be evaluated to
S3
-------
determine the limits on total releases from all such events over 10,000
years. Based upon the analyses described in Chapter 5, we would make the
release limits for "reasonably foreseeable" natural processes and events
10 times smaller than our proposed limits (which apply to both natural and
man-caused releases). The limits for "very unlikely" releases would not
change. The risk associated with these release limits would be about 200
excess cancer deaths over 10,000 years.
For releases caused by actions of humans, we could set numerical
limits for all expected events or for any single intrusion. The analyses
described in Chapter 5 indicate that a single intrusion could cause about
50 to 100 excess cancer deaths over 10,000 years. (This assumes that no
one would clean up the radioactivity released and that it would disperse
through the environment.) The limits would apply to those human actions
the implementing agencies judge to be credible. The limits would not
apply to recovery of the wastes for their resource value. We assume that
people knowing enough about the wastes to want to recover them would also
know about their dangers and would take appropriate precautions.
If the limits on releases caused by human intrusion were stated for
all such expected events over the 10,000-year period, the standard would
simply be the sum of limits on releases caused by natural events and by
human intrusion, and the total limits would be essentially unchanged. The
only real difference would be that the containment requirements would have
two components, and each limit would have to be met separately. It would
84
-------
not resolve the uncertainty of the frequency of possible human intrusion.
This approach could introduce unnecessary complexity and would not provide
any additional public health protection.
If the release limits for human intrusion were stated on a
single-event basis, it would be unnecessary for regulatory or implementing
agencies to estimate the number of intrusions that would be expected over
the assessment period. This would eliminate a major uncertainty in
assessing the risks and in determining whether a proposed disposal system
meets the containment requirements in the standards. The separate release
limit standards would therefore be easier to implement.
Standards with release limits for each separate human intrusion do
not, however, place a direct limit on the total estimated harm to future
generations. If a disposal system were located so that human intrusion
was quite likely, the risks from disposal of the wastes from 100,000 MTHM
could be much larger than the limits proposed in our standards. (For
example, our risk assessment assumes about 200 intrusions into the salt
repository over 10,000 years. Release limits corresponding to 50 to 100
excess deaths per intrusion would then be associated with 10,000 to 20,000
excess deaths over 10,000 years). Implementation of this alternative
would rely on qualitative judgments by the regulatory agencies;
quantitative estimates of overall risk would not be required.
85
-------
There are definite advantages and disadvantages to either of these
two possible choices, and they represent two viable options. We decided
to select the single set of containment requirements for our proposed
standards. We believe the simplicity of the single set of values, which
adequately protect the public health, outweigh the possible advantages of
the dual requirements.
86
-------
Chapter 5
PROJECTED HEALTH EFFECTS FROM DISPOSAL
This chapter is a synopsis of our assessment of the risks from
disposal of high-level radioactive wastes in mined geologic repositories.
We describe the model geologic repository and show how the released
radionuclides reach the accessible environment and people. We also
describe how we assess the risks to future generations. The results of
the risk assessment are given in terms of the number of fatal cancers and
of the number of genetic effects to be e'xpected in the exposed population
over 10,000 years or more. A more detailed description of the models
used, with additional references, is provided in Appendix B.
Our assessment consisted of a number of steps. First, we defined the
reference mined geologic repositories. We then analyzed various events or
processes that could lead to releases of radionuclides from a repository
and modeled the movement of those radionuclides through the geosphere to
the accessible environment. From the probability of release and the
associated consequence, we then calculated the resulting risk to the model
population (assumed rather similar in demography to the present United
States population) and summed the risks from all these releases over the
first 10,000 years after repository sealing.
The results of the assessment depend on our assumptions about the
performance of the repository and about the movement of radionuclides
through the environment. To describe the behavior of each part of the
87
-------
disposal system, we chose performance values that we believe would be
achievable by a carefully selected site and a well-designed repository.
The numerical values that describe each physical, chemical, or biological
process are given in Smith, et al. (SMC 32). The assessment then shows
how the results depend on the various characteristics of the disposal
system and which characteristics are most important.
We evaluated the risks from model repositories in five different
geologic media: bedded salt, dome salt, granite, basalt, and shale. Our
detailed discussion here concentrates on repositories in salt and granite,
since these were found to give lower risks. For comparison, some results
for basalt and shale repositories are also given.
5.1 MODEL HEPOSITORIES
Our model repositories contain radioactive waste consisting of
100,000 metric tons of unreprocessed spent fuel, aged 10 years after
removal from reactors. We chose spent fuel for our assessment because
this requires examination of the behavior of both fission products and
transuranium nuclides. The waste is assumed to be contained in canisters
that would last 100 years in salt and 500 years in granite.
Figure 5-1 shows a model of the repository in bedded salt; Figure 5-2
shows a model of the repository in granite. Both reference repositories
cover an area of 8 square kilometers, about one-fourth of which is
88
-------
Surface
Surface
Deposits
"_T_- Shale
Salt
.Repository
:n_" Shale
Aquifer
410 Meters
460 Meters
510 Meters
560 Meters
590 Meters
FIGURE 5-1 REFERENCE REPOSITORY IN BEDDED SALT
89
-------
Surface
200 Meters
230 Meters
460 Meters
FIGURE 5-2 REFERENCE REPOSITORY IN GRANITE
-------
actually mined. The reference repository in bedded salt is 460 meters
below the surface. The salt layer is 100 meters thick, so there are
50 meters of impervious salt above and below the repository level. There
are an additional SO meters of relatively impermeable shale immediately
above and below the salt; then there are overlying and underlying
water-bearing permeable layers 30 meters thick. For convenience, we refer
to these permeable layers as aquifers, although they may not be useable
sources of groundwater. Above the upper aquifer is another 330 meters of
overburden. The groundwater discharges into a stream 1600 meters from the
repository site. The reference granite repository is also 460 meters
below the surface. The granite formation continues indefinitely below
it. Above it there is a 230-meter thickness of granite, and above this is
a permeable stratum identical with the one modeled for the bedded-salt
repository. Above this permeable stratum are 200 meters of overburden.
We assume that the material removed to construct the rained cavity
will be returned as backfill after the wastes have been emplaced. This
backfilled material cannot be compressed to its original density; we
assume there will be void spaces amounting to 20 percent of the mined
volume. Because the repositories are below the water table and cannot be
completely sealed from groundwater, we expect that the void spaces will
fill with water. In granite, this water enters from the upper aquifer
through fractures in the rock and eventually fills all the void spaces in
the backfill. In salt, it enters from the upper aquifer through
91
-------
imperfections in shaft and borehole seals. Much of the void space is
eliminated by plastic movement of the salt, and only a small amount of
water is trapped in the salt in the form of brine pockets.
As the canisters fail, the waste is exposed to water in the
repository and is gradually leached from the matrix and dissolved. This
contaminated repository water is the largest source from which
radioactivity reaches the accessible environment. We selected a release
* -4
rate of 0.01 percent per year (10 per year) for the waste. We
assumed that the uptake of nuclides of uranium, plutonium, neptunium, and
technetium by groundwater was limited by their solubility, which was taken
to be 1 milligram per cubic meter of water. We also assumed that the
solubility of americium was 50 grams per cubic meter of water and that of
tin was 1 gram per cubic meter. Nuclides of these elements represent the
most important long-term concern in terms of movement through the
environment and possible hazard to people.
*"Release rate" is defined as the fraction of material
dissolved from the high-level waste matrix by the surrounding liquid
(water) during a given period of time. This rate generally determines
the maximum rate at which any nuclide can enter the repository water.
92
-------
We modeled the aquifer as a porous permeable rock. Water flows in
the aquifer slowly, the speed being determined by the porosity and
**
permeability of the rock and by the hydraulic pressure gradient in the
aquifer. We assumed that the aquifer permeability is 10 centimeters
per second (about 30 meters per year), the aquifer pressure gradient is
0.01, and the effective porosity is 0.15. This combination of
permeability, flow gradient, and porosity corresponds to an interstitial
***
water velocity of about 2 meters per year. Nuclides reaching the
aquifer will move in the direction of groundwater flow. A few nuclides
will move at the same speed as the groundwater, but most will move more
slowly. The extent to which a nuclide is slowed in comparison to the
Tlfllf "Jfllf
groundwater is expressed as a retardation factor. We selected
conservative retardation factors for the significant nuclides. These
factors were 100 for actinides, 10 for tin, and 1 for other nuclides.
The choice of these relatively small retardation factors means that the
calculated values for movement along the aquifer flow are'likely to be
faster than will actually be observed.
*"Porosity" is the fraction of the rock formation available to
water (void fraction).
**"Permeability" is the ability of a geologic formation to
transmit water.
***"Interstitial water velocity" is the speed of water flow
through the rock or other medium.
"Retardation factor" is the ratio of the interstitial water
velocity to the velocity of the radionuclide. For example, a
retardation factor of 100 means that the nuclide moves at 1/100 the
speed of the water. Retardation is caused by adsorption or other
mechanisms which slow the movement of chemical compounds through the
environment.
93
-------
5.2 RELEASE AND TRANSPORT MECHANISMS
Under an EPA contract, Arthur D. Little, Inc. (ADL) identified and
analyzed possible processes by which the waste could be released from a
repository (ADL 79). Once released, radionuclides could enter the
environment via air, land surface, or groundwater pathways. The ADL study
provides the basic information needed to describe the quantities of
radionuclides released from the repository by each process or event, and
the probabilities of release. The report includes estimates of the
fraction of the repository affected by an event and of the groundwater
flow rate through the affected area of the repository.
Radionuclides released directly to air or land surface reach the
accessible environment quickly. Releases into groundwater can reach the
accessible environment only after considerable delay. We applied a simple
one-dimensional transport model of radionuclides in the groundwater. The
model neglects dispersion, but includes the velocity of groundwater flow,
retardation of nuclides, and solubility limits for certain nuclides.
Once the radionuclides enter the accessible environment, they can
reach people through a number of pathways (SMJ 82). We considered a total
of 30 of these pathways. For radionuclides reaching surface water, for
example, we considered: (a) drinking water; (b) eating fish; (c) eating
three types of foods grown by irrigation—surface crops, milk and beef;
(d) breathing air contaminated by resuspension after irrigation; and
94
-------
(e) standing on ground contaminated by irrigation. For each nuclide, the
fraction eaten or breathed by humans from each pathway was calculated.
Doses to those organs believed susceptible to radiation induced cancer
were calculated, as well as doses to the reproductive organs. The doses
to each organ were then summed over all pathways, including the external
dose pathways. We then estimated the total numbers of fatal cancers and
first generation genetic defects to the population. Risk conversion
factors were derived from the report of the Committee on the Biological
Effects of Ionizing Radiation (BEIR) of the National Academy of Sciences
&
(NAS 72). Table 5-1 summarizes our estimates of the fatal cancers
caused by each curie of radioactivity released to the environment for each
of the three release modes: to surface water (via groundwater), to the
land surface, or to the air (SMJ 82).
We expect population distributions, food chains, and technology to
change dramatically over 10,000 years. Unlike geologic processes, they
can be realistically predicted only for short times. Accordingly, our
estimates of health effects used general models of environmental pathways,
populations, and living habits (SMJ 82). Rather than attempt to predict
future changes in populations, cancer cure rates, etc., they assume
present values for these parameters.
We have not recalculated these risk estimates on the basis of
the July 1980 report of the BEIR committee (NAS 80). The risk estimates
using the 1980 report would be somewhat lower, but not by as much as a
factor of 10.
95
-------
Table 5-1
Potential Health Effects (Fatal Cancers) Caused
per Curie Released to the Environment by Different Modes
Nuclide
Am-241
Am- 24i3
C- 14
Cs-135
Cs-137
1-129
Np-237
Pu-238
Pu-239
Pu-240
Pu-242
Ra-226
Sr- 90
Tc- 99
Sn-126
Releases to
Surface Water
0.73
2.77
0.05
0.004
0.02
0.01
0.6
0.02
0.07
0.07
0.07
3.17
0.12
0.0003
0.12
Releases to
Land Surface
0.09
1.03
0.00003
0.0004
0.0006
0.00002
0.003
0.003
0.06
0.05
0.06
0.08
0.001
0.00000006
0.04
Releases to
the Air
0.16
1.14
0.0002
0.0007
0.007
0.001
0.08
0.02
0.05
0.05
0.05
0.49
0.02
0.00004
0.11
96
-------
The risk from an event (e.g., movement of a fault or inadvertent
human intrusion by exploratory drilling) is the probability of that event
multiplied by its consequences. For each type of possible event, we
estimated the probability of its occurence within several different time
periods after disposal. We did this because the same event or process
will release different amounts of radionuclides—because of radioactive
decay—depending on the time when it occurs. We then estimated the
consequences for each event occurring within each time period after
repository sealing. We then added the risks from all individual events to
obtain the total projected risk to the population.
5.3 RESULTS OF RISK ASSESSMENTS
5.3.1 Population Risks
For the reference bedded-salt repository, the projected number of
fatal cancers in the first 10,000 years is about 100. Almost all of these
are due to releases of contaminated water from the repository to the land
surface as a result of drilling during resource exploration. For the
reference granite repository, the projected number of fatal cancers in the
first 10,000 years is about 700. Almost all of these result from transfer
of contaminated granite repository water to the surface during drilling in
the same way as for bedded salt. About 10 cancers are due to routine
releases from natural processes, in which the heat generated by the waste
is the driving force for a buoyancy pump, which raises repository water
through the fractures in the granite to the permeable stratum, through
97
-------
which it is carried to surface water. Figure 5-3 shows the projected
health effects over 10,000 years for each type of event that may affect
granite or bedded salt repositories. Figure 5-4 provides a comparison of
projected health effects for repositories in different geological media.
We also evaluated the health risks over longer periods, considering
only the events and processes occurring in the first 10,000 years. In the
granite repository, many nuclides remain in transit in the groundwater
during the first 10,000 years, so that the expected number of fatal
cancers increases to about 900 in 50,000 years and to 2,800 in 100,000
years. The corresponding number of fatal cancers from a bedded-salt
repository increases only to 150 in the first 20,000 years and remains
constant at this figure for 100,000 years.
To evaluate the effect of changing our assumptions, we also
calculated results for different choices of the more important variables.
Resistance of the waste to leaching and the insolubility of uranium,
neptunium, plutonium, and technetium are important in limiting the risks
from disposal. Both the nature of the waste form and the geochemical
characteristics of the site determine leaching resistance and
insolubility. Leaching and solubility change the availability of the
radionuclides to the groundwater in the repository. Reduction of the
leach rate by a factor of 100 per year from the reference case reduces the
projected fatal cancers in granite to about 10; increasing the leach rate
by a factor of 100 increases them to 2,500. Effects for the bedded-salt
98
-------
10 5-
10 4 -
10 3-
a io2-
I lo1'
* 10°-
-------
5000-
7000-
6000-
FIGURE 5-4:
PROJECTED HEALTH EFFECTS
OVER 10,000 YEARS FOR
REFERENCE REPOSITORIES
IN DIFFERENT
GEOLOGIC MEDIA
to
0)
o
o 5000.
s.
O)
o
o 4000-
<*-
-------
repository are much smaller. The effect of varying the leach rates is
shown in Figure 5-5. The effects of assuming infinite solubility are
shown in Figure 5-6. Eliminating the solubility limit for selected
actinides and for technetium would increase the projected number of
fatal cancers in the granite repository to almost 2,000, and in the
bedded salt to 9,400.
The stability and durability of canisters are less important than
resistance of the waste form. Modeling canister lifetime at 1,000
years produces relatively small decreases in the projected number of
fatal cancers; modeling canister lifetime at 5,000 years reduces the
number of projected fatal cancers by a factor of about five.
Long-lived canisters, on the order of 50,000 to 100,000 years, wbuld
substantially reduce the number of fatal cancers because of decay of
important nuclides, especially americium-243, while the containment
existed. The effect of different assumptions on canister lifetime is
shown in Figure 5-7.
Modeling all nuclides with retardation factors equal to one, so
that all nuclides move with the speed of groundwater, increases the
number of projected fatal cancers from the granite repository to over
30,000. Figure 5-6 shows the effects of assuming no retardation.
Higher retardation factors, such as 10,000 for the actinides (except
neptunium), do not significantly reduce the number of projected fatal
cancers, since these radionuclides move so slowly through the
environment.
101
-------
8000-
S-
to
OJ
o
o
o
OJ
o
in
o
OJ
-------
16,000-
14,000-
12,000-
j_
-------
8000-
7000-
6000-
S-
«3
CD
O
5000-
S-
CU
o
to
« 4000
•a
O)
+j
o
01
•"-J
o
Q-
3000-
2000-
1000-
FIGURE 5-7:
PROJECTED HEALTH EFFECTS
OVER 10,000 YEARS
VS.
DIFFERENT CANISTER LIFETIMES
(years)
1000
2000 3000
canister lifetime (years)
4000
5000
104
-------
Most of the fatal cancers for the reference case are associated with
intrusion into the repository by drilling; therefore, the number of fatal
cancers is roughly proportional to the frequency of drilling. The
frequencies suggested for such drilling events by ADL are one in 50 years
for the bedded-salt repository and one in 400 years for the granite
repository.
For the reference cases, the calculated first generation genetic
effects are lower than the number of fatal cancers by a factor of more
than 10. This is because most of the important nuclides involved are bone
seekers and none concentrate in gonads or soft tissue.
5.3.2 Risks to Individuals
Without effective institutional controls, the probability that
someone exploring for resources will drill into the repository at least
once in the first 5,000 years is high, even at drilling frequencies
significantly smaller than those estimated by ADL. The groundwater
immediately above the repository would be contaminated by repeated
drilling events, and some of it would remain contaminated for long periods
of time.
