&EPA
United States
Environmental Protection
Agency
Office of
Radiation Piogiams
Washington, D.C. 20460
EPA B2Qn -844)22-1
October 1984
Radiation
Radionuclides
Background information
Document For Final Rules
Volume I
r
-------
TECHNICAL REPORT DAT A
fPlcaie read Insiruciions on the reverse before completing}
[.REPORT NO. 2,
EPA 520.1-84-022-1
J, TITLE AND SUBTITLE
Radionuclides. Background Information Document for
Final Rules. Volume I.
1. AUTHOFUS)
Office ol Radiation Programs, EPA. PEI Associates.
Battelle Columbus Laboratories. Kilkelly Associates
>. PERFORMING ORGANIZATION NAME AND ADDRESS
PEI Associates, Inc.
11499 Chester Road
Cincinnati, Ohio 45246-0100
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Radiation Programs
Washington, D.C. 20460
3. RECIPIENT'S ACCESSION NO. m-
PISS 16575HA>
5. REPORT DATE
October 1984
8. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
1O. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-3878
Work Assignment. No. 2
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
1fi. ABSTRACT
On October 31, 1984, EPA published a notice in the Federal Register withdrawing
proposed standards for radionuclide emissions from four sources: 1} DOE facilities,
2) NRC-licensed facilities and non-DOE Federal facilities, 3) underground uranium
mines and 4) elemental phosphorus plants. This Background Information Document
supports the Agency's final actions on radionuclides. Vo]'ime I is an integrated Isk
assessment. It addresses historical and current regulatory programs and strategies,
hazard identifications (health effects), radionuclide emissions, reduction of dose
and risk, movement of radionuclides through environmental pathways, radiation dosime-
try, and estimating the risk of health effects resulting from radionuclide air
emissions. Volume II examines the source categories and presents the following
information for each category: a general description of the source category, a
brief description of the processes that lead to the emissions of radionuclides into
air, a summary of emissions data, and estimates of the radiation doses and health
risks to both individuals and populations.
IV. KEY WORDS AND DOCUMENT ANALYSIS
1. DESCRIPTORS
Air Pollution
Clean Air Act
Radionuclides
Radiobiology
Radiation dosimetry
18. DISTRIBUTION STATEMENT
Unlimited
b.lDENTIFlERS/OPEN ENDED TERMS
National Emission
Standards for Hazardous
Air Pollutants (NESHAP)
Radionuclides
Clean Air Act
Health Physics
19. SECURITY CLASS fTliis Repivtl
Unclassified
2D, SECURITY CLASS i Tliis pagci
Unclassified
c. COSATI Held/Group
21. NO. OF PAGES
291
22. PBiCE
EPA Form 2220-1 IS*-?. 4-77} PREVIOUS EDITION is OBSOLETE
" 11,S. 50VERKXSCT PR.STIXG OFFICE: I9B4-66I-2ZIj
-------
40 CFR Part 61 EPA 520/1-84-022-1
National Emission Standards
for Hazardous Air Pollutants
BACKGROUND INFORMATION DOCUMENT
(INTEGRATED RISK ASSESSMENT)
FINAL RULES FOR RADIOMUCLIDES
VOLUME I
October 22, 1984
U.S. Environmental Protection Agency
Office of Radiation Programs
Washington, D- :. 20460
-------
CONTENTS
Page
Figures viil
Tables x
1. INTRODUCTION 1-1
l.l History of Standards Development 1-1
1.2 Purpose of the Final Background Information Document 1-2
1.3 Scope of the Final Background Information Document 1-2
1.4 EPA's Computer Codes 1-4
2. CURRENT REGULATORY PROGRAMS AND STRATEGIES 2-1
2.1 Introduction 2-1
2.2 The International Cotmnission on Radiological Pro-
tection and the National Council on Radiation
Protection and Measurements 2-2
2.3 Federal Guidance 2-8
2.4 The Environmental Protection Agency 2-10
2.5 Nuclear Regulatory Commission 2-12
2.5.1 Fuel Cycle Licenses 2-12
2.5.2 Byproduct Material Licenses 2-13
2.6 Department of Energy 2-13
2.7 Other Federal Agencies 2-14
2.7.1 Bureau of Radiologies! Health 2-14
2.7.2 Mine Safety and Health Administration 2-14
2.7.3 Occupational Safety and Health
Administration 2-14
2.7.4 Department of Transportation 2-14
2.8 State Agencies 2-15
3. HAZARD IDENTIFICATION 3-1
3.1 Evidence That Radiation Is Carcinogenic 3-1
3.2 Evidence That Radiation Is Mutagenlc 3-6
3.3 Evidence That Radiation Is Teratogenic 3-9
3,4 Uncertainties 3-10
3.5 Summary of Evidence That Radiation is a Carcinogen,
Mutagen, and Teratogen 3-11
4. EMISSION OF RADIONUCLIDES INTO THE AIR 4-1
4.1 Introduction 4-1
4.2 Sources of Radionuclide Releases into the Air 4-2
iii
-------
CONTENTS (continued)
Page
4.2,1 Department of Energy (DOE) Facilities 4-2
4.2.2 Nuclear Regulatory Commission (NRC)
Licensed Facilities and non-BOE
Federal Facilities 4-13
4.2.3 Coal-Fired Utility and Industrial
Boilers 4-19
4.2.4 Underground Uranium Mines 4-19
4.2.5 Phosphate Sock Processing and Wet-
Process Fertilizer Plants 4-20
4.2.6 Elemental Phosphorus Plants 4-22
4,2.7 Mineral Extraction Industry Facilities 4-23
4.2.8 Uranium Fuel Cycle Facilities, Uranium
Mill Tailings, High-Level Waste
Management 4-27
4.2.9 Low-Energy Accelerators 4-31
4.3 Radionuclide Releases 4.3I
4.3.1 Department of Energy Facilities 4-32
4.3.2 NRC-Licensed Facilities and Non-DOE
Federal Facilities 4-32
4.3.3 Coal-Fired Utility and Industrial
Boilers 4-35
4.3.4 Underground Uranium Mines 4-35
4,3.5 Phosphate Rock Processing and Wet-
Process Fertilizer Plants 4-36
4.3.6 Elemental Phosphorus Plants 4-36
4,3.7 Mineral Extraction Industry 4-37
4.3.8 Uranium Fuel Cycle Facilities, Uranium
Tailings, High-Level Waste Management 4-38
4.4 Uncertainties $_4j
5. REDUCTION OF DOSE AND RISK 5_!
5.1 Introduction 5_l
5.1.1 Emission Control Technology 5-1
5.1.2 Work Practices 5-1
5.1.3 Impact of Existing Regulations on
Strategies for Reducing Emissions 5-2
5,2 Emission Control Technology 5-2
5.2.1 Scrubbers 5_3
5.2.2 Filters 5.4
5.2.3 Mechanical Collectors and Electrostatic
Precipitators 5_|Q
5.2,4 Charcoal Adsorbers 5-10
5.2.5 Miscellaneous Emission Control Equipment 5-13
5.3 Work Practices 5_l5
5.4 Summary of Emission Reduction Strategies 5-18
5.5 Uncertainties in Evaluation of Control Technology
Efficiencies 5-18
iv
-------
CONTENTS (continued)
Page
6. MOVEMENT OF RADIQNUCLIDES THROUGH ENVIRONMENTAL PATHWAYS 6-1
6.1 Introduction 6-1
6.2 Dispersion of Radionuclides through the Air 6-3
6.2.1 Introduction 6-3
6.2.2 Air Dispersion Models 6-4
6.2,., Uncertainties in Atmospheric Dispersion
Modeling 6-8
6.3 Deposition on Atmospheric Radionuclides 6-9
6.3,1 Introduction 6-9
6.3.2 Dry Deposition Model 6-9
6.3.3 Wet Deposition Model 6-9
6.3.4 Soil Concentration Model 6-10
6.3.5 Uncertainties 6-11
6.4 Transport through the Food Chain 6-11
6.4.1 Introduction 6-11
6.4.2 Concentration in Vegetation 6-12
6.4.3 Concentration in Meat and Milk 6-13
6.4.4 Summary 6-14
6.5 Calculating the Environmental Concentration of
Radionuclides: The AIRDOS-EPA Code 6-14
6.5.1 Introduction 6-14
6.5.2 AIRDOS-EPA 6-15
7. RADIATION DOSIMETRY 7-1
7.1 Introduction 7-1
7.2 Definitions 7-1
7.2.1 Activity 7-1
7.2.2 Exposure and Dose 7-2
7.2.3 External and Internal Exposures 7-2
7.2.4 Dose Equivalent 7-3
7.3 Dosimetry Models 7-3
7.3.1 Internal Doses 7-3
7,3.2 External Doses 7-8
7.3.3 Effects of Decay Products 7-8
7.3.4 Dose Rate Estimates 7-9
7.4 EPA Dose Calculation 7-9
7.4.1 Dose Rates 7-9
7.4.2 Exposure and Usage 7-10
7.5 Uncertainty Analysis 7-11
7.5.1 Dose Uncertainty Resulting from Indi-
vidual Variation 7-12
7.5.2 Dose Uncertainty Resulting from Age 7-14
7.5.3 Dose Uncertainty Caused by Measurement
Errors 7-23
7.6 Distribution of Doses in the General Population 7-23
7.7 Summary 7-29
-------
CONTENTS (continued)
Page
8. ESTIMATING THE BISK OF HEALTH EFFECTS RESULTING FROM RADIO-
NUCLIDE AIR EMISSIONS 8-1
8.1 Introduction 8-1
8.2 Cancer Risk Estimates for Low-LET Radiation 8-3
8,2.1 Afsutap'.icms Needed to Make Risk
t.-v'.nates 8-4
8.2.2 DC Response Functions 8-4
8.2.3 The Possible Effects of Dose Rate on
Radiocarciiwgenesis 8-6
8.2.4 Risk Projection Models 8-7
8.2.5 Effect of Various Assumptions on the
Numerical Risk Estimates 8-9
8.2.6 Comparison of Cancer Risk Estimates for
Low-LET Radiation 8-10
8.2.7 EPA Assumptions about Cancer Risks
Resulting from Low-LET Radiations 8-12
8.2.8 Methodology for Assessing the Risk of
Radiogenic Cancer 8-13
8.2.9 Organ Risks 8-14
8.2.10 Methodology for Calculating the Pro-
portion of Mortality Resulting
from Leukemia 8-19
8.2.11 Cancer Risks Due to Age-Dependent Doses 8-20
8.3 Fatal Cancer Risk Resulting from High-LET Radiations 8-21
8.3.1 Quality Factors for Alpha Particles 8-22
8.3.2 Dose Response Function 8-22
8.3.3 Assumptions Mac? by EPA for Evaluating
the Dose from Alpha Particle Emitters 8-23
8.4 Estimating the Risk Resulting from Lifetime
Population Exposures from Radon-222 Progeny 8-25
8.4.1 Characterizing Exposures to the General
Population vis-a-vis Underground
Miners 8-26
8.4.2 The EPA Model 8-28
8.4.3 Comparison of Risk Estimates 8-29
8.5 Uncertainties in Risk Estimates for Radiogenic
Cancer 8-33
8.5.1 Uncertainty in the Dose Response Models
Resulting from Bias in the A-bomb
Dosimetry 8-33
8.5.2 Sampling Variation 8-37
8.5.3 Uncertainties Arising from Model
Selection 8-37
8.5.4 Summary 8-39
8.6 Other Radiation-Induced Health Effects 8-41
8.6.1 Types of Genetic Harm and Duration of
Expression 8-41
vi
-------
CONTENTS (continued)
Page
8.6.2 Estimates of Genetic Ham Resulting from
Low-LET Radiations 8-44
8.6.3 Estimates of Genetic Harm for High-LET
Radiations 8-50
8.6.4 Uncertainty in Estimates of Radiogenetic
Ham 8-50
8.6.5 The EPA Genetic Risk Estimate 8-54
8.6.6 Teratogenic Effects 8-56
8.6.7 Nonstochastic Effects 8-63
8.7 Radiation Risk - A Perspective 8-63
9. SUMMARY OF DOSE AND RISK ESTIMATES 9-1
9.1 Introduction 9-1
9.2 Doses and Risks for Specific Facilities 9-2
9.3 Overall Uncertainties 9-2
9.3.1 Emission and Pathway Uncertainties 9-2
9.3.2 Dose Uncertainties 9-5
9.3.3 Risk Uncertainties 9-5
9.3.4 Overall Uncertainty 9-5
Addendum A - Computer Codes Used by EPA to Assess Doses from
Radiation Exposure A-l
Addendum B - Mechanics of the Life Table Implementation of the
Risk Estimates B-l
vii
-------
FIGURES
Number Page
5,2-1 Wet scrubber particulate control devices 5-5
5.2-2 Pilot-plant sintered-metal filter 5-7
5.2-3 Fabric filters 5-8
5.2-4 Multilayered sand filter 5-9
5.2-5 Open-face HEPA filter 5-9
5.2-6 Mechanical collectors 5-li
5.2-7 Electrostatic precipitators 5-12
5.3-1 Bulkheading of mine stopes 5-17
5.3-2 Mine pressurization 5-17
6.1-1 Pathways of airborne radionuclides into the environment 6-2
6.1-2 Vertical concentration profiles for plume versus down-
wind distance from release 6-6
6.5-1 Circular grid system used by AIRDOS-EPA &-16
7.3-1 Typical pattern of decline of activity of a radionuclide
in an organ, assuming an initial activity in the organ
and no addition uptake of radionuclide by the organ 7~4
7.3-2 The ICRP Task Group lung model for particulates 7-6
7,3-3 Schematic representation of radioactivity movement among
respiratory tract, gastrointestinal tract, and blood 7-7
7.5-1 Dose from chronic inhalation of iodine-131 in water at a
concentration of 1 microcurie/1 7-16
7.5-2 Dose from chronic inhalation of iodine-131 in air at a
concentration of 1 oicrocurie/m3 7-17
7,5-3 Compartments and pathways in model for strontium in
skeleton 7-19
7,5~4 Dose from chronic ingestion of 8trontium-90 in water
at a concentration of 1 nicrocurie/1 7-20
7.5-5 Dose from chronic inhalation of s£rontium-90 in air at
a concentration of 1 mierocurie/ia3 7-21
vili
-------
FIGURES (continued)
Number Page
7.5-6 Dose from chronic inhalation of plutonium-239 in air at
a concentration of 1 microcurie/ms 7-22
7.5-7 Compartments and pathways in model for plutoniuo in
skeleton 7-24
7.5-8 Dose from chronic ingestlon of plutonium-239 in water
at a concentration of microcurie/ms 7-25
7.5-9 Dose from chronic inhalation of plutonium-239 in air at
a concentration of microcurie/ns 7-26
ix
-------
TABLES
Number Page
3.1 Estimates of genetic detriment in a developed
country 3-9
4,2-1 Department of Energy facilities 4-3
4.2-2 Radionuclide emission points (stacks) at Brookhaven
National Laboratories 4-5
4,2-3 U.S. Naval Shipyard facilities 4-17
4.2-4 Parameters of-reference underground uranium mine 4-20
4.2-5 Parameters of reference phosphate-rock drying and
grinding plant 4-21
4.2-6 Reference wet-process phosphate fertilizer plant 4-22
4.2-7 Calciner stack emission characteristics 4-23
4.2-8 Parameters of reference aluminum reduction plant 4-24
4.2-9 Parameters of reference copper smelter 4-25
4.2-10 Parameters of reference zinc plant 4~26
4.2-11 Parameters of reference lead smelter 4-2?
4.2-12 Parameters of reference uranium conversion facility 4-28
4.2-13 Parameters for reference uranium fuel fabrication
facility 4-28
4.2-14 Parameters for reference light-water reactors 4-29
4.2-15 Parameters for reference uranium mill and tailings
impoundment 4-30
4.2-16 Parameters for reference fuel storage facility 4-30
4.2-17 Parameters of reference accelerator facilities 4-31
4.3-1 Summary of radionuclide emissions from DOE facilities 4-33
4.3-2 Summary of radionuclide emissions from NRC-licensed
facilities id other Federal facilities 4-34
4.3-3 Summary of radionuclide emissions from reference
coal-fired boilers 4-35
-------
TABLES (continued)
Number Page
4.3-4 Radionuclide emissions from the reference underground
uranium mine 4-36
4.3-5 Summary of radionuclide emissions from reference
fertilizer plant and reference phosphate rock
processing facility 4-36
4.3-6 Estimated annual radionuclide emissions from elemental
phosphorus plants 4-3?
4.3-7 Summary of emissions from reference mineral-extraction
facilities 4-38
4.3-8 Atmospheric emissions of radlonuclldes from the reference
uranium conversion facility 4-39
4.3-9 Radionuclide emissions from the reference fuel fabrication
facility 4-39
4.3-10 Atmospheric emissions of radionuclides from the reference
BWR and PWR facilities 4-40
4.3-11 Radionuclide emissions from the reference uranium mill 4-40
4.3-12 Radionuclide emissions from reference storage facility 4-41
4.3-13 Radionuclide releases from reference low-energy accel-
erators 4-41
5.4-1 Summary of emission reduction strategies 5-19
7.3-1 Organs for which dose rates are calculated 7-9
7.5-1 Age-dependent parameters for iodine metabolism in the
thyroid 7-15
7.6-1 Distributions of organ doses from inhalation and inges-
tion,of metals 7-28
8.2-1 Range of cancer fatalities induced by 10 rad low-LET
radiation 8-10
8.2-2 A comparison o£ estimates of the risk of fatal cancer
from a lifetime exposure at 1 rad/year 8-11
8.2-3 Proportion of the total risk of fatal radiogenic cancer
resulting from cancer at a particular site
xi
-------
TABLES (continued)
Number Page
8.2-4 UNSCEA1 estimates of cancer risks at specified sites 8-17
8.2-5 Comparison of proportion of the total risk of radiogenic
cancer fatalities by body organ 8-18
8.3-1 Estimated number of cancer fatalities from a lifetime ex-
posure to internally deposited alpha particle emitters 8-24
8.4-1 Potential alpha energy inhaled during one year of expo-
sure to one working level as a function of age by a
member of the general population 8-27
8.4-2 Age-dependent risk coefficients and minimum induction
period for lung cancer resulting from inhaling radon-
222 progeny 8-30
8,4-3 Risk estimate for exposures to radon progeny 8-31
8.5-1 A ranking of causes of uncertainty in estimates of the
risk of cancer 8-40
8.6-1 1CRP task group estimate of number of cases of serious
genetic,ill health in liveborn from parents irradiated
with 10 man-rem in a population of constant size 8-'6
8.6-2 BIER-3 estimates of genetic effects of an average
population exposure of 1 rem per 30-year generation 8-47
8.6-3 UNSCEM 1982 estimated effect of 1 rad per generation of
low dose or low dose rate, low-LET radiation on a pop-
ulation of 10 liveborn according to the doubling dose
method 8-48
8.6-4 Summary of genetic risk estimates per 10 liveborn for an
average population exposure of 1 rad of low dose or low
dose rate, low-LET radiation in a 30-year generation 8-49
8.6-5 EPA estimates for genetic risk in a population of 10
liveborn per rad exposure of the parents per generation 8-55
9.2-1 Doses and risks to nearby individuals 9-3
9.2-4 Doses and risks to regional population 9-4
xii
-------
CHAPTER 1: INTRODUCTION
1.1 History of Standards Development
In 1977, Congress amended the Clean Air Act (the Act) to address
airborne emissions of radioactive materials, Before 1977, these emis-
sions were either regulated under the Atomic Energy Act or unregulated.
Section 122 of the Act required the Administrator of the U.S.
Environmental Protection Agency (EPA), after providing public notice and
opportunity for public hearings (44 FR 21704, April 11, 1979), to
determine whether emissions of radioactive pollutants cause or
contribute to air pollution that may reasonably be expected to endanger
public health. On December 27, 1979, EPA published a notice in the
Federal Register listing radionuclides as hazardous air pollutants under
Section 112 of the Act (44 FR 76738, December 27, 1979). To support
this determination, EPA published a report entitled "Radiological Impact
Caused By Emissions of Radionuclides into Air in the United States
Preliminary Report" (EPA 520/7-79-006, Office of Radiation Programs,
U.S. EPA, Washington, B.C., August 1979).
On June 16, 1981, the Sierra Club filed suit in the U.S. District
Court for the Northern District of California pursuant to the citizens'
suit provision of the Act (Sierra Club v Gorsuch, No. 81-2436 WTS), The
suit alleged that EPA had a nondiscretionary duty to propose standards
for radionuclides under Section 112 of the Act within 180 days after
listing them. On September 30, 1982, the Court ordered EPA to publish
proposed regulations establishing emissions standards for radionuclides,
with a notice of hearing within 180 days of the date of that order.
On April 6, 1983, EPA published a notice in the Federal Register
proposing standards for radionuclide emission sources in four categories:
(1) DOE facilities, (2) NRC-licensed facilities and non-DOE Federal
facilities, (3) underground uranium mines, and (4) elemental phosphorus
plants. Several additional categories of sources that emit radionuclides
were identified, but it was determined that there were good reasons for
not proposing standards for them. These source categories were (1)
coal-fired boilers; (2) the phosphate industry; (3) other mineral-extrac-
tion industries; (4) uranium fuel-cycle facilities, uranium mill tailings,
and high-level waste management; and (5) low-energy accelerators (48 FI
1-1
-------
15077, April 6, 1983). To EPA's knowledge, these comprised all the
source categories that release potentially regulative amounts of radlonu-
clides to the air. To support these proposed standards and determina-
tions, EPA published a draft report entitled "Background Information
Document, Proposed Standards for Sadionuclides" (EPA 520/1-83-001,
Office of Radiation Programs, U.S. EPA, Washington, B.C., March 1983).
Following publication of the proposed standards, EPA held an infor-
mal public hearing in Washington, D.C., on April 28 and 29* 1983. The
comment period was held open an additional 30 days to receive written
comments. Subsequently, EPA received a number of requests to extend the
time for submission of public comments and to conduct a public hearing
on the proposed standards in the West to accommodate persons who were
unable to attend the first public hearing. In response to these requests,
EPA published a notice in the Federal Register that extended the comment
period by an additional 45 days and held an additional informal public
hearing in Denver, Colorado, on June 14, 1983 (48 FR 23665, May 26,
1983).
On February 17, 1984, the Sierra Club again filed suit in the U.S.
District Court for the Northern District of California pursuant to the
citizens* suit provision of the Act (Sierra Club v Ruekelshaus, No.
84-0656 WHO). The suit alleged that EPA had a. nondiscretionary duty to
issue final emissions standards for radionuclides or to find that they
do not constitute a hazardous air pollutant (i.e., "de-list" the pollu-
tant). In August 1984, the court granted the Sierra Club motion and
ordered EPA to take final actions on radionuclides by October 23, 1984.
1.2 Purpose of the Final Background Information Document
This Background Information Document supports the Agency's final
actions on radionuclides. It contains an integrated risk assessment
that provides the scientific basis for these actions.
1.3 Scope of the Final Background Information Document
Volume I contains background information on radiation protection
programs and a detailed description of the Agency's procedures and
methods for estimating radiation dose and risk due to radionuclide
emissions to the air. This material is arranged as shown in the follow-
ing descriptions of the chapters:
0 Chapter 2 - A summary of regulatory programs for radiation
protection and the current positions of the various national
and international advisory bodies and State and Federal Agen-
cies in regard to radiation,,
0 Chapter 3 - A description of what makes radiation hazardous,
the evidence that proves the hazard, and the evidence that
relates the amount of radiation exposure to the amount of
risk.
1-2
-------
0 Chapter 4 - A summary of sources that release radionuclides to
the air, the physical and chemical forms of these releases,
and the quantity of radionuclides that are released.
0 Chapter 5 - A description of how radionuclide emissions to the
air are controlled by means of emission control devices and
work practices.
Chapter 6 - A description of how radionuclides, once released
into the air, move through the environment and eventually
cause radiation exposure of people. This chapter also con-
tains a description of how EPA estimates the amounts of radio-
nuclides in the environment, i.e., in the air, on surfaces, in
the food chain, and in exposed humans.
0 Chapter 7 - A description of how radionuclides, once inhaled
and ingested, move through the body to organs and expose
these organs. This chapter also contains a description of how
EPA estimates the amounts of radiation dose due to this
radiation exposure of organs. It also describes how the
amount of radiation dose is estimated when the source of
radiation is gamma rays from a source outside of the body.
0 Chapter 8 - A description of how the risk of fatal cancers and
genetic effects is estimated once the amount of radiation dose
is known.
0 Chapter 9 - A summary of dose and risk estimates of source
categories emitting significant amounts of radionuclides,
which were made by using the procedures and information in the
previous chapters. Associated uncertainties are discussed in
the appropriate chapter, but overall uncertainties are discussed
in this chapter.
Volume II contains detailed risk estimates for each source of
emissions, which were performed according to the procedures given in
Volume I. Each chapter contains a general description of the source
category, a brief description of the processes leading to emissions of
radionuclides to the air, a summary of emissions data, and estimates of
radiation doses and health risks to both individuals and populations.
Except for DOE facilities, each chapter also contains a brief description
of emission control technology. Control technology for DOE facilities
is discussed "in a separate document entitled "Control Technology for
Radioactive Emissions to the Atmosphere at U.S. Department of Energy
Facilities" (PNL-4621, Pacific Northwest Laboratories, October 1984).
Volume II was originally issued in draft form in April 1983, when
emission standards for radionuclides were proposed. In response to
public comments, it has been revised and is now issued in final form.
1-3
-------
1.4 EPA's Computer Codes
The EPA calculates doses and risks due to facilities emitting
radionuclldes to the air using three computer codes: AIRDOS-EPA,
RABRISK, and DARTAB. These codes calculate, respectively, the resulting
concentrations of radionuclides in the environment, the dose and risk to
persons resulting from a given quantity of each of these radionuclides,
and the total lifetime risk to individuals and the total health impact
on populations. These computer codes are briefly summarized here to
describe how they fit together. Details of the calculations are pre-
sented later.
The AIRDOS-EPA computer code estimates radionuclide concentrations
in the air, rates of deposition on the ground, concentrations on the
ground, and the amounts of radionuclides taken into the body via
inhalation of air and ingestion of meat, milk, and fresh vegetables. A
Gaussian plume equation predicts the atmospheric dispersion of radionu-
clides released from stacks or area sources. The amounts of radionuclides
that are inhaled are calculated from these air concentrations and a
knowledge of how much air is inhaled by an average person. The amount:
of radionuclides Ingested in the meat, milk, and fresh produce that
people consume are estimated by coupling the output of the atmospheric
transport models with the same terrestrial food chain models used by the
U.S. Nuclear Regulatory Commission in Regulatory Guide 1.109.
Working-level exposures are also calculated for inhalation of Rn-222
short-lived decay products.
The RADRISK code computes dose rates to organs resulting from a
given quantity of a radionuclide that is Ingested or inhaled. These
dose rates are then used to estimate the risk of fatal cancers in an
exposed cohort of 100,000 persons. All persons in the cohort are assumed
to be born at the same time and to be at risk of dying from competing
causes (including natural background radiation). Estimates of potential
health risk due to exposure to a known quantity of approximately 500
different radionuclides are tabulated and stored until needed. These
risks are summarized in terms of the probability of premature death for
a member of the cohort due to a given quantity of each radionuclide that;
is ingested or inhaled.
The DARTAB computer code then provides estimations of the impact of
radionuclide emissions from a specific facility by combining the informa-
tion on the amounts of radionuclides that are ingested or Inhaled (as
provided by AIRDOS-EPA) with dosimetric and health effec;s data for a
given quantity of each radionuclide (as provided by RABRISK).
The DARTAB code estimates dose and risk for Individuals at user-
selected locations and for population groups. Radiation doses and risks
can be broken down by radionuclide, exposure pathway, and organj or they
can be summarized by direction and distance from the facility.
1-4
-------
Chapter 2: CURRENT REGULATORY PROGRAMS AND STRATEGIES
2,1 Introduction
People have always been exposed to ionizing radiations from the
cosmic rays and naturally-occurring radionuclldes in the earth that make
up the natural radiation background. Awareness of radiation and radio-
activity dates back only to the end of the last century—to the discov-
ery of x-rays i.i 1895 and the discovery of radioactivity in 1896. These
discoveries mark the beginning of radiation science and the deliberate
use of radiation and radionuclides in science, medicine, and industry.
The findings of radiation science rapidly led to the development of
medical and industrial radiology, nuclear physics, and nuclear medicine.
By the 1920's, the use of x-rays in diagnostic medicine and industrial
applications was widespread, and radium was being used by industry for
luminescent dials and by doctors in therapeutic procedures. By the
1930's, biomedical and gejietic researchers were studying the effects of
radiation on living organisms, and physicists were beginning to under-
stand the mechanisms of spontaneous fission and radioactive decay. By
the 1940's» a self-sustaining fission reaction was demonstrated, which
led directly to the construction of the first nuclear reactors and
atomic weapons.
Developments since ths end of World War II have been rapid. Today
the use of x-rays and radioactive materials is widespread and includes:
0 Nuclear reactors, and their supporting fuel-cycle facilities,
generate electricity; power ships and submarines; produce
radioisotopes for research, space, defense, and medical appli-
cations; and are used as research tools for nuclear engineers
and physicists.
0 Particle accelerators produce radioisotopes and are used as
research tools for studying the structure of materials and
atoms.
0 The radlopharmaceutlcal industry provides the radioisotopes
needed for bioaedical research and nuclear medicine.
2-1
-------
° Nuclear medicine has developed as a recognized medical specialty
in which radioisotopes are used in the diagnosis and treatment
of numerous diseases.
0 X-rays are widely used as a diagnostic tool in medicine and in
such diverse industrial fields as oil exploration and nonde-
structive testing,
0 Radionuclides are used in such common consumer products as
luminous-dial wristwatches and smoke detectors.
The following sections of this chapter provide a brief history of
the evolution of radiation protection philosophy and an outline of the
current regulatory programs and strategies of the government agencies
responsible for assuring that radiation and radionuclides are used
safely.