Individuals may be exposed to nuclides released from the repository
by breathing air contaminated as a result of releases of nuclides to land
surfaces or by drinking water contaminated as a result of introduction of
radionuclides into the groundwater (GO 82). Doses from breathing air were
105
-------
calculated as relatively small at distances of 100 meters or more from the
point of discharge to the surface. Radionuclides move slowly in
groundwater and exist in rather high concentrations over fairly small
areas. The most important nuclides in groundwater are the americium
isotopes 241 and 243. These nuclides, particularly the long-lived
americium-243, persist for a long time in the environment.
We analyzed a scenario in which an initial drilling event directly
contacted contaminated groundwater in a granite repository. This scenario
results in the greatest contamination of the groundwater in the adjacent
aquifer (GO 82). For each such hypothetical drilling event, contamination
would spread slowly in the aquifer. By 1,000 years after the drilling, an
aquifer area of about 4,000 square meters would be contaminated enough to
give a dose of more than 500 millirems per year to people who might
subsequently drill into the aquifer and drink the water. By 10,000 years
after the initial drilling, an aquifer area of about 80,000 square meters
would be contaminated to this level.
The probability that an individual would subsequently drill a water
well into an area contaminated by resource exploration depends on
conditions at the site, particularly the relative availability of
contaminated and uncontaminated water. People are not likely to drill
wells into a deep aquifer if adequate water is otherwise available. Using
the following assumptions, we made a generic estimate of the probability
that a person might encounter a contaminated area: about 500,000 water
106
-------
weLLs are drilled per year (GE 73), of which some 2 percent are deeper
than 150 meters (500 feet). The U.S. Geological Survey (NW 79) estimates
that about one-third of one percent of all wells are low-yield (less than
10 gallons per minute), deep (greater than 500 feet) wells. To allow for
increased use of ground water in the future, we assume that 5,000 deep,
low-yield wells are drilled in the United States each year. The number
of new wells, divided by the U.S. land area of 7,800,000 km , is
6.4x10 new wells per square meter per year. The probability of a
person drilling into an 80,000 square meter area (the area contaminated
above 500 millirem/year at 10,000 years) would be about one in two-hundred
during any 100-year period.
5.4 CONCLUSIONS
Our assessment of disposal in mined geologic repositories identifies
the major factors in estimating risks to future generations. The
following conclusions can be drawn from the assessment:
The release of a large fraction of the repository inventory should be
extremely unlikely. In a deep-mined geologic repository, only a volcanic
eruption or the impact of a large meteorite should produce a very large
release of radionuclides (more than a few percent of the inventory). The
chance of such releases occurring within 10,000 years is vanishingly small
if the wastes are placed deep enough below land or water in a nonvolcanic
area.
107
-------
The movement of important radionuclides in groundwater from the
repository to the accessible environment should be slow enough to provide
substantial time for decay of radioactivity. Low permeability and a small
hydraulic gradient in the groundwater stratum will result in slow
groundwater movement. High adsorption is necessary to further slow the
movement of such moderately long-lived radionuclides as plutonium-239,
americium-241, and americium-243.
The entry of radionuclides into groundwater should be slow. Some
radionuclides, particularly those of uranium, neptunium, plutonium, and
technetium, can be kept essentially insoluble in groundwater by proper
selection of their chemical form in the waste. We can select repository
sites where the geochemistry will keep these elements in their insoluble
forms. For other elements, selection of waste forms that are leached
slowly should limit the rate of entry into groundwater even though the
chemical forms of these elements in the waste are more soluble.
The chance that any individual will receive high doses should be
small. High doses are associated with use of contaminated groundwater or
air close to a point where radionuclides have been released. The
probability of an individual receiving high doses can be made small by
selecting sites where groundwater is not likely to be used, where there
are no resources likely to be exploited, and where natural intrusive
events are improbable.
108
-------
In general, we have shown that it is reasonable to expect disposal of
high-level radioactive wastes in a mined geologic repository (containing
the wastes from 100,000 MTHM) to result in fewer than an average of one
fatal cancer every 10 years for the first 10,000 years, provided the
repository is reasonably well designed and well sited. Few of these
health effects are expected to occur in the first 1,000 years after
emplacement. The most important cause of radionuclide release appears to
be human intrusion in the course of exploration for resources. Health
effects may be reduced by using waste forms resistant to leaching and
dissolving. More importantly, adverse effects will be reduced by
selection of sites where: (1) the geochemistry will maintain technetium
and most actinides in an insoluble form, (2) the hydrology will retain
many nuclides in the groundwater for long times, and (3) exploration for
resources is improbable.
109
-------
110
-------
Chapter 6
PROPOSED STANDARDS FOR WASTE DISPOSAL
The proposed standards for disposal contain three sections:
containment requirements, assurance requirements, and procedural
requirements. The containment requirements set limits on potential
releases of radioactive materials from disposal systems to the accessible
environment for 10,000 years after disposal. These requirements are based
on the assumption that our predictions of projected risks from disposal
systems are good enough to use in selecting, designing, and implementing
disposal methods. These predictions can then be used to assess the
long-term releases from a disposal system and to decide whether it will
provide adequate protection to present and future generations.
Closely coupled with the numerical containment requirements are seven
qualitative assurance requirements that are needed to have confidence that
the long-term release limits will be met. The assurance requirements
address and compensate for the uncertainties that necessarily accompany
plans to isolate high-level and transuranic wastes from the environment
for a very long time. No matter how promising analytical projections of
disposal system performance appear to be, these wastes should be disposed
of in a cautious manner that reduces the likelihood of unanticipated
releases. Our assurance requirements provide the context necessary for
application of our containment requirements, and they should ensure very
good long-term protection of the environment.
Ill
-------
Finally, our procedural requirements provide instructions that must
be followed to ensure that the containment requirements are properly
implemented for particular disposal systems. These requirements describe
the assumptions that are appropriate when assessing the performance of
disposal systems to determine compliance with our long-term release limits.
In the following sections, we discuss the various features of all
three sets of requirements, explain why we selected them, and describe how
they should be used.
6.1. CONTAINMENT REQUIREMENTS
The primary requirements of our proposed environmental standards are
numerical limits on the amount of radioactive waste that may be released
from a disposal system to the accessible environment for 10,000 years
after disposal. These containment requirements should provide excellent
protection of public health and the environment. In fact, they should
limit the risks to present and future generations to a level no greater
than the risks from equivalent amounts of unmined uranium ore.
The following sections indicate how we selected the form of these
containment requirements, why we chose the specific level of risk they
are based on, and how the release limits are to be used. Because they
address such a long time period, and because they include unplanned
releases, these containment requirements can be implemented only through
analytical projections of disposal system performance. There will be many
112
-------
uncertainties in making such long-term performance projections.
Accordingly, our proposed standards require only a "reasonable expectation"
that these containment requirements will be met; this determination must
be made by the implementing agencies.
6.1.1 The Accessible Environment
The proposed containment requirements apply to releases to the
"accessible environment," which we identify by a set of definitions:
"'Accessible environment1 includes (1) the atmosphere,
(2) land surfaces, (3) surface waters, (4) oceans, and
(5) parts of the Lithosphere that are more than ten kilometers
in any direction from the original location of any of the
radioactive wastes in a disposal system."
'"Lithosphere1 means the solid part of the Earth,
including any groundwater contained within it."
"'Groundwater1 means water below the land surface in a
zone of saturation."
Through these definitions we intend to protect all portions of the
environment that may be in direct contact with man. (For example, the
atmosphere, land surface, and any surface water directly over a geologic
repository is included within the "accessible environment".) In addition,
our definition of the "accessible environment" includes all groundwater
formations that are more than 10 kilometers away from a disposal system.
Groundwater is an important resource; currently, more than 20 percent of
the fresh water used in the United States comes from underground sources.
However, geologic formations that contain relatively small amounts of
113
-------
groundwater can serve as an important part of the containment system for a
geologic repository. Thus, we need to allow for contamination of some
groundwater while we are ensuring the long-term isolation of these wastes.
Accordingly, our approach does not provide any direct protection
for the relatively small amount of groundwater that could be within
10 kilometers of a geologic repository. However, since the amount of
groundwater left unprotected should be kept as small as possible,
consistent with other requirements, we expect that the Federal
environmental impact statement for each disposal system will identify all
sources of groundwater within 10 kilometers of the disposal system, will
describe the potential long-term environmental effects of possible
contamination of these sources of groundwater, and will consider these
effects as one of the factors in evaluating alternative sites.
On the other hand, our definition of "accessible environment"
includes all groundwater sources—regardless of their quality—that are
more than 10 kilometers away from a disposal system because our analyses
indicate that this should be an ample distance to provide the protection
expected from the geologic barriers of a repository. (In fact, all of the
analyses described in Chapter 5 assume that the accessible environment,
with respect to groundwater, is only 1 mile away from the repository—the
longer distance of 10 kilometers would tend to reduce the residual risks
identified in Chapter 5.)
114
-------
6.1.2 Radionuclide Release Limits
The proposed containment requirements set limits on the projected
releases of radionuclides to the accessible environment for 10,000 years
after disposal. Because of the uncertainties of such long-term
projections, we require a disposal system design to provide a "reasonable
expectation" that the release limits will be met. A comprehensive
performance assessment would be the basis for making this judgment.
These requirements would limit "reasonably foreseeable" releases to
the quantities derived from Table 2 of the proposed standards: "Release
Limits for Containment Requirements" (see Appendix A of this document).
"Very unlikely" releases would be limited to 10 times the quantities for
"reasonably foreseeable" releases. "Reasonably foreseeable" releases are
those estimated as having a greater than 1 in 100 chance of occurring
within 10,000 years. "Very unlikely" releases are those with a chance of
occurring in 10,000 years estimated to be less than 1 in 100 but more than
1 in 10,000. We selected the qualitative terms "reasonably foreseeable"
and "very unlikely" to emphasize that these are not precise estimates.
The implementing agencies may use some discretion in assessing the
probabilities of releases.
The values in Table 2 of the proposed standards are the quantities of
each radionuclide that would result in 10 projected fatal cancers over
10,000 years (SMJ 82) if released into surface water—the most dangerous
of the three release modes shown in Table 5-1. The values in Table 2 are
115
-------
expressed in terms of releases from wastes generated from 1000 metric tons
of heavy metal (MTHM) originally placed in the reactor; they correspond to
1,000 projected fatal cancers in 10,000 years from the 100,000 MTHM
geologic repository used in our risk assessment. We estimated that the
long-term risks from disposal of wastes from 100,000 MTHM in our model
repositories for salt and granite would be about 100 to 800 fatal cancers
in 10,000 years. We choose release limits for Table 2 of the proposed
standards that correspond to somewhat higher long-term risks.
We selected the value of 1 million curies of transuranic elements as
a reference value for transuranic wastes to require about the same degree
of control for the long-lived alpha-emitting radionuclides in transuranic
wastes that we are requiring for those radionuclides in high-level
wastes. Since a 100,000-MTHM repository contains about 300 million curies
of transuranic nuclides, the 1,000,000-curie reference quantity for
transuranic wastes is in about the same ratio to this total as the
1,000-MTHM reference quantity of spent fuel is with respect to the
100,000-MTHM model repository. Or, in other words, the reference values
were selected so that about the same fraction of transuranic radionuclides
would be retained for either high-level or transuranic wastes.
Since more than one nuclide would be released in any actual
situation, the limitation for each nuclide must be lowered (made more
stringent) to allow for the risks attributable to other nuclides. This is
done through summing the terms:
a/A + b/B + c/C + ...
116
-------
where a, b, c... are the cumulative releases of the various nuclides and
A, B, C... are the appropriate Table 2 values. The total releases are
summed over all radionuclides as well as over all release pathways.
The containment requirements provide for less stringent limitations
on "very unlikely" releases of radionuclides to the accessible
environment. They would set release limits that are higher by a factor of
10, with the requirement that the probability of such a release must be
lower by a factor of 100. The requirement for a greater decrease in
probability than the increase in released quantity reflects our aversion
to large consequences. This provision amounts to about a 10-percent
increase in effects on health.
These containment requirements do not permit large projected
cumulative releases, which would have severe consequences, unless their
probabilities are less than 1 in 10,000 over 10,000 years. Such large
releases could result only from extremely disruptive events, such as a
volcanic eruption or the impact of a large meteorite, occurring when the
repository still contained a significant inventory of undecayed waste
radionuclides. We do not believe that releases that are so highly
improbable should be regulated.
117
-------
6.1.3 The Level of Protection
We used the information in Chapters 3 and 5 to select a limit on harm
to future generations that we believe is very small and that we believe
good geologic repositories can meet. The limit on which we based our
proposed containment requirements is 1,000 excess cancer deaths over
10,000 years for a 100,000-MTHM repository.
As shown in Figure 6-1 (which reproduces Figures 5-4 through 5-7 from
Chapter 5, with the proposed level of protection indicated), there are
several combinations of geologic media, engineered barriers, and site
geochemistry that should provide at least this much protection. We
believe, with careful site selection and with the technology available to
develop engineered barriers, mined geologic repositories can keep risks
below this level.
118
-------
8000-
7000-
6000-
5000 •
4000-
3000-
2000-
1000 •
n _
PROJECTED HEALTH EFFECTS
OVER 10,000 YEARS FOR
REFERENCE REPOSITORIES
IN DIFFERENT
GEOLOGIC MEDIA
PROPOSED STANDARDS
i 1 I 1
°ciET B
-------
6.2 ASSURANCE REQUIREMENTS
Our seven qualitative assurance requirements address and compensate
for the uncertainties that necessarily accompany plans to isolate
high-level and transuranic wastes from the environment for a very long
time. They call for a cautious and "common-sense" approach to disposal
ttiat provides the necessary context for application of our containment
requirements. We expect that the specific steps taken by the implementing
agencies to comply with each of the following criteria will be described
in the Federal environmental impact statement—and other decision
documents—for each disposal system:
6.2.1 Criterion 1:
"Wastes shall be disposed of promptly once disposal systems
are available and the wastes have been suitably conditioned for
disposal."
Waste storage systems require active human controls for periodic
inspection and maintenance. To reduce dependence on active human
controls, this assurance requirement specifies that the wastes should be
removed from storage and disposed of promptly once adequate disposal
systems are available and the wastes have been appropriately processed for
disposal (i.e., placed into a suitably stable chemical and physical form).
We have not established a specific time period in this requirement,
because the appropriate length of storage may depend on details of the
disposal system design. For example, it may be desirable to store
120
-------
high-level wastes for several decades to allow for decay of most of the
short-lived radionuclides. The primary intent of this criterion is to
prevent wastes from being stored indefinitely to avoid ultimate disposal.
6.2.2 Criterion 2;
"Disposal systems shall be selected and designed to keep
releases to the accessible environment as small as reasonably
achievable, taking into account technical, social, and economic
considerat ions."
As discussed in Chapter 2, we assume that any exposure to radiation
can produce some harm. Accordingly, this assurance requirement stipulates
that releases of radioactive wastes be kept as small as reasonably
achievable. It instructs the regulatory agencies to reduce releases below
the limits of the containment requirements on the basis of site- or
method-specific information.
This principle has been used in radiation protection for a long
time. It is not a totally quantitative concept, and we have not placed
numerical specifications on it here. The technical judgment of the
implementing agencies would determine when this criterion has been
satisfied.
121
-------
6.2.3 Criterion 3:
"Disposal systems shall use several different types of
barriers to isolate the wastes from the accessible
environment. Both engineered and natural barriers shall be
included. Each such barrier shall separately be designed to
provide substantial isolation."
This assurance requirement emphasizes the need for caution in
designing a disposal system by requiring multiple barriers. We will
always be uncertain about how well a barrier may perform over the long
lifetimes required of disposal systems. Furthermore, if a barrier in a
disposal system fails, the failure would probably not be detected for a
long time, if at all.
Therefore, the barriers should retain the radioactivity by different
chemical and physical mechanisms. We believe that both engineered and
natural barriers should be included. Each of the barriers should be
designed separately to provide substantial isolation, even if the other
barriers perform poorly. Extra barriers should be added when they can be
provided at reasonable cost, even if they do not appear to be needed to
meet the performance requirements for the overall disposal system.
122
-------
6.2.4 Criterion 4;
"Disposal systems shall not rely upon active institutional
controls to isolate the wastes beyond a reasonable period of
time (e.g., a few hundred years) after disposal of the wastes."
This assurance requirement would limit reliance on "active"
institutional controls to a reasonable period after disposal, which should
be no more than a few hundred years. "Active" institutional controls
include guarding a disposal site, performing maintenance operations or
remedial actions at a disposal site, or controlling or cleaning up
releases from a disposal site.
This requirement does not mean we think society will lose all
knowledge of radioactivity, nuclear energy, radioactive wastes, or even
specific disposal sites after a few hundred years. On the contrary, we
believe that such information is likely to survive, even without the
extensive markers and records called for by the next assurance
requirement. However, merely having this knowledge does not guarantee
that it will be widely disseminated or effectively acted upon. We believe
it is prudent to assume that society may not retain active controls over
disposal systems for very long, and that unrelated activities may resume
at a disposal site even though the presence of radioactive wastes is
documented. The assumptions that we believe are appropriate when
considering the effectiveness of "passive" institutional controls are
described under our procedural requirements.
123
-------
6.2.5 Criterion 5:
"Disposal systems shall be identified by the most permanent
markers and records practicable to indicate the dangers of the
wastes and their location."
During our public workshops on radioactive waste criteria, we
explored whether the location of radioactive wastes should be concealed
from future generations or should be clearly marked (EPA 77). We became
convinced that concealment was impractical because the inevitable
anomalies of a disposal site would eventually attract attention.
Therefore, trying to limit future knowledge would probably only make the
situation more dangerous.
Accordingly, this assurance requirement calls for comprehensive
actions to be taken to pass on information about the wastes and the way we
disposed of them. However, we do not assume that this information can be
guaranteed to prevent disruption of the site if active controls are lost.
Unfortunately, warning signs of many kinds are often ignored today; we
cannot assume they would be much more effective in the future.
124
-------
6.2.6 Criterion 6;
"Disposal systems shall not be located where there has been
mining for resources or where there is a reasonable expectation
of exploration for scarce or easily accessible resources in the
future. Furthermore, disposal systems shall not be located
where there is a significant concentration of any material
which is not widely available from other sources."
This assurance requirement is a logical complement to our limitation
on active institutional controls. Resources will attract exploration, and
the people doing the exploring may not know of the locations or dangers of
radioactive wastes. Since active controls may not be available to limit
access, disposal systems should not be located where people are likely to
explore for scarce or easily accessible resources.
Obviously, exploration may occur anywhere, and careful judgment must
be used in applying this criterion. Past and current mining is an
important indicator, but future trends must also be considered. Some
resources that are not economical to recover today may become a
substantial part of our resource base in the future; oil shales are an
example of such a resource. Also, minerals not important now might become
so in the future.
Because of these considerations, we believe that disposal systems
should not be located where there is a substantial concentration of any
material not widely available elsewhere. Nor should they be located where
exploration or mining is likely in the future, particularly if related to
125
-------
resources that can be neither renewed nor significantly recycled (such as
energy resources). This assurance requirement could also be important
even if active institutional controls survive for a long time. A poorly
located disposal system could deny or seriously hinder use of critical
resources to future generations.
This assurance requirement should discourage the use of geologic
formations frequently associated with resources. For example, salt domes
are often mined either for their relatively pure salt or for use as
effective, inexpensive storage caverns. Of the 130 domes in the Gulf
Coast area judged suitable for such exploitation, 48 were in use as of
1965 (GR 81). As discussed under Criterion 4, we do not believe that
active institutional controls should be relied upon to prevent exploration
for resources for more than a few hundred years after disposal.
Therefore, the frequent use of salt domes as resources would argue against
locating a repository in this type of structure.
These particular concerns generally would not apply to bedded-salt
deposits, which are far more extensive. A specific site in a bedded-salt
deposit appears much less vulnerable to inadvertent intrusion than a
particular site in a salt dome.
126
-------
6.2.7 Criterion 7:
"Disposal systems shall be selected so that removal of most
of the wastes is not precluded for a reasonable period of time
after disposal."
This assurance requirement would permit corrective action if
development of technical and scientific knowledge should indicate that a
disposal system poses much greater risks than had originally been
expected. High-level radioactive waste disposal is a new technology.
Future generations may find it necessary or desirable to recover the
wastes and change the disposal site or methodology in light of new
knowledge.
This requirement is different from the more familiar concept of
retrievable storage. Recovering the wastes after the conventional
retrieval period of a few decades is likely to be expensive and, perhaps,
even dangerous. It could require, for example, mining the entire
repository volume to recover the wastes. This criterion merely requires
that this option be left open to our successors.
Current plans for mined geologic disposal would comply with this
assurance requirement. However, methods involving irreversible
incorporation of radionuclides into disposal sites would not. Deep-hole
placement, rock-melting, or hydrofracturing are examples of disposal
methods that do not appear to comply with this assurance requirement.
127
-------
6.3 PROCEDURAL REQUIREMENTS
Compliance with our containment requirements will be determined
through long-term projections of disposal system performance. The
"performance assessments" used to make these judgments should be done in
accordance with the proposed definition included in our disposal standards:
'"Performance assessment* means an analysis which
identifies those events and processes which might affect the
disposal system, examines their effects upon its barriers, and
estimates the probabilities and consequences of the events.
The analysis need not evaluate risks from all identified
events. However, it should provide a reasonable expectation
that the risks from events not evaluated are small in
comparison to the risks which are estimated in the analysis."
When our containment requirements are applied to a particular
disposal system, some of the procedures we used in our performance
assessments (described in Chapter 5) must be retained so that the intent
of our standards is met. On the other hand, some of the assumptions we
made should be replaced with specific information appropriate for each
disposal system. Our procedural requirements set forth three instructions
needed to ensure that performance assessments for particular disposal
systems are properly evaluated against the containment requirements. Each
of these three procedural requirements is discussed here:
128
-------
"The assessments shall consider realistic projections of
the protection provided by all of the engineered and natural
barriers of a disposal system."
In developing the containment requirements, we considered the overall
protection that should be achievable by the combination of barriers in a
geologic repository. Accordingly, the analyses used by NRG and DOE to
evaluate compliance with our requirements should consider realistic
assessments of the protection provided by all of the engineered and
natural barriers of a disposal system. For example, performance
assessments of a geologic repository system should include the protection
afforded by geochemical retardation of radionuclides and by the limited
solubility of radionuclides in groundwater, provided that reasonable
evidence is developed to support such mechanisms for that particular site.
"The assessments shall not assume that active institutional
controls can prevent or reduce releases to the accessible
environment beyond a reasonable period (e.g., a few hundred
years) after disposal. However, it should be assumed that the
Federal Government is committed to retaining passive
institutional control of disposal sites in perpetuity. Such
passive controls should be effective in deterring systematic or
persistent exploitation of a disposal site, and it should be
assumed that they can keep the chance of inadvertent human
intrusion very small as long as the Federal Government retains
such passive control of disposal sites."
The assumptions we made in our performance assessments about the
frequency of human intrusion were conservative because they ignored the
substantial protection that passive institutional controls should offer.
The performance assessments made for specific sites by the implementing
agencies do not need to be as pessimistic with regard to human intrusion.
Because of the uncertainties of controls requiring the active participation
129
-------
of people over a long time, performance assessments should not assume that
active institutional controls can prevent or reduce releases beyond a
reasonable period of time (e.g., a few hundred years) after disposal.
However, because the Federal Government is committed to retaining control
over these disposal sites in perpetuity, passive institutional controls
should substantially reduce the chance of inadvertent human intrusion well
beyond this period. These passive controls should not be assumed to
prevent all possibilities of inadvertent intrusion, because there is
always a chance that the controls will be overlooked or misunderstood.
However, such measures should be effective in deterring systematic or
persistent exploitation of a disposal site. Furthermore, the chance of
human intrusion should be very small as long as the Federal Government
retains passive control of disposal sites.
"The assessments shall use information regarding the
likelihood of human intrusion, and all other unplanned events
that may cause releases to the accessible environment, as
determined by the implementing agency for each particular
disposal site."
We based our performance assessments on relatively simple generic
models of geologic repositories and the data that was available for such
models. Where information was uncertain, we made conservative assumptions
that should tend to overestimate the long-term risks of disposal.
However, we do not intend that the implementing agencies should use all of
the same models, data, and assumptions that we did in making performance
assessments. Instead, the implementing agencies generally should use the
best information available for each particular disposal site.
130
-------
Chapter 7
PROPOSED STANDARDS FOR WASTE MANAGEMENT AND STORAGE
Subpart A of the proposed standards establishes limits on the
annual dose equivalent to any member of the public as the result of the
management and storage of spent fuel and high-level and transuranic
wastes. Management and storage operations for these materials include all
treatment and handling preparatory to disposal, including storage,
processing, packaging, and emplacement in the final disposal configuration.
Transportation is not included (see Section 2.6.1).
Selection of a disposal technology defines the nature of the
operations necessary before disposal. Several disposal technologies have
been proposed, such as geologic, extraterrestrial and transmutation.
These technologies have been reviewed for us by the MITRE Corporation
(AL 79a) and for the President by the Interagency Review Group (IRG 79).
7.1 WASTE MANAGEMENT OPERATIONS FOR VARIOUS DISPOSAL OPTIONS
7.1.1 Operations for Disposal of Unprocessed Spent Fuel
Disposal of unprocessed spent fuel is applicable only to commercial
power generation; defense fuel is reprocessed for recovery of weapons
grade plutoniura and for its highly enriched uranium content. Initial
storage of commercial spent fuel at the reactor site is required to reduce
radioactivity and heat generation by decay before further handling is
131
-------
attempted. Additional storage after packaging may or may not take place,
depending on the availability of a disposal site. The management of
unprocessed spent fuel prior to terrestrial disposal includes: storage
before packaging (if required), packaging, additional storage after
packaging, and emplacement.
Storage Before Packaging
Almost all spent fuel is now being stored, and considerable quantities
have been stored for a number of years. This storage experience has
provided considerable information on environmental impacts.
The Uranium Fuel Cycle standards (40 CFR 190) regulate storage of
spent fuel at nuclear power stations. These proposed standards apply to
storage at independent spent fuel storage facilities (ISFSF), which are
not located at the reactor sites. This is sometimes referred to as
away-from-reactor (AFR) storage. The Department of Energy (DOE) estimates
the major releases from an ISFSF, in which 3000 metric tons of heavy metal
(MTHM) have been stored in water basins for 6.5 years, to be about
890 curies of krypton-85 and 2.4 curies of tritium, per year. Almost all
of the krypton releases and about half of the tritium releases occur at
the time the spent fuel is received (HO 80a; DOE 79). The projected
krypton-85 releases are small compared to the 50,000 curies per gigawatt
(electrical)-year (GWe-y) permitted under the standards of 40 CFR 190,
since 3000 MTHM corresponds to about 90 GWe-y, based on 33,000 megawatt
(thermal)-days per metric ton and an electrical efficiency of 32 percent.
132
-------
We estimated that the annual dose to a maximally exposed individual
from the ISFSF (including dose associated with packaging) is
0.0006 millirem (mrem) to the whole body; the highest dose to any organ
of an individual is 0.015 mrem per year to the thyroid. DOE estimates the
aggregate annual doses to the population within 80 kilometers of a
generic ISFSF, including the dose from a packaging operation, to be
0.47 organ-rems to the thyroid and 0.025 person-rems, whole body (DOE 79).
Packaging
Packaging spent fuel for final disposal is assumed to be carried out
at the storage site rather than at the final disposal site. We estimate
the major releases from packaging are 810 curies of krypton-85 and
1.3 curies of tritium per year (HO 80a; DOE 79). The individual and
population doses are included in the ISFSF doses.
Extended Storage
We estimate that releases during recovery and handling of packaged
spent fuel stored for 30 years are about 0.01 curies of krypton-85 and
about 0.1 microcuries of tritium per year (HO 80a; DOE 79). Doses to any
organ of a maximally exposed individual are less than 1 microrem per year;
population doses are also negligible.
133
-------
Emplacement
Releases of waste materials during emplacement occur only during the
handling of waste packages. DOE estimates these releases to be about
—20
10 curies per year (DOE 79) . Resulting doses, both individual and
population, are below the 40 CFR 190 standards.
Naturally occuring radionuclides are released from repository
materials during construction and operation. However, the resulting doses
to individuals and populations are very small (DOE 79). These releases
are not considered further, since they are small and are not significantly
affected by these standards.
7.1.2 Operations Prior to Disposal of Processed High-Level Wastes
The management of processed high-level wastes prior to terrestrial
disposal includes five steps (HO 80a):
1. High-level liquid waste (HLW) storage
2. Waste solidification
3. Interim storage of solidified waste
4. Emplacement
5. Treatment and emplacement of retained krypton-85 and
iodine-12 9
High-Level Liquid Waste Storage
High-level liquid wastes result from defense activities and would
also result from the reprocessing of commercial spent fuel if reprocessing
were undertaken. The technology for storing high-level liquid wastes is
134
-------
based on 30 years' experience with defense wastes. Wastes are stored in
subsurface tanks of stainless or mild steel. It is expected that stainless
steel tanks will be required for the storage of high-level wastes from
reprocessing commercial spent fuel (DOE 79). Double containment reduces
the possibility of leakage to soil.
DOE has estimated the releases from a commercial reprocessing plant
with an annual capacity for treating 1140 cubic meters (300,000 gallons)
of high-level waste from 2000 metric tons of heavy metal. This is the
spent fuel that corresponds to about 60 GWe-y. About 27,000 curies of
tritium per year are estimated to be released to the atmosphere, while
releases of any other nuclide would be about 1 millicurie per year or
less. DOE calculated the maximum annual dose equivalent to any individual
to be about 0.1 millirem and the population dose within an 80-kilometer
radius to be about 11 person-rems.
Waste Solidification
In the event that commercial reprocessing of spent fuel becomes a
part of the national program, Nuclear Regulatory Commission (NRG)
regulations (10 CFR 50, Appendix F) require that the high-level liquid
wastes generated from this reprocessing be converted to a solid form
within 5 years and be transferred to the Federal Government within 10
years of its generation. Some alternatives proposed for long-term
management of defense wastes also include waste solidification (ERDA 77a;
ERDA 77b; ERDA 77c). Calcination is the most likely method of
solidification prior to incorporation into a solid matrix, while
135
-------
glassification provides a solid matrix for disposal. The Idaho National
Engineering Laboratory has calcined high-level wastes since about 1963,
and Hanford has been developing various glassification techniques since
the early 1960's.
We have assessed the environmental effects of calcination followed by
glassification (HO 80b). The study considered fluidized-bed or spray
calcination followed by in-can or continuous-tnelter glassification. Other
treatments are possible, including formation of crystalline (e.g.,
synthetic rock or metal) matrices. Effluent releases from a generic plant
using available or near-term available effluent cleaning technology were
calculated. The assessment covered the seven radionuclides (tritium,
iodine-129, ruthenium-106, cesium-134, cesium-137, strontium-90, and
plutonium-239) most likely to be released during normal operations of the
four potentially useful technologies available for calcination and
glassification. We reviewed the control technology for particulates and
for off-gas cleanup, estimated decontamination factors for typical off-gas
control equipment for the radionuclides under consideration, and we
estimated the overall system decontamination factors for iodine,
ruthenium, and particulates to be 1x20 , 1x10 , and 1x10 ,
respectively.
Radioactive effluent releases from the generic solidification plants
were calculated for two locations, a rural site typified by Barnwell,
South Carolina, and an urban site typified by St. Louis. Dose equivalents
136
-------
for whole-body and several organs were calculated for the maximally
exposed adult at each location. Results are summarized in Tables 7-1
and 7-2. The largest organ dose was determined to be to the lower large
intestine (LLI). Ruthenium-106 is the major dose contributor for
materials up to one year old. As ruthenium-106 decays, tritium become the
most significant nuclide in the older material.
Population doses were calculated for the areas within an 80-kilometer
radius of the generic plant at each site. Results are provided in
Table 7-3. Seven pathways were considered: air submersion, water
submersion, surface contamination, inhalation, and ingestion of
vegetables, meat and milk.
A number of factors contribute to the high population dose at the
urban site. There are about 2.5 million persons within 80 kilometers of
the urban site and about 0.5 million within the same distance of the rural
site. There also are larger numbers of meat animals (about 700,000 versus
200,000) and dairy animals (about 38,000 versus 14,000) and more vegetable
food crops produced (about 120,000 metric tons versus 30,000 metric tons)
within 80 kilometers of the urban site. Meteorological factors also
contribute to the higher dose at the urban site.
Interim Storage of Solidified Waste
Interim storage of solidified wastes would cause releases similar to
those for extended storage of spent fuel, which was found to have
negligible*impact. DOE and EPA have estimated that the storage of
137
-------
Table 7-1
Estimated Dose Equivalents to the Maximally Exposed Individual from
3H
106Ru
ALL
Total
(millirem
1 yr 5 yr
0.5 **
1 **
2 **
Dose Equivalents
Solidification
(Rural Site)
Body
per year)
10 yr
0.3
0.0
0.4
Lower Large Intestine
(millirem per year)
1 yr 5 yr 10 yr
0.5 ** 0.3
16 ** 0.1
20 ** 0.4
Table 7-2
to Maximum Individual from Solidification
(Urban Site)
Total Body
(millirem per year)
1 yr 5 yr 10 yr
3H
106Ru
ALL
1.6 **
8.7 **
11 **
1.6
0.1
2.2
Table 7-3
Population Doses from Waste
1 yr
5 yr
10 yr
Rural
Total Body
(Person-rem)
23
8
6
Site
LLI
( Organ- rem)
100
11
6
Lower Larger Intestine
(millirem per year)
1 yr 5 yr 10 yr
1.6 ** 1.9
130 ** 0.3
140 ** 2.4
Solidification
Urban Site
Total Body LLI
(Person-rem) (Organ-retn)
200 800
90 100
40 50
** not calculated
138
-------
solidified waste would have even smaller impact, since krypton-85 and
tritium, the only nuclides released in any quantity from spent-fuel
storage, are not contained in the solidified waste.
Emplacement
Emplacement of solidified wastes is essentially the same process as
emplacement of spent fuel. The doses estimated by DOE are extremely small
(DOE 79).
7.1.3 Collection and Disposal of Krypton-85 and Iodine-129 Wastes
Under our environmental standards for the Uranium Fuel Cycle
(40 CFR 190), krypton-85 and iodine-129 are considered to be wastes that
must be collected for disposal. The doses from krypton-85 and iodine-129
releases occurring at fuel reprocessing plants have been examined in
detail in connection with the development of the standards in 40 CFR 190.
Therefore, we undertook further analysis of only the doses associated with
these nuclides after removal from the reprocessing plants.
Krypton-85 Collection and Disposal Options
Our review indicates a number of collection methods suitable for
krypton-85 (AL 79b) . A model reprocessing plant would separate
12.7 million curies of krypton-85 from spent fuel as krypton gas. This
gas would be contained in cylinders, either as pressurized gas or adsorbed
on zeolite molecular sieves. Immobilization by ion implantation is being
139
-------
investigated but has not been as fully developed as the other techniques.
Either engineered storage or geologic disposal is suitable for final
disposal of the collected krypton.
DOE has estimated krypton leakage of 0.1 percent per year from
storage in cylinders as a pressurized gas (DOE 79). The gas may be stored
onsite for eventual release to the atmosphere. For this disposal option,
these releases would be in addition to discharges from normal operation.
Under these conditions it would require about 52 years to meet the Uranium
Fuel Cycle Standard limit of 50,000 curies per GWe-y. Disposal in a mined
geologic repository would provide smaller total releases, provided the
containers were sufficiently stable in the mine environment to prevent
increased leakage.