2-2 The International Commission on RadiologicalProtectionand the
National Council on Radiation Protection and Measurements
Initially, the dangers and risks posed by x-rays and radioactivity
were little understood. By 1896, however, "x-ray burns" were being
reported in the medical literature, and by 1910, it was understood that
such "burns" could also be caused by radioactive materials. By the
I920*s, sufficient direct evidence (from the experiences of radium dial
painters, medical radiologists, and miners) and indirect evidence (from
biomedical and genetic experiments with animals) had been accumulated to
persuade the scientific community that an official body should be estab-
lished to make recommendations concerning human protection against
exposure to x-rays and radium.
At the Second International Congress of Radiology meeting in
Stockholm, Sweden, in 1928, the first radiation protection commission
was created. Reflecting the uses of radiation and radioactive materials
at the time, the body was named the International X-Ray and Radium
Protection Commission and was charged with developing recommendations
concerning protection from radiation. In 1950, to reflect better its
role in a changing world, the commission was reconstituted and renamed
the International Commission on Radiation Protection (ICRP).
During the Second International Congress of Radiology, the newly
created coomission suggested to the nations represented at the Congress
that they appoint national advisory committees to represent their view-
points before the ICRP, and to act in concert with the Commission in
developing and disseminating recommendations on radiation protection.
This suggestion led to the formation, in 1929, of the Advisory Committee
on X-Ray and Radium Protection as the U.S. advisory group. This Advis-
ory Committee, after a series of reorganizations and name changes,
emerged in 1964 in its present form as the Congressionally chartered
2-2
-------
National Council on Radiation Protection and Measurements (NCRF). The
Congressional charter provides for the HCRP to:
0 Collect, analyze, develop, and disseminate in the public
interest information and recommendations about radiation
protection and radiation quantities, units, and measurements.
0 Develop basic concepts about radiation protection and radiation
quantities, units, and measurements, and the application of
these concepts.
° Provide a means by which organizations concerned with radiation
protection and radiation quantities, units, and measurements
may cooperate to effectively use their combined resources, and
to stimulate the work of such organizations,
0 Cooperate with the ICRP and other national and International
organizations concerned with radiation protections and radiation
quantities, units, and measurements.
Throughout their existence, the ICRP and the NCRP have worked together
closely to develop radiation protection recommendations that reflect the
current understanding of the dangers associated with exposure to ioniz-
ing radiation.
The first exposure limits adopted by the ICRP and the NCRP (ICRP34,
ICRP38, and NCEP36) established 0.2 roentgen/day* as the "tolerance
dose" for occupational exposure to x-rays and gamma radiation from
radium. This limit, equivalent to approximately 25 rads/year as mea-
sured in air, was established to guard against the known effects of
ionizing radiation on superficial tissue, changes in the blood, and
"derangement" of internal organs, especially the reproductive organs.
At the time the recommendations were made, high doses of radiation were
known to cause observable effects and even to induce cancer. However,
no such effects were observed at lower doses, and the epidemiological
evidence at the time was inadequate to even imply the carcinogenic
induction effects of moderate or low doses. Therefore, the aim of
radiation protection was to guard against known effects, and the "toler-
ance dose" limits that were adopted were believed to represent the level
of radiation that a person in norma? health could tolerate without
suffering observable effects. The concept of a tolerance dose and the
recommended occupational exposure limit of 0.2 R/day for x- and gamma
radiation remained In effect until the end of the 1940's. The recom-
mendations of the ICRP and the NCRP made no mention of exposure of the
general populace.
By the end of World War II, the widespread use of radioactive
materials and scientific evidence of genetic and somatic effects at
lower doses and dose rates suggested that the radiation protection
* The NCRP's recommendation was 0.1 roentgen/day measurad in air. This
limit is roughly equivalent to the ICRP limit, which was conventional-
ly measured at the point of exposure and included back—scatter.
2-3
-------
recommendations of the NCRP and the ICRP would have to be revised
downward.
By 1948, the NCRP had formulated Its position on appropriate new
limits. These limits were largely accepted by the ICRP in its recommen-
dations of 1950 and formally issued by the NCRP in 1954 (ICRP51, NCRP54).
Whereas the immediate effect was to lower the basic whole-body occupa-
tional dose limit to 0.3 rad/week (approximately 15 rads/year), the
revised recommendations also embodied several new and important concepts
in the formulation of radiation protection criteria.
First, the recommendations recognized the differences in the ef-
fects of various types and energies of radiation; both ICRP's and NCRP's
recommendations included discussions of the weighting factors that
should be applied to radiations of differing types and energies. The
NCRP advocated the use of the "ren" to express the equivalence in bio-
logical effect betveen radiations of differing types and energy.*
Although the ICRP noted the shift toward the acceptance of the rem, it
continued to expiess its recommendations in terms of the rad, with the
caveat that neutrons should carry a quality factor of ten.
Second, the recommendations of both organizations introduced the
concept of critical organs and tissues. The intent of this concept was
to assure that no tissue or organ, with the exception of the skin, would
receive a dose in excess of that allowed for the whole body. At the
time, scientific evidence was lacking on which to base different recom-
mended limits for the various tissues and organs. Thus, all blood-
forming organs were considered critical organs and were limited to the
same exposure as the whole body. The skin was allowed an exposure of 30
rad/year and the extremities were allowed a limit of 75 rads/year.
Third, the recommendations of the NCRP included the suggestion that
individuals under the age of 18 receive no more than one-tenth the expo-
sure allowed for adults. The reasoning behind this particular recommen-
dation is interesting as it reflects clearly the limited knowledge of
the times. The scientific evidence indicated a clear relationship
between accumulated dose and genetic effect. However, this evidence was
obtained exclusively from animal studies that had been conducted with
doses ranging from 25 to thousands of rads. There was no evidence from
exposures less than 25 rads accumulated dose, and the interpretation of
* The exact relationship between roentgens, rads, and rems is beyond the
scope of this work. In simple terms, the roentgen is a measure of the
degree of ionization induced by x- and gamma radiations in air. The
rad (radiation absorbed dose) is a measure of the energy imparted to
matter by radiation. And the rein (roentgen equivalent man) is a mea-
sure of equivalence for the relative biological effect of radiations
of different types and energies on man. Over the range of energies
typically encountered, the relationship of roentgens to rads to rems
for x- and gamma radiation is essentially equality. For beta radia-
tion, rads are equivalent to rems. And for alpha radiation one rad
equals 10 to 20 rems.
2-4
-------
the animal, data and the Implications for humans was unclear and did not
support a specific permissible dose. The data did suggest that genetic
damage was more dependent on accumulated dose than previously believed,
but experience showed that exposure for prolonged periods to the permis-
sible dose (1.0 R/week) did not result in any observable genetic ef-
fects. The NCRP decided that it was not necessary to change the occupa-
tional limit to provide additional protection beyond that provided by
the reduction in the permissible dose limit to 0.3 P ' ,aek. At the same
time, it recommended limiting the exposure of individuals under the age
of 18 to assure that they did not accumulate a genetic dose that would
later preclude their employment as radiation workers. The factor of ten
was rather arbitrary, but was believed to be sufficient to protect the
future employability of all individuals (NCRP54).
Fourth, the concept of a tolerance dose was replaced by the concept
of a maximum permissible dose. The change in terminology reflected the
increasing awareness that any radiation exposure might involve some risk
and that repair mechanisms might be less effective than previously be-
lieved. Therefore, the concept of a maximum permissible dose (expressed
as dose per unit of time) was adopted because it better reflects the
uncertainty in our knowledge than does the concept of tolerance dose.
The maximum permissible dose was defined as the level of exposure that
entailed a small risk compared with those posed by other hazards in life
(ICRP51).
Finally, in explicit recognition of the inadequacy of our knowledge
regarding the effects of radiation and of the possibility that any expo-
sure might have some potential for harm, the recommendations included an
admonition that every effort should be made to reduce exposure to all
kinds of ionizing radiation to the lowest possible level. This concept,
known originally as ALAP (as low as practicable) and later as ALARA (as
low as reasonably achievable), would become a cornerstone of radiation
protection philosophy.
During the 1950's, a great deal of scientific evidence on the
effects of radiation became available from studies of the radium dial
painters, radiologists, and the survivors of the atomic bombs dropped on
Japan. This evidence suggested that genetic effects and long-term
somatic effects were more important than previously considered. Thus,
by the late 1950's, the ICRP and NC1P recommendations were again revised
(ICRP59, NCRP59). These revisions include the following major changes;
the maximum permissible dose for whole-body exposure and the most criti-
cal organs (blood-forming organs, gonads, and the lens of the eye) was
lowered to 5 rems/year, with a quarterly limit of 3 reins; the limit for
exposure of other organs was set at 30 reos/year; internal exposures
were controlled by a comprehensive set of maximum permissible concentra-
tions of radionuclides in air and water based on the most restrictive
case of a young worker; and recommendations were included for some
nonoccupatlonal groups and for the general Copulation (for the first
time).
2-5
-------
The lowering of the maximum permissible whole-body done from 0.3
rad/week to 5 reins/year, with a quarterly limit of 3 Terns, reflects both
the new evidence and the uncertainties of the time. Although no adverse
effects were observed among workers who had received the maximum permis-
sible dose of 0.3 rad/week, there was concern that the lifetime accumu-
lation of as much as 750 rads (15 rads/year times 50 years) was too
much. Lowering the maximum permissible dose by a factor of three was
believed to provide a greater margin of safety. At the same time,
operational experience showed that 5 reins/year could be met in most
instances, particularly with the additional operational flexibility
provided by expressing the limit on an annual and quarterly basis.
The recommendations given for nonoccupational exposures were based
on concerns of genetic effects. The evidence available suggested that
genetic effects were primarily dependent on the total accumulated dose.
Thus, having sought the opinions of respected geneticists, the 1CRP and
the NCRP adopted the recommendation that accumulated gonadal dose to age
30 be limited to 5 rems from sources other than natural background and
medical exposure. As an operational guide, the NCRP recommended that
the maximum dose to any individual be limited to 0.5 rem/year, with
maximum permissible body burdens of radionuclides (to control internal
exposures) set at one-tenth that allowed for radiation workers. These
values were derived from consideration of the genetically significant
dose to the population, and were established "primarily for the purpose
of keeping the average dose to the whole population as low as reasonably
possible, and not because of the likelihood of specific injury to the
individual" (NCRP59).
During the 1960's, the ICRP and NCRP again lowered the maximum
permissible dose limits (ICRP65, NCRP71). The considerable scientific
data on the effects of exposure to ionizing radiation were still incon-
clusive with respect to the dose-response relationship at low exposure
levels; thus, both organizations continued to stress the need to keep
all exposures to the lowest possible level.
The NCRP and the ICRP made the following similar recommendations:
0 Limit the dose to the whole-body, red bone marrow, and gonads
to 5 rems in any year, with a retrospective limit of 10 to 15
rems in any given year as long as total accumulated dose did
not exceed 5X(N-18), where H is age in years.
° Limit the dose to the skin, hands, and forearms to 15, 75, and
30 rems per year, respectively.
0 Limit the -lose to any other organ or tissue to 15 rems per
year.
0 Limit the dose to any non-occupatlonally exposed individual in
the population to 0.5 rein per year.
0 Limit the average dose to the population to 0.17 rem per year.
2-6
-------
The scientific evidence and the protection philosophy on which the
above recommendations were based were set forth in detail in NCRP71. In
the case of occupational exposure limits, the goal of protection was to
ensure that the risks of genetic and somatic effects were small enough
to be comparable to the risks experienced by workers in other safe
industries. The conservatively derived numerical limits recommended
were based on the linear, no-threshold, dose-response model, and were
believed to represent a level of risk that was readily acceptable to an
average individual. For nonoccupational exposures, the goal of protec-
tion was to ensure that the risks of genetic or somatic effects were
small compared with other risks encountered in everyday life. The deri-
vation of specific limits was complicated by the unknown dose-response
relationship at low exposure levels and the fact that the risks of radi-
ation exposure did not necessarily accrue to the sane individuals who
benefited from the activity responsible for the exposure. Therefore, it
was necessary to derive limits that gave adequate protection to each
member of the public and to the gene pool of the population as a whole,
while still allowing the development of beneficial uses of radiation and
radionuclides.
In 1977, the ICRP made a fundamental change in its recommendations
when it abandoned the critical organ concept in favor of the weighted
whole-body dose equivalent concept for limiting occupational exposure
(ICRP77). The change, made to reflect our increased understanding of
the differing radiosensitivity of the various organs and tissues, did
not affect the overall limit of 5 reins per year, and is not intended to
be applied to nonoccupational exposures.
Also significant is the fact that ICRP's 1977 recommendations
represent the first explicit attempt to relate and justify permissible
radiation exposures with quantitative levels of acceptable risk. Thus,
average occupational exposures (approximately 0.5 rem/year) are equated
with risks in safe industries, given as 10 ** annually. At the maximum
limit of 5 reos/year, the risk is equated with that experienced by some
workers in recognized hazardous occupations. Similarly, the risks
implied by the nonoccupational limit of 0.5 rem/year are equated to
levels of risk of less thai? 10 z in a lifetime; the general populace's
average exposure is equivalent to a lifetime risk on the order of 10~3
to ID'1*. The ICRP believed these levels of risk were in the range that
most individuals find acceptable.
The NCRP has not formally changed its recommendations for occupa-
tional exposure to correspond to the 1977 recommendations of the ICRP.
It has been working diligently, however, to review its recommendations,
and has circulated a draft of proposed changes to various interested
scientists and regulatory bodies for their comment. The relevant non-
occupational exposure limits are:
0 0.5 rem/year whole-body dose equivalent, not including back-
ground or medical radiation, for individuals in the population
when the exposure Is not continuous.
2-7
-------
0 0.1 rem/year whole-body dose equivalent, not including back-
ground or medical radiation, for individuals in the population
when the exposure is continuous.
0 Continued use of a total dose limitation system based on
justification of every exposure and application of the "as low
as reasonably achievable" philosophy to every exposure.
The NCRP equates continuous exposure at the level of 0.I rem/year
to a lifetime risk of developing cancer of about one in a thousand. The
NCRP has not formulated exposure limits for specific organs, but it
notes that the permissible limits will necessarily be higher than the
whole-body limit in inverse ratio of the risk for a particular organ to
the total risk for whole-body exposure. In response to EPA's proposed
national emission standards for radionuclides, the NCRP suggested that
since the 0.1 rem/year limit is the limit for all exposures from all
sources (excluding natural background and medical radiation), the opera-
tor of any site responsible for more than 25 percent of the annual limit
be required to assure that the exposure of the maximally exposed indi-
vidual is less than 0.1 rem/year from all sources (NCRP 84).
2.3 Federal Guidance
The ICRP and the NCRP function as nongovernmental advisory bodies.
Their recommendations are not binding on any user of radiation or radio-
active materials. The wealth of new scientific information on the
effects of radiation that became available in the 1950's prompted the
President to establish an official government entity with responsibility
for formulating radiation protection criteria and coordinating radiation
protection activities. Thus, the Federal Radiation Council was estab-
lished in 1959 by Executive Order 10831. The Council included repre-
sentatives from all of the Federal agencies concerned with radiation
protection, and acted as a coordinating body for all of the radiation
activities conducted by the Federal government. In addition to Its
coordinating function, the Council's major responsibility was to
"...advise the President with respect to radiation matters, directly or
indirectly affecting health, including guidance for all Federal agencies
in the formulation of radiation standards and in the establishment and
execution of programs of cooperation with States..." (FRC60),
The Council's first recommendations concerning radiation protection
standards for Federal agencies were approved by the President in 1960.
Based largely on the work and recommendations of the ICRP and the NCRP,
the guidance established the following limits for occupational expo-
sures:
0 Whole body, head and trunk, prtive blood forming organs,
gonads, or lens of eye—not to exceed 3 rents in 13 weeks and
total accumulated dose limited to 5 times the number of years
beyond age 18.
0 Skin of whole body and thyroid—not to exceed LO rems in 13
weeks or 30 rems per year.
2-8
-------
0 Hands, forearms, feet, and ankles—not to exceed 25 rems in 13
weeks or 75 rems per year.
0 Bone—not to exceed 0.1 microgram of Radium-226 or its biolog-
ical equivalent.
0 Any other organ—not to exceed 5 rems per 13 weeks or 15 rents
per year.
Although these levels differ slightly from those recommended by
RP and 1CRP at the time, the differences do not represent any greater
lesser protection. In fact, the FRC not only accepted the levels
commended by the NCRP for occupational exposure, it adopted the NCRP's
ilosophy of acceptable risk for determining occupational exposure
nits. Although quantitative measures of risk were not given in the
idance, the prescribed levels were not expected to cause appreciable
dily injury to an individual during his or her lifetime. Thus, while
e possibility of some injury was not zero, it was so low as to be ac—
ptable if there was any significant benefit derived from the exposure.
The guidance also established exposure limits for members of the
blic. .These were set at 0.5 rem per year (whole body) for an indivi-
al, and an average of 5 rems in 30 years (gonadal) per capita. The
idance also provided for developing a suitable sample of the popula-
on as an operational basis for determining compliance with the limit
en doses to all individuals are unknown. Exposure to this population
niple was not to exceed 0.17 rem per capita per year. The population
nit of 0.5 rem to any individual per year, was derived from consldera-
on of natural background exposure. Natural background radiation
ries by a factor of two"to four from location to location.
In addition to the formal exposure limits, the guidance also estab-
shed as Federal policy that there should be no radiation exposure
thout an expectation of benefit, and that "every effort should be made
encourage the maintenance of radiation doses as far below this guide
practicable." The inclusion of the requirements to consider benefits
d keep all exposure to a minimum was based on the possibility that
ere is no threshold dose for radiation. The linear non-threshold dose
sponse was assumed to place an upper limit on the estimate of radia-
on risk. However, the FRC explicitly recognized that it might also
present the true level of risk. If so, then any radiation exposure
rried some risk, and it was necessary to avoid all unproductive expo-
res and to keep all productive exposures as "far below this guide as
acticable."
In 1967, the Federal Radiation Council issued guidance for the
ntrol of radiation hazards in uranium mining (FRC67). The need for
ch guidance was clearly indicated by the epldemiologieal evidence that
owed a. higher incidence of lung cancer in adult males who worked in
anium mines compared with the incidence in adult males from the same
cations who had not worked in mines. The Guidance established specif-
exposure limits and recommended that all exposures be kept as far
2-9
-------
below the guide limits as possible. The limits chosen represented a
trade-off between the risks incurred at various exposure levels, the
technical feasibility of reducing the exposure, and the benefits of the
activity responsible for the exposure. The guidance also applied to
nonuranium mines.
In 1970, the functions of the Federal Radiation Council were trans-
ferred to the U.S. Environmental Protection Agency. In 1971, the EPA
revised the Federal Guidance for the control of radiation hazards in
uranium mining (EPA71), Based on the risk levels associated with the
exposure limits established in 1967, the upper limit of exposure was re-
duced by a factor of three. The EP
-------
The development of this standard took into account both the maximum
risk to an individual and the overall effect of releases from fuel-cycle
operations on the population and balanced these risks against the costs
of effluent control in a. primarily qualitative way.
Under the authority of the Uranium Mill Tailings Radiation Control
Act, the EPA has promulgated standards limiting public exposure to
radiation from uranium tailings piles (EPA83a, EPA83b). Whereas the
standards for inactive and active tailings piles differ, a consistent
basis is used for these standards. Again, the Agency sought to balance
the radiation risks imposed on individuals and the population in the
vicinity of the pile against the feasibility and costs of control.
Under the authority of the U.S. Atomic Energy Act of 1954, as
amended, the EPA has proposed standards for disposal of spent fuel,
high-level wastes, and transuranic elements (EPA82). The proposed stan-
dard establishes two different limits: (1) during the active waste-
disposal phase, operations must be conducted so that no member of the
public receives a dose greater than that allowed for other phases of the
uranium fuel cycle; and (2) once the repository is closed, exposure is
to be controlled by limiting releases. The release limits were derived
by summing, over long time periods, the estimated risks to all persons
exposed to radioactive materials released into the environment. The
uncertainties involved in estimating the performance of a theoretical
repository led to this unusual approach, and the proposed standard
admonishes the agencies responsible for constructing and operating such
repositories to take steps to reduce releases below the upper bounds
given in the standard to the extent reasonably achievable.
Under the authority of the Safe Drinking Water Act, the EPA has
issued interim regulations covering the permissible levels of radium,
gross alpha and manmade beta, and photon emitting contaminants in com-
munity water systems (EPA76). The limits are expressed in picocuries/
liter. The limits chosen for manmade beta and photon emitters equate to
approximately 4 mrems/year whole—body or organ dose to the most exposed
individual. In the background information for the standard, the 4
mrems/year exposure through a single pathway that the standard permits
is explicitly compared with the overall population standard of 170
mrems/year, and the conclusion is expressed that the roughly 40-fold
decrease is appropriate for a single pathway.
Section 122 of the Clean Air Act amendments of 1977 (Public Law
95-95) directed the Administrator of EPA to review all relevant informa-
tion and determine if emissions of hazardous pollutants into air will
cause or contribute to air pollution that may reasonably be expected to
endanger public health. In December 1979, EPA designated radionuclides
as hazardous air pollutants under Section 112 of the Act. On April 6,
1983, SPA published proposed National Emission Standards for radionu-
clides for selected sources in the federal Register (48 CFR 15076).
2-11
-------
2.5 Nuclear Regulatory Commission
Under the authority of the Atomic Energy Act of 1954, as amended,
the Nuclear Regulatory Commission (NEC) is responsible for licensing and
regulating the use of byproduct, source, and special nuclear material,
and for assuring that all licensed activities are conducted in a manner
that protects public health and safety. The Federal guidance on radia-
tion protection applies directly to the NRCj therefore, the NRC must
assure that none of the operations of its licensees exposes an Individual
of the public to more than 0.5 rein/year. The dose limits imposed by the
EPA's standard for uranium fuel-cycle facilities also apply to the
fuel-cycle facilities licensed by the NRC. These facilities are prohib-
ited from releasing radioactive effluents in amounts that would result
in doses greater than the 25 mrems/year Unit imposed by that standard.
The NRC exercises its statutory authority by imposing a combination
of design criteria, operating parameters, and license conditions at the
time of construction and licensing. It assures that the license condi-
tions are fulfilled through inspection and enforcement. The NRC licens-
es more than 7000 users of radioactivity. The regulation of fuel-cycle
licensees is discussed separately from the regulation of byproduct
material licensees.
2.5.1 Fuel Cycle Licenses
The NRC does not use the term "fuel cycle facilities" to define its
classes of licensees. The tena is used here to coincide with the EPA
use of the term in its standard for uranium fuel cycle facilities. As a
practical matter, this term includes the NRC's large source and special
nuclear material, and production and utilization facilities. The NRC's
regulations require an analysis of probable radioactive effluents and
their effects on the population near fuel cycle facilities. The NRC
also assures that all exposures are as low as reasonably achievable by
imposing design criteria and specific equipment requirements on the
licensees. After a license has been issued, fuel-cycle licensees must
monitor their emissions and take environmental measurements to assure
that the design criteria and license conditions have been met. For
practical purposes, the NRC adopted the maximum permissible concen-
trations developed by the NCRP to relate effluent concentrations to
exposure.
In the 1970's, the NRC formalized the implementation of as low as
reasonably achievable exposure levels by issuing a regulatory guide for
as low as reasonably achievable design criteria. This coincided with a
decision to adopt, as a design criterion, a maximum permissible dose of
5 mrems/year from a single nuclear electric generating station. The
5-mrem liiait applies to the most exposed individual actually living in
the vicinity of the reactor, and refers to whole-body doses from ex-
ternal radiation by the air pathway (NRC77).
2-12
-------
2.5,2 Byproduct Material Licenses
The NRC's licensing and inspection procedure for byproduct material
users is less uniform than that imposed on major fuel-cycle licensees
for two reasons: (1) the much larger number of such licensees, and (2)
the much smaller potential for releasing significant quantities of
radioactive materials into the environment. The prelicensing assurance
procedures of imposing design reviews, operating practices, and license
conditions prior to construction and operation are similar. The amount
of protection that is afforded the public from releases of radioactive
materials from these facilities can vary considerably because of three
factors. First, the requirements that the NRG imposes for monitoring
effluents and environmental radioactivity are much less stringent for
these licensees. If the quantity of materials handled is small enough,
the NEC might not impose any monitoring requirements. Second, and more
important, the level of protection can vary considerably because where
the licensee must meet the effluent concentrations for an area of
unrestricted access is not consistently defined. Depending on the
particular licensee, this area has been defined as the nearest inhabited
structure, as the boundary of the user's property line, as the roof of
the building where the effluents are vented, or as the mouth of the
stack or vent. Finally, not all users are allowed to reach 100 percent
of the permissible concentrations in their effluents. In fact, the NRC
has implemented as low as reasonably achievable considerations on many
of these licensees by limiting them to 10 percent of the maximum permis-
sible concentration in their effluents.
2.6 Department of Energy
The U.S. Department of Energy (DOE) operates ? comple-x of national
laboratories and weapons facilities. These facilities are not licensed
by the NRC. The DOE is responsible, under the U.S. Atomic Energy Act of
1954, as amended, for assuring that these facilities are operated in a
manner that does not jeopardize public health and safety.
For practical purposes, the DOE has adopted the NCRP's maximum
permissible concentrations in air and water as a workable way to assure
that the dose limits of 0.5 rem/year whole-body and 1.5 rems/year to any
organ are being observed. The DOE also has a requirement that all doses
be kept as low as is reasonably achievable, but the contractors that op-
erate the various DOE sites have a great deal of latitude in implement-
ing policies and procedures to assure that all doses are kept to the
lowest possible level.
The DOE assures that its operations are within its operating guide-
lines by requiring its contractors to maintain radiation monitoring
systems around each of its sites and to report the results in an annual
summary report. New facilities and modifications to existing facilities
are subject to extensive design criteria reviews (similar to those used
by the NRC). During the mid-1970's, the DOE Initiated a systematic
effluent-reduction program that resulted in the upgrading of many facil-
ities and effected a corresponding reduction in the effluents (including
airborne and liquid radioactive materials) released to the environment.
2-13
-------
2.7 Other Federal Agencies
The Department of Defense operates several nuclear installations,
including a fleet of nuclear-powered submarines and their shore support
facilities. The DOD, like other Federal agencies, must comply with the
Federal radiation guidance, fhe BOD has not formally adopted any more
stringent exposure limits for members of the public than the 0,5 rem/year
allowed by the Federal guidance.
2.7.1 Center for Medical Devices andRadiological Health
Under the Radiation Control Act of 1968, the major responsibility
of the Center for Medical Devices and Radiological Health in the area of
radiation protection is the specification of performance criteria for
electronic products, including x-ray equipment and other medical devices.
This group also performs environmental sampling in support of other
agencies, but no regulatory authority is involved.
2.7.2 Mine Safety and Health Administration
The Mine Safety and Health Administration (MSHA) has the regulatory
authority to set standards for exposures of miners to radon and its
decay products and other (nonradiological) pollutants in mines. The
MSHA has adopted the Federal guidance for exposure of uranium miners
(EPA71). It has no authority or responsibility for protecting members
of the general public from the hazards associated with radiation.
2.7.3 Occupational Safety and Health Administration
The Occupational Safety and Health Administration (OSHA) is respon-
sible for assuring a safe work place for all workers. This authority,
however, does not apply to radiation workers at government-owned or
NRC-licensed facilities. This group does have the authority to set
exposure limits for workers at unlicensed facilities, such as particle
accelerators, but it does not have any authority to regulate public
exposure to radiation, OSHA has adopted the occupational exposure
limits of the NEC, except it has not imposed the requirement to keep all
doses 'as low as" is reasonably achievable,
2.7.4 Department of Transportation
The Department of Transportation (DOT) has statutory responsibility
for regulating the shipment and transportation of radioactive materials.
This authority includes the responsibility to protect the public from
exposure to radioactive materials while they are in transit. For prac-
tical purposes, the DOT has implemented its authority through the speci-
fication of performance standards for shipment containers, and by set-
ting maximum exposure rates at the surface of any package containing
radioactive materials. These limits were set to assure compliance with
the Federal guidance for occupational exposure, and they are believed to
2-14
-------
be sufficient to protect the public from exposure. The DOT also con-
trols potential public exposure by managing the routing of radioactive
shipments to avoid densely populated areas.
2.8 State Agencies
States have important authority for protecting the public from the
hazards associated with ionizing radiation. In 26 states, the states
have assumed NEC's inspection, enforcement, and licensing responsibili-
ties for users of source and byproduct materials and users of small
quantities of special nuclear material. These "NEC-agreement states,"
which license and regulate more than 11,500 users of radiation and radi-
oactive materials, are bound by formal agreements to adopt requirements
consistent with those imposed by the NRG. The NEC continues to perform
this function for all licensable uses of source, byproduct, and special
nuclear material in the 24 states that are not agreement states.
Nonagreement states, as well as SRC-agreement states, regulate the
exposures to workers from electronic sources of radiation. Also, all
states retain the authority to regulate the use of naturally occurring
(i.e., radium) and accelerator-produced radioactive materials.
2-15
-------
REFE1ENCES
EPA71 U.S. Environmental Protection Agency, Radiation Protection
Guidance for Federal Agencies: Underground Mining of Uranium
Ore, Federal Register _36_(132), July 9, 1971.
EPA76 U.S. Environmental Protection Agency, National Interim Primary
Drinking Water Regulations, EPA-570/9-76-Q03, 1976.
EPA77 U.S. Environmental Protection Agency, Environmental Radiation
Protection Standards for Nuclear Power Operations, 40 CFR 190,
Federal Register ^(9), January 13, 1977.
EPA78 U.S. Environmental Protection Agency, Radiation Protection
Guidance to Federal Agencies for Diagnostic X-Rays, Federal
Register 43_(22), February 1, 1978.
EPA81 U.S. Environmental Protection Agency, Federal Radiation
Protection Guidance for Occupational Exposure, Federal Register,
j»6_{15), January 23, 1981.
EPA82 U.S. Environmental Protection Agency, Environmental Standards
for the Management and Disposal of Spent Nuclear Fuel, High-
Level and Transuranic Radioactive Wastes, 40 CFR191, Federal
Register _47 (250), December 29, 1982.
EPA83a U.S. Environmental Protection Agency, Standards for Remedial
Actions at Inactive Uranium Processing Sites, Federal Register
48(590), January 5, 1983.
EPA83b U.S. Environmental Protection Agency, Environmental Standards
for Uranium Mill Tailings at Licensed Commercial Processing
Sites; Final Rule, Federal Register 4B(196), October 7, 1983.
FRC60 Federal Radiation Council, Radiation Protection Guidance for
Federal Agencies, Federal Register _44_(02), May 18, 1960.
FRC67 Federal Radiation Council, Guidance for the Control of
Radiation Hazards in Uranium Mining, Report No. 8, September
1967.
ICRP34 International X-Ray and Radium Protection Commission,
International Recommendations for X-Ray and Radium Protection,
British Journal of Radiology _7, 695-699, 1934.
2-16
-------
ICRP38 International X-Ray and Radium Protection Commission,
International Recommendations for X-Ray and Radium Protection,
Aaer. of Roent and Radium 40, 134-138, 1938.
ICRP51 International Commission on Radiological Protection,
International Recommendations on Radiological Protection 1950,
British Journal of Radiology 24, 46-53, 1951.
ICRP59 International Commission on Radiological Protection, Recommen-
dations of the ICRP 1958, ICRP Publication 1, Pergamon Press,
Oxford, 1959.
ICRP65 International Commission on Radiological Protection, Recommen-
dations of the ICRP 1965, ICRP Publication 9, Pergamon Press,
Oxford, 1965.
ICRP77 International Commission on Radiological Protection, Recommen-
dations of the International Commission on Radiological Protec-
tion, ICRP Publication 26, Pergaoon Press, Oxford, 1977.
NCRP36 Advisory Committee on X-ray and Radium Protection, X-ray
Protection, NCRP Report No. 3, 1936.
NCRP54 National Committee on Radiation Protection, Permissible Dose
From External Sources of Ionizing Radiation, National Bureau
of Standards Handbook 59, 1954.
NCRP59 National Committee on Radiation Protection, Maximum Permissible
Body Burdens and Maximum Permissible Concentrations of Radionu-
clides in Air and in Hater for Occupational Exposure, National
Bureau of Standards Handbook 69, 1959.
NCRP71 National Council on Radiation Protection and Measurements,
Basic Radiation Protection Criteria, NCRP Report No, 39, 1971.
NCRP 84 National Council on Radiation Protection and Measurements.,
Control of Air Emissions of Radionuclides, September 18, 1984.
NRC77 U.S. Nuclear Regulatory Commission, 1977, Appendix I: 10 CFR
50, Federal Register 44, September 26, 1979.
2-17
-------
Chapter 3; HAZARD IDENTIFICATION
The adverse biological reactions associated with ionizing radia-
tions, and hence with radioactive materials, are careinogenicity, muta-
genieity, and teratogenicity. Carcinogenicity is the ability to produce
cancer. Mutagenicity is the property of being able to induce genetic
mutation, which may be in the nucleus of either somatic (body) or germ
(reproductive) cells. Teratogenicity refers to the ability of an agent
to induce or increase the incidence of congenital malformations as a
result of permanent structural or functional deviations produced during
the growth and development of an embryo (these are more commonly re-
ferred to as birth defects).
Ionizing radiation causes injury by breaking constituent body
molecules into electrically charged fragments called "ions" and thereby
producing chemical rearrangements that may lead to permanent cellular
damage. The degree of biological damage caused by various types of
radiation varies according to how close together the ionizatlons occur.
Some ionizing radiations (e.g., alpha particles) produce intense regions
of ionization. For this reason they are called high-LET (linear energy
transfer) particles. Other types of radiation [such as high-energy
photons (x-rays)] that release electrons that cause ionization and beta
particles are called low-LET radiations because of the sparse pattern of
ionization they produce. In equal doses, the carcinogenicity and muta-
genicity of high-LET radiations are generally an order of magnitude or
more greater than for low-LET radiations.
Radium, radon, radon daughters, and several other naturally occur-
ring radioactive materials emit alpha particles; thus, when these mate-
rials are ingested or inhaled, they are a source of high-LET particles
within the body. Man-made radionuclides are usually beta and photon
emitters of low-LET radiations. Notable exceptions to this generaliza-
tion are plutonium and other transuranium radionuclides, most of which
emit alpha radiation.
3.1 Evidence That Radiation Is Carcinogenic
The production and properties of x-rays were demonstrated within
one month of the public reporting of Roentgen's discovery of x-rays.
The first report of acute skin injury was made in 1896 (Mo67). The
first human cancer attributed to this radiation was reported in 1902
(Vo02). By 1911, 94 cases of radiation-related skin cancer and 5 cases
of leukemia in man had been reported in the literature (Up75). Efforts
3-1
-------
to study this phenomenon through the use of experimental animals pro-
duced the first reported radiation-related cancers in experimental ani-
mals in 1910 and 1912 (MaalO, Maal2). Since that time, an extensive
body of literature has evolved on radiation earcinogenesls in man and
animals. This literature has been reviewed most recently by the United
Nations Scientific Committee on the Effects of Atomic Radiation
(UNSCEAR), and by the National Academy of Sciences Advisory Committee on
the Biological Effects of Ionizing Radiations (NAS-BEIR Committee)
(UNSCEA182, NAS80).
Identification of the carcinogenicity of radioactive emissions fol-
lowed a parallel course. The first association of inhaled radioactive
material and carcinogenesis in man was made by Uhlig in 1921 in a study
of radon exposure and lung cancer in underground miners in the Erz Moun-
tains (Uh21). This association was reaffirmed by Ludewig and Lorenser
in 1924 (Lu24)). Ingestion of radioactive materials was also demon-
strated to be a pathway for carcinogenesis in man. As early as 1925
ingested radium was known to cause bone necrosis (Ho25), and in 1929 the
first report was published on the association of radium ingestion and
osteogenic sarcoma (Mab29).
The expected levels of exposure to radioactive pollutants in the
environment are too low to produce an acute (immediate) response. Their
effect is more likely to be a delayed response, in the form of an in-
creased incidence of cancer long after exposure. An increase in cancer
incidence or mortality with increasing radiation dose has been demon-
strated for many types of cancer in botb boinan populations and labora-
tory animals (UNSCEAR77»82), Studies of :. '.mans exposed to internal or
external sources of ionizing radiatioi have shown that the incidence of
cancer increases with increased radiation exposure. This increased
cancer, however, is usually associated with appreciably greater doses
and exposure frequencies than those encountered in the environment.
Malignant tumors most often appeared long after the radiation exposure,
usually 10 to 35 years later (NAS80, UNSCEAR82). The tumors appeared in
various organs. In the case of internal sources of radiation due to
radioactive materials, the metabolism of the materials generally leads
to their deposition in specific organs and results in a higher—than—
normal risk of cancer in these organs.
Whereas many, if not most, chemical carcinogens appear to be organ-
or tissue-specific, ionizing radiation can be considered pancarcinogen-
ic. According to Storer (Stb75): "Ionizing radiation in sufficiently
high dosage acts as a complete carcinogen in that it serves as both
initiator and promoter. Further, cancers can be induced in nearly any
tissue or organ of man or experimental animals by the proper choice of
radiation dose and exposure schedule." Radiation-induced cancers in
humans have been reported in the following tissues: thyroid, female
breast, lung, bone marrow (leukemia), stomach, liver, large intestine,
brain, salivary glands, bone, esophagus, small intestine, urinary blad-
der, pancreas, rectum, lymphatic tissues, skin, pharynx, uterus, ovary,
mucosa of cranial sinuses, and kidney (UNSCEAR77.82; NAS72.80; Be77,
Ka82, Wa83).
3-2
-------
A number of studies of populations exposed to high levels of radia-
tion have identified which organs are at greatest risk following radia-
tion exposure. Brief discussions of these findings follow.
1. Atomic Bomb Survivors - The survivors of the atomic bomb explo-
sions at Hiroshima and Nagasaki, Japan, were exposed to whole-body
external radiation doses of 0 to more than 200 rads.* An interna-
tional group has been observing the population since 1950. The
most recent reports published by this group (Ka82» Wa83) indicate
that an increase in cancer mortality has been shown for many can-
cers, leukemia, thyroid, breast, lung cancer, esophogeal and stom-
ach cancer, colon cancer, cancer of urinary organs, and multiple
myeloma.
2. Ankylosi.ig Spondylitics - A large group of patients were given
x-ray therapy for ankylosing spondylitis of the spine during the
years 1934 to 1954, X-ray doses usually exceeded 100 rad. British
investigators have been following this group since about 1957. The
most recent review of the data shows excess cancers in irradiated
organs, including leukemia, lynphoma, lung and bone cancer, and
cancer of the pharynx, esophagus, stomach, pancreas, and large
intestine (UNSCEAR77, NAS80).
3. Mammary Exposure - Sewral groups of women who were exposed to
x-rays during diagnostic radiation of the thorax or during radio-
therapy for conditions involving the breast have been studied.
Although most of the groups have been followed only a relatively
short time (about 15 years), a significant increase in the Inci-
dence of breast cancer has been observed (UNSCEAR77). The dose
that produced these effects averaged about 100 rads,
4. Medical Treatment of Benign Conditions - Several groups of
persons who were medically treated with x-rays to alleviate some
benign conditions have been studied. Excess cancer has developed
in many of the organs irradiated (e.g., breast, brain, thyroid, and
probably salivary glands, skin, bone, and pelvic organs) following
doses ranging from less than 10 to more than 100 rads (UNSCEAR77).
Excess leukemia has also occurred in some groups. The followup
period for most groups has been short, often less than 20 years.
5. Underground Miners - Studies of excess cancer mortali-iy in U.S.
underground miners exposed to elevated levels of radon started in
the 1950's and 1960's. Groups that have worked in various types of
mines, including uranium and fluospar, are being studied in the
United States, Canada, Great Britain, Sweden, China, and Czechoslo-
vakia. Most of the miners studied have been subjected to high
rates of exposure; however, a recent review indicates increased
incidence of lung cancer has been observed in some miners exposed
The rad is the unit of dose In common use; 1 rad equals 100 ergs of
absorbed energy per gram of material.
3-3
-------
at cumulative levels approximating those that can occur wherever
high environmental concentrations of radon are present (NAS80),
The dose response shown in all the study groups Is nearly propor-
tional to the dose (NAS80).
6. Ingested or Injected Radium - Workers who ingested Ra-226 while
painting clock dials have been studied for 35 to 45 years, and
patients who received injections of Ra-226 or Ra-224 for medical
purposes have been studied for 20 to 30 years (NAS72.80). Excess
incidence of leukemia and osteosarcoma related to Ra-224 exposure
has been observed. Calculated cumulative average doses for these
study groups ranged from 200 to 1700 rads. A study now under way
that deals with exposure levels under 90 rads should provide addi-
tional data (NAS80),
7. Injected Thorotrast - Medical use of Thorotrast (colloidal
thorium dioxide) as an x-ray contrast medium introduced radioactive
thorium and its daughters into a number of patients. Research
studies have followed patients in Denmark, Portugal, Japan $ and
Germany for about 40 years and patients in the United States for
about 10 years (UNSCEAR77, NAS80). An increased incidence of
liver, bone, and lung cancer has been reported in addition to
increased anemia, leukemia, and multiple myeloma (In79). Calcu-
lated cumulative doses range from tens to hundreds of rads.
8. Diagnostic X-ray Exposure During Pregnancy - Effects of x-ray
exposure of the fetus during pregnancy have been studied in Great
Britain since 1954, and several retrospective studies have been
made in the United States since that time (HAS80, UNSCEAR77).
Increased incidence "of leukemia and other childhood cancers may be
induced in populations exposed to absorbed doses of 0.2 to 20 rads
in utero (NAS80, UNSCEAR77).
Not all of the cancers induced by radiation are fatal. The
fraction of fatal cancers is different for each type of cancer. The
BEIR-3 committee estimated the fraction of fatal cancers by site and sex
(NAS80). Estimates of cancers by site ranged from about 20 percent
fatal in the case of thyroid cancer to 100 percent fatal In the case of
liver cancer. They concluded that, on the average, females have 2.00
times as many total cancers as fatal cancers following radiation
exposure, and males have 1.5 times as many (NAS80). Although many of
the radiation-induced cancers are not fatal, they still are costly and
adversely affect the persons life style for the remainder of his or her
life span. Just how these costs and years of impaired life should be
weighed in evaluating the hazards of radiation exposure is not certain.
In this assessment, only the risk of fatal carcinogenesis is addressed.
In addition to the evidence that radiation is a pancarcinogen and
as such can Induce cancers in nearly ar.y tissue or organ, it can also
induce cancer by any route of exposure (dermal, inhalation, ingestlon,
and injection).
3-4
-------
Inhalation is likely to be the major route of environmental expo-
sure to airborne radioactive pollutants, and the principal organ at risk
is likely to be the lung. Some radiation exposure to airborne pollut-
ants by the ingestion route is possible, however, as these pollutants
are deposited on soil, on plants, or in sources of water. Ingestion of
inhaled particulate also occurs. Some radionuclides may also cause
whole-body gamma radiation exposure while airborne or after deposition
on the ground.
Estimates of cancer risk are based on the absorbed dose of radia-
tion in an organ or tisoue. Given the sane type of radiation, the risk
for a particular dosage would be the same, regardless of the source of
the radiation. Numerical estimates of the cancer risk posed by a unit
dose of radiation in various organs and tissues ?_re presented in Chapter
8. The models used to calculate radiation doses from a specific source
are described in Chapters, 6 and 7.
The overwhelming body of epidemiological (human) data makes it
unnecessary to base major conclusions concerning the risk of radiation-
induced cancers on evidence provided by animal tests; however, these
data are relevant to the interpretation of human data (NAS80) and con-
tribute additional evidence to the epidemiological data base for humans.
Radiation-induced cancers have been demonstrated in several animal
species, including rats, nice, hamsters, guinea pigs, cats, dogs, sheep,
cattle, pigs, and monkeys. Induced through multiple routes of adminis-
tration and at multiple dose levels, these cancers have occurred in
several organs or tissues. These animal studies have provided infor-
mation on the significance of dose rate compared with the age of the
animals at exposure, the sex of the animals, and the genetic character-
istics of the test strain. They have shown that radiation-induced
cancers become detectable after varying latent periods, sometimes several
years after exposure. The studies further show that the total number of
cancers that eventually develop varies consistently with the size of the
dose each animal receives. Experimental studies in animals have also
established that the carcinogenic effect of high-LET radiation (alpha
radiations or neutrons) is greater than that of low-LET radiation (x-
rays or gamma rays).
A number of researchers have induced transformations in mammalian
tissue culture. Including the embryo cells of mice and hamsters (Bo84,
Ke84, Ha84, Gu84). Researchers have found that the DNA molecule is the
carrier of radiation-induced transformations and that the radiation
causes alterations in specific segments of genetic information (Bo84),
Kennedy and Little have postulated that radiation-induced cell transfor-
mation is a two-step process (Ke84). In the first step, an alteration
frequently occurs in a large fraction of the cells exposed to a large
dose (600-rad) or to a low dose (100-rad) and a promoting agent. The
second step is a rare event that occurs in one cell out of the million
cells that are produced from the irradiated cells and involves the
malignant transformation of that cell. This transformation occurs
randomly during the growth stage of irradiated cultures. A significant
finding of this research is that the process involved in the malignant
3-5
-------
:ransformation of mouse embryo cells caused by radiation is similar to
:hat caused by chemical carcinogens. Another major finding of recent
research (Gu84) is that DNA from radiation-induced mouse tumors contains
in activated oncogens that can transform specific types of cultured
:ells when introduced into these cells. The researchers also found that
2 difference in only one base in the oncogene was responsible for the
transformation. Thus, radiation can induce tumors even when only a
small change in the DNA occurs as a result of irradiation.
In like concentrations, radioactive materials are quite potent when
compared with chemical carcinogens. Chromosome aberrations in cultured
raman peripheral lymphocytes have been demonstrated at Rn-222 alpha
loses of about 48 mrads/y with an external gamma dose cf about 100
nrads/y (Po77), Use of the dose conversion factoi of these same inves-,
tlgators (Fi71) translates to a continuous exposure of about 0.042 pg/m
af Rn-222 and its daughters. Moreover, studies of underground miners
have demonstrated significant increases in the Incidence of lung cancer
it 50 cumulative working level months of Rn-222 exposure occurring
across a 17-year average period of exposure.* This is equivalent to
about 0.1 of the working level of Rn-222 and its daughters in resi-
dential atmospheres. An equivalent air concentration would be about 20
iCi/m of ln-222 or 0.130 pg/m of Rn-222 and its daughters. (For a
lore detailed discussion of working level exposures, see Chapter 8.)
3.2 Evidence That Radiation is Mutagenic
Radiation can change the structure, number, or genetic cor tent of
the chromosomes in a cell nucleus. These genetic radiation effects are
classified as either gene mutations or chromosomal aberrations. Gene
mutations refer to alterations of the basic units of heredity, the
genes. Chromosomal aberrations refer to changes in the normal number or
structure of chromosomes. Both gene mutation and chromosomal aberra-
tions are heritable; therefore, they are considered together es genetic
effects. Mutations and chromosomal aberrations can occur in somatic
(body) or germ (reproductive) cells. In the case of germ cells, the
atttagenic effect of radiation is not seen in those persons exposed to
the radiation, but in their descendents.
Mutations often result in miscarriages or produce such undesirable
changes in a population as congenital malformations that result in
nental or physical defects. Mutations occur in many types of cells; no
tendency toward any specific locus or chromosome has been identified.
For this reason, they can affect any characteristic of a species, A
relatively wide array of chromosome aberrations occur in both humans and
inlmals.
Early experimental studies showed that x-radiation is mutagenic.
En 1927, H. J. Muller reported radiation-induced genetic changes were
reported in animals, and in 1928, L. J. Stadler reported such changes in
^Personal cotmBtUiieatlon from E. P. Radford to Dr. Neal Nelson (ORP),
1981.
3-6
-------
plants (Ki62). Although genetic studies were carried out in the 1930's,
mostly in plants and fruit flies (Drosophila), the bulk of the studies
on mammals started after the use of nuclear weapons in World Mar II
(UNSCEM58).
Very little quantitative data are available on radiogenic mutations
in humans, particularly from low-dose exposures, for the following
reasons: these mutations are interspersed over many generations, some
are so mild they are not noticeable, and some mutagenic defects that do
occur are similar to nonmutagenic effects and are therefore not neces-
sarily recorded as mutations. The bulk of data supporting the mutagenic
character of ionizing radiation comes from extensive studies of experi-
mental animals, mostly mice (UNSCEAR77.82; NAS72,80). These studies
have demonstrated all forms of radiation mutagenesls—lethal mutations,
translocations, inversions, nondisjunction, point mutations, etc.
Mutation rates calculated from these studies are extrapolated to humans
(because the basic mechanisms of mutations are believed to be the same
in all cells) and form the basis for estimating the genetic impact of
ionizing radiation on humans (NAS80, UNSCEAR82). The vast majority of
the demonstrated mutations in human germ cells contribute to both in-
creased mortality and illness (NAS80, UNSCEAR82). Moreover, the radia-
tion protection community is generally in agreement that the probability
of inducing genetic changes increases linearly with dose and that no
"threshold" dose is required to initiate heritable damage to germ cells.
A considerable body of evidence has been documented concerning the
production of mutations in cultured cells exposed to radiation. Such
mutations have been produced in Chinese hamster ovary cells, mouse
lymphoma cells, human diploid fibroblasts, and human blood lymphocytes.
Many of the radiation-induced specific types of mutations produced in
human and Chinese hamster cultured cells are associated with structural
changes in the X chromosome. Evidence suggests that these mutations may
be largely due to deletions in the chromosomes. Thacker, Stretch, and
Stephens found that human, mouse, and Chinese hamster cells all exhibit
the same fixed probability of radiation-induced mutations (Th77).
Analysis of published data on x- or gamma radiation-induced mutations in
cultured cells of humans and mice show that when the induced mutation
frequencies are plotted against log of survival, the relationship is
linear. This relationship suggests that mutation frequency curves can
be predicted from a knowledge of survival curves for each cell type.
Mutagenicity in human somatic cells has been demonstrated on the
basis of chromosome aberrations detected in cultured lymphocytes.
Chromosome aberrations in humans have been demonstrated in lymphocytes
cultured from persons exposed to ingested Sr-90 and Ra-226 (Tu63); in-
haled/ingested ln-222, U-nat, or Pu-239 (Br77); or inhaled Sn-222 (Po78);
and in atomic bomb survivors (Aw78). Although no evidence of health
impact currently exists, these chromosome aberrations demonstrate that
mutageneeis is occurring in somatic cells of humans exposed to ionizing
radiation.
Evidence of mutagenesis in human germ cells (cells of the ovary or
testis) Is less conclusive. Studies have been made of several populations
3-7
-------
exposed to medical radiation, atomic bomb survivors, and a population in
an area of high background radiation in India (UNSCEAR77). Although
these studies suggest an increased incidence of chromosomal aberrations
in germ cells following exposure to ionizing radiation, the data are not
convincing (UNSCEAR77).
Investigators who analyzed the data on children born to survivors
of the atomic bombings of Hiroshima and Nagasaki found no statistically
significant genetic effects due to parental exposure (Sc81). They did
find, however, that the observed effects are in the direction of genetic
damage from the bomb radiation exposure. They also were able to calcu-
late that an average doubling dose* of 156 reins of ionizing radiation
will produce a 100 percent increase over the spontaneous mutation rate,.
The average doubling dose in mice is generally estimated to be much
lower, about 30 to 40 rems. These doses apply to acute radiation ex-
posure. Extensive experiments with mice indicate that the genetic yield
from low-level, chronic exposures to radiation is about one-third that
of acute radiation (Sc81). In a later report, the same researchers
estimated an acute doubling dose of 250 rems (Sa82).
The incidence of serious genetic disease due to mutations and
chromosome aberrations induced by radiation is referred to as genetic
detriment. Serious genetic disease includes inherited ill health,
handicaps, or disabilities. Genetic disease nay be manifest at birth or
may not become evident until some time in adulthood.
Researchers have attempted to measure genetic detriment due to
radiation exposure by using indices such as years of life lost, relative
length of hospitilizatiqn or medical care necessary, or time lost from
work. Measures of genetic detriment have several shortcomings. For
example, they do not differentiate with regard to the range of severi-
ties of a disease; nor do they include a measure of the impact of a
disease on the family, health care centers, schools, and society in
general. For example, measures of genetic detriment based on years of
life lost is much higher for Down's syndrome than for Huntington's
disease, largely because of the much higher incidence of Down's syn-
drome. The difficulty experienced by the families of those suffering
from each genetic disease is not accounted for, however. Those genetic
diseases that necessitate long-term stays in institutions may pose
burdens on society that are inversely related to mortality.
Carter and the U.N. Committee (Ca80,82; UNSCEAR82) have provided
approximate estimates of genetic detriment in a developed country. As
shown in Table 3-1, dominant genetic diseases usually rank relatively
low because their onset is late in life.
Using 'utilization of hospital services as an index of genetic
detriment, researchers have found that children with dominant or reces-
sive diseases or congenital malformations are, on the average, admitted
A doubling dose is one that will produce a 100 percent increase over
the spontaneous mutation rate.
3-8
-------
to hospitals 5 to 7 times more often In their first year of life (Tr77).
Children with any of these three types of genetic diseases spend con-
siderably more time in the hospital than other children.
Radiation-induced genetic detiiment thus includes Impairment of
life, shortened life span, and increased hospitalization. Only
estimates of the frequency of radiation-induced genetic impairment are
presented in Chapter 8 of this document. Although the numbers represent
rough approximations, they are relatively small in comparison with the
magnitude of detriment associated with spontaneously arising genetic
diseases (OTSCEAR82).
Table 3-1. Estimates of genetic detriment in a
developed country (UNSGEAR82)
Criteria forGenetic diseases, listed in the
genetic determinant order of severity (greatest to least)
lears of impaired life Chromosomal
X-linked
Recessive
Dominant
Irregularly inherited
Years of life lost Recessive
Irregularly inherited
X-linked
Dominant
Degree of life impairment Recessive
Chromosomal
X-linked
Dominant
Impaired life weighted for Recessive
degree of impairment Chromosomal
X-linked
Dominant
3.3 Evidence That Radiation Is Teratogenic
Teratogenicity is the malformation of cells, tissues, or organs of
a fetus resulting from physiologic and biochemical changes. Radiation
is a well-known teratogenic agent. Case reports of radiation-induced
teratology were made as early as 1921 (Sta21). By 1929, an extensive
review of a series of pregnancies yielded data indicated that 18 of the
children born to 76 irradiated mothers had abnormally small heads (mi-
crocephally) (Mu30). Although the radiation dose in these cases is not
known, it was high.
3-9
-------
Early experimental studies (primarily In the 1940's and 1950's) dem-
onstrated the teratogenip properties of x-rays in fish, amphibia, chick,
mouse, and rat embryos (Ru53). These experiments showed that the devel-
oping fetus is much more sensitive to radiation than the mother and
provided data on periods of special sensitivity and dose-response. The
malformations produced in the embryo depend on which cells, tissues, or
organs in the fetus are most actively differentiating at the time of
radiation. Embryos are relatively resistant to radiation-induced tera-
togenic effects during the earliest stages of their development, and are
most sensitive during development of the neuroblast (these cells eventu-
ally become the nerve cells). These experiments showed that different
malformations could be elicited by irradiating the fetus at specific
times during its development.
Substantial evidence points to the ability of radiation to induce
teratogenic effects in human embryos as well. In a recent study of
mental retardation in children exposed in_ utero to atomic bomb radiation
in Hiroshima and Nagasaki, researchers found that damage to the child
appears to be related linearly to the radiation dose that the fe.tus
receives (Ot84). The greatest risk of damage occurs at 8 to 15 weeks,
which is the time the nervous system Is undergoing the most rapid dif-
ferentiation and proliferation of cells. They concluded that the age of
the fetus at the time of exposure is the most Important factor in deter-
mining the extent and type of damage from radiation. A numerical esti-
mate of mental retardation risk due to radiation is given in Chapter 8.
3,4 Uncertainties
Although ouch is known about radiation dose-effect relationships at
high-level doses, uncertainty exists when dose-effect relationships
based on direct observations are extrapolated to lower doses, partic-
ularly when the dose rates are low. As described in Chapter 8, the
range of extrapolation varies depending on the sensitivity of the organ
system. For breast cancer, this may be as small as a factor of four.
Uncertainties in the dose-effect relationships are recognized to relate
to such factors as differences in quality and type of radiation, total
dose, dose distribution, dose rate, and radiosensitlvity (including
repair mechanisms, sex, variations in age, organ, and state of health).
The range of uncertainty in the estimates of radiation risk is examined
in some detail In Chapter 8.
The uncertainties in the details of mechanisms of carclnogenesis,
mutagenesis, and teratogenesis make it necessary to rely on the consid-
ered judgments of experts on the biological effects of ionizing radia-
tion. These findings, which are well documented in publications by the
National Academy of Sciences and the United Nations Scientific Committee
on the Effects of Atomic Radiation, are used by advisory bodies such as
the International Council on Radiation Protection and Measurements
(ICRP) in developing their recommendations. The EPA has considered all
such findings in formulating its estimate of the relationship between
radiation dose and response.
3-10
-------
Estimates of the risk from ionizing radiation are often limited to
fatal cancers and genetic effects. Quantitative data on the incidence
of nonfatal radiogenic cancers are sparse, and the current practice is
to assume that the total cancer incidence resulting from whole-body
exposure is 1.5 to 2.0 times the mortality. In 1980, the NAS-BEIR
Committee estimated the effects of ionizing radiation directly from
epidemiology studies on the basis of both cancer incidence and the
number of fatal cancers induced per unit dose (NAS80). The lifetime
risk from chronic exposure can be estimated from these data, either on
the basis of (1) relative risk (i.e., the percentage of increase in
fatal cancer), or (2) absolute risk (i.e., the number of excess cancers
per year at risk following exposure). The latter method results in
numerically smaller estimated risks for common cancers, but a larger
estimated risk for rare cancers.
3,5 Summary of Evidence That Radiation is a Carcinogen, Mutagen, and
Teratogen
Radiation has been shown to be a carcinogen, a mutagen, and a
teratogetv. At sufficiently high doses, radiation acts as a complete
carcinogen, serving as both initiator and promoter. With proper choice
of radiation dose and exposure schedule, cancers can be induced in
nearly any tissue or organ in both humans and animals. At lower doses,
radiation produces a delayed response in the form of increased incidence
of cancer long after the exposure period. This has been documented
extensively in both humans and animals. Human data are extensive and
include atomic bomb survivors, many types of radiation-treated patients,
underground miners, and radium dial workers. Animal data include
demonstrations in many mammalian species and in mammalian tissue cul-
tures. A significant finding from tissue culture studies is that radi-
ation induces cancers by a process that is similar to that of chemical
carcinogens. Further, DNA altered by radiation can cause transformation
of other cultured cells when introduced to normal cells, even when the
change in the DNA is very small.
Evidence of mutagenic properties of radiation comes mostly from
animal data, in which all forms of radiation-induced mutations have bean
demonstrated, mostly in mice. Tissue cultures of human lymphocytes have
also shown radiation-induced mutations. Data on humans are less conclu-
sive; however, estimates of genetic detriment due to radiation exposure
have been made by the use of measures such as years of life lost or
years requiring hospitalization.
Evidence that radiation is a teratogen has been demonstrated in
animals and in humans. A fetus is most sensitive to radiation during
the early stages of organ development (between 8 and 15 weeks for the
human fetus). The radiation-induced malformations produced depend on
which cells are most actively differentiating.