Iodine-129 Collection and Disposal Options
We also reviewed collection and disposal methods for iodine-129
(AL 79b). A generic reprocessing plant separates about 66 curies of
iodine-129 per year in a total of about 600 to 650 kilograms of iodine.
The iodine can be immobilized either as insoluble barium iodate in
concrete or by sorption on zeolites. The immobilized iodine would then be
suitable for terrestrial disposal, although extraterrestrial or seabed
disposal options are being considered.
No detailed characterization of the effects associated with
preparation of iodine-129 for final disposal is available. However, these
140
-------
effects can be compared with the effects of similar treatments of
Low-level and intermediate wastes, which are insignificant (DOE 79).
7.1.4 Extraterrestrial Disposal
Chemical processing is required if spent fuel is to be reprocessed
and is also required for transmutation and extraterrestrial disposal.
Both transmutation and extraterrestrial disposal of entire spent-fuel
elements appear to be impractical because of the amount of material
involved. Chemical processing is required to separate the long-lived
radionuclides from the shorter-lived elements. This reduces the material
to be transmuted or disposed extraterrestrially. DOE has prepared
environmental assessments of conventional disposal technologies for spent
fuel and for chemically reprocessed high-level liquid wastes (DOE 79,
DOE 80). Limited descriptions of the environmental effects of
technologies for extraterrestrial disposal and transmutation are available
(AL 79a, DOE 80).
DOE and EPA analyses of the operations required to prepare high-level
waste for extraterrestial disposal indicate that separation of the
actinide fraction of the wastes (partitioning) would be required to avoid
an excessive number of launches of space vehicles (AL 79a). After
separation, nonactinide wastes would be disposed by terrestial methods.
The separated actinide wastes would be packaged to make a waste form
resistant to dispersion in the event of a launch accident or failure to
achieve proper orbit. In some schemes, iodine-129 and other long-lived
fission products would also be sent into space.
141
-------
There is insufficient information for a detailed environmental
analysis of the release of radionuclides during preparation for
extraterrestrial disposal and in the event of accidents during launching
(AL 79a; DOE 80). Since partitioning of actinides is a chemical operation
similar to fuel reprocessing, it is reasonable to expect somewhat similar
releases.
7.1.5 Transmutation
Transmutation is the conversion of radionuclides into other stable or
radioactive nuclides, usually by irradiation with neutrons. Our review of
this disposal option indicates that it is in the early conceptual design
stage and that there is no firm information on potential releases or doses
(DOE 79). The process requires that the nuclides to be transmuted be
separated from the other materials. The actinides must be partitioned,
and it is usually necessary to further separate individual actinides
(fractionation). The actinides are then incorporated into fuel rods for
the irradiation. The requirements for transmutation are similar to known
processes in the nuclear fuel cycle—partitioning and fractionation are
similar to fuel reprocessing, and incorporation of actinides into fuel
rods is similar to fuel fabrication.
7.1.6 Other Separations
There have been many proposals for separation of various nuclides
preparatory to waste disposal (DOE 80): separation for special disposal
of long-lived nuclides (e.g., iodine-129, carbon-14, technetium-99) and
142
-------
separation of short-lived nuclides, particularly strontium-90 and
cesium-137, to minimize heat loading for geologic disposal. Other
proposals include more complete separations of lanthanides and actinides
for space disposal or chemical synthesis of a waste form. None of these
proposals is beyond the conceptual stage, and no information on
environmental releases or effects is available. All of these separation
techniques are chemical processes similar to fuel reprocessing.
7.2 DERIVATION OF THE STANDARDS
Our review of the waste management operations to be undertaken prior
to terrestrial disposal reveal that it is feasible, with available
technology and proper siting of facilities, to limit maximum individual
doses to those established by the Uranium Fuel Cycle Standard
(40 CFR 190). Many of the processes to be undertaken are similar to
processes regulated under 40 CFR 190. Therefore, we are confident these
processes can be regulated within these limits. The individual doses from
some chemical processes have not been fully evaluated, but we believe that
they can be expected to fall in the range of those associated with fuel
reprocessing plants. Storage operations for collected krypton-85 and
iodine-129 are also found to be within 40 CFR 190, although special
provisions must be made for adequate time of decay before releasing
krypton to the atmosphere.
143
-------
Although we have little information on the waste management
operations required for extraterrestrial disposal and transmutation, we
believe the limit established under 40 CFR 190 will not preclude these
options.
7.2.1 Selection of Standards Format
The proposed standards for waste management operations are expressed
in terms of limits on annual radiation dose to individuals and to body
organs. This is the format used in our environmental standards for the
Uranium Fuel Cycle, 40 CFR 190.10(a). For a number of reasons, it is
desirable to express the waste management operations standard in the same
format, unless further investigation shows this format is unsuitable:.
1. The Uranium Fuel Cycle standards are conceptually simple and
relatively easy to implement.
2. Many of the procedures covered by the waste management operations
standard are similar to operations covered by the Uranium Fuel
Cycle standards.
3. It is possible that facilities regulated by the waste management
operations standard will be sited adjacent to, or in proximity
to, facilities regulated under the Uranium Fuel Cycle standards.
It would simplify procedures to use the same format in regulating
these facilities.
144
-------
7.2.2 Selection of Numerical Limits
Our evaluations of waste management operations indicate that
population doses are relatively small and that limiting doses to
individuals is the optimal method to control radionuclide releases.
Estimated releases are small enough that they conform to the 25 mrem/year
limit established in 40 CFR 190.
Our estimates show that the highest individual doses for the waste
management operations preparatory to terrestrial disposal of either spent
fuel or reprocessed wastes are about 15 to 20 millirem per year at a rural
site and about 135 millirem per year at an urban site. These maximum
doses are caused by the solidification of 1-year-old liquid wastes. The
highest individual doses have not been estimated for operations serving
extraterrestrial waste disposal and transmutation. These operations are
similar to some of the operations in the Uranium Fuel Cycle and are,
therefore, expected to have similar doses.
Dose Limits Higher than those in 40 CFR 190
Setting permissible dose limits higher than those in 40 CFR 190
(e.g., a level of 250 millirem per year) would accomodate the doses
estimated for solidification at an urban site and would allow latitude for
other operations. We do not believe, however, that there is justification
for setting limits above those established by 40 CFR 190. Our
investigations indicate that there is no need for latitude in dose
limitations for any operations other than solidification. We estimate
145
-------
that only a few solidification facilities will be required to carry out
the waste management program and that most of these are expected to be
operated in conjunction with fuel reprocessing. Since fuel reprocessing
requires large exclusion areas, we see little or no problem with
restricting waste solidification to rural sites.
Therefore we reject dose limits higher than those in 40 CFR 190.
Dose Limits Lower than those in 40 CFR 190
Our assessment is that the waste management operations required for
terrestrial disposal could be carried out with annual dose limits to the
maximally exposed individual below those required by 40 CFR 190. The
principal requirement is that extended storage (for periods greater than
1 year) be provided before solidification of reprocessed wastes. There is
insufficient information to establish that the operations in support of
extraterrestrial disposal or transmutation could meet 40 CFR 190.
In the environmental impact statement for the 40 CFR 190 standards
(EPA 76), we considered the alternatives of reducing the annual whole-body
dose limits from 25 millirem to 15 millirem, then to 5 millirem, with
comparable reductions of permissible dose limits to thyroid and other
organs. We judged the improvements to be "not significant" and "small"
respectively. The improvements in public health from dose limits below
25 millirem for waste management operations would be even smaller and
would affect a smaller industry.
146
-------
Implementation of standards with lower limits than those in
40 CFR 190 would be difficult. Doses could be estimated by establishing
environmental transport models for measured effluent releases, but
confirmation of doses by direct measurement of environmental
concentrations of nuclides would be at or beyond the state of the art for
environmental monitoring.
Establishment of dose limits substantially more restrictive than the
40 CFR 190 limits might preclude operations supporting extraterrestrial
disposal and transmutation. Environmental protection must include all
parts of the management and disposal of spent fuel and high-level and
transuranic wastes. The possible benefits that might result from further
investigation of extraterrestrial disposal and transmutation outweigh the
small benefits of reducing the limits for the operational standards. The
Interagency Review Group has recommended further research on
extraterrestrial disposal options (IRG 79).
We reject dose limitation standards below those of 40 CFR 190,
because they would not provide significant reductions in risk. They would
be more difficult to implement, and they would preclude consideration of
ottier potentially useful technologies.
Limits Equal to those of 40 CFR 190
As discussed in section 7.2.1, administration of the individual dose
limits for the waste management operations standard is simplified if its
format is the same as that for the Uranium Fuel Cycle standards. There
147
-------
is, in our judgment, no justification for setting the individual dose
limits above those established by 40 CFR 190. There is little to be
gained in setting the limit lower than 40 CFR 190. Therefore, we have
chosen individual dose limits equal to those of 40 CFR 190 as the proposed
standard.
The proposed standard establishes limits of 25 millirem per year to
the whole body, 75 millirem per year to the thyroid, and 25 millirem to
any other organ of any member of the general population. These limits are
known to be readily itnplementable, are protective of the public, and do
not impose prohibitive costs. These dose limits would permit the use of
available monitoring capabilities and would require the use of the best
available technology to prevent excessive radiation exposure of the
public. We estimate that these limits would'not preclude extraterrestrial
disposal or transmutation.
7.2.3 Implications of the Operational Standards
The proposed waste management operations standards extend the
protection of 40 CFR 190 to similar activities not included under that
regulation. With the exception of transportation, generally applicable
environmental standards will apply to all operations in the generation of
commercial electrical power, from milling of uranium ore through
emplacement of wastes in their final disposal site.
148
-------
The proposed standards provide that doses to individuals from
combined operations covered by 40 CFR 190 and 40 CFR 191 shall not exceed
the limits of either part. This provision is analagous to the provision
that the dose limits for any individual from any combination of the
activities covered by 40 CFR 190 shall be the same as those from any
single activity. Together, the two standards prevent an individual from
receiving excessive doses from any combination of nuclear activities.
149
-------
150
-------
Chapter 8
ENVIRONMENTAL IMPACTS
Our proposed standards address the operations required to store,
prepare, and dispose of high-level and transuranic wastes, as well as the
long-term performance of their disposal systems over 10,000 years. The
intent of these standards is to minimize releases of radionuclides to the
environment and to assure that the risks to the public health and safety
are small.
8.1 HEALTH IMPACT
8.1.1 Standards for Waste Management Operations
The proposed standards for waste management operations limit the
maximum dose to an individual in the general population to 25 millirem per
year. Evaluations of emissions from existing facilities indicate that
actual doses can be expected to be much lower than 25 millirem per year,
and will generally range between 5 and 10 millirem per year. However, the
possible proximity of waste management facilities to other operations of
the uranium fuel cycle already governed by a limit of 25 millirem per year
under 40 CFR 190—and the difficulty of measuring such small emissions—
make it impractical to impose a smaller limit. In addition, we do not
want to preclude development of alternative disposal methods that, while
they may lead to slightly larger doses to some people during the
operational phase, could lead to a significantly lower long-term risk to
151
-------
future generations. The overall risk to populations, as opposed to the
risk to individuals, from releases from management operations for
high-level and transuranic waste disposal is expected to be small, because
we anticipate that only a few such facilities will be needed to process
both existing and future inventories. Thus, we conclude that the risk to
both individuals and to populations from management operations for
high-level waste disposal carried out according to these standards will be
small and will present no significant hazard to the public health and
welfare.
8.1.2 Disposal
The proposed standards for disposal set limits for releases of
specific radionuclides over the first 10,000 years after disposal. The
maximum impact expected over this time period from disposal in compliance
with these containment requirements is estimated to be about 10 excess
premature deaths from cancer for each 1000 metric tons of heavy metal
(MTHM) charged to the reactor. In making this estimate, we have used an
approach we believe makes the estimates conservatively large.
We expect that the first method to be utilized for disposal of
high-level radioactive and transuranic wastes will be disposal in a mined
repository deep in a continental rock formation. To obtain a basis for
judging whether such disposal will meet our standards, we have made risk
assessments for a model repository in different geologic media. The model
repository is similar to current concepts of repositories discussed in
152
-------
recent literature. The model assumes release of radionuclides might occur
by any of several modes after disposal in the repository. The
radionuclides considered are those that our evaluation indicates have a
potential to make a significant contribution to the total exposure
expected from the release of a fraction of the radioactive waste. Our
estimates show about 700 expected health effects from a repository in
granite and about 100 from a repository in bedded salt over 10,000 years.
We judge this impact to be small.
8.2 CONTAMINATION OF THE ENVIRONMENT
The limits that the standards for disposal set for releases to the
accessible environment generally prevent contamination of any part of that
environment to a level that would interfere with its use. We made the
health impact estimates discussed in -section 8.1.2 on the assumption that
the accessible environment was used without restriction once 100 years had
passed after disposal.
The standards for disposal do not, however, limit releases to sources
of groundwater within 10 kilometers from the emplacement of these
radioactive wastes. Our assessment of individual doses has shown that
inadvertent human intrusions for resource exploration could significantly
contaminate groundwater areas of a few tens of thousands of square
meters. More serious contamination could occur as a result of severe
improbable natural events such as faulting.
153
-------
However, the standards for disposal do tend to restrict the nature of
the sources of groundwater that could be contaminated. Because the
standards impose limits on the amounts of radioactive nuclides that can
reach groundwater more than 10 kilometers from the point of emplacement of
wastes, any unprotected groundwater must move very slowly. Slowly moving
groundwater is usually contained in relatively impermeable formations that
are poor producers of water and, therefore, are unlikely to be exploited.
Although a certain amount of groundwater may become contaminated—and lost
for potential future use—we believe that this minor resource loss would
be justified by the very good long-term protection that would be provided
by disposal systems complying with our standards.
8.3 CONCLUSION
The proposed standards for waste management operations limit doses to
persons in the general population to 25 mrem/year. In actual practice,
few people are expected to be exposed, and doses are expected to be
substantially below this limit. Therefore, we conclude that these waste
management operations do not represent a significant risk to the health
and safety of the population.
The maximum impact expected from a disposal system in compliance with
this standard is estimated to be less than 1,000 premature cancer deaths
over the first 10,000 years for disposal of the high-level wastes from
100,000 MTHM of reactor fuel (about as much as all currently operating
reactors should produce over their lifetimes). The environmental impacts
154
-------
of the alternatives we considered range up to about 10 times as large as
this; these impacts are discussed in Chapter 4. Our analyses indicate
that construction of repositories in bedded salt or granite can achieve
our proposed containment requirements, and repositories in other geologic
media may as well. Comparison with other environmental radiation hazards
indicates that disposal in compliance with our proposed standards would
represent a very small added risk to the population over such an extended
period of time.
155
-------
156
-------
Chapter 9
REGUIATORY IMPACT
This chapter reviews the projected costs associated with management
and disposal of high-level radioactive waste, and it summarizes the
potential regulatory impacts of our proposed standards. The details of
the analyses described here are presented in our Regulatory Impact
Analysis (EPA 82), prepared in accordance with Executive Order 12291.
The situation regarding the disposal of high-level waste is unusual
from a regulatory standpoint. In most cases, a regulation concerns an
ongoing activity. Any modifications that the regulation causes in the
activity may be considered to be costs that should be outweighed by the
regulatory benefits. For high-level waste disposal, however, the
appropriate regulations must be developed well before the activity to be
regulated can even begin. Thus, the typical perspectives about balancing
regulatory costs and benefits do not apply.
To investigate the potential impacts of this proposed action, we
evaluated how the costs of high-level waste management and disposal might
change due to alternative stringency levels of our containment
requirements—or due to changes in our assurance requirements. Because
there is no "baseline" program to consider, we could not quantify the
costs and benefits of our proposed action compared to the consequences of
no regulation.
157
-------
The most important benefit of our action should be the assurance that
these wastes will be disposed of with adequate protection of public health
and the environment. This assurance, in turn, should allow the Federal
program to proceed expeditiously to develop acceptable disposal methods at
appropriate sites. It may be argued that a further benefit would be the
resolution of a key issue that might lead to expanded commercial use of
nuclear power. This would be a benefit if nuclear power has clear
advantages, economic and otherwise, compared to alternative methods of
generating electricity; however, we have not analyzed this issue.
Our containment requirements consist of limits on potential releases
of radioactivity from a disposal system; these limits are to be used as
overall design objectives. These requirements are stated in terms of
projected releases for 10,000 years after disposal of the wastes. To
judge the risks associated with these release limits, we have used
generalized environmental pathway models to assess the potential health
impacts of the releases that would be allowed by our standards. However,
calculations of these "residual risks" are clearly not reliable as
absolute values, since projections of population distributions, ways of
life, and human behavior over 10,000 years cannot be meaningful. Rather,
these calculations are valuable only for understanding the relative
"residual risks" from different sources of radiation exposure (such as
risks from different disposal system designs, or risks from natural ore
bodies).
158
-------
For the release limits we have proposed as part of our environmental
standards, the residual risks projected by these models would be less than
1,000 premature deaths from cancer over the 10,000 year period, an average
of one premature death every 10 years. To judge the effects on disposal
costs of changing the level of protection, we also compared release limits
corresponding to residual risk values of: 100, 1,000, 5,000, and 10,000
premature deaths over the 10,000 year period. We chose this range of
residual risks because it appears to represent the range of performance
that may be expected of mined geologic repositories.
To do this analysis, we evaluated the long-terra performance of
generic models of geologic repositories in three different geologic media:
bedded salt, granite, and basalt. We did the analysis in two steps:
First, we used our performance projections to assess the quality of
the engineering controls that would be needed in each of the three model
repositories to meet each of the four different levels of protection.
In doing so, we encountered the problem that development of specific
engineered barriers (e.g., waste forms and canisters) has not yet
progressed enough to clearly associate the costs of manufacturing these
engineered barriers with their performance levels. Thus, we had to make
some rather speculative judgments to associate disposal costs with
alternative stringency levels. The results of this analysis are displayed
in Figure 9-1.