In conclusion, evidence of the carcinogenic, mutagenic, and tera-
togenic properties cf radiation is very substantial. These health
effects pose a detrimental risk to exposed persons.
3-11
-------
REFERENCES
Aw,8 Awa A. A. et al., Relationship Between Dose and Chromosome
Aberrations in Atomic Bomb Survivors, Hiroshima and Nagasaki,
RERF TR 12-77, Radiation Effects Research Foundation, Japan,
L9 / o.
Be77 Beebe G W., Kato H. and Land C. E. Mortality Experience of
Atomic Bomb Survivors, 1950-1974, Life Span Study Report 8,
RERF TR 1-77, Radiation Effects Research Foundation, Japan,
Bo84 ik>rek C., Ong A., Morgan V. and Cleaver J. E., Inhibition of
X-ray and Ultraviolet Light-Induced Transformation in Vitro by
Modifiers of Poly (ADP-ribose) Synthesis, Radiation~Resel^c"h
.99:219-227, 1984.
Br77 Brandom K. F. et al., Somatic Cell Chromosome Changes in
Humans Exposed to 239Plutoniura and 222ladon; Contract No.
E(29-2)-3639, Progress Report July 1, 1976, through September
30, 1977, Department of Energy, Washington, B.C., 1977.
Ca80 Carter C. 0., Some Rough Estimates of the "Load" From Spontan-
eously Arising Genetic Disorders, paper submitted to the U.N.
Scientific Committee on the Effects of Atomic Radiation,
September 1980.
Ca82 Carter C. 0., Contributions of Gene Mutations to Genetic
Disease in Humans, Progress in Mutation Research, Vol. 3,
Elsevier/North-Holland Biomedical Press, Amsterdam, 1982, pp.
1~8«
F171 Fischer P., Pohl-Ruling J. and Pohl E., Chromosome Studies on
Persons Exposed to Increased Levels of Radon in the Environ-
ment, Abstract No. 233, 4th International Congress of Human
Genetics, Paris, September 1971,
Gu84 Guerrero I., Villasante A., Corces V, and Pelllcer A., Activa-
tion of a c-K-ras Oncogene by Somatic Mutation in Mouse
Lymphoisas Induced by Gamma Radiation, Science 225, 1159-1162
September 14, 1984. *
3-12
-------
Ha84 Han A., Hill, C. K. and Elkind M. M., Repair Processes and
Radiation Quality in Neoplastic Transformation of Mammalian
Cells, Radiation Research 99, 249-261, 1984.
Ho25 Hoffman F. L., Radium (Mesothorium) Necrosis. J.A.M.A 85
961-965, 1925. —
In79 International Meeting on the Toxieity of Thorotrast and Other
Alpha-Emitting Heavy Elements, Lisbon, June 1977, Environ-
mental Research 18, 1-255, 1979.
Ka82 Kato H. and Schull W. J., Studies of the Mortality of A-Bomb
Survivors, Report 7 .?art 1, Cancer Mortality Among Atomic Bomb
Survivors, 1950-78, Radiation Research 90, 395-432, 1982,
Ke84 Kennedy A. R. and Little J. B., Evidence That a Second Event
in X-ray Induced Oncogenic Transfomation in Vitro Occurs
During Cellular Proliferation, Radiation Research 99_, 228-248,
1984.
Ki62 King R.C., Genstics, Oxford University Press, New York, 1962.
Lu24 Ludewig P. and Lorenser E., Untersuchung der Grubenluft in den
Schneeberger Gruben auf den Gebalt an Radiumemanation, Zschr
f. Phys. ^2, 178-185, 1924.
MaalO Marie P., Clunet J. and Raulot-Lapointe G., Contribution a
Letude du Developpement des Tumeurs Malignes sur les Ulceres
de Roentgen, Bull. Assoc. Franc. Etude Cancer, 3, 404, 1910,
cited in UNSCEAR77. ~~
Maal2 Marie P., Clunet J. and Raulot-Lapointe G., Nouveau cas de
Tumeur Maligne Provoquee par une Radlodermite Experimentale
Chez let Rat Blanc, Bull. Assoc. Franc. Itude Cancer 5, 125,
1912, cited in UNSCEAR77. ~
Mab29 Martland H. S. and Humphries E. E., Osteogenic Sarcoma in Dial
Painters Using Luminous Paint, Arch. Pathol. _7, 406-417, 1929.
Mo67 Morgan K. Z. and Turner J. E., Principles of Radiation Protec-
tion, John Wiley and Sons, Inc., New York, 1967.
Mu30 Murphy D. P. and DeRenyi M., Post-Conception Pelvic Irradia-
tion of the Albino Rat (Mug norvegicus); Its Effect on the
Offspring, Surgery, Gynecology aad Obstetrics 50, 861-863,
1930. —
NAS72 National Academy of Sciences - National Research Council, The
Effects on Populations of Exposures to Low Levels of Ionizing
Radiation, Report of the Committee on the Biological Effects
of Ionizing Radiations (BEIR Report), Washington, D.C., 1972.
3-13
-------
NAS80
Ot84
Po77
Po78
Ru53
Sa82
Sc81
Sta21
Stb75
Th77
Tr77
National Academy of Sciences - national Research Council, The
Effects on Populations of Exposures to Low Levels of Ionizing
Radiation, Committee on the Biological Effects of Ionizing
Radiation, Washington, D.C., 1980,
Otake M. Ph.D. and Schull W. Ph.D. , Mental Retardation in
Children Exposed in Utero to the Atomic Bombs: A Reassess-
ment, Technical Report RERF TR 1-83, Radiation Effects Research
Foundation, April 1984.
Pohl-Ruling J. , Fischer P. and Pohl E. , Einf luss Erhohter
Umweltradio-Aktivitat and Beruflicher Strahlen-Belastung auf
die Chromosomen-Aberrationen in den Lyophocyten des Peripheren
Blutes. Tagungsaber, Osteir.-Unga.r, Tagung uber biomedlzin,
Forschung, Seibersdorf, September 1977
Pohl-Ruling J.t Fischer P. and Pohl E., The Low-Level Shape of
Dose Response for Chromosome Aberrations, IAEA-SM-224/403,
presented at Internatinal Symposium on the Latent Biological
Effects of Ionizing Radiation, IAEA, Vienna, 1978.
Rugh R. , Vertebrate Radiobiology :
Sci. J3, 271-302, 1953.
Embryology, Ann. Rev. Kucl.
Satoh C. , Awa A. A., Neel J. V., Schull W. J., Kato H. , Hamil
ton H. B., Otake M. and Goriki K, , Genetic Effects of Atomic
Bombs, Hunan Genetics, Part A, The Unfolding Genome, Alan R.
Liss, Inc., »ew York, 1982, pp. 267-276.
Schull W. J., Otake M. and Neel J. V., Genetic Effects of the
Atomic Bombs: A Reappraisal, Science 213, 1220-1227, Septem-
ber 1981.
Stettner E. , Bin tfeiterer Fall einer Schadingung einer Men-
schlchen Frucht durch Roentgen Bestrahlung, Jb. Kinderheitk,
Phys. Irzieh. 95, 43-51, 1921.
Storer J. B., Radiation Carclnogenesis, Cancer 1, F. F.
Becker, editor, Plenum Press, New York, 1975, pp. 453-483.
Thacker J., Stretch A. and Stephens M. A., The induction of
Thioguanine-Reslstant Mutants of Chinese Hamster Cells by
Gamma Rays, Mutation Research 42 » 313-326, 1977.
Trliflble B. K. and Smith M. E. , The Incidence of Genetic Dis-
ease and the Impact on Man of an Altered Mutation Rate, Cana-
dian Journal of Genetic Cytology, 1!?, 375-385, 1977.
3-14
-------
Iu63 Tuscany R. and Klener V.» Pokles Euploldie v Bunkach Kostni
Drene osob s Vnitrni Kontaminaci Nekterymi Radioisotopy, Clsk.
Fysiol. ^2, 391, 1963.
Uh21 Uhlig M., Uber den Schneeberger Lungenkrebs, Virchows Arch.
Pathol. Anat. 230, 76-98, 1921.
UNSGEAR58 United Nations Scientific Conmittee Report on the Effects of
Atomic Radiation, Official Records: Thirteenth Session,
Supplement No. 17(A/3838), United Nations, New York, 1958.
UNSCEAR77 United Nations Scientific Committee on the Effects of Atonic
Radiation, Sources and Effects of Ionizing Radiation, United
Nations, New York, 1977.
UNSCIM82 United Nations Scientific Conmittee on the Effects of Atonic
Radiation, Ionizing Radiation: Sources and Biological Ef-
fects, United Nations, New York, 1982.
Up75 Upton A. C., Physical Carcinogenesis: Radiation—History and
Sources, Cancer j_f 387-403, 1975, F. F. Becker, editor.
?o02 Von Frieben A., Demonstration Lines Cancroids des Rechten
Handruckens, das sich nach Langdauarnder Einwirkung von
Rontgen-strahlen Entwickelt Hatte, Fortschr. Geb. Rontgenstr.
j>» 106, 1902, cited in Up75.
Wa83 Wakabayashi T., Kato H., Ikeda T. and Schull W. J.f Studies of
the Mortality of A-Bomb Survivors, Report 7, Part III, Inci-
dence of Cancer in 1959-1978, based on the Tumor Registry,
Nagasaki, Radiat. Res. 93, 112-146, 1983.
3-15
-------
Chapter 4; EMISSION OF MDIONUCLIDES INTO THE AIR
4.1 Introduction
Radionuclides are used or produced in thousands of locations
throughout the United States, including national defense weaponry
facilities, nuclear powerplants, industrial plants, research and devel-
opment laboratories, and medical facilities. Fossil-fuel combustion
processes, such as large coal-fired boilers, make some contribution to
the exposure of the general public. Certain kinds of mining and milling
also substantially increase the local concentration of radionuclides in
the air,
Although air cleaning equipment is usually used in these facilities,
some radionuclides are still released into the air and can disperse
into populated areas.
Sources of emissions of radionuclides to the air can be divided
into the following groups;
(1) Department of Energy facilities
(2) Nuclear Regulatory Commission licensed facilities (exclusive
of commercial nuclear power generating facilities) and non-
DOE Federal facilities
(3) Coal-fired utility and industrial boilers
(4) Underground uranium mines
(5) Phosphate rock processing and wet-process fertilizer plants
(6) Elemental phosphorus plants
(7) Other mineral extraction and processing facilities
(8) Uranium fuel cycle facilities, including uranium mill tailing
and high-level waste disposal facilities and commercial
nuclear power generating facilities
(9) Low-energy accelerators.
4-1
-------
These sources are described in the chapter, including the physical
characteristics of the radionuclide releases (i.e., particle size and
physical state) and the amount of release. This information is used in
subsequent chapters to evaluate the movement of radlonuclides through
the environment.
Because of the large number of facilities within certain source
categories (e.g., coal-fired industrial boilers), conducting a risk
assessment for each plant is not practicable. Therefore, in some
cases, it was necessary to develop a model or reference facility upon
which to base a risk assessment. The reference facilities were devel-
oped from data obtained from multiple facilities within a source cate-
gory. These data reflect the range in operating parameters and radio-
nuclide emissions that are representative of the particular source
category. The operating parameters and radionuclide emission rates of
those source categories for which reference facilities were developed
are discussed in the following subsections of this chapter.
4.2 Sources of Radionuclide Releases into the Air
Naturally occurring and manmade radionuclides are emitted to air
from a variety of sources. Sources of manmade radionuclides include
nuclear powerplants and other facilities that use nuclear fuel, research
and development laboratories, medical facilities, and national defense
facilities. The type and quantity of radionuclide emissions from these
sources are typically well defined. Mining, mineral processing! and
fossil-fuel combustion are also potential sources of naturally occurring
radionuclides.
The discussions of the radionuclide emission sources provided in
the following subsections include descriptions of the various facilities,
specific emission release points (or areas), and a general description
of the types of emissions. The quantities emitted are provided in
Section 4.3.
4.2.1. Department of Energy (DOE) Facilities
The DOE owns or directs under contract many facilities that amit
radionuclides into the air. The largest of these facilities and their
locations are listed in Table 4.2-1, These facilities support weapons
production and numerous research and development programs for the
Department of Defense (DOD), including biomedical studies, studies of
environmental and safety aspects of nuclear energy, and investigations
concerning nuclear waste processing.
The diversity of operations among the various sites makes it
difficult to assess DOE facilities on a generic basis. The major
emissions from the various facilities, however, are similar and consist
largely of inert gases such as argon (Ar-41), krypton (Kr-85 and 88),
and xenon (Xe-133). These gases are heavier than air and only slightly
soluble in water. Tritium (H-3) and oxygen (0-15) are also commonly
emitted. A site-by-slte review of each sovirce follows. Volume II,
Chapter 2, of this document discusses each of these facilities in
greater detail.
4-2
-------
Table 4.2-1. Department of Energy facilities
Laboratory
Location
Argonne National Laboratory
Brookhaven National Laboratory
Feed Materials Production Center
Fermi National Accelerator Laboratory
Hanford Reservation
Idaho National Engineering Laboratory
Lawrence Livermore National Laboratory
Los Alamos National Laboratory
Oak Ridge Reservation
Paducah Gaseous Diffusion Plant
Portsmouth Gaseous Diffusion Plant
Rocky Flats Plant
Savannah River Plant
Ames Laboratory
Bettis Atomic Power Laboratory
Knolls Atomic Power Laboratory
Lawre,ice-Berkeley Laboratory
Mound Facility
Nevada Test Site
Pantex Plant
Pinellas Plant
Rockwell International
Sandia National Laboratories
Stanford Linear Accelerator Center
Reactive Metals, Inc.*
Argonne, Illinois
Long Island, New York
Fernald, Ohio
Batavia, Illinois
Richland, Washington
Upper Snake River, Plain, Idaho
Livermore, California
Los Alamos, New Mexico
Oak Ridge, Tennessee
Paducah, Kentucky
Piketon, Ohio
Jefferson County, Colorado
Aiken, South Carolina
Ames, Iowa
West Mlflin, Pennsylvania
Schenectady, New York
Berkeley, California
Miamisburg, Ohio
Nya County, Nevada
Amarillo, Texas
Pinellas County, Florida
Santa Susana, California
Albuquerque, New Mexico
Stanford, California
Ashtabula, Ohio
See Volume II for description.
Argonne National Laboratory
Argonne National Laboratory is an energy research and development
center that performs investigations In basic physics, chemistry, mate-
rials science, the environmental sciences, and biomedicine. Argonne
also plays an Important role as a nuclear and nonnuclear engineering
center. The laboratory complex is located in Dupage County, Illinois,
43 kilometers southwest of Chicago.
Argonne National Laboratory has the following principal nuclear
facilities:
(1) 10- and 200-kW research reactors
(2) A critical assembly reactor
(3) A 60-inch cyclotron
4-3
-------
(4) A prototype, superconducting, heavy ion linear accelerator
(5) Van de Graaff and Dynaoitran-type charged-partiele accelera-
tors
(6) A high-energy neutron source
(7) Cobalt-60 irradiation sources
(8) Laboratories engaged in work with aiulticurie quantities of
the actinide elements
The 200-kW JANUS research reactor and the laboratory handling area (hot
cells) are the main sources of radionueiide releases from the Argonne
complex.
Specific details of the site activities and emissions are availa-
ble from annual emission reports prepared by the laboratory (G082), the
DOE Effluent Information System (DOESla), and environmental monitoring
studies conducted by DOE (ERDA77a).
Brookhaven RationalLaboratory
Studies conducted at Brookhavan Laboratories pertain to the use,
environmental effects, and transport of both nuclear and nonnuclear
energy materials. Other research programs include applied nuclear
studies involving various radioisotopes and investigations of the
physical, chemical, and biological effects of radiation. Brookhaven
Laboratory is located in the canter of Long Island, about 113 kilome-
ters from New York City.
The equipment and facilities used to support the research projects
conducted at Brookhaven include several reactors, particle accelera-
tors, and laboratories. Point and area sources of radionueiide releas-
es at Brookhaven include:
(1) The 40->fW High-Flux Beam Reactor (HFBR)
(2) The Alternating Gradient Syncrotron, a proton accelerator
used in ultra-high energy particle physics research
(3) The Brookhaven Linac Isotope Production Facility (BLIP)
(4) The Chemistry Linac Irradiation Facility (CLIF)
(5) The Brookhaven Medical Research Reactor
(6) The Van de Graaff accelerator
(7) Various chemistry and medical research laboratories
Most of the airborne radionueiide emissions from Brookhaven originate
from the High-Flux Beam Reactor, the Brookhaven Linac Isotope Produc-
tion Facility, and the Van de Graaff research generator. Lesser
emissions are from the chemistry and medical research centers.
4-4
-------
During 1981, emissions were Identified from seven stacks, as list-
ed in Table 4,2-2. Because very snail quantities of radionuclides are
released from the Hazardous Waste Management Area, the assessments of
exposure and health risk at the Brookhaven site are based on airborne
releases from the remaining six effluent stacks. Process descriptions,
effluent data, and site information were obtained from reports prepared
by Brookhaven Laboratories (Na82) and DOE studies (BOEBla, ERDA77a).
Table 4.2-2.
Radionuclide emission points (stacks) at Brookhaven
National Laboratories
Stack
Location height (m)
Irookhaven Llnac Isotope Production Facility, Buildlng-931 46
High-Flux Beam Reactor Hot Laboratory 98
Hazardous Waste Management Area 10
Medical Research Reactor Building-491 Unknown
Chemistry Building-555 Unknown
Medical Research Center Unknown
Van de Graaff Accelerator Iuilding-901 18
Fermi National Accelerator Laboratory
The Fermi National Accelerator Laboratory is principally involved
with basic research in high-energy physics. Another important activity
involves the treatment of cancer patients with neutrons released by the
second stage of the accelerator. The Fermi complex is located east of
Batavia, Illinois, in the greater Chicago area.
The accelerator at the Fermi Laboratory, a proton synchrotron,
routinely operates at energies up to 400 GeV (billion electron volts).
The proton beams produced in the accelerator are used in three differ-
ent onsite experimental facilities: (1) the Meson area, (2) the
Neutrino area, and (3) the Proton area. Production of radlonuclides in
these areas and by the accelerator occurs when either the proton beam
Itself or secondary particles interact with air.
Another source of radionuclides at Fermi Laboratory is a magnet-
debonding oven, where failed magnets for the accelerator are baked at
high temperatures to break down the adhesiveg that help form the mag-
nets.
4-5
-------
Hanford Reservation
The Hanford Reservation was established in 1943 as a plutonium
production facility for nuclear armaments. Information used to evalu-
ate the facility was obtained from DOE and Hanford reports (DOESia,
ERDA75, ERDA77b» Su82). Plutonium production has decreased, and other
programs filled the gap, such as management and storage of radioactive
wastes, reactor operations, fuel fabrication, energy research and
development, and biophysical and biomedical research. The reservation,
which is located 270 kilometers south of Seattle, Washington, is sepa-
rated into four areas, which are designated the 100, 200, 300, and 400
Areas. The activities of each area are described briefly.
100 Area. The 100 Area contains the nine plutonium production
reactors for which the site was originally developed. Eight of these
reactors are currently on standby. Operating facilities include the
N-Reactor and the 1706 Laboratory, which provides support services for
the reactor.
200 Area. Activities conducted in the 200 Area include fuel
processing, nuclear waste treatment and storage, equipment decontamina-
tion, and research. Plutonium reclamation from spent fuel is performed
at the PUREX Plant in this area.
300Area. The major facilities in the 300 Area are the Hanford
Engineering Development Laboratory, the fuel fabrication facility, and
the Life Sciences Laboratory. The Hanford Engineering Development
Laboratory, the largest operation in this area, supports all activities
of the development program for the fast breeder reactor. Life science
research in this area includes plutonium Inhalation studies and other
programs investigating the physiological effects of radioactive materi-
als.
400 Area. The only facility currently In operation in the 400
Area is the Fast Flux Test Facility. When the Fuel Materials Examina-
tion Facility currently under construction is completed, the_4QQ Area
will be the center of the Hanford breeder reactor research program.
Idaho National Engineering Laboratory
The Idaho National Engineering Laboratory is a reactor testing
facility in southeastern Idaho, about 56 kilometers west of Idaho
Falls. The following four contractors operate facilities here: EG&G
Idaho, Inc.; Allied Chemical Corporation} Argonne National Laboratory;
and Westinghouse Eleciric Corporation.
EG&G Facilities. EG&G, Inc., operates several test reactors.
These reactors provide operating information for the development of
reactor safety programs, for determination of the performance of reac-
tor materials and equipment, and occasionally, for use in research per-
formed fay private organizations. Other activities include disassembly
4-6
-------
and reassembly of large radioactive reactor components, preparation of
test specimens for use in various operating reactors, and waste han-
dling.
Allied Chemical Corp., Idaho ChemicalProcessing Plant. Fuel
processing is the major operation that Allied conducts at this site.
The Idaho Chemical Processing Plant stores irradiated fuel and re-
processed fuel, and converts high-level-radioactive liquid waste to
solid form.
ArgonneNational Laboratory, West Facility. The Argonne National
Laboratory complex currently has five operational facilities: the
Experimental Breeder Reactor, the Transient Reactor Test Facility, the
Zero Power Plutonium Reactor, the Hot Fuels Examination Facility, and
the Laboratory and Office Support Complex, Each of these facilities
provides research and physical support for the DOE fast breeder reactor
project,
Westinghouse Electric Corporation. Westinghouse operates the
Naval Reactor Facility at the Idaho Laboratory. This facility serves
as a testing area for prototype naval reactors and as a disassembly and
inspection area for expended reactor cores (DOESla, DOE82a, ERDA77a,
ERDA77c).
Lawrence Livermore National Laboratory
The Lawrence Livermore National Laboratory, situated 64 kilometers
east of San Francisco, California, is primarily a nuclear weapons
research and development center. Other activities, however, include
research programs in laser isotope separation, laser fusion, magnetic
fusion, biomedical studies, and nonnuclear energy.
Two accelerators, the Insulated Core Transfer Accelerator and the
Electron Positron Linear Accelerator, are used in support of the fusion
and neutron physics research programs. The Light Isotope Handling
Facility supports research in the area of light isotopes. The remain-
ing facilities at this site deal with equipment decontamination and
waste disposal (DOE81a, UC82).
Los Alamos National Laboratory
Los Alamos National Laboratory is one of the prime research and
development centers for DOE's nuclear weapons program. This facility
is located about 100 kilometers north-northeast of Albuquerque, New
Mexico. In addition to defense-related activities, programs include
research in the physical sciences, energy resources, environmental
studies, and biomedical applications of radiation.
Radionuclides are released from 13 technical areas at this site.
These areas contain research reactors that produce materials for use in
high-temperature chemistry applications, weapons systems development,
nuclear safety program development, accelerator operations, bioiaedical
4-7
-------
research, development of Isotope separation processes, and waste dis-
posal (DOESla, LANL82).
Oak Ridge Reservation
Oak Ridge National Laboratory, located about 35 kilometers west of
Knoxville, Tennessee, is a multldiselplinary research facility that
conducts basic and applied research into all aspects of energy produc-
tion. Three major facilities are located on the Oak Eidge Reservation:
Oak Ridge National Laboratory, Oak Ridge Gaseous Diffusion Plant, and
the Y-12 plant.
The equipment facilities used to support research activities at
Oak Ridge Laboratory include nuclear reactors, chemical pilot plants,
research laboratories, and waste disposal and handling areas. Radionu-
clide emissions are released during isotope preparation and chemistry
laboratory operations. Emissions from the Gaseous Diffusion Plant and
Y-12 plant generally consist of particulates released during fuel
processing and enrichment (DOESla, EPA79a, UCC82a).
Paducah Gaseous Diffusion Plant
The DOE operation at the Paducah Gaseous Diffusion Plant consists
of a uranium enrichment facility and a uranium hexafluoride
manufacturing complex. The plant is located 6 kilometers south of the
Ohio River in McCrasken County, Kentucky.
The primary activity at this site is the diffusion cascade for the
enrichment of uranium in fissionable uranium-235 content. All stages
of the enrichment cascade take place in five buildings on the site.
The manufacturing facility produces uranium hexafluoride from uranium
oxide feedstocks (DOESla, UCC82b).
Portsmouth Gaseous Diffusion Plant
The Portsmouth Gaseous Diffusion Plant, situated in Pike County,
Ohio, about 1.6 kilometers east of the Scioto River, is operated by
Goodyear Atomic Corporation. Its primary function is the production of
enriched uranium. Operations at this plant are similar to those de-
scribed for the Paducah Gaseous Diffusion Plant. The most significant
release point, which accounts for about 84 percent of total emissions,
is the X326 Top Purge Vent (DOESla, EPA?9a, GAC82).
Rocky Flats Plant
Activities at the Rocky Flats Plant, located in Jefferson County,
Colorado, about 26 kilometers from Denver, are restricted to fabrica-
tion and assembly of components for nuclear weapons and the support of
these operations.
Fabrication operations include reduction rolling, blanking, form-
ing, and heat treating. Assembly operations include cleaning, brazing,
4-8
-------
marking, welding* weighing, matching* sampling, heating, and monitor-
ing. Because of the toxicity of plutonium, all material-handling
activities that involve plutonium are performed under strictly con-
trolled conditions.
Solid residue generated during plutoniuns-related operations is
recycled through one of two plutoniuo-recovery processes. Process
selection depends on the purity and plutonium content of the residue.
Both processes produce a plutonium nitrate solution from which the
metal can be extracted. The recovered plutonium is returned to the
storage vault for use in foundry operations. A secondary objective of
the process is the recovery of americium-241.
Radionuclides are, released from short stacks and building vents at
this plant. Quantities are similar at several of the release points.
Building 771, Main Plenum, was selected for comparison purposes and
calculations. This point releases 54 percent of the plutonium-239 and
-240 and 3 percent of the uranium-233, -234, and -235 emitted at Rocky
Flats. The most significant release point for uranium is from a single
duct in Building 383, which releases approximately 19 percent of the
total uranium emissions from the plant (DOE81a, EPA79a).
Savannah River Plant
The facilities at the Savannah River Plant are used primarily to
produce plutonium and tritium, the basic materials required for nuclear
weapons. Materials for medical and space applications are also manu-
factured here, however. The Savannah River Plant is situated along the
Savannah River at a site 35 kilometers southeast of Augusta, Georgia.
The site covers about 770 square kilometers.
Operations are grouped into five major areas (designated the 100,
200, 300, 400, and 700 Areas) according to their operational function
in the plutonium manufacture/recovery process.
100 Area - Nuclear Production Reactors. These production reactors
are currently in operation; a fourth is being upgraded. The three
operating reactors produce plutonium and tritium by Irradiation of
uranium and lithium. Heavy water is used both as a neutron moderator
and as a primary coolant.
200 Area - Separations and Waste Management Facilities. Nuclear
fuel reprocessing occurs in this area. Plutonium is recovered from
irradiated uranium by the PUREX solvent-extraction process. Enriched
uranium and plutonium-238 are recovered from other irradiated materials
by a solvent-extraction procedure similar to the PUREX process.
300 Area - Fuel and Target Fabrication. Tubular fuel and target
elements are produced by cladding depleted uranium fuel In aluminum or
aluminum/lithium shells. A low-power reactor and a subcritical test
reactor are then used to test for assembly defects.
4-9
-------
400 Area -Heavy Water Production and Recovery. Heavy water is
produced fron river water by distillation and extraction. Heavy water
is also recovered from contaminated reactor coolant. Heavy water is
transported from this area to the 100 Area for use in the production
reactors.
700 Area - The Savannah River Laboratory. Research and process
development work is performed at the Savannah River Laboratory, Major
activities in this area include fabrication of fuel element and target
prototypes; fabrication of radioisotopic sources for medical, space,
and industrial applications; thermal and safety studies of reactor
operations; and applied research in the areas of physics and the envi-
ronmental sciences (DOESla, DOESlb).
Feed Material Production Center
The Feed Material Production Center, located 32 kilometers north-
west of Cincinnati, Ohio, produces uranium metal and other materials
for DOE facilities. The uranium may be natural, depleted, or enriched
with respect to uranium-235.
Raw materials are processed in the following manner. The material
is first dissolved in nitric acid and separated by liquid organic
extraction. The recovered uranium is reconverted to uranyl nitrate,
heated to form uranium trioxide, reduced to uranium dioxide with hydro-
gen, and reacted with hydrogen fluoride to form uranium tetrafluoride.
Purified metal is made by reacting the uranium tetrafluoride with me-
talic magnesium in a refractory-lined vessel (DOE81a, EP.479a, ERDA7?d).
Ames Laboratory
Until 1978, the Ames Laboratory, which is operated by Iowa State
University, was used as a neutron source for the production of byprod-
uct materials and the neutron irradiation of various materials for
research. The reactor was fueled with enriched uranium, moderated and
cooled by heavy water (DO), and operated continuously at 5000 watts
thermal. Operation of tne Ames Laboratory Research Reactor was termi-
nated on December 1, 1977. Decommissioning began January 3, 1978, and
was completed on October 31, 1981. A waste processing and disposal
facility that is still located at the site serves the campus reactor
and research laboratories.
Prior to its decommissioning, the major airborne releases from the
research reactor were tritium and argon-41. Tritium, the major radlo-
nuclide released during the 1981 decommissioning activities, was emit-
tpd from the 30-meter reactor stack, which is 215 meters from the
nearest property boundary. Monitoring has indicated that no airborne
emissions from the research laboratories have reached the main campus
(EPA79a, DOESla, DOE82b, ERDA77e).
4-10
-------
Bettls Atomic Foyer Laboratory
The Bettis Atomic Power Laboratory is situated on an 0,8-square-
kilometer tract in West Mifflift, Pennsylvania, approximately 12 kilome-
ters south of Pittsburgh. This facility designs and develops nuclear
power reactors. The most significant program currently in progress is
the fabrication of fuel assemblies for DOE's light-water breeder reac-
tor program (BAPL82, DOESla, DOE82c, ERDA77a).