159
-------
c*
o
600
5 soo
S 400
VI
" 300
i
S 100
S
SALT REPOSITORY
(Engineering Barrier Costs Only)
600
400
200
GRANITE REPOSITORY
600'
400
200
BASALT REPOSITORY
100 1000 5000 10000 100 1000 5000 10000 100 1000 5000 10000
— . _ level of protection (health effects over 10,000 years)
FIGURE 9-1: VARIATIONS IN WASTE MANAGEMENT COST vs. LEVEL OF PROTECTION
-------
Second, we tried to allow for the possible effect of alternative
stringency levels on site selection. This is particularly relevant
because our analyses indicate that the most important part of the
protection offered by a mined geologic repository comes from the
hydrological and geochemical characteristics of the site itself.
The costs of usin8 a "good" site rather than a "bad" site (within the same
type of geologic media) do not involve differences in construction cost.
Instead, they involve the difficulty of finding a site that is "good
enough." Since there are so few data on site characterization, we have no
good basis for judging how many sites might have to be studied to meet
different levels. However, we did made some assumptions about how site
selection costs might increase in order to meet more stringent standards.
We then combined these assumptions with our evaluations of the variations
in engineered barrier costs to arrive at our second set of disposal cost
estimates. The results from this analysis are shown in Figure 9-2.
The results of these assessments of disposal costs and alternative
stringency levels indicate that the costs are not very sensitive to
different levels of protection, particularly for the geologic media
(bedded salt and granite) that are better at reducing long-term risks.
Even with our hypothesis about increased site selection costs with more
stringent levels, the difference in costs for different levels are much
smaller than the overall uncertainties in waste management costs. For
example, consider the increased costs of complying with the release limits
we have proposed, rather than release limits 10 times less stringent.
161
-------
(Engineering Barrier Costs and Site Selection Costs)
600
l| 500-
o>
.*
-v.
S 400
+j
VI
o
^ 300
i
in
s
200
100
SALT REPOSITORY
600 GRANITE REPOSITORY 600l BASALT REPOSITORY
400
200
400-
200
100 1000 5000 10000 100 1000 5000 10000 100 1000 5000 10000
. ... level Of protection (health effects over 10,000 years)
FIGURE 9-2: VARIATIONS IN WASTE MANAGEMENT COST vs. LEVEL OF PROTECTION
-------
The potential increase ranges from zero to 50 million (1981) dollars per
year. For comparison, the total costs of high-level waste management and
disposal (independent of our action) have been estimated as between
700 million and almost 1.5 billion (1981) dollars per year. Electrical
utility revenues were about 100 billion dollars in 1980.
These analyses, while indicating that disposal costs appear to be
relatively insensitive to differences in the level of protection, do not
provide a way to determine the acceptability of the residual risks from a
societal perspective, nor do they indicate a level of protection that is
preferable from a balancing of costs and benefits. One possible approach
to balancing costs and benefits would be to judge the cost per life saved
by different levels of protection, perhaps taking into account some method
of discounting costs and benefits. However, our calculations of residual
risks are not reliable as absolute values. Thus, we have no meaningful
way to calculate an absolute value of the cost per life saved by different
levels of protection.
In the absence of the ability to make meaningful cost and benefit
comparisons, we have used other tests of economic feasibility and
acceptability of risk to judge the appropriateness of the level of
protection we have proposed. As discussed above, setting the release
limits at the level we chose—as opposed to a level 10 times less or
10 times more stringent—appears to cause only very minor effects on the
costs of high-level waste disposal. To judge the acceptability of the
remaining long-term risk, we considered the risks that would otherwise be
163
-------
caused if the uranium ore used Co produce Che wastes had not been mined.
The magnitude of the risks from these unmined ore bodies is very uncertain
due, in part, to the wide variety of settings in which uranium ore is
found—many of which are closer to the surface than a geologic repository
would be. Using the same generalized environmental pathway models that
were used to assess the risks from our models of geologic repositories,
the risks from a comparable amount of untained uranium ore are estimated to
range from a few hundred to more than 1 million health effects over
10,000 years (WI 80). The lower end of this range is roughly equal to the
residual risk associated with our proposed release limits. Thus, the
upper limit of the risk that our standards would allow from the disposal
of high-level wastes appears to pose a threat very close to the minimal
risk posed by nature, had the uranium ore never been mined and the
high-level wastes never been generated.
Our assurance requirements provide seven qualitative criteria that
should reduce some of the uncertainties inherent in disposing of wastes
that must be isolated for a very long time. The specific provisions of
these assurance requirements are described in Chapter 6 of this document.
Only three of the criteria have a significant potential to increase the
costs of high-level waste disposal. These are:
Criterion 2, which calls for disposal systems to keep radioactive
releases as small as reasonably achievable;
Criterion 3, which calls for disposal systems to use multiple
barriers, botti engineered and natural; and
Criterion A, which restricts reliance on active institutional
controls to a resonable period after disposal (e.g., a few hundred
years).
164
-------
Each of these three assurance requirements might have the effect of
requiring better engineered barriers than would otherwise be needed to
meet our environmental standards. This would be particularly true for a.
repository sited in a relatively good geologic media (such as our generic
models for bedded salt or granite). However, even if no engineered
barriers at all appeared to be needed for long-term protection after
disposal, fairly protective canisters and waste forms would be needed for
other phases of waste management, such as transportation to and
emplacement in a repository. Therefore, we believe that these assurance
requirements would necessitate—at most—only moderate improvements in
waste form performance, and we judged that the impact that these
improvements might have on disposal costs should be less than 10 million
(1981) dollars per year. Since this impact concerns improvements to
engineered barriers, the potential cost increase would be duplicative of
any engineered barrier impacts caused by our environmental standards.
Thus, the potential cost effects of our containment requirements and our
assurance requirements should generally not be added together. (For some
unusual possibilities, adding the effects of the containment requirements
and the assurance requirements might be appropriate, but these
possibilities would tend to involve relatively small impacts.)
165
-------
166
-------
Appendix A
THE PROPOSED STANDARDS
167
-------
168
-------
A new Part 191 is proposed to be added to Title 40, Code of Federal
Regulations, as follows:
SUBCHAPTER F - RADIATION PROTECTION PROGRAMS
PART 191 - ENVIRONMENTAL RADIATION PROTECTION STANDARDS FOR
MANAGEMENT AND DISPOSAL OF SPENT NUCLEAR FUEL, HIGH-LEVEL AND
TRANSURANIC RADIOACTIVE WASTES
Subpart A - Environmental Standards for Management and Storage
191.01 Applicability
191.02 Definitions
191.03 Standards for Normal Operations
191.04 Variances for Unusual Operations
191.05 Effective Date
Subpart B - Environmental Standards for Disposal
191.11 Applicability
191.12 Definitions
191.13 Containment Requirements
191.14 Assurance Requirements
191.15 Procedural Requirements
191.16 Effective Date
AUTHORITY: The Atomic Energy Act of 1954, as amended; Reorganization Plan
No. 3 of 1970.
169
-------
SUBPART A - ENVIRONMENTAL STANDARDS FOR MANAGEMENT AND STORAGE
191.01 Applicability
This Subpart applies to radiation doses received by members of the
public as a result of the management (except for transportation) and
storage of spent nuclear fuel, high-level, or transuranic radioactive
wastes, to the extent that these operations are not subject to the
provisions of Part 190 of Title 40.
191.02 Definitions
Unless otherwise indicated in this Subpart, all terms shall have the
same meaning as in Subpart A of Part 190.
(a) "Spent nuclear fuel" means any nuclear fuel removed from a
nuclear reactor after it has been irradiated.
(b) "High-level radioactive wastes" means any of the following that
contain radionuclides in concentrations greater than those identified in
Table 1: (1) liquid wastes resulting from the operation of the first cycle
solvent extraction system, or equivalent, in a facility for reprocessing
spent nuclear fuels; (2) tJ-<=>. concentrated wastes from subsequent
extraction cycles, or equivalent; (3) solids into which such liquid wastes
have been converted; or (4) spent nuclear fuel if disposed of without
reprocessing.
(c) "Transuranic wastes," as used in this Part, means wastes
containing more than 100 nanocuries of alpha emitting transuranic
isotopes, with half-lives greater than one year, per gram of waste.
170
-------
(d) "Storage" means placement of radioactive wastes with planned
capability to readily retrieve such materials.
(e) "Management and storage" means any activity, operation, or
process, except for transportation, conducted to prepare spent nuclear
fuel, high-level or transuranic radioactive wastes for storage or
disposal, the storage of any of these materials, or activities associated
with the disposal of these materials.
(f) "General environment" means the total terrestial, atmospheric,
and aquatic environments outside sites within which any operation
associated with the management and'storage of spent nuclear fuel,
high-level or transuranic radioactive wastes is conducted.
(g) "Member of the public" means any individual who is not engaged
in operations involving the management, storage, and disposal of materials
covered by these standards. A worker so engaged is a member of the public
except when on duty at a site.
191.03 Standards for Normal Operations
Operations covered by this Subpart should be conducted so as to
reduce exposures to members of the public to the extent reasonably
achievable, taking into account technical, social, and economic
considerations. As an upper limit, except for variances in accordance
with 191.04, these operations shall be conducted in such a manner as to
provide reasonable assurance that the combined annual dose equivalent to
any member of the public due to: (a) operations covered by Part 190,
171
-------
(b) planned discharges of radioactive material to the general environment
from operations covered by this Subpart, and (c) direct radiation from
these operations; shall not exceed 25 millirems to the whole body,
75 millirems to the thyroid, or 25 millirems to any other organ.
191.04 Variances for Unusual Operations
(a) The implementing agency may grant a variance temporarily
authorizing operations which exceed the standards specified in 191.03 when
abnormal operating conditions exist if: (Da written request justifiying
continued operation has been submitted, (2) the costs and benefits of
continued operation have been considered to the extent possible, (3) the
alternatives to continued operation have been considered, and (4)
continued operation is deemed to be in the public interest.
(b) Before the variance is granted, the implementing agency shall
announce, by publication in the Federal Register and by letter to the
governors of affected States: (1) the nature of the abnormal operating
conditions, (2) the degree to which continued operation is expected to
result in doses exceeding the standards, (3) the proposed schedule for
achieving conformance with the standards, and (4) the action planned by
the implementing agency.
191.05 Effective Date
The standards in this Subpart shall be effective 12 months from the
promulgation date of this rule.
172
-------
SUBPART B - ENVIRONMENTAL STANDARDS FOR DISPOSAL
191.11 Applicability
This Subpart applies to radioactive materials released into the
accessible environment as a result of the disposal of high-level or
transuranic radioactive wastes, including the disposal of spent nuclear
fuel. This Subpart does not apply to disposal directly into the oceans or
ocean sediments.
191.12 Definitions
Unless otherwise indicated in this Subpart, all terms shall have the
same meaning as in Subpart A of this Part.
(a) "Disposal" means isolation of radioactive wastes with no intent
to recover them.
(D) "Barriers" means any materials or structures that prevent or
substantially delay movement of the radioactive wastes toward the
accessible environment.
(c) "Disposal system" means any combination of engineered and
natural barriers that contains radioactive wastes after disposal.
(d) "Groundwater" means water below the land surface in a zone of
saturation.
(e) "Lithosphere" means the solid part of the Earth, including any
groundwater contained within it.
173
-------
(f) "Accessible environment" includes (1) the atmosphere, (2) land
surfaces, (3) surface waters, (4) oceans, and (5) parts of the lithosphere
that are more than ten kilometers in any direction from the original
location of any of the radioactive wastes in a disposal system.
(g) "Reasonably foreseeable releases" means releases of radioactive
wastes to the accessible environment that are estimated to have more than
one chance in 100 of occurring within 10,000 years.
(h) "Very unlikely releases" means releases of radioactive wastes to
the accessible environment that are estimated to have between one chance
in 100 and one chance in 10,000 of occurring within 10,000 years.
(i) "Performance assessment" means an analysis which identifies
those events and processes which might affect the disposal system,
examines their effects upon its barriers, and estimates the probabilities
and consequences of the events. The analysis need not evaluate risks from
all identified events. However, it should provide a reasonable
expectation that the risks from events not evaluated are small in
comparison to the risks which are estimated in the analysis.
(j) "Active institutional controls" means (1) guarding a disposal
site, (2) performing maintenance operations or remedial actions at a
disposal site, or (3) controlling or cleaning up releases from a disposal
site.
174
-------
(k) "Passive institutional controls" means (1) permanent markers
placed at a disposal site, (2) public records or archives, (3) Federal
Government ownership or control of land use, or (4) other methods of
preserving knowledge about the location, design, or contents of a disposal
system.
(1) "Heavy metal" means all uranium, plutoniutn, or thorium placed
into a nuclear reactor.
191.13 Containment Requirements
Disposal systems for high-level or transuranic wastes shall be
designed to provide a reasonable expectation that for 10,000 years after
disposal:
(a) Reasonably foreseeable releases of waste to the accessible
environment are projected to be less than the quantities calculated
according to Table 2.
(b) Very unlikely releases of waste to the accessible environment
are projected to be less than ten times the quantities calculated
according to Table 2.
191.14 Assurance Requirements
To provide the confidence needed for compliance with the containment
requirements of 191.13, disposal of high-level or transuranic wastes shall
be conducted in accordance with the following requirements:
175
-------
(a) Wastes shall be disposed of promptly once disposal systems are
available and the wastes have been suitably conditioned for disposal.
(b) Disposal systems shall be selected and designed to keep releases
to the accessible environment as small as reasonably achievable, taking
into account technical, social, and economic considerations.
(c) Disposal systems shall use several different types of barriers
to isolate the wastes from the accessible environment. Both engineered
and natural barriers shall be included. Each such barrier shall
separately be designed to provide substantial isolation.
(d) Disposal systems shall not rely upon active institutional
controls to isolate the wastes beyond a reasonable period of time (e.g., a
few hundred years) after disposal of the wastes.
(e) Disposal systems shall be identified by the most permanent
markers and records practicable to indicate the dangers of the wastes and
their location.
(f) Disposal systems shall not be located where there has been
mining for resources or where there is a reasonable expectation of
exploration for scarce or easily accessible resources in the future.
Furthermore, disposal systems shall not be located where there is a
significant concentration of any material which is not widely available
from other sources.
(g) Disposal systems shall be selected so that removal of most of
the wastes is not precluded for a reasonable period of time after disposal.
176
-------
191.15 Procedural Requirements
Performance assessments to determine compliance with the containment
requirements of 191.13 shall be conducted in accordance with the following:
(a) The assessments shall consider realistic projections of the
protection provided by all of the engineered and natural barriers of a
disposal system.
(b) The assessments shall not assume that active institutional
controls can prevent or reduce releases to the accessible environment
beyond a reasonable period (e.g., a few hundred years) after disposal.
However, it should be assumed that the Federal Government is committed to
retaining passive institutional control of disposal sites in perpetuity.
Such passive controls should be effective in deterring systematic or
persistent exploitation of a disposal site, and it should be assumed that
they can keep the chance of inadvertent human intrusion very small as long
as the Federal Government retains such passive control of disposal sites.
(c) The assessments shall use information regarding the likelihood
of human intrusion, and all other unplanned events that may cause releases
to the accessible environment, as determined by the implementing agency
for each particular disposal site.
191.16 Effective Date
The standards in this Subpart shall be effective immediately upon
promulgation of this rule; however, this Subpart does not apply to wastes
disposed of before promulgation of this rule.
177
-------
TABLE 1 - CONCENTRATIONS IDENTIFYING HIGH-LEVEL RADIOACTIVE WASTES
Radionuclide Concentration
(curies per gram of waste)
Carbon-14 8x 10~6
Cesium-135 8x 10~4
Cesium-137 5x 10~3
Plutonium-241 3x 10~6
Strontium-90 7x 10~3
Technetium-99 3x 10~6
Tin-126 7x 10"7
Any alpha-emitting transuranic
radionuclide with a half-life Ix 10~7
greater than 20 years
Any other radionuclide with a half-life
_o
greater than 20 years 1x10
NOTE: In cases where a waste contains a mixture of radionuclides, it
shall be considered a high-level radioactive waste if the sum of the
ratios of the radionuclide concentration in the waste to the concentration
in Table 1 exceeds one.
178
-------
For example, if a waste containing radionuclides A, B, and C in
concentrations Ca, C^, and Cc, and if the concentration limits from
Table 1 are GLa, CL^, and CLC, then the waste shall be considered
high-level radioactive waste if the following relationship exists:
.ca cb cc
CLa CLb CLC
179
-------
TABLE 2 - RELEASE LIMITS FOR CONTAINMENT REQUIREMENTS
(Cumulative Releases to the Accessible Environment
for 10,000 Years After Disposal)
Radionuclide
Release Limit
(curies per 1000 MTHM)
Americiura-241 --------------------- 10
Americium-243 --------------------- 4
Carbon-14 200
Cesium-135 2000
Cesium-137 500
Neptunium-237 20
Plutoniura-238 400
Plutoniura-239 100
Plutonium-240 • 100
Plutoniura-242 100
Radiura-226 3
Strontium-90 80
Technetium-99 10000
Tin-126 80
Any other alpha-emitting
radionuclide -------------------- 10
Any other radionuclide which does
not emit alpha particles -------------- 500
180
-------
NOTE 1: The Release Limits in Table 2 apply either to the amount of
high-level wastes generated from 1,000 metric tons of heavy metal (MTHM),
or to an amount of transuranic (TRU) wastes containing one million curies
of alpha-emitting transuranic radionuclides. To develop Release Limits
for a particular disposal system, the quantities in Table 2 shall be
adjusted for the amount of wastes included in the disposal system.
For example:
(a) If a particular disposal system contained the high-level wastes
from 50,000 MTHM, the Release Limits for that system would be the
quantities in Table 2 multiplied by 50 (50,000 MTHM divided by 1,000 MTHM).