Knolls Atomic Power Laboratory
Knolls Atomic Power Laboratory has facilities at three separate
sites: Knolls, Kesselring, and Windsor. Development of nuclear
reactors and training of operating personnel are the major efforts at
the Knolls Laboratory. The Knolls and Kesselring complexes are located
near Schenectady, New York, and the Windsor site is near Windsor,
Connecticut. Pressurized water reactors are located at both Kesselring
and Windsor sites, where operating personnel are trained.
All releases of radionuclides from the Knolls sites are from
elevated stacks (DOIBla, DOE82c, ERDA?7a).
Lawrence-Berkele}' Laboratory
The research facilities of Lawrence-Berkeley Laboratory are locat-
ed on the Berkeley campus of the University of California. These
facilities include four large accelerators, several small accelerators,
several radiochemical laboratories, and the Tritium Labeling
Laboratory, The large accelerators include the Bevatron, the Super
H1LAC, the 224-centimeter Sector~Focused Cyclotron, and the
467-centimeter Cyclotron.
The Tritium Facility was designed to accommodate kilocurle quanti-
ties of tritium as a labeling agent for chemical and biomedical re-
search, Radiochemical and radiobiological studies in many laboratories
typically use millicurie quantities of various radionuclides (DOESla,
LBL81).
Mound Facility
The Mound Facility, located in Miamisburg, Ohio, about 16 kilome-
ters southwest of Dayton, Ohio, has a variety of active programs.
These Include -research and development, processing of solid wastes for
tritium recovery, fabrication and testing of weapons components, pro-
duction of stable isotopes for the market, and manufacture of radioiso-
topic heat sources for military and aerospace applications.
The principal eaisslons of tritium and plutoniura emanate from nine
buildings, designated as HH, SW, H, PP, R, SM, WD, WDA, and 41. Build-
ings HH and SW, which contain the tritium recovery and reprocessing
facilities, are the sole release points of tritium. Plutonium is
released from the other facilities as a result of heat source produc-
tion and waste disposal operations {DOEBla, EPA79a, Fa82).
4-11
-------
Nevada Test Site
The Nevada Test Site lies about 100 kilometers northwest of Las
Vegas, Nevada, in Nye County, This facility, which is part of DOE's
weapons research and development complex, is responsible for design,
maintenance, and testing of nuclear weapons. Other activities at this
site Include development of new nuclear energy technologies and radio-
active waste disposal.
Radionucllde emissions result primarily from underground tests of
nuclear weapons. Sources of these releases include drill-back opera-
tions, tunnel ventilation, leakage of gases from underground test
sites, and resuspension of contaminated soils (DOE81a, ERDA77a,
ERDA77f).
Pant ex Plant.
The Pantex Plant, located 30 kilometers northeast of Amarillo,
Texas, is a nuclear weapons assembly and disassembly plant. Because
most radioactive materials handled during the assembly of nuclear
weapons are contained in sealed vessels, normal operations Involving
these materials do not result in major releases of radionuclides
(DQESla, BOE82c, ERDA77a, Mab82).
Pinellas Plant
The Pinellas Plant, located 10 kilometers northwest of St. Peters-
burg, Florida, is a major facility engaged in the production of nuclear
weapons. Although descriptions of the principal operations resulting
in atmospheric releases of radioactive materials could not be found in
the literature, they are neutron generator development and production,
testing, and laboratory operations. Small, sealed, plutonium capsules
are used as heat sources in the manufacture of radioisotopic thermoe-
lectric generators. The heat sources are triple-encapsulated to pre-
vent release of plutonium to the atmosphere.
Emissions of radionuclides were identified from three sources:
the Main stack, Laboratory stack, and Building stack (BOESla, EPA79a).
Rockwell International
Rockwell International operates two facilities, one near Los
Angeles and one near Santa Susana, California. These facilities con-
duct research and development and also manufacture nuclear reactor
components. The Los Angeles facility performs uranium fuel processing
operations and conducts research involving gamma radiation. The Santa
Susana facility uses neutron radiography to Inspect nuclear reactor
components. This facility also serves as a materials handling labora-
tory and waste processing operation for other DOE facilities.
Radlonuclide emissions originate from the materials handling
laboratory and the waste processing facilities at the Santa Susana site
(DOESla, EPA79a, ESG82).
4-12
-------
Sandia National Laboratories
The operations at Sandia National Laboratories near Albuquerque,
New Mexico, include weapons testing, arming and fusing nuclear weapons,
and developing modifications to delivery systems. The major facilities
include the Sandia Pulsed Reactor, the Annular Core Pulsed Reactor
(both of which are used to irradiate test materials), and the Relativ-
istic Electron Beam Accelerator. Support facilities include the Neu-
tron Generator Facility, the Tube Loading Facility, the Fusion Target
Loading Facility, the Tritium Laboratory, and the Nondestructive Test
Facility, all of which are located in Technical Areas (TA) 1 and V.
TA-I, in the northwest corner of the site, also houses research and
design laboratories. Technical Area III is the site of the Sandia
low-level radioactive waste dump (DOESla, DOE82c, ERDA77a, SNL82).
Stanford Linear Accelerator Center
*ie Stanford Linear Accelerator Center is a large research labora-
tory devoted to theoretical and experimental research in high-energy
physics and to the development of new techniques in high-energy accel-
erator particle detectors. This accelerator complex is located about
halfway between San Francisco and San Jose, California. The main tool
of the laboratory is a linear accelerator that is used to accelerate
electrons and positrons. Two storage ring facilities, SPEA1 and PEP,
are used to generate high-energy particles by the collision of the
opposing particle beams. Colliding beam storage rings such as SPEAR
and PEP truly "recycle" the beam particles. The same particles are
brought into collision over and over again, rather than striking a
target only once. For this reason, colliding-beam devices generate
much less radiation and residual radioactivity than do conventional
accelerators.
No immediate venting of the accelerator facility occurs. A wait-
ing period allows all isotopes (with the exception of argon-41) to
decay before they are exhausted from the facility. Therefore, the
release of radioactivity is infrequent and limited to argon-41 for
brief periods of 30 to 60 minutes. Airborne releases are from the
accelerator vent stack (DOE81a, DOESlc).
4.2,2 Nuclear Regulatory Commission (NRG) Licensed Facilities and
non-DOE Federal Facilities
This source category encompasses six different classifications of
facilities: Research and Test Reactors, Accelerators, Radiopharmaeeu-
tical Manufacturing, Department of Defense Activities, Radiation Source
Manufacturing, and other NRC Licensees. These facility classifications
include sources licensed by NRC and sources licensed by States under an
agreement with NRC. Most of the emissions are gases containing iso-
types of argon, hydrogen, oxygen, nitrogen, krypton, and xenon. Small
amounts of iodine and technetium occur from radiopharmaceutical
4-13
-------
applications. These sources are discussed in more detail in Volume II,
Chapter 3. Brief discussions of the characteristics of the major
sources follow.
Research and Test Reactors
This category consists of land-based, NRC-licensed reactors that
are operated for purposes other than commercial power production, i.e.,
for basic and applied research and for teaching. Seventy such reactors
are currently licensed to operate In the United States. Design types
include heavy water, graphite, tank, pool, homogeneous solid, and
uranium-zirconium hydride. These reactors are operated to test reactor
designs; to test reactor components and safety features; to support
basic and applied research in the fields of physics, biology, and
chemistry; and to educate. Power levels range from near zero to 10 MW.
Airborne emissions from research and test reactors are usually
limited to argon-41 and tritium. Because the emissions from each of
the facilities in this classification generally vary only in the quanti-
ty of radionuclides emitted, a reference facility was chosen to be
representative of the entire group for the purpose of determining
emission characteristics for dose and health risk assessments.
The reference facility used in the risk assessments was a univer-
sity heavy-water reflected reactor. This reactor can achieve a power
level of 5 MW with atmospheric emissions equal to the highest levels
reported to the NEC (NRC77, Ki79). A stack height of 50 meters was
used for assessments of the radionuclide releases.
Radiopharmaceutical Industry
The use of radioactive materials in medical treatments and re-
search has been steadily increasing. Manufacturers/suppliers, users,
and waste receivers are sources of radionuclide emissions from this
source category.
Manufacturers/Suppliers« Radiopharoaceutlcal manufacturing in-
volves numerous chemical processes with the potential for releasing
radioactive materials into the air. Most materials used in the manu-
facture of radiopharmaceuticals are produced in nuclear reactors.
Reactor-produced materials account for up to 80 percent of the marketed
Pharmaceuticals.
In a reactor, the main steps in radionuclide production are as
follows (EPA80):
(1) A suitable target is prepared and irradiated with neutrons
(2) For removal of undesirable impurities or for concentration of
the desired product, irradiated targets are processed by
dissolution or by more complicated processes, such as ion
exchange, precipitation, or distillation.
4-14
-------
(3) Radionuclides are placed in inventory, dispensed, and pack-
aged for shipment.
Smaller amounts of certain radionuclides are also produced in
particle accelerators, such as the cyclotron. Cyclotrons can be used
to produce nuclides with decay characteristics that are preferable to
those of other isotopes of the same element produced in reactors.
Cyclotrons can also be used to produce Isotopes of elements for which
no reactor-produced nuclides exist.
The emissions from the reference facility represented the highest
values reported. Airborne releases are from a single 15-meter-high
stack (TRI79a, EPA79a, NRC81, Coa82, Fra82a, Fra82b, Le79, Ro82b).
Users. Radionuclides are used extensively for medical diagnosis,
therapy, and research. The number of medical facilities using radioac-
tive materials has grown from 38 in 1946 to more than 10,000 at the
present time. With the exception of radioactive gases such as xenon,
radionuclides are generally handled by hospitals in solid or liquid
form, which tends to decrease the likelihood of airborne radiomielide
releases.
For assessments of dose and health risk, a reference facility was
designated as a typical effluent source for subsequent modeling. The
parameters assigned to the reference facility are typical of & large
hospital with nuclear medicine capabilities. Principal airborne re-
leases are from building ventilation exhausts with an effective height
of 10 meters (Le80, NRC81, Ro82a, TRI79a).
Waste Receiving Faci1itig s
Most radioactive emissions from radiophamaceutical users are re-
leased as liquids to sanitary sewer systems. When these liquid wastes
are treated, radionuclides may be emitted into the air. Radionuclide
releases at sewage treatment plants depend upon several factors. The
chemical and physical properties of wastewater and sludge influence the
potential amount of radioactivity released; for example, the potential
for release is greater at points in the treatment process where waste-
water pH is acidic. Other factors that affect radionuclide releases
include decay losses, evaporative losses, solids removal, degree of
system retention, and dilution.
Sludge drying and incineration are the greatest sources of radio-
nuclide emissions from sewage treatment plants because the high temper-
atures employed in these processes (typically 725°C) volatilize the
iodine and technetium. In addition, sludge incineration has the small-
est time delay (compared with other sludge treatment processes) and the
greatest potential for release of particulates due to mechanical agita-
tion of ash and combustion gases in the incinerator (TRI79a). The
selected reference treatment plant is one that dries and incinerates
sludge because these activities release the most radionuclides.
4-15
-------
Department of Defense Facilities
Facilities operated by the Department of Defense fall into three
categories: the Armed Forces Radiobiology Research Institute, U.S.
Army Reactors, and U.S. Navy Shipyards. The Radiobiology Research
Institute and U.S. Army reactor facilities research the effects of
radiation on health and on electronics used by the armed forces, U.S.
Navy releases of airborne radlonuclides are from reactors used for
propulsion in the submarine and surface fleets.
Armed ForcesRadiobiology Research Institute. The Armed Forces
Radiobiology Research Institute is located on the grounds of the Na-
tional Naval Medical Center in Bethesda, Maryland, approximately 20
kilometers northwest of Washington, B.C. In support of Department of
Defense radiation research, the Institute operates a TRIGA Mark-F
pool-type thermal research reactor and a linear accelerator (LINAC).
Most of this research involves the medical effects of nuclear radiation
and the effects of transient radiation on electronics and other equip-
ment.
The Mark-F reactor is licensed by the NRC to operate at steady-
state power levels up to 1.0 MW (thermal). This reactor is also capa-
ble of pulse operations. At peak power it can produce a 10-msec pulse
of about 2500 MW (thermal). The Institute's linac typically operates
in an energy range of 18 to 20 MeV, but it can operate at energies up
to 30 MeV. Emissions from the reactor and accelerator are released
from a common stack atop the Institute building (Sh81).
U.S. Army Facilities. The U.S. Army Test and Evaluation Command
operates two reactors: the Army Pulse Radiation Facility at Aberdeen
Proving Ground, Maryland, and the Fast Burst Reactor at White Sands
Missile Range, New Mexico. These reactors are similar ir. design and
are used to support Army and other Department of Defense (BOD) studies
in nuclear radiation effects.
Both Army reactors are bare, unreflected, and unmoderated and they
are fueled with enriched uranium. These reactors are capable of self-
emitting, superprompt, critical-pulse operations, as well as steady-
state operations at power levels up to 10 kW (EPA79a), The reactors
are used primarily by DOD and defense contractors to study the effects
of nuclear weapons on electronics and other DOD equipment.
The White Sands Laboratory is the principal source of radioactive
airborne emissions from Army reactors. Concrete structures around the
reactor at White Sands reflect, and thus lower, the energy of the
neutrons from the reactor. These low-energy neutrons produce airborne
radioactivity in the reactor building by activation of argon-40 in the
air. Concrete structures at Aberdeen are farther from the reactor;
hence, much less (essentially zero) argon-41 is produced at this
facility because there are very few low-energy neutrons (TRI?9a,
AMTE81, Aa82, De76).
4-16
-------
U.S. Navy Facilities. Almost all airborne emissions of radionu-
clides from U.S. Navy facilities result from activities at the nine
naval shipyards (listed in Table 4.2-3), where construction, overhaul,
refueling, and maintenance of the Navy's nuclear fleet of 133 subma-
rines and ships take place. Operations performed at these shipyards
include construction, startup testing, refueling, and maintenance of
the pressurized water reactors that power the nuclear fleet. Radioac-
tive wastes generated by these activities are processed and sealed at
the shipyards and then shipped to commercial waste disposal sites.
Table 4.2-3. U.S. Naval Shipyard facilities
Facility
Mare Island Naval Shipyard
Electric Boat Division, General Dynamics
Pearl Harbor Naval Shipyard
Portsmouth Naval Shipyard
Ingalls Shipbuilding Division
U.S. Naval Station and Naval Shipyard
Newport News Shipbuilding and Drydock
Norfolk Naval Shipyard
Puget Sound Naval Shipyard
Location
Vallejo, California
Groton, Connecticut
Pearl Harbor, Hawaii
Kittery, Maine
Pascagoula, Mississippi
Charleston, South Carolina
Newport News, Virginia
Portsmouth, Virginia
Bremerton, Washington
The primary sources of airborne radioactive emissions from naval
shipyards are the support facilities that process and package radioac-
tive waste materials for shipment to disposal sites. These facilities
handle solid, low-level radioactive wastes, such as contaminated rags,
paper, filters, ion exchange resins, and scrap materials.
During operation, shipboard nuclear reactors release small amounts
of radioactivity (carbon-14) into the atmosphere. Most of these re-
leases take place at sea, more than 12 miles from shore (Ri82 and
EPA77a).
Radiation Source Manufacturers
The term "radiation source" refers to radioactive material that is
enclosed in a sealed container or other matrix that prevents its dis-
persion. Radiation sources are used in a wide variety of industrial
and consumer products, including (1) radioisotope gauges, which measure
the thickness of industrial products; (2) static eliminators, which are
used to reduce static electricity in industrial machines; (3) nonde-
structive testing equipmen . (4) self-illuminating signs and watch
dials; and (5) smoke detect, rs (EPA79a).
Manufacturers of mala
radioactive materials recei*
,'ces process bulk quantities of
radionuclide production facilities
4-17
-------
reactors' Duri«8 Processing operations, the
are handled vlth remote manipulators and custom
«
enclos res, such as glove boxes (Cob83).
Other NEC Licenses (Cob83)
?8Kry includes three different groups of facilities:
laboratories, low-level waste disposal sites, and NRC-
licensed mineral and metal processing facilities.
Laboratories. The NRC-licensed laboratories include test, research
and development laboratories in industry, government agencies . anT '
Hr^iS K ««e«ch institutions. Approximately 700 laboratories are
Jorm rheyFpfeement "T*5 tO handlg ™^*°t*P** i» -n unsealed
,,n~'i f EPA assumes that an equal number of NEC licensees handle
unsealed radioisotopes.
Laboratory facilities at the various sites vary from a single
PTrM*b°rat0ry UP t0 3°° indi'-'idual ^oratories located
i J buildings at a maJ°r university. Both academic and
laboratories use byproduct materials in basic research «id
ape; "f?11 reS6arf laboratori" conduct basic epical atd
end ifJi^ dl°n"f lde r" earch wl«ted to a broad spectrum of diseases
and health problems. Government laboratories use radionuclides for
specify purposes, such as food and drug testing, water and air quality
measurements, and ocean and fisheries monitoring. quality
jfastejlsposal Sites. Of the six commercial low-level radioactive
waste disposal sites, only three are currently operational. These
Sevll ""I I"!? 3f 10Cated 3t Ba««-". SouthC-rolina; Beatty,
Nevada; and Richland, Washington.
ffati°"al sltes accePt low-level radioactive wastes in a
«r . f' I** n0t SPCCial nUClear Bater"ls, transuranlcs, or
spent .reactor fuels. Wastes accepted for disposal by shallow-land
burial must meet specific site acceptance criteria. The disposal sites
^ ° 3 large ^^^ a"a °f
usual! u a' ««es «e
usually buried in the transport containers in which they arrive at the
site to minimize atmospheric radioactive emissions.
ma^ M1"6"1 gnd Metal ProCessinfe_JacilitieS. Facilities that extract
metals from thorium- and uranium-bearing ^es are licensed by NRC or by
an Agreement State. Six facilities, located ln California, Florida
Illinois, New Mexico, and Pennsylvania (two facilities), are licensed
by NRC; and four facilities, located in Alabama, Colorado, Oregon? and
mT nr'Kr6 f "S ^ AgreGment States' ^ facilities licenced by
NRC columbium and tantalum followed by rare earth extraction processes
are the principal sources of radioactive materials that require control
under present provisions of 10 CFR AQ. "muruj.
Most Agreement-State- and NRC-licensed facilities are processing
uranium- and thorium-bearing ores for refractory metals, their oxides,
4-18
-------
rare earth metals. The industrial processes used in licensed facili-
s vary from wet chemical and solvent extraction to high-temperature
terlng and smelting.
,3 Coal-FiredUtility and Industrial Boilers
Coal-fired boilers are used to generate electricity, steam, and
water for public and industrial consumption. Electric utility
lers are generally larger than industrial boilers, and their design
ige is also much more limited. Coal contains traces of naturally
Burring radionuclides, which tend to be concentrated into the fly ash
•ing combustion. The radionuclide emissions are in the form of fine
ioluble particulate matter and consist largely of uranium, thorium,
I their decay products. The quantity of radionuclides released into
i atmosphere varies, depending on the radionuclide content of the
il, the furnace design, and the design and efficiency of the
rticulate control equipment.
In 1979 there were 1224 coal-fired utility boilers with a total
lerating capacity of 225 GW in service. By 1985 there will be
rroximately 1360 coal-fired units with a generating capacity of 307
in service (TRI79b), For an efficient assessment of the risks posed
radionuclide emissions from coal-fired utility boilers, a reference
;ility was developed. The reference facility is assumed to have a
ick height of 185 meters and a plume rise of 50 meters, typical of
•ge utility boilers.
Parameters for the reference industrial boiler were determined by
5 same general methods as those used for utility boilers, but they
re based on lower thermal capacities and coal consumption. The
ference industrial boiler facility is assumed to have a stack height
150 meters.
2.4 Underground Uranium Mines
In 1982, underground uranium mining accounted for about 46 percent
all U.S. uranium production (DOE83). All U.S. uranium is currently
led in the West, more than 65 percent of it in New Mexico, Wyoming,
i Texas.
A modified room-and-pillar method is generally used for under-
>und uranium mining. A large-diameter main entry shaft is drilled to
Level below the ore body, and a haulageway is established underneath
2 ore body. Vertical raises are then driven up from the haulageviay
the ore body. Development drifts are driven along the base of the
2 body connecting with the vertical raises. Mined ore is hauled
ang the development drifts to the vertical raises and is gravity-fed
the haulageway for transport to the main shaft for hoisting to the
rface.
4-19
-------
Ventilation shafts are installed at appropriate distances along
the ore body, and typical ventilation flow rates are on the order of
6000 m3/oin. The principal radioactive effluent in the mine ventila-
tion air is the large amount of gaseous radon-222 and is released
during mining operations. Additional radon-222 and particulate
(uranium and thorium) emissions result from surface operations at
underground mines (EPA79a, EPA82d).
Radon is a heavy gas that is only slightly soluble in water.
Because radon is a noble gas, it is chemically inactive. Radon cannot
be scrubbed or filtered from an exhaust gas stream and generally does
not adhere to particulate matter.
The major source of airborne radionuclides released from under-
ground uranium mines is the ventilation air exhausted from the mine.
Large underground mines usually have several vents, sometimes as many
as 15S spread out over a large area. Emissions from these vents vary
greatly and depend on factors such as the size of the working area,
mining practices, production rate, ore grade, and ventilation rates
(Ja80).
Several above-ground operations also may release radionuclides as
a result of waste rock storage, wind erosion, and ore dumping and load-
ing operations. These releases take the form of gas and particulate
emissions (EPA83a).
For determination of the effect of radionuclide releases from
underground uranium mines, a -reference facility was composited from
available site information for working mines (DOE83, EPA79a, Ja80).
The parameters for the reference mine used for estimating emissions to
the air are listed in Table 4.2-4. Emission data are presented in
Section 4.3.
Table 4.2-4. Parameters of reference underground uranium mine
Parameter Value
Ore grade 0.22% U308
Ore production 102,000 MT/ya
Days of operation 250 days/y
Number of vents 5
Vent height 3 meters
Radon emissions 11,000 Ci/y
£1
MT/y = metric tons/year
A.2.5 Phosphate Roc_k_Processing and Wet-Process Fertilizer Plants
Mining of phosphate rock is the fifth largest mining industry in
the United States. The southeastern United States is the center of
phosphate mining operations; over 90 percent of the U.S. phosphate is
4-20
-------
mined in Florida, Tennessee,.and North Carolina, Concentrations of
uranium and its decay products in phosphate rock can be as -i-j-.h as 100
tines greater than those in natural soil and rocks, and the handling
and processing of this rock can release elevated concentrations of
radionuclides in either gaseous or particulate form.
The as-mined phosphate rock is usually ground to a uniform size
and drisd. Both the drying and grinding operations can produce signif-
icant amounts of dust containing radionuclides. This step is frequent-
ly followed by calcining, which involves heating the rock to remove
unwanted organics. This process step can release additional partieu-
lates as well as gases containing vaporized radionuclides.
Most phosphate rock produced in this manner is used in the produc-
tion of agricultural fertilizers. This wet-process manufacture of
fertilizer produces phosphoric acid from the phosphate rock, which is
used to produce diammonium phosphate or triple superphosphate. Because
this process involves further crushing and extensive handling of the
phosphate rock, the radionuclides contained in the rock are released
into the atmosphere as particulates (TRW82a).
The parameters of a reference phosphate-rock drying and grinding
plant and a wet-process fertilizer plant are shown in Tables 4.2-5 and
4.2-6.
Table 4.2-5.
Parameters of reference phosphate-rock drying
and grinding plant
Parameter
(a)
Number of units
Phosphate rock processing rate (MT/y)
Operating factor (h/y)
Dryers
3
2.7E+6
6570
Grinders
4
1 . 21+6
6460
Uranium-238 content of phosphate
rock (pCi/g)W
Stack parameters
Height (meters)
Diameter (meters)
Exit gas velocity (m/see)
Exit gas temperature (°C)
40
20
2
10
60
Type of control system
Particulate emission rate [g/MT (lb/ton)] 130 (0.26)
Low-energy
scrubber
40
20
2
10
60
Medium-energy
scrubber
25 (0.05)
(a)
(b)
Dryer units process 145 MT/h; grinder units process 45 MT/h.
Uranium-238 is assumed to be in equilibrium with its daughter
products.
4-21
-------
Table 4.2-6. Reference wet-process phosphate fertilizer plant
(a)
DAP Diammonium phosphate.
GTSP Granular triple superphosphate,
(c)
Process
Parameter
Production rate (MT/y)
Operating factor (h/y)
Radionuclide content of product (pCi/g)^
Uranium-238, uranium-234, thorium-230
Radium-226
Lead-210, polonium-210
Stack parameters
Height (meters)
Diameter (meters)
Exit gas velocity (m/sec)
Exit gas temperature (°C)
Type of control system
Particulate emission rate (g/MT)
DAP
-------
uniform sizes for further processing. The rock, along with other
materials! is then fed into an electric furnace to produce phosphorus.
The phosphorus leaves the furnace as a gas, which is cooled and
collected in water condensers (EPA77b» EPA79b). The remaining gas is
vented or recycled to the calciner.
The stack parameters used in the assessments of health risks from
the elemental phosphorus plants are presented in Table 4.2-7.
Table 4.2-7. Calciner stack emission characteristics
Plant
Stack height
(meters)
Heat emission
(calories/sec)
FMC
Pocatello, Idaho
Monsanto
Soda Springs, Idaho
Monsanto
Columbia t Tennessee
Stauffer
Silver Bow, Montana
Stauffer
Mt. Pleasant, Tennessee
Occidental
Columbia, Tennessee
30
31
35
27
35
31
8.8E+5
2.0E+6
1 . OE+6
3.0E+4
6.01+5
1.2E+6
4.2.7 Mineral Extraction Industry Facilities
All industrial operations that are involved in the extraction and
processing of mineral ores release some quantity of radionuclides into
the atmosphere. The aluminum, copper, zinc, and lead industries have
the greatest potential for radionuclide releases because of the high
volume of material processed and because they all utilize high tempera-
ture smelting. These emissions are largely in the form of fine particles
of uranium, lead, polonium, and (at times) thorium. Most of these
radioactive metallic elements occur In the oxide and sulfate form.
Aluminum Industry Facilities^
The production of aluminum differs somewhat from other mineral-ex-
traction industries btcause contaminants in the ore are removed during
the milling of the ore rather than during smelting. The aluminum ore
(usually bauxite) is converted into aluminum oxide at the mines, and
4-23
-------
subsequently shipped to the smelters for final processing. Aluminum
metal is produced in electric reduction cells. Particulate emissions
from the process reflect the composition of the feed materials, which
includes alumina, carbon, aluminum fluoride, and cryolite (EPA79c,
EPA82a).
The parameters of a reference aluminum reduction plant are listed
in Table 4.2-8. These values are used to estimate the radionuclide
emissions to air.
Table 4.2-8. Parameters of reference aluminum reduction plant (TRI81)
Parameter
Value
Capacity
Capacity factor
Type of equipment
Main stack parameters
Height
Diameter
Exit gas velocity
Exit gas temperature
Roof monitor
Height
Width
Exit gas velocity
Exit gas temperature
Anode bake plant
Height
Diameter
Exit gas velocity
Exit gas temperature
136,000 MT/y aluminum
0.94
Center-worked prebake cells
-' m (4 stacks)
m
30 m/sec
160°C
10 m
1,2 m
0.01 m/sec
37°C
30 m
1.8 m
4.5 m/sec
96°C
Copper Industry Facilities
Copper ore is processed to yield a concentrate containing copper,
sulfur, iron, and other remaining impurities, which are removed by
smelting. The three major steps in copper production are roasting,
smelting, and converting. Each of these steps release sulfur oxides
and particulates that may contain radionuclides.
The purpose of roasting the concentrated copper ore is to remove
some of its sulfur content. Some particulate material is released
during this process. All domestic copper smelters produce an interme-
diate grade of copper by smelting the copper ore at high temperatures
with other materials to form two liquids that separate into a mixture
4-24
-------
of copper and Iron impurities and a layer containing a significant
fraction of the other materials in the ore. The converter process
removes the iron impurities from the copper and iron mixture at high
temperatures before its final purification in a refining furnace.
The parameters of a reference copper smelter that were used to
estimate the radioactive emissions to the air are shown in Table 4.2-9.
The copper output capacity of the reference plant is 56,000 MT/y» and
a capacity factor of 0.75 was chosen for this plant. Main stack heights
for facilities without roasters range fron 61 to 228 meters. The
control equipment applied to the reference facility was chosen to
represent typical equipment on actual copper smelters (EPA82a),
Table 4.2-9. Parameters of reference copper smelter (TRI81)
Parameter
Value
Capacity
Capacity factor
Type of equipment used
Stack parameters
Main stack
Height
Diameter
Exhaust gas velocity
Exhaust gas temperature
Acid plant
Height
Diameter
Exhaust gas velocity
Exhaust gas temperature
Particulate emission rate
Main stack
Acid plant
56,000 MT/y
0.75
Reverberatory furnace
183 m
2.6 m
28 ra/see
135°C
30.4 m
1.8 m
16.5 m/sec
79°C
247 kg/h
11 kg/h
Zinc Industry Facilities
A zinc smelter produces 99.99+ percent zinc from concentrate
containing approximately 62 percent zinc. The zinc concentrates are
first roasted at approximately 600°C to convert sulfur to sulfur dioxide
and to produce an Impure zinc oride or calcine. The calcine is then
transferred to tanks, leached trith dilute sulfuric acid, and treated to
remove such Impurities as lead, gold, and silver.
The leaching step varies somewhat from plant to plant, but the
basic process of selective precipitation of the impurities from the
leach solution remains the same. This solution is purified and piped
4-25
-------
„,
on aluninun
tanks at
Table 4.2-10. Paraneters of reference zinc plant (TRI81)
Process
Capacity
Capacity factor
Radionuclide concentration of input ore
Uranlum-238
Thorium-232
Stack parameters
Number
Height
Diameter
Exhaust gas velocity
Exhaust gas temperature
Electrolytic reduction
88,000 MT/y zinc
0.8
0.18 pCi/g
0.08 PCi/g
1
100 m
2 as
20 tn/sec
150°C
t0tal Produ«ion capacity of
un
Other
P anc parameters are based on actual measurements (EPA82b).
Lead Industry Facilities
'
4-26
-------
Table 4,2-11. Parameters of reference lead smelter (T1I81)
Parameter Value
Capacity 220,000 MT/y lead
Capacity factor 0.92
Radionuclide concentration of input ore
Uranium-238 0.9 pCi/g
Thorium-232 0.5 pCi/g
Stack parameters:
Number 1
Main stack
Height 30 m
Diameter 1 m
Exit gas velocity 9 m/sec
Exit gas temperature 90°C
Acid plant stack
Height 30 m
Diameter 1.8 m
Exit gas velocity 1.7 m/sec
Exit gas temperature 93°C
The reference lead smelter has a capacity of 220,000 MT lead per
year, typical of existing plants. The plant operates at a load factor
of 0.92, which was the industrywide average for 1979 (DOC80). Other
plant parameters are based on a composite of data taken at an operating
facility.
4.2.8 Uranium Fuel Cycle Facilities. Uranium Mill Tailings, High-Level
Waste Management
Uranium Fuel Cycle Facilities
Uranium fuel cycle facilities are involved in chemical conversion
of uranium, isotopic enrichment, fabrication of fuel, and generation of
electricity.
Uranium Conversion. Two industrial processes are used for uranium
hexafluoride production, the dry hydrofluoride (hydrofluor) method, and
the solvent extraction method (EPA73a). The hydrofluor process con-
sists of reduction, hydrofluorination, and fluorinatlon of the ore con-
centrates to produce crude uranium hexafluoride, followed by fractional
distillation to obtain a pure product. The dry hydrofluor process sep-
arates Impurities either as volatile compounds or as solid constituents
of ash. The solvent extraction process employs a wet chemical solvent
extraction step at the start of the process to prepare high-purity
uranium for the subsequent reduction, hydrofluorination, and fluorina-
tion steps. The wet solvent extraction method separates Impurities by
extracting the uranium into organic solvent and leaving the impurities
dissolved in an aqueous solution.
4-27
-------
Table 4.2-12,
Parameters of reference uranium conversion facility
Type
Ore grade
Fluorination-fractionation (dry hydrofluor) UF6 plant
Low-impurity plant feed containing 2800 pCi of
20°PCI of
Annual
capacity 10,000 MT of uranium
Emission control
Stack
Height
Plume rise
Primary treatment, secondary bag filters on dust con
trol streams, and secondary or tertiary scrubbers on
process off-gas streams *<-ruDDers on
10 m
0.0
developed for assessing
plants are presented in
Parameters for reference uranium fuel
fabrication facility
Table 4.2-13.
Table 4.2-13.
Type of facility
Ammonium diuranate (ADU)
Direct conversion (DC)
Capacity
Fixed stack height, no plume rise 10 m
4-28
UF6 feed to plant hydrolyzed in
water, uranium precipitated in
ammonia to form ADU. ADU calcined
to form U02
UF6 feed to plant reacted with water
vapor and hydrogen to form U02
1500 MT/y
-------
Nuclear Power Plants
Nuclear power plants operate on the same general principles as
fossil-fuel-flred generating stations. The only significant difference
Is that a nuclear reactor, rather than a fossil-fuel-fired boiler, sup-
plies the heat to generate steam. The fission process in the reactor
produces radioactive gases that may enter the coolant. These contami-
nants are periodically removed from the coolant and subsequently re-
leased in the form of gaseous isotopes such as argon, xenon, and krypton,
which are largely Inert.
Two basic types of light-water-cooled reactors are currently in
use in the United States: boiling-water reactors (BWR) and pressur-
ized-water reactors (PWR). Reference facilities for the two types of
commercial reactors, boiling-water and pressurized-water reactors, were
developed for the impact analysis of the nuclear power industry (param-
eters are listed in Table 4.2-14). The reference facilities use a
recirculating u-tube steam generator, and their characteristics were
developed by the NRC in its environmental statement on light-water-
cooled reactors (NRC76, EPA73b).
Table 4,2-14. Parameters for reference light-water reactors
Parameter Value
Type Boiling-water reactor and pressurized-water
reactor
Capacity 1000 MW(e)
Fuel Uranium only
Fixtd stack height, 20 m
no plume rise
Uranium Mill Tailings
As with any ore-processing operation, uranium milling produces
large quantities of waste rock. Uranium mill wastes, or tailings, are
usually stored in an impoundment located on the mill site. Tailings
are usually discharged to the impoundment area as a liquid slurry.
Tailings impoundment areas consist of a pond and a dry beach, the sizes
of which are based on water recycle rates and evaporation rates.
Radionuclide emissions, which are primarily from the dry beach areas,
result from wind erosion and diffusion of radioactive gases out of the
tailings. The largest radlonuclide emission is radon-222 gas.
For purposes of estimating the emissions and health impacts from
uranium mill tailings, a reference model was developed and values were
assigned to the important parameters (Maa78). These are presented in
Table 4.2-15. Because the activity of the mill itself is important for
4-29
-------
an assessment of the impact of the tailings impoundment, parameters for
a model mill are also included in this table (NRC?9a, EPA83a» EPA82d).
Table 4.2-15,
Parameters for reference uranium mill and
tailings impoundment
Parameter
Value
Type of process
Ore process rate
Operating days per year
Mill lifetime
Ore grade
Uranium recovery
Ore activity
Ore storage area
Ore storage time
Effective suack height
Area of tailings impoundment
Dry beach
Pond and wet beach
Average depth of tailings
Acid-leach solvent extraction
2000 metric tons per day
300 days
20 years
0.2% U308
95%
560 pCi/g, uranium-238 and daughter
products in secular equilibrium
1 hectare
10 days
15 meters
60 hectares
15 hectares
45 hectares
12 meters
High-Level Waste Management
In normal operation, uranium fuel-cycle facilities, specifically
nuclear reactors and other NRC-licensed facilities, generate high-level
radioactive waste, primarily in the form of spent reactor fuels. The
option selected for disposal of the spent fuels determines the kind of
facilities required for their management. In the interim period, the
spent fuels are stored in pools of water, often located at the power-
plant or DOE facility.
The reference plant for nuclear generating stations includes re-
leases from spent fuel storage in the form of gaseous krypton and
smaller amounts of tritium. For the assessment of uranium fuel-cycling
releases, an offsite fuel storage facility was selected as the refer-
ence facility. The site parameters (EPA82c) are listed in Table 4.2-16.
Table 4.2-16. Parameters for reference fuel storage facility (EPA82c)
Parameter
Value
Capacity
Facility life remaining
Percentage of release respirable
Source type
Discharge height
Distance to site boundary
5,000 tons
30 years
100%
Point source
100 meters
500 meters
4-30
-------
4.2.9 Lpw-TVigrgy Accelerators
Particle accelerators not operated by DOE are generally low-energy
medical and research facilities. The equipment, operational energies,
particles accelerated, and target materials used at these facilities
vary greatly. Possible sources of radionuclide emissions include loss
of target integrity, handling of irradiated targets, and activation of
air and dust by the particle beam. The radionuclide emissions are in
the form of relatively small quantities of isotopes of oxygen,
nitrogen, argon, carbon, and tritium.
Three reference accelerator facilities were developed to assess
the health impacts from low-energy accelerators. The parameters as-
signed to the reference facilities are listed in Table 4.2-17.
Table 4.2-17, Parameters of reference accelerator facilities
Parameter
Type of accelerator
Emission control
Roof-type stack height
Value
6 MeV Van de Graaff with tritium
target, operated 3000 h/y
18 MeV electron LINAC, operated
2000 h/y
100 MeV research cyclotron,
operated 1000 h/y
None
16.8 meters
4.3 Radjonuclide Releases
The emission data used in the health impact assessments are summa-
rized in the following subsections. Insofar as possible, measured
radionuclide emission data have been used. In the absence of measured
data, however, estimates are based on calculated or extrapolated val-
ues. The emission data for DOE facilities were obtained from DOZ's
Effluent Information System for the calendar year 1981 (DOEaSl); the
data for NRC-licensed facilities were obtained from NRC annual effluent
reports; and the data for the other categories, such as coal-fired
utility and industrial boilers, uranium and nonuranium mines, and the
various extraction industries, were obtained from various reports
prepared for the EPA.
More detailed source data for the individual source categories are
available in Chapters 2 through 1 in Volume II of this document.
4-31
-------
4.3.1 Department of Energy Facilities
The individual TOE facilities were briefly described in the pre-
ceding section. Only the largest DOE emission sources are presented in
Table 4,3-1 because so many sources are involved. Volume II, Chapter
2» of this document provides detailed emission data for all of the DOS
facilities.
Radio-'uelide emissions from DOE facilities result from three types
of operations: (1) nuclear reactor operations, (2) nuclear fuel and
weapons materials processing, and (3) accelerator operations. The
radionuclide releases resulting ffom the operation of nuclear reactors
are in the gaseous state. The principal radionuclides released are
noble gases [argon (AR-41), krypton (Kr-85 and 88), and xenon (Xe-133)]
and isotopes of hydrogen (H-2 and H-3). These releases occur during
routine purging of radioactive decay products from reactor cooling
systems and refueling operations.
Radionuclide releases from nuclear materials processing are pri-
marily particulates, which are released during solid materials handling;
however, very small quantities of gaseous radionuclides are released
during the processing of spent nuclear reactor fuel elements. The
primary particulate emissions from DOE production facilities are ura-
nium (U-234 and li-238). Gaseous releases include tritium and deuterium
(H-3 and H-2) and the noble gases listed previously—xenon, argon, and
krypton.
Accelerator facilities, the third category of DOE emission sources,
release radionuclides in the gaseous state. These emissions result
from high-energy particles reacting with air and from the radioactiva-
tion of air by secondary particles generated in the accelerator. The
primary radionuclides emissions from accelerators are oxygen (0-15),
nitrogen (N-'3), argon (Ar-41), and carbon (C-ll).
4.3.2 NRC-Licensed Facilitiesand Kon-DOE Federal Facilities
As an aid co consistent analysis of NRC-licensed facilities, a
reference source was developed for the individual types of facilities
included in this category. Radionuclide emission data for the reference
NRC-licensed facilities are summarized in Table 4.3-2. Annual radionu-
clide emission rates are referred to as "source terms" in the computer-
ized models used to estimate health impact.
4-32
-------
Table 4.3-1. Summary of radlonucllde emissions from DOE facilities
Facility
i
rgonne National
Laboratory
rookhaven National
Laboratory
eed Materials Production
Center
'ertni National Accelerator
I aboratory
lanford Reservation
[daho National Engineering
Laboratory
..awrence Livermore National
Laboratory
Los Alamos National
Laboratory
Oak Ridge Beservation
Savannah River Plant
Radionuelide
Ar-41
Kr-85
H-3
0-15
Ar-41
U-238
U-234
C-ll
H-3
Ar-41
Cs-138
H-3
Ar-i 1
Kr-85
H-3
N-13
0-15
H-3
C-ll
N-13
0-15
Ar-41
H-3
H-3
Kr-85
Xe-133
H-3
Ar-41
Kr-85
Xr-88
Xe-133
Amount
released
(Cl/y)
0.4
6.7
660
36,000
170
0.11
0.11
1500
18
65,000
11,000
400
2,500
59,000
2,600
170
170
1,100
130,000
25,000
200,000
1,400
6,100
11,000
6,600
32,000
350,000
62,000
840,000
1,500
3,900
4-33
-------
Table 4.3-2. Summary of radionuclide emissions from NRC-lIcensed
facilities and other Federal facilities
Facility
Radionuclide
Amount
released
(Ci/y)
Research Reactor
Reference Facility
Accelerator Reference
Facility
Cyclotron
Radiopharmaceutical Industry
Reference Supply
Facility
Reference User
Facility
Reference Sewage
Treatment Plcnt
Armed Forces Radio-
biology Research
Institute
U.S. Army Pulse Reactors
U.S. Navy
Reference Nuclear
Shipyard
Manufacturers of
Radiation Sources
Reference Facility
Other NRC Licensees
Laboratories
Waste Disposal Sites
Mineral and Metal
Processing Facilities
Ar-41
H-3
N-13
0-15
C-ll
1-125
1-131
Xe-133
Te-99m
1-125
1-125
Xe-133
1-131
Te-99m
Ar-41
N-13
0-15
Ar-41
Ar-41
C-14
Kr-87
Xe-135
H-3
Kr-85
C-14
H-3
H-3
Rn-222
8,560
22
0.04
1.0
2.0E-3
0.02
0.076
23
4.5E-3
9.5E-3
0.05
6.4
5.0E-4
8.0E-4
1.3
3.5E-2
3.5E-2
13.3
0.41
0.10
0,05
0.25
1,060
61.8
4.3
29
6,000
Not available
4-34
-------
4.3.3 Coal-Fired Utility and Industrial Boilers
Both industrial and utility coal-fired boilers emit radionuclldes
in fly ash, A primary factor influencing the radlonuclide content in
the fly ash generated during combustion is the type of coal, i.e., its
mineral content and the concentrations of uranium, thorium, and their
decay products. Other factors affecting radlonuclide emissions are the
furnace design and capacity, the capacity factor, the heat rate, propor-
tion of fly ash to bottom ash, enrichment factors, and emission control
efficiency.
Measurements have shown that trace elements, such as uranium,
lead, and polonium, are distributed unequally between bottom ash and
fly ash (Be78, Wa82). Although the concentration mechanism is not
fully understood, the preferential concentration of certain volatile
elements on particle surfaces results in depletion of these elements in
the bottom ash and their enrichment in the fly ash (Sm80). The highest
concentration of the trace elements in fly ash is found in particulates
in the 0.5- to 10.0-micrometer diameter range, the size range that can
be inhaled and deposited in the lung. Particulate control devices are
less efficient in removing these fine particles than larger particles.
Based on measured data, typical enrichment factors are 2 for uranium,
1.5 for radium, 5 for lead and polonium, and 1 for all other radionuclides
(EPA81). The radlonuclide emissions for the reference utility and
industrial boilers are listed in Table 4.3-3. These sources are
discussed further In Volume II, Chapter 4, of this document.
Table 4.3-3. Summary of radlonuclide emissions from reference
coal-fired boilers
Facility
Utility boiler
Industrial boiler
Radionuclide
U-238
Th-230
Rn-222
Pb-210
Po-210
U-238
Th-230
Rn-222
Pb-210
Po-210
Emissions (Ci/y)
0.1
0.05
0.96
0.25
0.25
0.01
0.005
0.25
0.025
0.025
4.3.4 Underground Uranium Mines
Radon-222 is the predominant radlonuclide released from
underground uranium mines. Emissions of uranium and thorium also
4-35
-------
have been detected at uranium nines* but at levels so low as to be
insignificant compared with those of radon. The radon-222» uranium-238,
and thorium-232 emissions from the reference underground uranium mine
are shown in Table 4.3-4.
Table 4.3-4. Radionuclide emissions from the reference
underground uranium mine (EPA83a, Ja80)
Emissions (Ci/y)
Source Radon-222 Uranium-238 Thoriuu-232
Mine vents 11,000
Ore, subore, and waste rock piles 500 0,02 3E-4
4.3.5 Phosphate RockProcessing and Wet-Process Fertilizer Plants
The radlonuclide stack emissions from the reference fertilizer
plant and reference rock drying and grinding plant are listed in Table
4.3-5. More extensive discussions of these facilities and their
emissions appear in Volume II, Chapter 6, of this document.
Table 4.3-5. Summary of radionuclide emissions from reference
fertilizer plant and reference phosphate rock processing plant
Facility Radionuclide Emissions (Ci/y)
Wet-process fertilizer plant
(DAP and GTSP
Phosphate rock
combined)
drying and grinding
U-238
Th-230
Pb-210
Po-210
U-238
Th-230
Pb-210
Po-210
0.007
0.007
0,003
0.003
0,015
0.015
0.015
0,015
4.3.6 Elemental Phosphorus Plants
Polonium-210 and lead-210 are the radionuclides emitted from
elemental phosphorus plants in the most significant quantities. More
than 95 percent of the polonium-210 and lead-210 are released from the
calciner stacks. The high temperature of the calciners volatilizes the
polonium-210 and lead-210 from the phosphate rock and results in the
release of much greater quantities of these radionuclides than of the
uranium, thorium, radium radionuclides. The EPA conducted extensive
4-36
-------
testing at elemental phosphorus plants, and these data were used to
develop the annual radionuclide emission estimates shown in Table
4.3-6.
Table 4.3-6. Estimated annual radionuclide emissions from
elemental phosphorus plants
Plant
FMC(a)
Pocatello, Idaho
Monsanto
Soda Springs, Idaho
(a)
Monsanto
Columbia, Tennessee
(a)
Stauffer^ ;
Silver Bow, Montana
Stauffer(b)
Mt. Pleasant, Tennessee
Occidental^
Columbia, Tennessee
Uranium-238
41-3
6E-3
2E-3
6E-4
2E-4 ,
2E-4
Emissions (Ci/y)
Lead-210
0.1
5.6
0.4
0.1
0.05
0.05
Polonium-210
9
21
0.6
0.7
0.1
0.1
(a)
(b)
Based on measured emission rates.
Based on estimated emission rates.
4,3.7 Mineral Extraction Industry
Most of the radionuclide emissions from mineral-extraction facili-
ties are in the form of fine particulates. Lead, copper, zinc, and
aluminum facilities were chosen as the reference facilities because
each uses high-temperature smelters with the potential for significant
releases of particulates.
Radionuclide concentrations in particulates emitted from a smelter
are similar to or greater than the concentrations in the materials
processed. The radionuclide concentrations greater than those in the
original ore are due to the enrichment that takes place when nuclides
4-37
-------
volatilize during the high-temperature phase of production. Calcula-
tions of the releases for the reference smelters are based on the
assumption that the radionuclide content in the partlculates released
is the same as that in the input ore and the application of appropriate
enrichment reactors frt volatile radionuclides. Multiplying the con-
centrations of radionuclides in the ore by the total annual particulate
release yields the total annual radionuclide release. The radionuclide
emissions for the reference facilities in this category are listed in
Table 4,3-7. More detailed discussions of the emissions from each
facility can be found in Volume II, Chapter 7.
Table 4.3-7. Summary of emissions from reference
mineral-extraction facilities
Facility Radionuclide
Lead smelter U-238
Pb-210
Po-210
Copper smelter U-238
Th-230
Pb-210
Po-210
Zinc smelter U-238
Pb-210
Po-210
Aluminum reduction plant U-238
Th-230
Pb-210
Po-210
Emissions (Ci/y)
8.6E-3
2.61-2
2. 1E-2
0.04
2. 1E-3
6.5E-2
0,03
5.6E-4
2.5E-2
1 . 5E-3
1.5E-4
2.8E-4
5.2E-4
4.7E-4
4,3.8 Uranium Fuel Cycle Facilities, Uranium Tailings,
Waste Management
High-Level
Uranium Conversion Facilities
Conversion facilities handle no irradiated material; therefore,
all radionuclides present also occur in nature. These radionuclides
are radium, thorium, uranium, and their respective decay products.
Uranium is the major source of radioactivity in the gaseous effluents.
Possible chemical species of uranium effluents include U.Og, UO , UF,,
UF-, and (NH,)JJ70 . In the wet solvent extraction method, uranium is
present as uranyl nitrate, which may also appear in gaseous effluents.
Thus, the uranium may be released as both soluble and insoluble aero-
sols. The emissions from the reference facility are listed in Table
4.3-8
-------
Table 4,3-8. Atmospheric emissions of radionuclldes from the
reference uranium conversion facility
Radionuelide Emissions (Ci/y)
U-238 0.083
Th-234 0.082
Rn-222 9.2
Uranium Fuel Fabrication Facilities
Particulate emissions account for all radionuclides released from
fuel fabrication facilities. The radionuelide emissions from the
reference fuel fabrication facility are shown in Table 4.3-9.
Table 4.3-9. Radionuelide emissions from the
reference fuel fabrication facility (EPA78b» NEC76)
Radionuelide Emissions (Ci/y)
U-234
U-235
U-236
U-238
Th-231
Th-234
Pa-234
0.013
4.6E-4
7.0E-4
1.7E-3
4.6E-4
1 . 7E-3
I . 7E-3
Light-Water Reactors
Radionuelide emissions from boiling-water reactors (BWR) and pres-
surized-water reactors (PWR) usually result from contaminants released
from the fissionable fuels into the cooling water. These contaminants
are removed from the cooling water and periodically released to the
atmosphere. A summary of the airborne releases from the reference
reactors is presented in Table 4.3-10.
4-39
-------
Table 4.3-10. Atmospheric emissions of radionuclides from the
reference BMR and PWR facilities (EPA77c)
Emissions (Ci/y)
Radionuelide BWR PWE
1-131
Kr-85
Xe-133
H-3
0.2
300
12,000
60
0.2
150
10,000
400
Uraru w Mill Tailings
The amount of airborne emissions from tailings disposal areas
depends upon the aize of dry tailings beach areas that are subject to
wind erosion and radon-222 diffusion. When tailings impoundment ar«as
are almost completely covered by water, radionuclide emissions will be
low (NRC79b). The airborne radioactive emissions for the reference
uranium mill tailings area due to wind erosion and gaseous diffusion
are listed in Table 4.3-11.
Table 4.3-11. Radionuclide emissions from the reference
uranium mill (EPA83b)
Radionuclide Emissions (Ci/y)
U-238
U-234
Th-230
Ra-226
Pb-210
Po-210
Rn-222
8.9E-3
8.9E-3
1.2E-1
1.2E-1
1.2E-1
1.2E-1
4.4E+3
(a)
During the operational phase of the mill.
High-Level Waste Management
Airborne emissions from the reference fuel storage facility result
from venting the cask gases and from activity in tlia cask unloading and
fuel storage pools. The radionuclide emissions for the reference fuel
storage facility are shown in Table 4.3-12.
4-40
-------
Table 4.3-12. Radionuclide emissions from reference storage facility
(EPA82c)
Radlonuclide Emissions (Ci/y)
H-3 2.4
Kr-85 890
4.3.9 Log-Energy Accelerators
Emissions of radioactive materials at accelerator facilities are
produced by two principal mechanisms; 1) the activation of air by
accelerated particles or secondary radiation, which results in radioac-
tive carbon, nitrogen, oxygen, or argon; and 2) the loss of radioactive
material (usually tritium) from a target into the air. Airborne radlo-
nuclide releases from the three reference accelerator facilities are
presented in Table 4,3-13,
Table 4.3-13. Radionuclide releases from reference
low-energy accelerators (EPA79c)
(Cl/y)
Radionuclide
C-ll
N-13
0-15
H-3
C-14
Ar-41
100 MeV cyclotron
2.0E-3
0.04
1.0
—
—
—
Type of accelerator
18 MeV
electron linac
minim- immm
-._
1 . OE-9
l.OE-4
6 MeV Van
de Graaff
«P«
—
—
1.0
—
~
4.4 Uncertainties
Quantifying fadionuclide emissions from the source categories
addressed in this report necessitated the review and summarization of
significant amounts of data collected by numerous agencies. The
emission data presented for facilities in this chapter were gathered
from a variety of information sources. These information sources
include reports prepared by individual facilities, tests conducted for
standards development not associated with radionucllde emissions,
engineering determinations of expected releases from physical and
chemical processes, and tests conducted specifically for determination
of radionuclide emissions.
4-41
-------
It was neither practical nor feasible to evaluate every single
facility for individual contributions to health risk; therefore, for
certain source categories, a single reference facility that best char-
acterized the industry was either selected or developed from existing
source emission data. The use of a single facility to represent a
large number of sources simplifies the assessment of health risks, but
the potential for errors increases because of the necessary assumption
that all relevant factors used In the analysis of the reference
facility are in fact representative of the industry.
In those cases where measured emission data were not available for
a facility or process; an assessment of the expected releases was based
on an engineering analysis of the process generating the release.
These mass and chemical balances require assumptions regarding process
parameters. As with the selection of reference facilities, airborne
emissions determined through the use of mass balances may include an
expected level of uncertainty due to the required assumptions. Simi-
larly, some annual radionuclide emission rates are based partially on
fugitive particulate emission factors. The fundamental nature of fugi-
tive emissions makes them extremely difficult to quantify precisely,
and emission factors represent tHe mean estimate of emissions, which
can vary substantial1^ due to wind, humidity, material handling prac-
tices, and other factors.
Even annual radionuclide emission rates based on physical and
radiological measurements are not exact. Any physical measurement is
subject to uncertainties imposed by the accuracy and precision of the
sampling methodology and analytical procedures used. Of these two
factors, imprecision in sampling (and sample handling/preparation)
generally presents the greater uncertainty. Determination of the
radioactivity of a sample is fairly straightforward; the significant
uncertainties result from the random and systematic errors of an
instrument or method. Analytical problems can occur when- several
different radionuclides are collected in one sample and must be
determined individually. Considering the uncertainty in both sampling
and analysis, emission measurements for radionuclides are generally
accepted as being accurate within approximately ±20 percent at best.
In general, the range of uncertainty in annual radionuclide emission
rates based on physical and radiological measurements are expected to
be comparable.
Other factors that can increase the overall uncertainty of the
emission data are as follows:
(1) The use of enrichment and partitioning factors that were
determined from a single source for a particular radionu-
clide.
(2) The use of data not specifically collected to quantify radi-
onuclide emissions.
(3) The adequacy of quality control and quality assurance proce-
dures followed during the collection and analysis of samples,
4-42
-------
REFERENCES
Aa82
AMTE81
AnSla
An81b
BAPL82
Be78
Coa82
Cob 8 3
De76
Aaserude R. A., Dubyoski H. G.» Harrell D. R. and Kazi A. H.,
Amy Pulse Radiation Division Reactor, Annual Operating
Report, Material Testing Directorate, Aberdeen Proving
Ground, 1982. B
Array Material Test and Evaluation Directorate, White Sands
Missile Range Fast Burst Reactor, Annual Operating Report,
Applied Sciences Division, White Sands Missile Range, N. M.,
1981.
Andrews V. E., Emissions of Naturally Occurring Radioactivity
From Stauffer Elemental Phosphorus Plant, ORP/LV-81-4, EPA
Office of Radiation Programs, Las Vegas, Nevada, August 1981.
Andrews ¥. E., Emissions of Naturally Occurring Radioactivity
From Monsanto Elemental Phosphorus Plant, ORP/LV-81-5, EPA,
Office of Radiation Programs, Las Vegas, Nevada, August 1981.
Bettis Atomic Power Laboratory, Effluent and Environmental
Monitoring Report for Calendar Year 1981, WAPD-RC/E(ESE)-576
West Misslin, Pennsylvania, 1982.
Beck H. L., et al, Perturbations of the National Radiation
Environment Due to the Utilization of Coal as an Energy
Source, Paper presented at the DOE/UT Symposium on the Natu-
ral Radiation Environment III, Houston, Texas, April 23-28,
!.!# 78 •
Cole L. W., Environmental Survey of the Manufacturing Facili-
ty, Medi-Physics, Inc., Arlington Heights, Illinois, Oak
Ridge Associated Universities, Oak Ridge, Tennessee, January
1982.
Corbet C. D., et al. Background Information on Sources of
Low-Level Radionuclide Emissions to Air, PSL-467G (Draft
Report), Pacific Northwest Laboratory, Richland, Washington,
March 1983.
De La Paz A. and Dressel R. W,, White Sands Missile Range
Fast Burst Reactor Facility, Annual Operating Report, Army
Material Test and Evaluation Directorate, White Sands Missile
Range, N. M., 1976.
4-43
-------
DOC80
DOESla
DOESlb
DOESlc
DOE82a
DOE82b
DOES 3
DOI76
Dr80
EPA73a
EPA73b
EPA75
Department of Commerce, U.S. Industrial Outlook for 200
Industries With Projections for 1984, Washington, D.C., 198C.
Department of Energy, Effluent Information System, Washing-
ton, B.C., 1981. s
Department of Energy, Environmental Monitoring in the Vicini-
ty of the Savannah River Plant, Annual Report for 1981
EPA77a
Department of Energy, Environmental Monitoring Report for
?o!? °ed Llnear Accelerator Center, Annual Report for CY
1981, Stanford University, Stanford, California, 1981.
Department of Energy, Idaho Operations Office, 1981 Environ-
mental Monitoring Program Report for Idaho National Engineer
ing Laboratory Site IDO-12082 (81), 1982.
Department of Energy, Environmental Monitoring Summary for
Ames Laboratory, Calendar Year 1981, Milo D. Voss, Ames
Laboratory, Ames, Iowa, 1982,
Department of Energy, Statistical Data of the Uranium Indus-
try, GJO-100(83), Grand Junction, Colorado, January J983.
Department of Interior, Preprint from the 1976 Bureau of
Mines Minerals Yearbook: Zinc, Washington, D.C., 1976.
Droppo, *J. G., et al. , An Environmental Study of Active and
Inactive Uranium Mines and Their Effluents, Part I, Task 3
Pacific Northwest Laboratory, PNL-3069, Part I, August 1980.
Environmental Protection Agency, Environmental Analysis of
the Uranium Fuel Cycle - Part I - Fuel Supply, EPA-520/9-73-
003c, Office of Radiation Programs, Washington, D.C. 1973.
Environmental Protection Agency, Environmental Analysis of
the Uranium Fuel Cycle - Part II, Nuclear Power Reactors,
f r~52?a?;73~°03C' °"iCe °f ******* Programs, Washington,
V , L. 1973 .
Environmental Protection Agency, Development for Interim
Final Effluent Limitations Guidelines and Proposed New Source
Performance Standards for the Lead Segment of the Nonferrous
Metals Manufacturing Point Source Category, EPA-440/1-75/032-
9, Washington, B.C., February 1975.
Environmental Protection Agency, Radiological Survey of Puget
Sound Naval Shipyard, Bremerton, Washington, and Environs.