(b) If a particular disposal system contained five million curies of
alpha-emitting transuranic wastes, the Release Limits for that system
would be the quantities in Table 2 multiplied by five (five million curies
divided by one million curies).
(c) If a particular disposal system contained both the high-level
wastes from 50,000 MTHM and 5 million curies of alpha-emitting transuranic
wastes, the Release Limits for that system would be the quantities in
Table 2 multiplied by 55:
50,000 KTHM 5,000,000 curies TRU
+ ; „ 55
1,000 KTHM 1,000,000 curies TRU
181
-------
NOTE 2: In cases where a mixture of radionuclides is projected
to be released, the limiting values shall be determined as follows:
For each radionuclide in the mixture, determine the ratio between the
cumulative release quantity projected over 10,000 years and the limit
for that radionuclide as determined from Table 2 and Note 1. The sum
of such ratios for all the radionuclides in the mixture may not exceed
one.
For example, if radionuclides A, B, and C are projected to be
released in amounts Qa, Qb, and Qc, and if the applicable Release
Limits are RLa, RL^, and RLC, then the cumulative releases over
10,000 years shall be limited so that the following relationship exists:
Qa Qb Qc
KL RL RL
182
-------
Appendix B
RISK ASSESSMENTS OF GEOLOGIC REPOSITORIES
183
-------
184
-------
Our assessments of the health risks associated with disposal in mined
geologic repositories are intended to provide conservative estimates of
the potential risks from our model repository systems. (The repository
characteristics were chosen to be stringent enough for adequate
protection, but lenient enough to be reasonably achievable.) More
sophisticated risk assessments of these model repositories—using more
realistic repository performance characteristics—should predict less harm
than our estimates. Some potential repository sites should be more
environmentally acceptable than our models; others could be less
environmentally acceptable. The Department of Energy and the Nuclear
Regulatory Commission will perform assessments of specific candidate
repository sites.
Our risk assessments include estimates of the potential harm to a
representative population and estimates of the annual dose equivalents to
highly exposed individuals. We chose to express the harm to the
population as numbers of fatal cancers and numbers of genetic effects—and
the individual dose assessments in annual dose equivalent—rather than as
the quantities or concentrations of radioactive materials released to the
environment. This approach allows us to combine the effects of the
various radionuclides and to represent more clearly the potential risks of
high-level waste disposal.
By varying the hypothetical conditions of our generic model
repositories, we can explore a wide variety of circumstances. These
projections then become a tool for comparing various conditions as they
185
-------
affect repository performance. Some postulated conditions may produce
risks too nigh to be considered acceptable; others may greatly reduce the
risks but may be technologically or economically unattainable. The full
report of our population risk assessment and our individual dose
assessment are presented in Smith, et al. (SMC 82) and Goldin, et al.
(GO 82), respectively.
B.I GENERAL FEATURES OF THE MODEL
The model system used for our assessment is made up of five parts:
1. the waste container;
2. the physical and chemical characteristics of the waste itself
(the waste form);
3. the geologic medium into which the waste is placed (the host rock);
4. the geologic media between the host rock and man's
environment (the geosphere);
5. the environmental (i.e., above ground) pathways through which
people can come into contact with waste radionuclides
(the biosphere).
The first four parts are characteristics of the disposal system and
are common to both the population and the individual risk assessments; the
fifth part, while considered in both assessments, is treated differently
because of the different approaches needed to assess individual and
population exposures. These environmental pathways are discussed more
completely by Smith, Fowler, and Goldin (SMJ 82). The events and
processes leading to radionuclide releases are discussed in section B.2,
together with the population risk assessments. The individual dose
assessments are presented in section B.3.
186
-------
B.I.I Waste Characteristics
Table B-l lists the major radionuclides contained in the radioactive
inventory of our model repository. We selected these 15 radionuclides on
the basis of their quantity in spent fuel, their half-lives, and their
biological behavior as reflected in dose-equivalent conversion factors
(rem per curie ingested or inhaled). Table B-2 lists the geochemical
characteristics (i.e., retardation factors and solubility limits) of each
of these radionuclides.
Each model repository contains the waste from 100,000 metric tons of
heavy metal (MIHM) charged to reactors. We assume the waste is in the
form of unreprocessed spent fuel, since this assumption requires
examination of the behavior of both the fission products and the
transuranic nuclides. The composition by weight of the spent fuel is
assumed to be 95.5 percent uranium isotopes, 0.9 percent plutonium
isotopes, 0.1 percent other transuranic isotopes, and 3.5 percent fission
products. We assume all the spent fuel has aged for approximately 10
years after discharge from the reactors. More detailed information on the
repository inventory is given by Arthur D. Little, Inc. (ADL 79).
Continued integrity of the waste containers assures retention of the
radioactive wastes in the repository. If the containers fail, release of
the radioactive material depends on either the release rate of nuclides
from the fuel matrix (waste form) or the solubility of the radioactive
elements as they come into contact with the groundwater. For our
187
-------
Table B-l
Characteristics of Radioactive Waste
Radionuclide
Pu-238
Pu-239
Pu-240
Pu-242
Am-241
Am- 243
Np-237
Cs-135
Cs-137
1-129
Tc- 99
Sn-126
Zr- 93
Sr- 90
C- 14
(Spent Fuel)
Initial Quantity
in Repository (ADL 79)
(curies)
220,000,000
33,000,000
49,000,000
170,000
17,000,000
1,700,000
33,000
22,000
8,600,000,000
3,800
1,400,000
56,000
190,000
6,000,000,000
28,000
Half-Life
(years)
89
24,400
6,260
380,000
458
7,650
2,140,000
3,000,000
30
16,000,000
210,000
100,000
950,000
29
5,730
188
-------
Table B-2
Solubility Limits and Retardation Factors (SMC 82)
Solubility Limit
3
Radionuclide (Ci/m ) (ppm) Retardation Factor
Pu-239
Pu-240
Pu-242
Am-241
Am-243
Hp-237
Cs-135
Cs-137
1-129
Tc- 99
Sn-126
Zr- 93
Sr- 90
C- L4
6.0E-5
2.2E-4
4.0E-6
160
10
7.2E-7
na
na
na
2.0E-5
3 . OE-2
na
na
na
0.001
0.001
0.001
50.0
50.0
0.001
na
na
na
0.001
1,0
na
na
na
100
100
100
100
100
100
1
I
I
1
10
1
1
1
na = Solubility limit not used or not known
ppm = parts radionuclide per million parts solution by weight.
189
-------
reference cases, containers last 100 years in salt and 500 years in other
media. For the release rate, we assume that, after failure of the
container, a constant fraction of the remaining radionuclide dissolves per
unit time in the repository groundwater. For our reference cases, this
-4
fraction is taken to be 1 part in 10,000 (10 ) per year. The
solubility limits become important when there is not enough water to
dissolve the quantity of the radioactive element that could otherwise be
released. This is usually the case for technetium, uranium, neptunium,
and plutonium, and it also applies to americium in some situations.
B.I.2 Descriptions of Generic Repositories
Our population risk assessment report (SMC 82) and the Arthur D.
Little study (ADL 79) describe our model repositories for five different
geologic media—bedded salt, granite, basalt, shale, and salt dome. To
demonstrate our analytical method, we will discuss only the bedded salt,
granite, and basalt models, since, for our purposes, the model shale
repository behaves very similarly to basalt and the salt dome repository
behaves similarly to the bedded salt.
Each repository is 2 kilometers wide and 4 kilometers long. About
one-fourth of the repository is mined for the waste, the rest being left
as walls and pillars. The mined portion is 5 meters from floor to roof.
After the wastes have been placed in the repository, the mined areas are
backfilled. We assume that the backfill cannot be refilled to the
original density, which results in void space amounting to 20 percent of
its total volume, or 2,000,000 cubic meters.
190
-------
The model bedded salt repository (Figure 5-1 on page 89) is 460
meters (1500 feet) below the surface. Above and below the repository
layer are 50 meters of salt, 50 meters of impermeable rock, and a 30-meter
thick, porous, water-bearing medium (aquifer). The characteristics of the
overlying aquifer are shown in Table B-3. There are 330 meters of
undefined sedimentary overburden between the top of the upper aquifer and
the surface.
The granite repository (Figure 5-2 on page 90) is also modeled as
460 meters below the surface. The granite formation continues downward
indefinitely, so there is no lower aquifer. There are 230 meters of
granite above the repository and then an aquifer identical to the one
modeled above the salt repository. Above the aquifer are 200 meters of
overburden.
The oasalt repository (Figure B-l) is also positioned at 460 meters
underground, with 100 meters of basalt both above and below. The basalt
is bounded by overlying and underlying 30-meter thick aquifers at 330 and
560 meters, respectively.
In granite and basalt, the void spaces in the backfill fill with
groundwater moving from the aquifer through fractures in the bulk rock and
also through the sealed shafts and boreholes. We assume that all the
water becomes connected throughout the mined volume and shafts and can be
considered as one large volume. In the salt repository, the water may
191
-------
-Surface
Surface
— Deposits
*iTf>
«* Aquifer t^
460 meters
FIGURE B-l: REFERENCE REPOSITORY IN BASALT
192
-------
Table B-3
Aquifer Characteristics
Distance from repository
to overlying aquifer (meters)
Distance along aquifer to
man's environment (meters)
Permeability in aquifer
pathway (meters/year)
Transverse diffusivity (meters)
Gradient in aquifer pathway
Effective porosity in aquifer pathway
Thickness of aquifer (meters)
Width of aquifer over repository (meters)
230.0 (granite)
100.0 (bedded salt)
100.0 (basalt)
1600.0
31.5
6.0
0.01
0.15
30.0
4000.0
193
-------
enter through imperfections in shaft and borehole seals. Once the salt
repository has been closed, the salt begins to flow plastically under the
weight of the overlying salt and the other geologic formations, so that
most of the void space in the backfill is eliminated. Only a small amount
of the water may remain trapped in the salt in the form of brine pockets.
We assume that each brine pocket is in contact with two waste canisters
and contains 0.06 cubic meters of brine.
B.2 RISK ASSESSMENTS FOR POPULATIONS
In this section we estimate the number of serious health effects
(fatal cancers and genetic defects) that could result from releases of
radionuclides from a repository. The three objectives of our analysis
are: (1) to estimate the probabilities of accidental events, (2) to
address the overall impact on man, and (3) to provide information for
dec is ion-making.
We considered mechanisms and initiating events that would release
radionuclides from the model repository and their subsequent transport
through the geosphere and the biosphere (SMC 82). Releases were
categorized according to three general pathways to the biosphere: releases
to air, releases to land surface, and releases to aquifers (and ultimately
to surface water). A number of initiating events were then identified
that would allow radionuclides to enter one or more of these media:
194
-------
1. human intrusion (drilling for resources) — releases to land
surface and to aquifers
a. direct hit on a canister
b. intrusion into the repository without hitting a canister
2. faulting — releases to aquifers
a. destruction of canisters
b. disturbance of repository without destroying canisters
3. breccia pipe formation — releases to aquifers.
4. volcanoes — releases to air and land surface
5. meteorite impact — releases to air and land surface
For each repository type we examined a number of pathways in the
biosphere through which people may be exposed to radiation. The pathways
for releases to air were inhalation, eating contaminated foods, exposure
to external radiation from contaminated air, and exposure to external
radiation from contaminated land surface. The same pathways apply to
releases to land surface since the nuclides deposited on land surface will
be resuspended into the air. We also calculated the population risks from
use of the streams into which the aquifers discharge. The pathways for
releases to an aquifer and subsequent discharge from a stream are drinking
water, eating fish, eating milk, meat, and irrigated foods, breathing air
contaminated through resuspension, being expoaured to external radiation
from contaminated air, and being exposured to external radiation from
contaminated land surfaces. We also calculated risks from eating seafood
contaminated by entry of nuclides into the ocean from the rivers, but
these risks were very small. We calculated the fraction of each nuclide
ingested or inhaled by humans for each subpathway and then determined the
resulting doses to each of 10 organs: ovaries, testes, red marrow,
195
-------
thyroid, lower gastrointestinal tract (LLI), lung, liver, kidney, bone,
and total body (considered representative of other soft tissue). We used
the dose equivalent conversion factors (retn per curie) of INREM-II (KI 78,
DU 79) to calculate the organ doses. The ovary and testes doses were
converted to genetic health effects, and the doses to the other organs
were converted to fatal cancers. The genetic and fatal cancer
dose-to-health effect conversion factors are based on the work of the BEIR
Committee (NAS 72). For all the situations we considered, the number of
potential genetic effects was always small relative to the projected
number of fatal cancers; therefore, this discussion concentrates on the
latter category of health effects.
B.2.1 Assessment Methodology
The consequences we assessed for a population are the number of fatal
cancers produced by a release over the period of time considered. We
examined the probability and consequences of each event occuring at a
number of different times. The probabilities and consequences of all the
possible events and event times can be combined in two ways. First, we
can present the results in a graph in which the ordinate of a point on the
graph gives the probability of obtaining at least the number of
consequences shown on the abscissa. (For example, Figure B-2 illustrates
total fatal cancers caused by releases from a granite repository in the
first 10,000 years after repository sealing—it shows a probability greater
than 1 in 10 of a few hundred fatal cancers in this time and less than a
1 in 100,000 probability of much more than a thousand fatal cancers.)
196
-------
GRANITE
REPOSITORY
101 102 103 104 105
health effects over 10,000 years
10'
FIGURE B-2: PROBABILITY OF POPULATION RISKS OVER 10,000 YEARS
197
-------
Alternatively, we can multiply each consequence by its probability and sum
the products, obtaining an expected value for the total number of fatal
cancers resulting from all events over the time period considered.
(Figure 5-3 on page 99 displays the sums of the product of the
probabilities and consequences for each type of event for the bedded salt
and granite repositories—a total for each type of repository is then
obtained by summing over all events.) In the remainder of our discussion
of population risks, we will use the latter value to illustrate and
compare the harm associated with different characteristics of disposal
systems.
B.2.2 Releases
Table B-4 indicates the major repository characteristics that control
the amount and rate of radionuclide release. Releases to air and to land
surfaces are direct and rapid; the major controlling factor is the
fraction of the waste affected. Releases through an aquifer are indirect
and slow; the major controlling factors are the fraction of the waste
affected and the rate of flow of contaminated water through the aquifer.
The flow rate depends on the permeability, porosity, cross-sectional area
of the flow path and the hydraulic gradient driving the flow. In the
immediate vicinity of the repository, a major component in the hydraulic
gradient is the buoyancy resulting from heat produced by the waste; any
natural gradient between aquifers adds to the thermal buoyancy gradient.
The maximum value of the thermal buoyancy gradient is about 0.25, which
gradually decreases to zero as the waste cools (SMC 82). Table B-4 gives
the particular characteristics associated with each of the release modes
we analyzed.
198
-------
Table B-4
Reference Case Characteristics (SMC 82)
Normal Groundwater Flow (to surface water via aquifer)
Permeability of host rock (m/yr)
Porosity of host rock
2
Cross-sectional flow area (m )
Hydraulic gradient
Drilling-direct hit of a canister (to land surface)
Fraction of repository inventory affected
Probability (yr )
Drilling-does not hit canister (to land surface)
Fraction of repository inventory affected
3
Volume of repository water involved (m )
3
Volume of repository water released (m )
Probability (yr )
Faulting (to surface water via aquifer)
Fraction of repository directly disrupted
Fraction of repository inventory affected
•3
Void volume of repository affected (m )
Permeability in fault zone flow path (m/yr)
Porosity in fault zone flow path
2
Cross-sectional area of fault zone (m )
Hydraulic gradient
Granite
3xlO~4
lxlO~4
8xl06
thermal
4xlO~6
2.5xlO~6
1.0
2xl06
200.0
0.0025
0.003
1.0
2xl06
3150.0
0.1
4000.0
thermal
Bedded Salt
—
—
4xlO~6
2xlO~5
6xlO~5
0.06
0.06
0.02
0.003
0.05
'IxlO5
31.5
0.1
4000.0
0.1 plus
thermal
Basalt
3xlO~3
lxlO~4
8xl06
0.1 plus
thermal
4xlO~6
lxlO~5
1.0
2xl06
200.0
0.01
0.003
1.0
2xl06
3150.0
0.1
4000.0
0.1 plus
thermal
Probability (yr )
2x10
-8
2x10
-8
5x10
-7
199
-------
Table B-4 (continued)
Reference Case Characteristics (SMC 82)
Granite
Bedded Salt
Basalt
Breccia Pipe (to surface water via aquifer)
Fraction of repository directly disrupted
Fraction of repository inventory affected
Permeability in breccia zone flow path (ra/yr)
Porosity in breccia zone flow path
2
Cross-sectional area of breccia zone (m )
Hydraulic gradient
Probability after first 500 yr (yr )
0.004
0.016
3150.0
0.2
3xl04
0.1 plus
thermal
1x10
-8
Volcano (to air and land surface)
Fraction of repository inventory affected
Probability (yr )
0.004
0.004
0.004
6X10'10
Meteorite (to air and land surface)
Fraction of repository inventory affected
Probability (yr )
0.001
-11
4x10
0.001
-11
4x10
0.001
4x10
200
-------
Normal Groundwater Flow
The rock surrounding a granite or basalt repository is permeable, and
water moving through it could transport radioactive material to the
overlying aquifer. The buoyancy effect resulting from the decay heat of
the waste drives water through rock fractures in accordance with the
permeability shown in Table B-4. In this model, water flows upward into
the aquifer over the entire area of the repository and is replenished
through fractures in the rock around the repository.
The release mechanism we considered for a bedded-salt repository is
more complicated. Water may enter the repository from the upper aquifer
through imperfections in the sealed openings, especially the main shaft
seal. At the same time, plastic deformation of the salt caused by
lithostatic pressure from the overburden eliminates these void spaces.
This phenomenon tends to squeeze any water from the void spaces back up
the shaft seals. If the canisters have failed, the water squeezed back up
the shafts will contain some radionuclides from the waste. Contaminated
water could then be released from the shafts into the overlying aquifer.