EPA-520/5-77-001, Office of Radiation Programs, Washington,
D . C . , 19/7.
4-44
-------
EPA77b
EPA77c
EPA78a
ErA78b
EPA79a
EPA79b
EPA79c
EPA80
EPA81
EPA82a-
EPA82b
EPA82c
Environmental Protection Agency, Radiological Surveys of
Idaho Phosphate Ore Processing—The Thermal Plant, ORP/LV-77-
J, Office of Radiation Programs, Las Vegas, Nevada, 1977.
Environmental Profection Agency, Summary of Radioactivity
Released in Effluents from Nuclear Power Plants from 1973
through 1976, EPA-520/3-77-012, Washington, D.C. 1977.
Environmental Protection Agency, Office of Radiation Pro-
grans, Radiation Dose Estimates due to Air Particulate Emis-
sions from Selected Phosphate Industry Operations, ORP/EERF-
78-1, Montgomery, Alabama, 1978.
Environmental Protection Agency, A Radiological Emissions
Study at a Fuel Fabrication Facility, EPA-520/5-77-004,
Office of Radiation Programs, Washington, D.C. 1978,
Environmental Protection Agency, Radiological Impact Caused
™, ? ^10n °f ^dionuclides into Air in the United States,
EPA-520/7-79-006, Washington, B.C., 1979.
Environmental Protection Agency, Phosphate Rock Plants, Back-
ground Information for Proposed Standards, EPA-450/3-79-017
Office of Air Quality Planning and Standards, Research Trian-
gle Park, North Carolina, 1979.
Environmental Protection Agency, Primary Aluminum; Draft
Guidelines for Control of Fluoride Emissions From Existing
Primary Aluminum Plants, EPA-450/2-78-049, Research Triangle
Park, North Carolina, 1979.
Environmental Protection Agency, National Emissions Data
System Information, EFA-450/4-80-013, Office of Air Quality
Planning and Standards, Research Triangle Park, N.C. July
1980.
Environmental Protection Agency, The Radiological Impact of
Coal-Fired Industrial Boilers (Draft Report), Office of
Radiation Programs, Washington, D.C, 1981.
Environmental Protection Agency, Emissions of Naturally
Occurring Radioactivity for Aluminum and Copper Facilities,
EPA-520/6-82-018, Las Vegas, Nevada, November 1982.
Environmental Protection Agency, Emissions of Naturally
Occurring Radioactivity: Underground Zinc Mine and Mill,
EPA-520/6-82-020, Las Vegas, Nevada, November 1982.
Environmental Protection Agency, Draft Environmental Impact
Statement for 40 CFR 191: Environmental Standards for
Management and Disposal of Spent Nuclear Fuel, High-Level,
and Transuranic Radioactive Wastes, EPA-520/1-82-025
December 1?82. *
4-45
-------
EPA82d
EPA83a
EPA83b
Environmental Protection Agency, Final Environmental Impact
Statement for Remedial Action Standards for Inactive Uraniu
Processing Sites, EPA-520/4-82-013-1. October 1982.
Pote«lon
Potential Health and Envl-
ber
ERDA75
^ceasing, EPA 520/1-83-008-
T o Devel°Pnent Administration, Final Envi-
Inpact Statement, Waste Management "
ERDA77a
ERDA77b
Energy Research and Development Administration, Final EnvJ-
ronmantal Impact Statement, High Performance F^el LaSorftory
Hanford Reservation. Richland, Washington, ERDA-1550, UC-2?
11, Washington, D.C., 1977. 9
Energy Research and Development Administration, Final Fnvl-
ronmental Impact Statement, Waste Management
Laboratory«
ERDA77c
ERM77° friL*uc«r,:±^°-!^Ad"i°r'»T. ™ ««.*-
Monitoring Annual Report
ERDA77e
ERDA?7f
ESG82
Fa82
Energy Research and Development Administration, Environmental
Monxtorxng at ^es Laboratory, Calendar Year 1976, IS-™S
Milo D. Voss, Ames Laboratory, Ames, Iowa, 1977. '
Energy Research and Development Administration, Final Envi-
52- ?
Canoga Park, California, 1982.
torTn. R" M" andnCffagno D' G-. Annual Environmental Moni-
toring Report: Calendar Year 1981, Ueport No. MLM-2930
Facility, Mla»l£urg.
4-46
-------
Fra82a Frame P. W,, Environmental Survey of the New England Nuclear
Corporation, Billerica, Massachusetts, Oak Ridge Associated
Universities, Oak Ridge, Tennessee, April 1982,
Fra82b Frame P. W., Environmental Survey of the New England Nuclear
Corporation, Boston, Massachusetts, Oak Ridge Associated
Universities, Oak Ridge, Tennessee, April 1982.
FrbBl Franklin J. C., Control of Radiation Hazards in Underground
Mines, Bureau of Mines, in Proceedings of International Con-
ference on Radiation Hazards in Mining: Control Measurement
and Medical Aspects, Colorado School of Mines, Golden, Colo-
rado, October 1981.
GAC82 Goodyear Atomic Corporation, Portsmouth Gaseous Diffusion
Plant Environmental Monitoring Report for Calendar Year 1981,
Acox,, Anderson, Hary, Klein, and Vausher, Piketon, Ohio,
April 1982.
Go82 Golchert N, W., Duffy T. L. and Sedlet J., Environmental
Monitoring at Argonne National Laboratory -— Annual Report
for 1981 (ANL-82-12), March 1982.
Ja80 Jackson P. 0., et al,, An Investigation of Radon-222 Emis-
sions From Underground Uranium Mines—Progress Report 2,
Pacific Northwest Laboratory, Richland, Washington, February
1980.
Ki79 Kirkland R. S., Mnual Report for the Georgia Institute of
Technology, Facility License R-97, January 1, 1978, through
December 31, 1978, to the U.S. Nuclear Regulatory Commission,
March 7, 1979.
LANL82 Los Alamos National Laboratory, Environmental Surveillance at
Los Alamos During 1981, Los Alamos National Laboratory Re-
port, LA-9349-ENV (UC-41), April 1982.
LBL81 Lawrence Berkeley Laboratory, Annual Environmental Monitoring
Report of the Lawrence Berkeley Laboratory, Report No. LBL-
19553, University of California, Berkeley, California, 1981.
Le79 Leventhal L., et al., Radioactive Airborne Effluents From the
Radiopharmaceutical Industry, in Proceedings of the Health
Physics Society, 24th Annual Meeting, Philadelphia, Pennsyl-
vania, 1979.
Le80 Leventhal L., et al., A Study of Effluent Control Technolo-
gies Employed by Radiopharmaceutical Users and Suppliers, in
Book of Papers, International Radiation Protection Associa-
tion, 5th International Congress, Volume II, Jerusalem,
Israel, 1980.
4-47
-------
Maa78 Magno P., 1978, Radon-222 Releases From Milling Operations,
Testimony Before the Atomic Safety and Licensing Board in the
Matter of Perkins Nuclear Station, May 16, J978.
Mab82 Mason and Hanger-Silas Mason Company, Environmental Monitor-
ing Report for Pantex Plant Covering 1981, MHSMP-82-14,
Amarillo, Texas, 1982.
Na82 Naidn J. R. and Olmer L. L., editors, 1981 Environmental
Monitoring Report, Brookhaven National Laboratory, Safety and
Environmental Protection Division, April 1982.
NRC76 Nuclear Regulatory Commission, Final Generic Environmental
Statement on the Use of Recycle Plutonium in Mixed Oxide Fuel
in Light-Water Cooled Reactors, NUREG0002, Vol. 3, National
Technical Information Service, Springfield, Virginia, 1976.
NRC77 Nuclear Regulatory Commission, Operating Units Status Report,
Data as of July 31, 1977, NUREG-0020, Vol. 4, No. 8, August
1977.
NRC79a Nuclear Regulatory Commission, Generic Environmental Impact
Statement on Uranium Milling, NUREG-0511, Washington, D.C.,
1979.
NRC79B Nuclear Regulatory Commission, Final Environmental Statement
Related to Operation of the Sweetwater Uranium Project,
NUREG-0403, Washington, D.C., 1979.
NRC81 Nuclear Regulatory Commission, A Survey of Radioactive Efflu-
ent Releases From Byproduct Material Facilities, NUREG-0819,
Office of Nuclear Material Safety and Safeguards, Washington,
B.C., 1981.
Ri82 Rice P. D., Sjoblom G. L., Steel J. M. and Harvey B. F.,
Environmental Monitoring and Disposal of Radioactive Wastes
From U.S. Naval Nuclear-Powered Ships and Their Support
Facilities, Report NT-82-1, Naval Nuclear Propulsion Program,
Department of the Navy, Washington, D.C., 1982.
Ro82a Rocco B. P., Environmental Survey of the Medl-Physics Facili-
ty, South Plainfield, New Jersey, Oak Ridge Associated Uni-
versities, Oak Ridge, Tennessee, January 1982,
Ro82b Rocco B. P., Environmental Survey of the E. R. Squibb and
Sons Facility, New Brunswick, New Jersey, Oak Ridge Associ-
ated Universities, Oak Ridge, Tennessee, March 1982.
SNL82 Sandia National Laboratories, Environmental Monitoring Report
1981, SAND-82-0833, Albuquerque, New Mexico, 1982.
Sh81 Sholtis J. A. and More M. L., Reactor Facility, Armed Forces
Radiobiology Research Institute, AFRRI Technical Report
TR81-2, Bethesda, Maryland, 1981.
4-48
-------
Sm80 Smith R. D., The Trace Element Chemistry of Coal During
Combustion and the Emissions From Coal-Fired Plants, Progress
in Energy and' Combustion Science £, 53-119, 1980.
Su82 Sula M. J., McCormack, Dirkes R. L., Price K. 1. and Eddy P.
A.» Environmental Surveillance at Hanford for CY-81, PNL-
4211, May 1982.
TRI79a Teknekron Research, Inc., Information Base (Including Sources
and Emission Rates) for the Evaluation and Control of Radio-
active Materials to Ambient Air, Interim Report, Volume I,
EPA Contract No. 68-01-5142, McLean, Virginia, July 1979.
TRI79b Teknekron Research, Inc., 1979 Utility Simulation Model
Documentation, Vol. 1, R-001-EPA-79, Prepared for the U.S.
Environmental Protection Agency, Washington, D.C., July 1979.
TRI81 Teknekron Research, Inc., Partial and Supplemental Background
Information Document—Primary Pyrometallurgical Extraction
Process (Draft), Report to Environmental Agency under Con-
tract So. 68-01-5142, USEPA Docket Number A-79-11, McLean,
Virginia, May 1981.
TRW82a TRW, Particulate Emissions and Control Costs of Radionuclide
Sources in Phosphate Rock Processing Plants, a report pre-
pared by Stacy G. Smith for Office of Radiation Programs, Re-
search Triangle Park, North Carolina, December 1982.
TRW82b TRW, Industry and Particulate Matter Control Technology
Information for Diammonium Phosphate and Granular Trip Super-
phosphate Manufacture, a report prepared by TRW's Environmen-
tal Division for the Environmental Protection Agency, Re-
search Triangle Park, North Carolina, December 15, 1982.
UC82 University of California, Environmental Monitoring at the
Lawrence Livermore National Laboratory - 1981 Annual Report,
Publication No. UCRL-50027-81, University of California,
Livermore, California, 1982.
UCC82a Union Carbide Corporation, Environmental Monitor! g Report,
U.S. Department of Energy Oak Ridge Facilities, Calendar Year
1981, Report No. Y/UB-16, Union Carbide Corporation, Oak
Ridge, Tennessee, 1982.
UCC82b Union Carbide Corporation, Environmental Monitoring Report,
U.S. Department of Energy, Paducah Gaseous Diffusion Plant,
Paducah, Kentucky, May 1982.
Wa82 Wagner P. and Greiner N. R., Third Annual Report, Radioactive
Emissions From Coal Production and Utilization, October 1,
1980-September 30,1981, LA-9359-PR, Los Alamos National
Laboratory, Los Alamos, N, M,, 1982.
4-49
-------
Chapter 5: REDUCTION OF DOSE AND RISK
5.1 Introduce:'. -
Genetic and somatic health effects due to radlonuclide emissions
can be limited by two basic strategies: (1) the application of emission
control technology, and (2) the implementation of work practice require-
ments. These two control strategies are documented in this chapter.
The particular facilities for which each is applicable are identified,
and the factors that create uncertainties in the evaluation of the effi-
ciency of these and other procedures in the reduction of radionuclide
emissions are described.
5.1.1 Emission Control Technology
Emission control technology implies the installation of a piece of
equipment that removes radionuclides from flue gas prior to its dis-
charge to the air. The most widely used emission control devices are
scrubbers, filters, charcoal adsorbers, cyclonic collectors, and elec-
trostatic precipitatprs (ESPs). These and other less common control
devices are discussed in Section 5.2. Some of these devices are unique
and have only limited application, but all have been demonstrated to be
effective in reducing radionuclide emissions.
5.1.2 Work Practices
Work practice procedures are techniques that reduce radionuclide
emissions at the source by process modifications or refinements. Work
practices include procedures that reduce radionuclide emissions by
reducing the radionuclide content of the process, and processes that
minimize the amount of radionuclides entering the flue gases.
Fugitive emissions are emissions that escape from roof monitors,
doors, storage piles, exposed soil surfaces, etc., rather than from a
stack or vent. If necessary, fugitive emissions are usually reduced
through the implementation of specific work practices. Examples include
applying earth covers, wetting arid areas, and enclosing conveying
equipment. Brief descriptions of the various types of fugitive emission
control are presented in this chapter; more detailed information can be
found in the references.
5-1
-------
Impact of Existing Regulations on Strategies for Reducing
^^ , Partlculate emissions from several of the source cate-
gories discussed in this report are currently regulated by existing
2Ttr iStTLf a1ardS- A ^^ dlSCUS"ion °f -i-"- stands
is pertinent to this chapter because many of the existing techniques for
8
Ereat!r th Sj em^f °«* from coal-fired boilers with a heat input
SP r? ? A m«lion Bt" Per h°Ur are «8«lated under Section 111 of
T f" (M FR 24878' December 23» 1971). The Subpart D New
^rT 6 Standard
-------
The key parameter for evaluating the effectiveness of a control
technology is its collection efficiency. The efficiency of a control
device is the ratio of the amount of pollutant removed to th-j amount of
pollutant entering. Particulate control efficiency can be expressed in
terms of weight, particle number, or radioactivity of pollutant removed;
however, unless stated otherwise, collection efficiency is assumed to be
based on the weight. If the weight is measured over the entire particle
size range or distribution, the efficiency is referred to as the overall
collection efficiency. Collection efficiency can be computed for one or
more particle size ranges, however, and when this is done, efficiency is
reported as fractional collection efficiency.
Penetration is another term that is sometimes used In describing
the performance of a control device. Penetration is the ratio of the
amount of pollutant passing through the control device to the amount of
pollutant entering the device. The sum of penetration plus efficiency
for a control device must equal 1.
Several additional considerations merit discussion in the context
of evaluati ng the effectiveness of an emission control technology in
reducing radionuclide dose and risk. If a process involves high tem-
peratures (e.g., a combustion process), some radionuclides can be vola-
tilized during the process. As the flue gas cools before its discharge
to-the atmosphere, some of the radionuclides may condense on the surface
of nonradioactive particulate matter (i.e., the nonradioactive particles
function as condensation nuclei). Such condensation normally takes
place preferentially on particles with a high surface-to-volume ratio.
This phenomenon results in an increase in the concentration of condensed
volatile radionuclides on smaller-sized particulate emissions. This is
generally referred to as fine particle enrichment. Nevertheless, par-
ticulate matter with condensed radionuclides behave the same as other
particles and can be collected by regular particulate control equipment.
The focus of the remainder of this section is on descriptions of
various control devices available to reduce radionuclide emissions and
identification of the facilities where these control devices can be
used. Because most of these control devices were not designed specifi-
cally to remove radionuclides, explanations emphasize the operating
principles by which the devices collect nonradioactive particulate
matter and gases.
5.2.1 Scrubbers
Scrubbers can be installed on a variety of process exhaust streams
and can serve numerous functions. For example, in phosphate fertilizer
processes, scrubbers can serve economical purposes by recovering and
conserving ammonia (NH_). Scrubbers also efficiently reduce gaseous and
particulate emissions. For the latter purpose, scrubbers are currently
used on coal-fired boilers and in phosphate, elemental phosphorus, and
mineral extraction industries. They are also used by NRC facilities
(PNL83).
5-3
-------
Despite the many designs and applications, the fundamental process
of all scrubbers is the same. In each case, the gas and liquid phase
streams are mixed, and the gaseous and/or participate components of the
gas stream are absorbed and removed from the process by the liquid
stream. The process for disposal of the waste stream can be either
"wet" or "dry," depending on the liquid-recovery design. Spray-tower,
packed-bed, tray-tower, venturi, and wet centrifugal scrubbers are
examples of the types of scrubbers that are in commercial use.
For reduction of radionuclide emissions, most scrubbers function as
a particulate control device and often constitute only part of an over-
all control system that nay also include filters, scrubbers, mist elimi-
nators, charcoal adsorbers, and other devices (TRI79). Figure 5.2-1
illustrates two designs that have proven to be effective in particulate
control. These and other wet scrubbers reduce radionuclide emissions
from sewage treatment plants, light-water-reactor fuel-fabrication
facilities, uranium conversion plants, separation and waste calcining
facilities, uranium "yellowcake" processing and packaging, and elemental
phosphorus plants. (Yellowcake is the final precipitate formed in the
milling of uranium, consisting of various forms of triuranium octoxide,
U_0R). For example, a high-energy venturi scrubber applied to the
exhaust of one elemental phosphorus calciner provided about 97 percent
removal efficiency for polonium-210 (DM80).
Scrubbers are most effective in removing larger particulate matter
(greater than 1 micrometer in diameter) and can be more practical than
filters for exhaust streams with high moisture content. Typical scrub-
ber applications are exhausts from ore dryers and sewage treatment
plants (PNL83). Depending on particle size in the exhaust and the type
of scrubber, efficiencies can range from about 93 percent for a baffle-
type scrubber to 99+ percent for a high-energy venturi scrubber (DM80).
As previously discussed, some radionuclide sources are regulated by
EPA's New Source Performance Standards. Two such sources, coal-fired
boilers and the phosphate fertilizer industry, must operate scrubbers to
achieve sulfur dioxide and fluoride emission reductions, respectively,
and these scrubbers also reduce radionuclide emissions.
5.2.2 Filters
Filters are one of the most frequently used radionuclide emission
reduction devices. Various designs provide effective particulate con-
trol in each of the nine source categories except underground uranium
mines. The effluent characteristics (e.g., volume, temperature, and
type of particulate) of the source categories may differ greatly, but
most filter designs can accommodate a wide range of operating condi-
tions. Filters are extremely versatile and can be used to supplement
other control equipment, such as ESPs and mechanical collectors.
The types of filters used to reduce radionuclide emissions include
high-efficiency particulate air (HEPA), fabric, sintered-metal, and sand
filters. Efficiencies of HEPA filters, as reported by vendors, are
5-4
-------
SYMBOLS
A
B
C
D
E
F
WET CENTRIFUGAL SCRUBBER
PARTS
CLEAN-AIR OUTLET
ENTRAPMENT SEPARATOR
WATER INLET
IMPINGEMENT PLATES
DIRTY-AIR INLET
WET CYCLONE FOR COLLECT-
ING HEAVY MATERIAL
WATER AND SLUDGE DRAIN
VENTURI SCRUBBER
Figure 5.2-1. Wet scrubber participate control devices (PNL83),
5-5
-------
99.9? percent (DM80). Efficlencles of
(DM89). Except for the sintered
wany different applications.
.
and nave achieved QQ QQO ar,^ QQ n
removal efficiencies, respectively (PNLs)
Fabric filters (baghouses) , shown in Fieure 5 2-1
•
efficiency of 99 percent or areatPr H I? * a°"Ves a "mova
-- P
8hovm ^
°°""°1
DOE £aclUtles. Such fU«s a idea! for°M hf
volume, exhaust stream. Sand filters have hifh ^"f" "' l"ee~
equal to those of a single-staged HEPA f ^ hi8huremoval efficiencies
e emssons efective I v wii-h
- - ^ -: ---
applications. Fabric filters
operate lB
,
the dlaposal
5-6
-------
PRESSURE
TAPS
BLOWBACK
VALVES (3)
EXIT
MANIFOLD
©e EXIT
VIEWING
PORT
SINTERED METAL
FILTER
VALVE
Figure 5.2-2, Pilot-plant sintered-metal filter.
5-7
-------
OUTLET
PIPE
BAFFLE
PLATE
CLEAN-AIR
SIDE
HOPPER
WALKWAY
AIR REVERSAL CLEAN-AIR HANI FOLD ,
SHAKER Q VALVE 1 1'
& i—i N— i 1~ i n <— i 1 — i — — r~i -r~» — "
(1
V—
n
y \aC4
i
- SCREW
rnni/f YOB
n
L J
N
1
r~
i
tj
f
«
_i_
1 ' ^.- - Z
r~i
i i
L j
^
T
TT
i i
u J
i
y
\ ^
\ r
DISCHARGE INSPECTION
DOOR
rS r— i i
n
i i
L j
-A
\
_L
n
i
U J
y/
w^ • -
T ., \ .»:
f T
i i
C i
h
i
SCRE«
CONVEYOR
n
CLEAN AIR
TO FAN
Figure 5.2-3. Fabric filters.
5-8
-------
LAYER DEPTH, in.
A
B
C
D
E
F
12
12
12
6
12
36
INLET
PLENUM
AIR FLOW
SPECIFICATIONS
1-1/4-in. TO 3-!n. GRAVEL
5/8-in. TO 1-1/2-in. GRAVEL
1/4-in. TO 5/8-in. 6RAVEL
HO, 8 TO 1/4-in SAND
NO. 20 TO NO. 8 SAND
HO. 50 TO HO, 30 SAND
DISCHARGE
PLENUM
STATIC LINES
MONITORING TUBES
SAMPLING LINES
Figure 5.2-4, Multilayered sand filter.
STEEL-CASED HEPA FILTER
Figure 5.2-5. Open-face HEPA filter.
5-9
-------
so, the use of HEPA filters is limited to ambient air exhaust stream
iditions, which may require that moisture separators or other control
i?ices be installed upstream of these filters.
2.3 Mechanical Collectors and Electrostatic Precipitators
Mechanical collectors, illustrated in Figure 5,2-6, separate
cticles from the gas stream by centrifugal and gravitational forces.
s collection efficiency of a mechanical collector is a function of
^figuration and particle size. For a double-vortex cyclone, reported
paration efficiencies exceed 99 percent for particles greater than 6
crometers and 95 percent for particles greater than 1 micrometer (Ae),
nerally, mechanical collectors are used only as precleaners upstream
an electrostatic precipitator or a fabric filter.
Electrostatic precipitators (ESPs), shown in Figure 5.2-7, use
gh-voltage sources to charge the particles in the gas stream, which
e then collected on large metal plates. The collecting plates are
riodically cleaned by rapping. The efficiencies of ESPs can exceed
.9 percent, depending on the application.
Electrostatic precipitators and mechanical <: lectors are currently
ed •o reduce radionuclide emissions from coal-i. .
-------
GAS
IN-*
OUTLET T\JB£
OUST OUTLET TUBE
DUST Join
CHAN-SAC TUtI
CLEAN-GAS
OUTLET
DIHTY-GAS INLET
DUST PARTICLES
DROPPING INTO HOPPER
CAST IRON
COLLECTING TUBE
ASK HANDIJMG
VALVI
Figure 5.2-6. Mechanical collectors.
5-11
-------
TOP END _
PANEL GIRDER
GAS
DISTRIBUTION
PLATE -^
ROLL-FORMED
OUST PLATE -
!8-in. MODULES
CASING
PANEL
SUPPORTING
STEEL
BEL0K THIS
ELEVA*ION
Sn?
IONIZER
COLLECTOR
PLATES
LOW-VOLTAGE DESIGN
12,000 TO 13,000 VOLTS ON IONIZER
6,000 TO 7,000 VOLTS ON COLLECTING
PLATES
Figure 5.2-7. Electrostatic precipitators.
5-12
-------
of particulate filters. In fact, they are often Installed In series
with HEPA filters or other particulate control devices.
In most instances, charcoal adsorbers are used to reduce radio-
iodine emissions; however, these adsorbers can be used, along with
chilled traps, to collect and store krypton and xenon gases. Because
these gases have relatively short half-lives, radionuclide dose and risk
can be reduced by preventing the gases from being released until their
radioactivity has decreased. Manufacturers of radiation sources utilize
this method. Charcoal adsorbers can also be used to remove antimony
from hot-cell exhausts (PNL83).
Despite attractive (possibly 99.9 percent) radioiodine removal
efficiencies, some important limitations must be considered before char-
coal adsorbers are installed. Flow rate, humidity, temperature, iodine
concentration, adsorber bed age, and other parameters affect removal ef-
ficiencies and may reduce them to 90 percent. For example, as tempera-
ture increases, iodine desorbs from charcoal. Also, in certain effluent
streams, charcoal may ignite at temperatures as low as 180°C, and the
presence of nitrogen oxide-nitrogen dioxide (NO-NO-) may cause spontane-
ous ignition (PNL83).
5.2.5 Miscellaneous Emission ControlEquipment
A few unique control technologies are used to control radionu-
clides. These include silver-based sorbent systems, oxidation/adsorp-
tion processes, cryogenic distillation, and purge cascades, which —
although not widely used — have been shown to be effective in reducing
gaseous and particulate radionuclides.
Silver-Based Sorbent Systems
Silver-based sorbent systems can employ solid or liquid sorption
techniques. Solid sorbent systems use reactant pellets in a pellet bedj
liquid sorbent systems use a concentrated scrubbej: solution in a packed
tower. Despite the process differences, the emission control capabili-
ties of both systems are similar.
In general, both solid and liquid sorbent systems have proven to be
effective radioiodine control devices for certain processes. They are
currently used at waste-management and fuel-processing facilities.
Silver-based sorbent systems operate most effectively in low-flow ex-
haust streams or in the removal of trace amounts of iodine downstream of
other control devices. With few exceptions, solid and liquid sorbent
systems are chemically and thermally stable and therefore function
efficiently when charcoal adsorbers might be unsafe or ineffective.
Despite their similarities in application, the particular process mecha-
nisms and parameters of solid and liquid sorbent systems should be
discussed independently.
Most solid sorbent control devices are similar in design and oper-
ating parameters. Solid sorbent control devices use silver zeolite or
silver nitrate silica pellets in pellet beds that are 5 to 20 cm thick.
5-13
-------
Operating temperatures should be higher than 110°C to prevent moisture
interference. Regeneration of pellet beds has not been perfected;
therefore, the use of two pellet beds is recommended so that the spent
reagents in one pellet bed can be replaced while the other bed Is oper-
ated. Limitations of these sorbent systems are the instability of some
silver zeolite reagents when used on acidic exhausts from reprocessing
facilities, lack of knowledge regarding the chemistry of the sorption
processes, and uncertainties regarding decreased efficiencies caused by
organic vapors, other halogens, and sulfur compounds.
The only liquid sorbent control system in existence is one in which
iodine is removed by coating Berl saddles or other types of ceramic
packing with a concentrated silver nitrate solution. Exhaust gases pass
over the coated packing at about 190°Ct and radioiodine is collected as
silver iodide or silver iodate on the ceramic packing. As silver ni-
trate is consumed, the packing is recharged by flushing with a fresh
silver nitrate solution. Proper operation of this system requires
accurate temperature control and frequent packing regeneration. As with
solid sorbents, liquid sorbent iodine capacities are reduced in the
presence of other halogens (PNL83).
Oxidation/Adsorption Processes
The oxidation/adsorption process is a potential means of control-
ling tritium emissions from various reactors and processing facilities.
At low concentrations, tritium must be converted to tritiated water to
be effectively trapped. Thus, the process for removing tritium from
exhaust gases involves two steps: (1) oxidation of tritium to tritiated
water, and (2) removal of tritiated water. Oxidation of tritium is
accomplished by the use of catalysts or metal oxides. For example,
tritium from a process with an operating temperature of about 177°C can
be effectively oxidized with platinum or palladium catalysts. Similar
results can be obtained with a metal oxide, such as copper oxide, on
processes operating at temperatures between 500° and 700°C. In both
cases, hydrogen can be added to the exhaust stream to enhance tritium
oxidation. Furthermore, the metal oxide beds can be regenerated with
air at temperatures between 300° and 500°C.
The absorption of tritiated water involves cooling the exhaust,
passing it through a condenser, and then through a desiccant. The
exhaust cooler reduces the gas temperature to approximately 21°C (room
temperature). The condenser removes as much of the tritiated water as
possible before the gas stream enters the desiccator. This reduces the
dewatering required by the desiccator and also the frequency at which
new desiccant is needed or desiccant regeneration is required. The
desiccator uses silica gel, molecular sieve, or other desiccants to
remove moisture still entrained in the gas stream.
Overall removal efficiencies vary with design criteria and operat-
ing parameters (such as bed depth, flow rate, and temperature). Tritium
5-14
-------
emissions can be reduced below normal detection levels (0.1 to 1.0
uCi/iea ) of an ionization counter; however, the numerous operating condi-
tions involved make it impossible to state specific reduction efficien-
cies (PNL83, DM80). Tritium removal efficiencies as high as 99.9999
percent have been reported (PNL83).
Cryogenic Distillation
For several years, rare gases have been processed commercially by
cryogenic distillation. This same cryogenic process can be used to
reduce radioactive noble gas emissions from sources such as fuel re-
processing plants and reactors. Krypton and xenon are the most common
noble gases removed by cryogenic distillation. After removal, these
condensed gases are stored until their radioactivity has decreased to a
level that is safe for release.
The distillation process is a complex multistage system. Exhaust
gases must be pretreated before they enter the cryogenic distillery.