However, in the situations we felt were likely, too little water leaked
into the repository to cause any releases (ADL 79, SMC 82).
201
-------
Drilling
Human intrusion (drilling) involves either a direct hit on the waste
or intrusion into any water contacting the waste. A direct hit results in
transportation of a fraction of the waste directly to the land surface and
opens direct communication between the repository and aquifer(s) through
the borehole. Release mechanisms for drilling intrusions into the water
are similar to those for direct hits on the waste. For a granite or
basalt repository, the material brought to the surface is a small fraction
of the pool of contaminated water in the entire repository. For a
bedded-salt repository, we assume that only one of a large number of small
brine pockets around the waste would be affected. We further assume that
intrusion into a brine pocket would cause ejection of the entire contents
of the pocket directly to the land surface because of the high lithostatic
pressure of the salt (ADL 79).
Faulting
Fault movement can open a high-permeability pathway for release of
radioactive material to the overlying aquifer. The heat produced by the
decaying radioactive material in the repository (thermal buoyancy) then
drives the flow of water from the repository to the aquifer. In bedded
salt and basalt, the hydraulic gradient between aquifers also affects the
flow. In granite, recharge is from the surrounding bulk rock; in salt-and
basalt, recharge proceeds principally through the interconnection of the
lower aquifer with the upper aquifer.
202
-------
If waste is directly disturbed by the fault, radionuclides from
damaged canisters can enter the water flowing to the aquifer at a rate
that may be limited either by the release rate from the waste matrix or by
the solubility of the radionuclides. Also, if canisters fail for other
reasons—permitting radionuclides to leach or dissolve into the repository
water—and a fault occurs later, this contaminated water can be
transported to the overlying aquifer.
Breccia Pipes
A breccia pipe is a permeable formation resulting from the collapse
of rock into cavities formed by the dissolution of underlying rock.
Breccia pipes might occur near bedded salt-repositories, but they are not
relevant to the other geologic media we studied. The release mechanisms
to the overlying aquifer are the same as in faulting, but the flow area is
larger than the flow area of the fault, and the permeability of the flow
path is greater than that for faulting.
Meteorites and Volcanoes
Meteorites and volcanoes that disrupt a repository would pulverize
parts of the repository and its host rock and would eject radioactive
material into the atmosphere and onto the land surface. We assume that
material ejected into the air would become uniformly mixed into the
troposphere and would then settle onto the surface of the earth. Nuclides
would reach people through the pathways previously described in
Section B.I. Either type of event has an extremely small probability.
203
-------
B.2.3 Reference Case Results
Table B-5 presents the projected number of fatal cancers resulting
from the release of radionuclides by the release modes identified in
Table B-4. For the bedded-salt repository, drilling events that miss the
canisters but hit a brine pocket constitute the dominant type of
initiating event, causing about 200 fatal cancers, primarily from releases
of araericium-241 (44 percent) and americium-243 (55 percent). For the
granite repository, about 750 fatal cancers are projected in the first
10,000 years, due mainly to drilling into the repository water and to
releases due to normal groundwater flow. The major causes of health
effects are, again, americium-241 (10 percent) and americium-243
(&9 percent). For all model repositories the risks from faulting,
meteorites, volcanoes, and breccia pipes are insignificant. The dominant
initiating event in the basalt repository is also drilling into the
contaminated water in the repository; this could result in 3,000 fatal
cancers over the 10,000 years, but releases from normal groundwater flow
would add another 1,400 projected effects.
204
-------
Table B-5
Projected Population Risks Over 10,000 Years:
Reference Cases*
Projected Health Effects
Repository Routine Drilling
Type Release Faulting (No hit) (Hit)
Granite 10 + 750 +
Bedded Salt 0 + 180 8
Basalt 1,400 3 3,000 2
Breccia Volcano;
Pipe Meteorite Total
+ 760
+ + 190
+ 4,400
* From Table 7-4, SMC 82
# "No hit" means the drill does not hit solid waste but only repository water
while "hit" indicates the drill does hit solid waste.
+ = Less than 1 projected fatal cancer
— = not applicable
205
-------
B.2.4 Sensitivity Analyses
To this point, we have examined our reference cases for calculating
releases due to natural groundwater flow and several initiating events.
Using our assumed values, our risk assessment estimated numbers of health
effects for the reference cases. However, for a broader insight into the
possible consequences of radionuclide releases to the biosphere, we have
varied the values of several parameters of the repository models. These
variations involved changes in canister life, different release rates
(sometimes called "leach rates") from the waste form, different
retardation factors for radionuclides in the aquifer, and different
solubility limits of the radionuclides.
Different Canister Life and Waste Form Characteristics
We varied the values for canister failure time and waste matrix leach
rate from the reference repository characteristics. The projected health
effects in Table B-6 show the effect of these changes.
The primary effect of longer canister lifetimes is to lengthen the
period over which risks from human intrusion and normal groundwater flow
can be lowered or eliminated. Over 10,000 years, canister lifetime has
much less effect on the health risk than does the waste-form release
rate. There is a slight decrease in the risk value when increasing the
canister lifetime from the reference age to 1,000 years. However, a
canister lifetime of 5,000 years, rather than 1,000 years, significantly
reduces the projected number of fatal cancers over 10,000 years.
206
-------
Table B-6
Projected Population Risks Over 10,000 Years:
Various Canister Lives and Waste Form Release Rates
Canister Life
Reference Case (100 years)
(500 years)
1000 years
5000 years
Waste Form Release Rate
Reference Case
(10-4 yr'1)
High Estimate
(10~2 yr"1)
Low Estimate
(10~6 yr'1)
Projected Health Effects
Granite Bedded Salt Basalt
760
575
120
760
2,500
10
190
4,400
90 3,900
40 180
190 4,400
200 18,000
50
50
— = not applicable
207
-------
Release-rate changes have much larger effects on the resulting risks
than canister life changes, and release-rate changes have more effect for
the granite and basalt repositories than for the bedded-salt repository.
The risks for the bedded-salt repository do not increase much above those
for a release rate of 10 parts of remaining waste per year (ppy)
because of the solubility limits of the americium isotopes. Risks for the
granite and basalt repositories do not increase much as the release rate
_3
increases above 10 ppy because there is no more waste to dissolve
during the 10,000 years.
Finally, although not shown here, studies of simultaneously varying
the waste matrix leach rate and canister lifetimes reinforced the previous
observation that the leach rate has more effect on the overall risk than
does the canister lifetime.
Solubility of Radionuclides
We examined the model repositories for unlimited solubility of all
the radionuclides so that entry of all nuclides into groundwater would be
limited only by the leach rate. Table B-7 shows that solubility limits
are extremely important in limiting projected releases from the model
repositories and resulting fatalities, especially for bedded salt. In the
model bedded salt repository, the most significant solubility limit is
that of the americium isotopes in brine pockets.
208
-------
Table B-7
Projected Population Risks Over 10,000 Years:
Various Retardation Factors and Solubility Limits
Projected Health Effects
Granite Bedded Salt Basalt
Solubility Limits
Reference Case * 760 190 4,400
No Solubility Limits 2,800 14,700 12,600
Retardation Factors
Reference Case * 760 190 4,400
All Retardation Factors
Equal to 1 38,000 200 5 million
* see Table B-2
209
-------
Retardation Factors in the Aquifer
Table B-7 also presents the projected health effects for two
different choices of retardation factors: our reference case assumptions
about retardation factors (see Table B-2), and a case that assumes that
all retardation factors are equal to one (no retardation). Lower
retardation produces many more health effects from expected releases or
from faulting at granite and basalt repositories. Releases from bedded-
salt repositories depend only slightly on retardation, because releases to
an aquifer are much smaller than releases directly to the land surface.
Genetic Health Effects
In all types of repositories analyzed, the number of projected
genetic effects are at least one order of magnitude less than the number
of projected cancers, because most of the radionuclides involved do not
concentrate in the gonads. For the reference case release modes—over
10,000 years following repository closure—we project 15, 2, and 150
genetic effects from releases from granite, bedded salt, and basalt
repositories, respectively. In comparison, we project 760, 190, and 4,400
fatal cancers for these three repositories.
B.2.5 Conclusions
Several broad conclusions can be drawn from these performance
assessments. First, major changes in the geochemistry at a site can
affect long-term risks much more than major changes in the engineered
barriers. For example, neglecting geochemical retardation for a granite
210
-------
repository increases the consequences from about 800 health effects to
38,000. In comparison, assuming that the waste form dissolves very
quickly raises risks to a little more than 3,000, while assuming a zero
lifetime for the waste canister increases risks only to about 1,000.
Thus, it appears that efforts to identify a repository site with
appropriate characteristics can have greater benefits than efforts to
improve engineering controls.
Second, comparing the two types of engineering controls, variations
of waste-form release rate consistently have more effect on long-term
risks than variations of canister lifetime. Improvements in waste form
appear to provide more benefits than improvements in waste canisters.
Finally, good engineering controls, particularly good waste forms,
can overcome poor site characteristics. Our generic model of a basalt
repository assumes that relatively large amounts of groundwater are
available to dissolve and transport waste. In spite of this disadvantage,
our basalt model can achieve risks comparable to those at the low end of
the range for our granite model if the waste form used with basalt is
about an order of magnitude better than that used with granite.
211
-------
B.3 DOSE ASSESSMENT FOR INDIVIDUALS
We assessed the annual dose equivalent (hereafter referred to as
annual dose, for convenience) that could be received by individuals who
breathe contaminated air or drink contaminated water from a well drilled
close to a repository. These pathways are less complex than those in the
population risk assessment because they do not include the ingestion of
foodstuffs or water from a river. We used the model repositories and
geology described in Section B.I.2, and we examined several possible
pathways through which individuals could be exposed. For each case, we
calculated the annual doses to individuals and, for those cases involving
releases through a borehole, the extent (i.e., area) of the aquifer that
would be contaminated as a result of the releases. We chose four of the
most significant scenarios to describe in this Appendix:
1. releases of solid waste to the land surface from striking waste
while drilling;
2. releases to groundwater from striking contaminated repository
water while drilling;
3. releases to groundwater from a fault movement that allows
releases of water from within the repository, but that does not
directly disrupt the waste and;
4. releases to the aquifer from natural groundwater flow.
The probabilities and consequences of the four initiating events
differ markedly. Over 10,000 years, it is quite likely that someone will
inadvertantly drill into a repository (assuming that institutional
controls have no effect beginning 100 years after disposal), thereby
opening a small pathway through which relatively small quantities of
radionuclides could be released. Fault movement, on the other hand, is
212
-------
unlikely at a well-chosen site—but, if a fault should develop, it would
open a large pathway through which large quantities of radionuclides could
be released. Finally, natural groundwater flow is inevitable in the
non-salt repositories beginning at the time of closure. However, the
consequences will be small at an offsite well due to the slow movement of
the radionuclides.
B.3.1 Pathway Characteristics
The preceding section described the events and processes we
considered—for the individual dose analyses—that may allow radionuclides
to reach an aquifer or the earth's surface. In this section we trace the
pathways that those events and processes open from the waste to an
individual.
Geospheric Pathways
Drilling. Drilling in search of resources will leave porous material
in the resulting borehole through which contaminated water from the
repository can flow. This contaminated water flows upward to the
overlying aquifer mainly because of the heat produced by radioactive
decay. In the cases of the basalt and bedded-salt repositories, it is
assumed that the underlying aquifer is connected to the repository—
resulting in a hydraulic gradient, which (conservatively) is assumed to
drive water to the overlying aquifer. Upon reaching the aquifer, the
contaminated water moves downstream from its point of origin, the
borehole. Simultaneously, it very slowly spreads transversely across the
213
-------
aquifer, contaminating a parabolically shaped area. At some future time,
according to our model, an individual might unknowingly drill a well
within the contaminated area to obtain drinking water. At the time the
person has ingested the water, the pathways caused by drilling are
complete.
Faulting. The faulting event is postulated to open communication
between the repository and an aquifer along a line that runs the full
length of the repository—involving a much larger amount of radionuclides
than drilling. The faulting event does not release radionuclides to the
land surface, since we assume the implementing agencies would not permit
selection of a site where an aquifer is under such pressure as to be
artesian. (We also assume that the fault movement is much too small to
transport radioactive waste directly to the surface.)
We assumed the fault would create a 10-meter wide zone of high
permeability. The fault could cross the aquifer in any direction. A
fault in the direction of the aquifer flow was chosen for analysis since
it would produce the highest individual doses from faulting, because waste
from many individual canisters along the fault would contribute to the
dose rather than just a few if the fault were oblique to the canister row.
We considered two faulting scenarios. In one, the fault strikes and
breaks a row of canisters; in the other, the canisters have failed before
the fault develops, and the contents of two or more rows of failed
214
-------
canisters have combined in the porous volume of the repository backfill.
However, only the pathway from a fault intersecting repository water is
presented because these releases are much larger than releases restricted
to direct effects on canisters. This pathway then terminates when the
contaminated aquifer water is ingested by someone.
Normal Groundwater Flow. We also analyzed the natural groundwater
flow pathway through the repository to the overlying aquifer. We looked
only at the basalt repository, because this allowed us to estimate the
upper-bound dose rate from natural groundwater flow. (The groundwater
flow rate through the basalt repository is the greatest normal flow rate
associated with any of the types of repositories we analyzed.) There is
very little water flow through the bedded-salt repository since the
plasticity of the salt allows it to seal cracks and fissures that may
develop. In the case of the granite repository, there is no lower aquifer
to produce a hydraulic gradient—so the only driving force to take
nuclides to the overlying aquifer is the heat produced by the decaying
waste.
Atmospheric Pathway
This pathway is straightforward and also is a result of drilling.
Material reaches the surface either as a solid (a core sample) or a liquid
(contaminated water from the repository).
215
-------
The waste material brought to the surface may be removed by the
drilling crew, it may be covered by other residue from the borehole, or it
may lie at the top of the pile of residue. The latter scenario provides
the most conservative estimate and is the one we used. The material is
assumed to be a point source. We did not calculate the direct radiation
dose from this release, which may be very large, because it depends on the
way the material is handled, on the length of time the drillers remain
exposed, and on other factors. We considered this incident to be an
industrial accident not pertinent to this work. However, if the material
becomes airborne it could be spread beyond the drilling site and could
eventually expose people in the general population. In this analysis, the
pathway ends when the material is inhaled and has exposed an individual's
lungs. We also consider exposure from submersion in contaminated air and
direct exposure from the contaminated surface; however, these doses were
found to be very small compared to the inhalation pathway and are not
discussed in this Appendix.
B.3.2 Dose Analysis
A computer code called MAXDOSE-EPA was used to perform the individual
dose analyses (SE 81). The results of our study (GO 82) are complex
because of the various pathways, the many variations of the parameters,
and the number of data points corresponding to the variety of years after
the initiating event for which we analyzed. Besides this, there is much
data concerned with areas contaminated to various levels. Rather than
attempt to summarize all of these results in this section, we present
results for only those events that occur 1,000 years after the repository
216
-------
is sealed (except for shaft seal leakage that begins at 500 years, when
the canisters fail). The method used to calculate the annual individual
dose rates leads to a rate equal to the exposure in the 50th year of
chronic exposure, which is how all our individual dose equivalent rates
should be interpreted. Chapter 5 of Goldin, et al., (GO 82) presents the
entire dose rate and contaminated land area results for the granite and
bedded-salt repositories.
A drill that hits solid waste brings 15 percent of the waste of one
canister to the surface. The largest dose rate is incurred by breathing
contaminated air close to the drilling site (20 meters) 10 years after the
event. The dose rate is about 11 rera per year (rem/yr) to the lungs,
mostly from americium-241. At longer times after the drilling,
plutoniunr-239 and -240 become dominant as the dose rates fall from about
4 rem/yr at 1,000 years after the event to about 20 mrem/yr at 10,000
years.
The same direct-hit drilling event will also allow nuclides to be
released to the aquifer via the borehole. The largest dose rate from
drinking contaminated aquifer water is about 600 rem/yr to red bone
marrow, again 20 meters from the original point of release but 1,000 years
after the event. No areas are contaminated enough to give more than 0.5
rem/yr until 1,000 years when americium-241 has traveled the 20-meter
distance and contaminated 0.33 hectares (3,300 square meters). The
contaminated area then increases to 6 hectares as the nuclides spread and,
after 10,000 years, decreases because of radioactive decay.
217
-------
Releases caused by a fault intersecting a granite repository are—as
would be expected—much larger than those associated with drilling. The
doses range from 40 rem/yr at 10 years and 20 meters from the repository
to 3.4x10 rem/yr at 100 year.s and 20 meters. The latter dose rate
slowly decreases after 100 years to 2,600 rem/yr at 10,000 years.
Tin-126 dominates the dose rate for 20 years following the faulting event;
americium-241 then dominates until 5,000 years, when americium-243 takes
over. At 10,000 years, plutonium-239 delivers the greatest doses in this
scenario—which is very unlikely to occur (see section B.3.3).
Doses from natural groundwater flow are not considered in
MAXDOSE-EPA. Instead, we used an equation from section 4.1.1 of GO 82 to
describe the release rate from the repository, and then we used
equation 3-7 from the population risk assessment report (SMC 82) to relate
the radionuclide flow rate from the repository to the nuclide flow rate at
the well used for drinking water. We then related this flow rate to a
radionuclide concentration and converted the concentration to a dose rate,
using the dose conversion factors found in Table 3-4 of GO 82. The
results of this analysis are presented in Table B-8. Within 100 meters of
the repository there are potentially high dose rates. However, beyond 100
meters the potential annual doses are, with one exception, below 1 rem per
year. The dominant nuclide close to the repository is americum-243 while
cesium-135 generally dominates at greater distances—this is indicative of
the longer half-life and greater mobility of cesium-135.
218
-------
Table B-8
Dose Equivalent Rates (rem/yr) from Drinking Groundwater
Contaminated by Normal Groundwater Flow
Through a
Basalt Repository
Years after Closure
Distance from
Repository (m)
20
100
1000
2000
4000
8000
10,000
15,000
20,000
1000
1.5E+1*
94%
Sn-126
1.5E+1
94%
Sn-126
9.8E-1
74%
Cs-135
O.OE+0
0%
None
0 . OE+0
0%
None
O.OE+0
0%
None
O.OE+0
0%
None
O.OE+0
0%
None
O.OE+0
0%
None
2500
1.5E+5
68%
Am-243
1.3E+1
94%
Sn-126
8.1E-1
78%
Cs-135
8.0E-1
78%
Cs-135
7.8E-1
80%
Cs-135
O.OE+0
0%
None
O.OE+0
0%
None
O.OE+0
0%
None
O.OE+0
0%
None
5000
8.5E+4
75%
Am-243
l.OE+1
94%
Sn-126
5.9E-1
82%
Cs-135
5.9E-1
83%
Cs-135
5.8E-1
84%
Cs-135
5.6E-1
87%
Cs-135
O.OE+0
0%
None
O.OE+0
0%
None
O.OE+0
0%
None
10,000
3.4E+4
72%
Am-243
2.5E+4
70%
Am- 14 3
5.7E+0
94%
Sn-126
3.3E-1
89%
Cs-135
3.3E-1
90%
Cs-135
3.2E-1
92%
Cs-135
3.2E-1
92%
Cs-135
3.1E-1
94%
Cs-135
O.OE+0
0%
None
*1.5E+1 = 15 rem/yr
94% = 94% of the 15
Sn-126.
rem/yr is from
219
-------
B.3.3 Probabilities of Failure Modes
In this section we provide information useful for considering the
probabilities of some of the potential individual doses just discussed.
It should be noted, however, that the following probabilities of
inadvertant human intrusion assume that all institutional controls—both
active and passive—will be ineffectual beginning 100 years after
disposal. Thus, we believe that these probabilities for human intrusion
may be conservatively high.
Direct Hit on Waste
From the information given in the ADL report (ADL 79), the probability
of a drill hitting a waste canister is estimated to be 2x10 hits per
year in a bedded-salt repository, 3x10 hits per year in a granite
repository, and 1x10 hits per year in a basalt repository.
Drilling into Contaminated Repository Water
According to the ADL study (ADL 79), the frequency, and therefore the
probability, of drilling into the granite repository is once every 400
years, or 2x10 per year. The probability of drilling into a basalt
-2
repository is given as 1x10 per year. The probability of drilling
into a brine pocket in a salt repository may be conservatively estimated
to be the same as the drilling frequency into any part of a salt
repository—even though it would likely be lower because brine pockets do
not exist throughout the repository. However, using the more conservative
approach, ADL estimated 21 drilling attempts in the first 1,000 years—
_2
which results in a probability of 2x10 hits per year.
220
-------
Fault Movement
ADL (ADL 79) sets the probability of a fault occurring as 2xlO~8
per year for bedded salt and granite and 5xlO~7 per year for basalt.
However, we believe that only 10 percent of the faults would be lengthwise
(i.e., parallel or within 18 degrees of the longest repository axis).
These would produce the effects studied here, but there is a 90 percent
probability that the effects would be less than those given here.
B.3.4 Probability of Exposure
Air
The area around the point where the waste material is brought to the
surface following a direct hit on a waste canister—where the air and
ground are contaminated enough to result in a level of 500 mrem/yr—was
found to have a radius of 170 meters. Since the radioactivity persists
for hundreds or thousands of years and the radioactive material is assumed
not to be reburied, it is virtually certain that someone would eventually
encounter some small level of exposure.
Water
The probability of a person encountering a contaminated area of the
aquifer can be estimated by the probability of drilling for drinking water
in the contaminated area, the area of the contamination, and the length of
time the area will remain contaminated. There is no simple way to display
the many results of these calculations; however, an example will
illustrate the procedure.
221
-------
As was discussed in Section 5.3 of this document, the average well
-4
drilling density in the United States is 6.4x10 well per hectare per
century. Through calculations performed by the computer code MAXDOSE-EPA,
it was found that the area contaminated above 500 mrem/yr at 10,000 years
is 8 hectares. The probability of a well being drilled into the
contaminated area is then (6.4x10 ) times 8 — or about 5x10 per
year.
B.3.5 Conclusion
Considering the failure model probabilities and the associated dose
equivalents, we conclude that the chance of individuals receiving large
doses from disruptive events in a repository is small.
222
-------
REFERENCES
(ADL 79) Arthur D. Little, Inc., 1979. Technical Support of
Standards for High-Level Radioactive Waste Management.
Vol. A.-D and Addendum. U.S. Environmental Protection
Agency (EPA/520/4-79-007), Washington, B.C.
(AL 79a) Altomare, P.M., R. Bernardi, D. Gabriel, D. Nainen,
W. Parker and R. Pfundstein, 1979. Alternative
Disposal Concepts for High-Level and~Transuranic
Radioactive Waste Disposal.U.S. Environmental
Protection Agency (ORP-CSD 79-1), Washington, D.C.
(AL 79b) Altomare, P.M., M. Barbier, N. Lord and D. Nainen,
1979« Assessment of Waste Management of Volatile
Radionuclides. U.S. Environmental Protection Agency
(ORP-CSD 79-2), Washington, D.C.
(APS 78) American Physical Society, Study Group on Nuclear Fuel
Cycles and Waste Management, 1978. The Nuclear Fuel
Cycle: An Appraisal. Ch. VII, "High-Level Waste and
Transuranic Waste Management." Rev. Mod. Phys. 50,
S107-S142. —
(BNWL 76) Battelle-Pacific Northwest Laboratory, 1976.
Alternative Process for Managing Existing Commercial
High-Level Radioactive Wastes. U.S. Nuclear Regulatory
Commission (NUREG-0043), Washington, D.C.
(BO 80) Bogen, K. T. and A.S. Goldin, 1980. Natural Background
Radiation Exposure in the United States.
U.S. Environmental Protection Agency (ORP/SEPD 80-12),
Washington, D.C.
(CO 78) Cohen, J.J. and W.C. King, 1978. Determination of a
Radioactive Waste Classification System.University of
California Radiation Laboratory (UCRL-52535)
Livermore, California, '
(CO 79) Cochran, T.B., D. Rotow and A.R. Tamplin, 1979.
Radioactive Waste Management. Natural Resources
Defense Council, Inc., Washington D.C.
(DOE 79) U.S. Department of Energy, May 1979. Environmental
Aspects of Commercial Radioactive Waste Management.
U.S. Department of Energy (DOE/ET-0029),
Washington, D.C.
223
-------
(DOE 80) U.S. Department of Energy, October 1980. Final
Environmental Impact Statement: Management of
Commercially Generated Radioactive Waste.
U.S. Department of Energy (DOE/EIS-0046F),
Washington, D.C.
(DOE 81) U.S. Department of Energy, September 1981. Spent Fuel
and Radioactive Waste Inventories and Projections as of
December 31, 1980. U.S. Department of Energy
(DOE/NE-0017), Washington, D.C.
(DU 79) Dunning, Jr., D.E., S.R. Bernard, P.J. Walsh,
G.G. Killough, and J.C. Pleasant, 1979. Estimates of
Internal Dose Equivalent to 22 Target Organs for
Radionuclides Occurring in Routine Releases from
Nuclear Fuel Cycle Facilities, Vol. II. Oak Ridge
National Laboratory (NUREG/CR-0150, Vol. 2;
ORNL/NUREG/TM-190/V2), Oak Ridge, Tennessee.
(EL 77) Ellett, W.H. and A.C.B. Richardson, 1977. Origins of
Human Cancer, "Estimates of the Cancer Risk Due to
Nuclear Electric Power Generation." Cold Spring Harbor
Laboratory, Cold Spring Harbor, New York.
(EPA 76) U.S. Environmental Protection Agency, November 1976.
Final Environmental Statement, 40 CFR 190,
Environmental Radiation Protection Requirements for
Normal Operations of Activities in the Uranium Fuel
Cycle. U.S. Environmental Protection Agency
(EPA 520/4-76-016), Washington, D.C.
(EPA 77) U.S. Environmental Protection Agency, February 3-5,
1977. Proceedings: A Workshop on Issues Pertinent to
the Development of Environmental Protection Criteria
for Radioactive Wastes, Reston, Virginia.
U.S. Environmental Protection Agency (ORP/CSD 77-1),
Washington, D.C.
(EPA 78a) U.S. Environmental Protection Agency, 1978. State of
Geological Knowledge Regarding Potential Transport of
High-Level Radioactive Waste from Deep Continental
Repositories; Report of an Ad Hoc Panel of Earth
Scientists. U.S. Environmental Protection Agency
(EPA 520/4-78-004), Washington, D.C.
(EPA 78b) U.S. Environmental Protection Agency, March 30-April 1,
1978. Proceedings of a Public Forum on Environmental
Protection Criteria for Radioactive Wastes, Denver,
Colorado. U.S. Environmental Protection Agency
(ORP/CSD-78-2), Washington, D.C.
224
-------
(EPA 80)
(EPA 82)
(ERDA 75)
(ERDA 77a)
(ERDA 77b)
(ERDA 77c)
(FI 55)
(FO 71)
U.S. Environmental Protection Agency, 1980.
Radiological Quality of the Environment in the United
States. U.S. Environmental Protection Agency
(EPA 520/1-79-008), Washington, D.C.
U.S. Environmental Protection Agency, 1982. Draft
Regulatory Impact Analysis for 40 CFR 191;
Environmental Standards for Management and Disposal of
Spent Nuclear Fuel, High-Level and Transuranic
Radioactive Wastes. U.S. Environmental Protection
Agency (EPA 520/1-82-024), Washington, B.C.
U.S. Energy Research and Development Administration,
1975. Final Environmental Impact Statement: Waste
Management Operations, Hanford Reservation. Richland,
Washington. Vol. 1 of 2. U.S. Energy Research and
Development Administration (U.S. Department of Energy)
(ERDA 1538), Washington, D.C.
U.S. Energy Research and Development Administration,
Sept. 1977. Alternatives for Long-Term Management of
Defense High-Level Radioactive Wastes, Hanford
Reservation. Richland, Washington. U.S. Energy
Research and Development Administration
(U.S. Department of Energy) (ERDA 77-44),
Washington, D.C.
U.S. Energy Research and Development Administration,
Sept. 1977. Alternatives for Long-Term Management of
Defense High-Level Radioactive Wastes, Idaho Chemical
Processing Plant. Idaho Falls. Idaho. U.S. Energy
Research and Development Administration
(U.S. Department of Energy) (ERDA 77-43),
Washington, D.C.
U.S. Energy Research and Development Administration,
Sept. 1977. Alternatives for Long-Term Management of
Defense High-Level Radioactive Wastes, Savannah River
Plant, Aiken, South Carolina. U.S. Energy Research and
Development Administration (U.S. Department of Energy)
(ERDA 77-42), Washington, D.C.
Fix, P.P., 1955. Hydrogeochemical Exploration for
Uranium. U.S. Geological Survey Professional Paper
300, pp. 667-671. U.S. Geological Survey, Reston,
Virginia.
Foster, R.J., 1971. Physical Geology.
Publishing Co., Columbus, Ohio.
Charles Merrill
225
-------
(GE 73) Geraghty, James J., D.W. Miller, F. Van Der Leeden, and
F. L. Troise, 1973. Water Atlas of the United States.
Water Information Center, Inc., Port Washington, New
York.
(GO 82) Goldin, A.S., R.K. Struckmeyer, B. Serini, and
C-Y Hung, 1982. Draft Report, Potential Individual
Doses from Disposal of High-Level Radioactive Wastes in
Geologic Repositories. U.S. Environmental Protection
Agency (EPA 520/1-82-026), Washington, B.C.
(GR 81) Griswold, George B. 1981. Solution Mining in Salt
Domes of the Gulf Coast Embayment. Pacific Northwest
Laboratory, Battelle Memorial Institute, (PNL-3190),
Richland, Washington.
(HE 79) Healy, J.W. and J.C. Rodgers, 1979. Limits for the
Burial of the Department of Energy Transuranic Wastes.
Los Alamos Scientific Laboratory (IA-UR-79-100),
Los Alamos, New Mexico.
(HO 80a) Holconb, William F., 1980. Radiation Exposures from
Normal Operations in the Management and Disposal of
High-level Radioactive Wastes and Spent Nuclear Fuel.
U.S. Environmental Protection Agency (EPA/3-80-008),
Washington, D.C.
(HO 80b) Holcomb, William F., William N. Crofford, R.L. Clark
and F.C. Sturz, 1980. Radiation Exposures from
Solidification Processes for High-Level Radioactive
Liquid Wastes. U.S. Environmental Protection Agency
(EPA 520/3-80-007), Washington, D.C.
(ICRU 71) International Commission on Radiation Units and
Measurements, July 1, 1971. Radiation Quantities and
U"its. International Commission on Radiation Units and
Measurements (Report No. 19), Bethesda, Maryland.
(ICRU 73) International Commission on Radiation Units and
Measurements, Sept. 1, 1973. Dose Equivalent.
International Commission on Radiation Units and
Measurements (Supplement to Report No. 19), Bethesda,
Maryland.
(IRG 79) Interagency Review Group on Nuclear Waste Management,
March, 1979. "Report to the President". National
Technical Information Service (TID-28817), Springfield,
Virginia 22161.
226
-------
(KI 78) Killough, G.G., D.E. Dunning, Jr., S.R. Bernard, and
J.C. Pleasant, 1978. Estimates of Internal Dose
Equivalent to 22 Target Organs for Radionuclides
Occurring in Routine Releases from Nuclear Fuel Cycle
Facilities, Vol. 1. Oak Ridge National Laboratory
(NUREG/CR-0150; ORNL/NUREG/TM-190), Oak Ridge,
Tennessee.
(NAS 72) National Academy of Sciences, Advisory Committee on the
Biological Effects of Ionizing Radiation, 1972. The
Effects on Populations of Exposure to Low Levels of
Ionizing Radiation. National Academy of Sciences,
Washington, D.C.
(NAS 79) National Academy of Sciences, 1979. Implementation of
Long-Term Environmental Radiation Standards; The Issue
of Verification. National Academy of Sciences,
Washington, D.C.
(NAS 80) National Academy of Sciences, Advisory Committee on the
Biological Effects of Ionizing Radiation, 1980. The
Effects on Populations of Exposure to Low Levels of
Ionizing Radiation. National Academy of Sciences,
Washington, D.C.
(NCRP 75) National Council on Radiation Protection and
Measurements, 1975. Natural Background Radiation in
the United States. National Council on Radiation
Protection and Measurements (Report No. 45, Vol. A),
Washington, D.C.
(NE 81) Newcomb, W.E., 1981. Status of the Geologic Disposal
Programs in the United States. U.S. Department of
Energy, Washington, D.C.
(NRC 80) U.S. Nuclear Regulatory Commission, 1980. Advance
Notice of Proposed Rulemaking, 10 CFR Part 60:
"Technical Criteria for Regulating Disposal of
High-Level Radioactive Waste," Federal Register.
Vol. 45, No. 94, page 31393.
(NRC 81a) U.S. Nuclear Regulatory Commission, 1981. Draft
Environmental Impact Statement on 10 CFR Part 61:
' Licensing Requirements for Land Disposal of
Radioactive Waste." U.S. Nuclear Re^ilat-^ry Commission
(NUREG-0782), Washington, D.C.
227
-------
(NRC 81b) U.S. Nuclear Regulatory Commission, 1981. Notice of
Proposed Rulemaking, 10 CFR Part 60: "Disposal of
High-Level Radioactive Wastes in Geologic
Repositories," Federal Register, Vol. 46, No. 130,
page 35280.
(NU 77) Nuclear Energy Policy Study Group, 1977. Nuclear Power
Issues and Choices, Ch. 8, "Radioactive Waste."
Ballinger Publishing Co., Cambridge, Massachusetts.
(NW 79) National Water Association, 1979. Deep Low Yield Wells
in the United States; A Survey of Selected States.
National Water Well Association Research Facility,
Worthington, Ohio.
(OA 72) Oakley, D.T., 1972, Natural Radiation Exposure in the
United States. U.S. Environmental Protection Agency
(ORP/SID 72-1), Washington, D.C.
(SE 81) Serini, B.L. and C.B. Smith, 1981. Maxdose-EPA: A
Computerized Method for Estimating Individual Doses
from a High-Level Radioactive Waste Repository.
U.S. Environmental Protection Agency
(EPA 520/4-81-006), Washington, D.C.
(SMC 82) Smith, C.B., D.J. Egan, W.A. Williams, J.M. Gruhlke,
C.Y. Hung, and B. Serini, 1982. Population Risks from
Disposal of High-Level Radioactive Wastes in Geologic
Repositories. U.S. Environmental Protection Agency
(EPA 520/3-80-006), Washington, D.C.
(SMJ 82) Smith, J.M., T.W. Fowler and A.S. Goldin, 1982.
Environmental Pathway Models for Estimating Population
Risks from Disposal of High-Level Radioactive Waste in
Geologic Repositories. U.S. Environmental Protection
Agency (EPA 520/5-80-002), Washington, D.C.
(UN 77) United Nations Scientific Committee on the Effects of
Atomic Radiation, 1977. Sources and Effects of
Ionizing Radiation. Report to the General Assembly,
United Nations, New York, New York.
(WI 80) Williams, W.A., 1980. Population Risks from Uranium
Ore Bodies. U.S. Environmental Protection Agency
(EPA 520/3-80-009), Washington, D.C.
(WI 81) Wild, R.E., O.I. Oztunali, J.J. Clancy, C.J. Pitt, and
E.D. Picazo, 1981. Data Base for Radioactive Waste
Management. U.S. Nuclear Regulatory Commission
(NUREG/CR-1759), Washington, D.C.
228
------- |