Potentially hazardous or troublesome gases, such as carbon dioxide,
nitrogen oxides, and oxygen, must be removed from the exhaust stream.
After pretreatment, the exhaust stream enters a two-stage separation
process. In the first stage, krypton and xenon are separated from the
other gases by the use of liquid nitrogen in a countercurrent stripping
column. In the second stage, a fractioning column is used to separate
the krypton and xenon for storage (PNL83).
Purge Cascades
Purge cascades are a series of traps through which an exhaust
stream must pass before being vented to the atmosphere. Depending on
the filtering media being used, these traps can remove a wide range of
radionuclides. Purge cascade traps may contain particulate filtering
media or scrubbing media. Media such as sodium fluoride and alumina are
used to control particulate emissions related to uranium and thorium
processes. Caustic scrubbing solutions, such as sodium hydroxide or
potassium hydroxide, lower the radioiodine concentrations in exhaust
streams and also reduce particulate emissions. In addition, some traps
are refrigerated to increase their radionuclide reduction efficiency.
These cascades are used at several gaseous diffusion plants to reduce
gaseous and particulate emissions (DM80). Efficiencies range from 85 to
100 percent, depending on design, application, and filter media used
(DM80).
5.3 Work Practices
Work practices are process modifications, refinements, or tech-
niques that reduce radionuclide emissions at their source. Some work
practices were developed to improve process performance and have inci-
dentally proven to be effective in reducing radionuclide emissions and
dose and risk. Many of these practices are currently being implemented.
5-15
-------
8
™
H the
decay. Delayed venting techniques can be
facilities,
accelfratorf^Tf a?cejera*or tubinS ls a wk practice applied to
accelerators The air in the accelerator tubes ls evacuated to reduce
attiv!0™* 1 Elr ^e radi°actl™ d«ing operation and thus the radio-
active emissions. When evacuations are impractical, pure inert Rases
surfer Rel!11* ^ ^ ** ^ ^^ ^« "tivfted
to reduc"; rad, M £" ' ""^ °Perators i-plement this technique
to reduce radioactive noble gas emissions, primarily argon.
use o ,ln underSround «r.nl«» mining include the
fillii L!f fS6K K' bul^heading' mine Pressurization, and back-
filling. Sealants have been able to reduce radon emissions in under-
I™ raK1Um,f "eS by " PerC6nt' MthouSh seal»ts have not yet been
lor the Vll e"ecjlvj J6CauSe °f their hi8h ~-t. research continues
for the development of better sealants. Bulkheading, as shown in Figure
mini I- JV6S & °ff mlned-°ut st°Pe«. ^is practice can reduce
mine-air radon concentrations up to 60 percent. Pressurizing a mine as
shown in Figure 53-2, retards radon diffusion into mine air? howeve -,
cent L v«,r emissions has b"" estimated to be about 20 perl
cent. Backfilling entails refilling mined-out areas with waste or dirt-
this procedure can control up to 80 percent of the radon emissions.
er^ W- .<* *««* practices, including washing ore and wet
, are applicable to uranium and phosphate processing. Washing
process feed rock to reduce its initial dust concentration before the
milling process has proven to be effective in controlling particulate
S°"
extenwh h * °" ^^ or "Iclnera, or the
extent to which an ore is dryed or calcined, determines the amount of
fine dust in the product and, consequently, the amount of particulate
emissions. Wet-grinding systems, a viable alternative to dry-grindina
processes emit fewer particulates and eliminate the need to dry the
feedstock (TRI79). Wet-grinding phosphate systems are currently used to
reduce particulate emissions jt two facilities.
_ Mining and milling industries use work practices to reduce fugitive
emissions. The control of fugitive emissions from uranium, phosphate,
and other metal and nonmetal mining/milling processes primarily reduces
particulates and radon gas (Ko80) . These controls include earth covers,
wetting of arid areas, and covered transport facilities.
Earth covers which consist of layered soil approximately 3 meters
deep are frequently used on waste piles, reclaimed lands, or inactive
surface mining areas to reduce both particulate and radon emissions
5-16
-------
TIHBER
LAGGING
EPOXY
COATING"
BRATTICE
COATED
W/EPOXY
*
I
BLOWER
POLYETHYLENE
Figure 5.3-1. Bulkheadlng of mine scopes.
Po
P > PO
Figure 5.3-2. Mine pressurization.
I
P
* * *
i
1 AIR
!
I
FLOW |
STOPE
P P P |
P
i
P
5-17
-------
Earth covers may not be practical for mining areas and storage piles
that are only temporarily inactive because of the need for frequent
access. Fugitive emissions from arid storage piles and mining areas can
be controlled by wetting the exposed surfaces with water. Chemical
sprays (as opposed to water) are used occasionally, but only to coat
waste piles. Covering transport facilities (e.g., conveyor systems),
which are used throughout the mining/milling operations, not only re-
duces emissions, but also conserves and protects resources (TRI79,
DM80).
Controlled land use is another strategy that reduces population
exposure, but it is not classified as a work practice. By owning and
controlling the use of a buffer area of land surrounding a mine, mining
companies can reduce the radiation dose and risk to the population
without necessarily reducing actual radionuclide emission rates.
5.4 Summaryof Emission Reduction Strategies
Table 5.4-1 summarizes the major control technology applications.
Although certain control devices nay be applicable to a particular
source, a multitude of processes within that source may require inde-
pendent control devices; therefore, Table 5.4-1 also includes supple-
mental controls. Because fugitive and process techniques have limited
or no supplemental controls, however, most of the supplemental controls
shown are source control devices.
5.5 Uncertainties in Evaluation ofControl Technology Efficiencies
A key parameter for evaluating the overall performance of a control
technology is its removal or collection efficiency. A more important
parameter in terms of reducing radiation dose and risk, however, is its
radionuclide collection efficiency. Collection efficiencies are usually
determined in one of two ways; (1) direct measurement of pollutant
levels (e.g., stack testing), or (2) mass/material balance. Efficiency
calculations based on direct measurement require the simultaneous mea-
surement of the radioactivity of the pollutant entering the control de-
vice and the radioactivity of the pollutant exiting the control device.
Efficiency calculations that use a material balance do not directly
measure the amount of pollutant emitted to the atmosphere. For example,
the particulate collection efficiency of an ESP Installed on a coal-
fired boiler can be estimated by measuring the ash content of the coal,
the coal feed rate, and the amount of fly ash collected by the ESP. In
this example, the difference between the weight of ash entering the ESP
and the weight of fly ash collected by the ESP would be assumed to be
the amount discharged to the air.
Additional uncertainty is associated with quantifying radionuclide
collection efficiency. In the above example, determination of particu-
late collection efficiency was relatively straightforward. Determining
radionuclide collection efficiency is complicated by factors such as
fine particle enrichment and the physical and chemical forms of the
radionuclides.
5-18
-------
Table 5.4-1. Summary of emission reduction strategies
Source category
and/or affected facility
Research and test reactors
Hot cells
Commercial waste management
Plutonium fuel fabrication
Plutonium production reactor
Radiophannaceutlcal
Separation/waste calcining
Elemental phosphorus
High-level waste tank farm
Extraction industry
Plutoniui, glovebox/storage
vault
Phosphate industry
Accelerators
Mining
Coal-fired boilers
Uranium conversion
Light-water reactor fuel
fabrication
Pouer generating reactors
Uranium milling
Weapons test sites
Control UchnoloRita Work practices
Mechanical Fugitive
_ .. collectors Charcoal Process emission
Scrubber Filters and ESPs adsorbers techniques controls
p I NG
f l.Sb
P I P
P
p I
p p P I.NG m
p p p 1
p p p
p
p p p p p
p
p p p p p
P NG
Rn
P P p
P P
P P
p I.NG KG
P P p P.Rn
p I BG
iegend for types of mdionuclide gases reaoved:
P - Participates—uranium, plutonlum, and others
1 » Iodine
NG - Hoble gases—argon, krypton, and Kenon
Rn " Radon
Sb • Antinony
5-19
-------
If the actual percentage of fine particle enrichment is unknown or
is known to fluctuate with process changes, the use of particulate col-
lection efficiency to estimate radionuclide collection efficiency adds
still another degree of uncertainty. For example, a particulate control
device may be known to have an overall collection efficiency of 99 per-
cent; however, if a process is characterized by significant enrichment,
the radionuclide collection efficiency osay be considerably less than 99
percent because of the higher concentration of radionuclides on the fine
particles that are in the 1 percent fraction that is not removed by the
control device. Also, a high-temperature process can volatilize radio-
nuclides (e.g., Po~210) that are otherwise in the solid (particulate)
state. If a process is equipped with a particulate control device and
some fraction of radionuclides is volatilized, the radionuclide collec-
tion efficiency is uncertain and becomes dependent on quantifying the
amount of volatilization that has occurred.
All physical measurements required to calculate efficiency are
subject to uncertainties imposed by the precision and accuracy of the
sampling methodologies and the analytical procedures. Sampling uncer-
tainties not only include variabilities in the procedures used to col-
lect the sample (e.g., repeatability and reproducibility of the method),
but also such variabilities as the representativeness of the sample
collected and the representativeness of process operation conditions at
the time of sampling. Thus, the uncertainty associated with sample
collection is difficult to quantify. On the other hand, quantifying the
uncertainty associated with analytical procedures is more straightfor-
ward and can be accomplished by computing a 95 percent confidence inter-
val for each analysis.
Despite the uncertainties involved in determining control technol-
ogy performance, control efficiencies usually do not vary dramatically.
For example, a high-energy scrubber with a specified efficiency of 99
percent will normally operate within a percentage point of this value as
long as the equipment is operated in accordance with design specifi-
cations. Furthermore, when the uncertainty of all the elements in the
overall radionuclide risk assessment process is considered, the uncer-
tainty associated with quantifying control technology performance does
not appear to be a major contributor to the overall uncertainty in the
final assessment results.
5-20
-------
REFERENCES
Ae Aerodyne Development Corporation, Series "sv" Dust Collector,
Bulletin No. 1275-SV, undated.
DM80 Danes and Moore, Airborne Radioactive Emission Control Tech-
nology, unpublished report prepared under EPA Contract No,
68-01-4992, White Plains, New York, 1980.
Ko80 Kown B. T.» et al., Technical Assessment of Radon-222 Con-
trol, Technology for Underground Uranium Mines, Bechtel
National, Inc., Report prepared under EPA Contract No.
68-02-2616, Task 9, 1980.
PNL83 Pacific Northwest Laboratory, Control Technology for Radioac-
tive Emissions to the Atmosphere at U.S. Department of Energy
Facilities (Draft), PNL-4621, March 1983.
TRI79 Teknekron Research, Inc., Technical Support for the Evalua-
tion and Control of Emissions of Radioactive Materials to
Ambient Air, McLean, Virginia, 1979.
5-21
-------
Chapter 6; MOVEMENT OF RADIONUCLIDES
THROUGH ENVIRONMENTAL PATHWAYS
6*1 In tr oduct i on*
This chapter describes how airborne radionuclides are transported
in the environment from the point of release into the air up to the
final concentration in various compartments of the environment, where
these radionuclides can affect human beings. The objective of Chapter 6
is to describe the environmental pathways that are considered at the
U.S. Environmental Protection Agency in evaluating radionuclide concen-
trations in air, soil, and food that result from airborne releases of
radioactivity from various facilities. In this context, facilities are
not only those normally associated by the public with radioactivityi
such as national laboratories and uranium processing plants, but also
other mineral processing plants, fossil fuel combustion facilities, etc.
The airborne environmental pathways are shown in Figure 6.1-1.
Starting the process, the radionuclide sources release the materials in
the form of particulates or gases, footing a plume that disperses down-
wind (Section 6.2). Concentrations of these radionuclides in the air
can directly affect people in two ways: through external dose caused by
photon exposure from the plume, or through internal dose resulting from
radionuclide inhalation. As the airborne radionuclides move from the
point of release, they (especially those in particulate form) deposit on
ground surfaces and vegetation as a result of dry deposition and pre-
cipitation scavenging (Section 6.3). Photon radiation from the radionu-
clides deposited on the ground also contributes to the external doses.
Finally, small fractions of the radionuclides deposited on plant
surfaces and agricultural land enter the food chains, concentrating in
produce and in animal products such as milk and meat (Section 6.4).
Consumption of contaminated foodstuff then contributes to the internal
doses of radiation to individuals.
Radionuclide concentrations in air, on soil surfaces, and in food
products can be calculated by using the computer code AIRDOS-EPA. A
description of the code and some examples of its applications, with an
overview of the uncertainties associated with its predictions, appear in
Section 6.5.
*Technical terms such as radioactivity, exposure, dose, and photon
radiation are defined in Chapter 7,
6-1
-------
CONCENTRATION IN PLANTS
CONCENTRATION
IN ANIMALS
Figure 6.1-1. Pathways of airborne radionuclides into the environment
6-2
-------
This chapter gives an overview of the basic environmental processes
considered by EPA in assessing atmospheric releases of radionuclides.
See references Ha82, Ti83, and NCRP84 for a more detailed description of
the processes, modeling techniques, and uncertainty estimates.
6•2 Dispersion of Radionuclides Through the Air
6.2.1 Introduction
Radionuclides entering the atmosphere are transported away from
their point of release and are diluted by atmospheric processes. To
perform a radiological assessment, it is necessary to model the long-
term average dispersion resulting from these processes. This is because
the sources under consideration release radionuclides at rates that are
substantially uniform when considered over long periods of time, and
because the somatic and genetic effects on human health are generally
treated as being the result of chronic exposure over long periods of
time.
As large-scale winds move over the earth's surface, a turbulent
boundary layer, or mixed layer, is created that controls the dispersion
of the released radionuclides. The depth and dispersion properties of
the mixed layer, which are highly variable over short periods of time,
are controlled by two sources of turbulent effects: mechanical drag of
the ground surface and heat transfer into or from the boundary layer.
The mechanical drag of the ground surface on the atmosphere creates a
shear zone that can produce significant mechanical mixing. The mechani-
cal mixing is stronger when the wind is stronger and the roughness ele-
ments (water, grains of dirt, grass, crops, shrubs and trees, buildings,
etc.) are larger. The vertical scale (dimension or thickness) of the
mechanical mining zone is related to the size of these roughness
elements. Heat transfer into or from the boundary layer, the second
source of turbulent effects, also strongly affects the mixed layer's
turbulent structure and thickness. Solar heating creates huge rising
bubbles or thermals near the ground. These large bubbles produce
turbulent eddies of a much larger scale than those from the mechanical
drag of the ground surface. With strong solar heating on a clear day,
the mixing layer may be a few thousand meters deep. On a clear, calm
night the boundary layer virtually disappears, so that radionuclides
(and other pollutants) are dispersed with very little turbulent
diffusion.
The objective of the atmospheric transport models used by EPA is to
incorporate the essential physical data necessary to characterize an
extremely complex turbulent flow process into a simplified model that is
adequate to predict the long-term dispersion of radionuclide releases.
In general, the data necessary to implement a detailed theoretical model
of atmospheric dispersion are not available and would be impractical to
obtain. Apart from the data problem, the mathematical complexities and
difficulties of a direct solution to the turbulent dispersion problem
are profound and beyond the practical scope of routine EPA regulatory
assessments. The widely accepted alternative has been to incorporate
6-3
-------
experimental observations into a semi-empirical model such as outlined
below that is practicable to implement.
Three basic meteorological quantities govern dispersion: wind
direction, wind speed, and stability. Wind direction determines which
way a pluae will be carried by the wind: a wind from the northwest
moves the plume toward the southeast. Although wind direction is a
continous variable, wind directions are commonly divided into
16 sectors, each centered on one of the cardinal compass directions
(e.g., north, north-northeast, northeast, etc.). Since there are
16 sectors, each one co*~ers a 22-1/2-degree angle. Wind speed directly
influences the dilution of radionuclides in the atmosphere. If other
properties are equalt concentration is inversely proportional to wind
speed. This raises the question of what happens in a calm. A wind too
light to turn an anemometer (about 0.5 m/s) and therefore recorded as a
calm can still disperse an atmospheric release. Customary wind speed
categories include 0 to 3 knots* (lowest speed) to greater than 21 knots
(highest speed).
Atmospheric stability, the third meteorological quantity,
categorizes the behavior of a parcel of air when it is adiabatically
(without heat transfer) displaced in a vertical direction. If the
displaced parcel would be expected to return toward its original
position, the category is stable; if it would continue to move away from
its original position, the category is unstable. Under conditions of
neutral stability, the parcel would be expected to remain at its new
elevation without moving toward or away from its old one. Typically,
the conditions associated with the unstable classes are very little
cloud cover, low wind speeds, and a sun high in the sky. The atmosphere
is neutral on a windy, cloudy day or night, and is stable at the surface
at night when the sky is clear and wind speeds are low. Dilution due to
vertical mixing occurs more rapidly with increasing distance under
unstable conditions than under stable ones. Stability categories range
from A (very unstable) to D (neutral) to G (very stable).
A table of frequencies (fractions of time) for each combination of
stability, wind direction, and wind speed is the starting point for any
assessment of long-term atmospheric dispersion,
6.2.2 Air Dispersion Models
EPA uses a Gaussian model for most radionuclide dispersion
calculations. The model also includes consideration of such processes
as plume rise, depletion due to deposition, and radionuclide ingrowth
and decay.
*A knot is one nautical mile per hour. A nautical mile is 1852
meters.
6-4
-------
Gaussian Plume Model
The basic workhorse of EPA dispersion calculations is the Gaussian
model, Hanna et al. (Ha82) have listed several reasons why the Gaussian
model is one of the most commonly used. These are quoted below:
"(1) It produces results that agree with experimental data as well
as any model.
"(2) It is fairly easy to perform mathematical operations on this
equation.
"(3) It is appealing conceptually.
"(4) It is consistent with the random nature of turbulence.
"(5) It is a solution to the Fickian diffusion equation for
constants K and u.
"(6) Other so-called theoretical formulas contain large amounts of
empiricism in their final stages.
"(7) As a result of the above, it has found its way into most
government guidebooks, thus acquiring a 'blessed* (sic)
The long-term Gaussian plume model gets its name from the shape
presumed for the vertical concentration distribution. For a ground
level source, the concentration is maximum at ground level and decreases
with elevation like half of a normal or Gaussian distribution. For an
elevated release, the concentration is symmetrically distributed about
the effective height of the plume, characteristic of a full Gaussian
distribution. Actually the vertical dispersion is limited by the ground
surface below and any inversion lid* above the release (see Fig. 6.1~2).
At large distances from the point of the release, the concentration
becomes uniformly distributed between the ground and the lid. Within
each of the 16 direction sectors, the concentration is considered to be
uniform at any given distance from the release. For a ground-level
release, the ground-level concentration decreases monotonically with
distance from the release point; for an elevated release, the ground-
level concentration increases, reaches a maximum value, and then
decreases with increasing distance from the release point.
*An inversion lid is defined by the altitude in the atmosphere
where the potential temperature begins to increase with increasing
height, thus limiting the volume of air available for diluting releases,
6-5
-------
i
ELEVATION
tm\ 8£fllll
1
«,., , Hln..>».e \ IHTlRMtDtATi 1 \ !
SMALL DISTANCE \ OBTANCE \ \
%£™SS£'M" \SKSlVSfiS» ! \ !
V \ 1 '
^ i i 1 '
AII i i i i» ..1
ID 1 KVCJ •>! EAfiF
»IXINO LID HEIGHT
LAHOE DISTANCE
(UNIFOHMLV MIXED)
DISTANCE FROM
HELEASE
MIXING LID HEIGHT
ELEVATION
•H.
SMALL DISTANCE
(CAUSSJAN)
r?.
1 \ !
\ \ !
i\ INTIMMID1ATE DISTANCE \ I
]\ (OMOUNDANDMIXINS \ i
D LID AFFECT PMOFILEI \ *
)'\
LANE DISTANCE
(UNIFORMLY MIXED)
(bl ELEVATED RELEASE
Figure 6.1-2.
Vertical concentration profiles for plume
versus downwind distance from release.
6-6
-------
Mathematically, the long-term average dispersion calculation used
by EPA can be expressed as
2.03 exp[-0.5(h /„ )2]
X/Q =
H
u x g
where X/Q (s/m3) £s the concentration for a unit release rate at a
distance x(m) from the release point, he(m) is the effective height of
the release, az(ni) is the vertical dispersion parameter appropriate to
the stability category and distance x, and u(m/s) is the wind speed. At
distances where the release is uniformly nixed between the ground and
lid, the expression becomes
X/Q = -Li" (6-2)
M u x h
where hj(m) is the lid height and the other quantities are the same as
before.
Plume Rise Model
Vertical momentum or buoyancy can cause a plume to rise to an
effective height that is several times the physical height of the
release. The momentum flux of a release is proportional to the product
of the volume flow rate and the vertical exit velocity while the
buoyancy flux is proportional to the product of the volume flow rate and
the difference between the temperatures of the release gases and the
ambient air. Momentum rise is. initially dominant for most plumes, even
though buoyant rise may become the more important process at larger
distances. In any case, plume rise increases with distance from the
release point; the effective height of the plume may not reach a
limiting value until the plume is several kilometers from the point of
release.
Plume Depletion Model
As radionuclides in the plume are dispersed, their activity is
depleted by dry deposition and precipitation scavenging.
The rate of plume depletion due to dry deposition and precipitation
scavenging is proportional to the deposition rate (see 6.3). ORP uses a
source depletion model which considers the shape of the vertical con-
centration profile to be unchanged by depletion. Depletion due to
deposition generally does not cause more than half of the released
activity to be removed at a distance of 80 km. Depletion by
precipitation scavenging occurs only during periods of precipitation.
6-7
-------
Radiological Decay and Inprowth
alS° reduce the concentration in the plume.
^"^ betWCen the P°int of relea*« »d •
soraf-liv " ab°Ut 5 hOUr8' Thu8' only nuclid« *«>
For La* 1 Tl be.aPPreciably depleted by radiological decay.
S pS^l if'Ifr f-*i1Ch haS-a 1*8 h°Ur half~life> d«*y* ^ about
13 percent of its original activity in 5 hours.
a chaitha. radi°nucUde *• a P««* for other radionuclides in
even though rh 7 Pr°dUCtS WiU beCOme part °f the Pluffie's
y n°* releaS6d b thC 80
is th K ' P*' ce.
is the parent of barium- 137m, which has a half-life of about 2 6
mnutes The bariu»-137B activity would reach 90 percent of that of th
T*l7f I L/ ^°Ut 8'5 minUteS' the time "quired at a typical wind
speed of 5 «/. for the release to travel about 2.5 to. For many
nuclides, the radiological effects associated with exposure to decay
products are at least as important as those from exposure to the parent.
' ' *" h°
a , of "sua- s
due to photons from its decay product bariuin-137m.
6'2'3 ""certainties in Atmospheric Dispersion Modeling
EPA must deal with several uncertainties in its modeling of
,1™? "i- dlSP^ST' TW° basic """derations contribute to these
model and r- »' K"" lnV°1VeS the Paran»*«« that enter into the
situanf ThW they are ^°™ °r C3n be det«™i^d for a particular
situation The presumption is that the basic assumptions for which the
c±Lr% 7e !" satisfied and ^at the uncertainty of predicted
the «? f1^ PSnuS Primarily on the Certainty of the data used in
the calculations. The second consideration involves the use of a
modeling technique under conditions that do not satisfy the basir
Br»oH~Hr fST WhlCh the m?del W3S devel°P^. Such use may be the only
III ne f natlVe aV3ilable for «"e^ing atmospheric dispersion/
but the principle uncertanties are now related to evaluating the sig-
nificance of these effects that are not considered in the model. An
example of this would be the use of the Gaussian plume model, which was
developed for short distances over an open, flat Lrrain, to'Jsess
dispersion over large distances or in a complex terrain dominated by
hills and valleys. J
In regard to the first consideration, the authors of NCRP84
£?!£ HdeVh*i ^e de5erTnination of appropriate basic parameters such as
wind speed and direction can be accomplished so that they are not major
contributions to model uncertainty. However, the uncertainties
associated with derived parameters (such as stability class) or lumped
parameters (such as those used to characterize deposition, resuspension,
or building wake effects) can dominate the model uncertainties.
The effect of the uncertainty of an input variable can strongly or
weakly influence the model output depending upon circumstances, for
example, the effective height of a release, he, can be estimated using
6-8
-------
a plume rise model to within a factor of about 1.4 (NCRP84). From
equations (6-1) and (6-2), it is clear that when ffz is much smaller than
he that the effect of this uncertainty on equation (6-1) is strong;
whereas at large distances where equation (6-2) is appropriate, the
value of he has little effect on the calculated concentration at all.
Little and Miller (Li79 and Mi82) have surveyed a number of
validation studies of atmospheric dispersion models. Although these
studies provide limited data, they indicate an uncertainty of
approximately a factor of 2 for annual average concentrations for
locations within 10 km of the release and approximately a factor of 4
(77 percent of their samples) to 10 (92 percent of their samples) for
locations between 30 and 140 km of the release. The validation studies
were for fairly complex terrain, i.e., substantial hills and valleys but
not extreme conditions of either terrain or meteorology.
6.3 Deposition of Atmospheric Radionuclides
6.3.1 Introduction
Atmospheric deposition includes a complex set of processes that
result in the transfer of radionuclides from the plume to the ground
surface and vegetation. Processes are categorized as dry when they
result in the direct transfer from the plume to the surfaces in contact
with it and wet when the transfer is first from the plume to precipita-
tion and then from the precipitation to the ground or vegetation
surfaces,
6.3.2 Dry Deposition Model
Dry deposition models generally relate the surface deposition flux
to the air concentration at some reference height, typically 1 meter
above the ground. The resulting equation is
¥ = vd X0 (6-3)
where W is the deposition flux to the surface (Ci/m^s), Xo is the
reference height air concentration (Ci/m^), and v,j is the deposition
velocity (m/s). Although v
-------
from the plume by an element of precipitation is presumed to remain with
detour*? " element "ntU reacMn8 the «r°U°d 8urfa«» ^e
deposition flux is proportional to the total wetted activity in a
IxpressedT *" "* ^ ^ ^^^ ™*
W • *sc X L
(6_4)
where W is the surface flux (Ci/m2.), X is the average wetted air
concentration ci/»3>, L £ the depth of thfi wefcted f f gj , and Xflc
is the scavenging rate (.-1). X8C i. a variable that lumps together the
complex xnteractions between precipitation and the plume. Bee Le the
HoTf, T ^ ^ P™P°r^?nal to «« vertically integrated concentra-
tion U.e., the total activity in a column of unit ground surface area),
it » independent of the effective height of the release. Raising the
effective height of a release lowers the dry deposition flux but leaves
the tlux resulting from precipitation scavenging unchanged.
6-3.4 Soil Concentration Model
The deposited radionuclides accumulate in the surface soil until
they are removed either by radiological decay or by processes such as
leacning. The areal concentration can be expressed as
_ „ l-exp(-XB cb)
3 "
where Ca is the areal concentration (Ci/nO, W is the radionuclide flux
to the ground surface Ci/m2s), tb(.) is the time for radionuclide
buildup in soils, and XB is the effective removal rate from soil (8-l)
When the deposited radionuclide is the parent of other radionuclides,
their soil concentrations at time tb due to ingrowth from the parent
must also be calculated.
For calculating root transfer to crops, the radionculide concentra-
tion in the surface soil layer can be expressed as
(6-6)
where C? is the soil concentration (Ci/kg) and P is the areal density of
dry soil (kg/m^) for the plowed or mixed soil layer.
The value of tfci the deposition accumulation time, is typically in
the range of 20 to 100 years. For nearby individual assessments, tu is
chosen to correspond to the expected operational life time of the
facility. If EPA considers it likely that the facility would be
6-10
-------
replaced by another similar one at that time, then tj, is increased
accordingly up to a, maximum value of 100 years. Of course, only those
environmental concentrations which depend on soil deposition are
affected by the choice of tD. For collective (population) assessments,
a value of 100 years is used for t^. This value corresponds to
establishing a 100-year cutoff for the time following a release when any
significant intake or external exposure associated with deposition on
soil might take place. Since radionuclide inhalation is generally the
dominant risk pathway, total risk is not sensitive to the choice of tD.
The value of Ag is the sum of the radiological decay constant, A,
and an environmental removal rate for deposited radionuclideo from soil,
Xs. Hoffman and Baes (Hob79) considered a simplified leaching-loss
model appropriate to agricultural soil for calculating Xs. Their range
of values for the parameter KJJ (the equilibrium distribution coefficient
relating the ratio of the radionuclide concentration in soil water to
that on soil particles) for Cs is from 36,5 to 30,000 ml/g. The
corresponding ratio of Xs is 820:1. The uncertainty in Xs is also
significantly affected by the uncertainty in the other parameters as
well. Although their model is a reasonable one, adequate studies for
its validation do not exist. Since the choice of appropriate values for
Xs is so uncertain, EPA has used 0.2 y~"l as a general nominal value (the
geometric mean of Xs for Pu*, I~, Cs4, and Sr^* ions is 1.2 10~2 y-1
using Hoffman and Baes median data values) and a value of 0.1 y~l for
urban settings where strong surface runoff would be expected to increase
the effective removal rate.
6.3.5 Uncertainties
Uncertainties in v
-------
Concentration in Vegetation
can
£^ Tv (l-exp(-XEte)
Y
W «• v
V
, d
wnere Cy is the crop concentration (Ci/kg) at harvest, W is the
deposition flux (Ci/m28), fr is the fraction of the deposition flux
winch the vegetation intercepts, Yv is the vegetation yield (kg/m*), Tv
J«f.f^tra?*i°C*j"n ff?tor» XE is the effective removal rate of the
intercepted radionuclide from the vegetation (s'l), and t. is the
r^nfrf tMe °f the veeetation to the radionuclide flux (s). Miller
IM1/9J has observed thai- Aat-a fnr- f i « •, -,
expression r Snd YV are Wel1 «Pre"nted by the
f = l-exp(-tY ) (6-8)
r v
where Y was found to range between 2.3 and 3.3 m2/kg when Y £
expressed in kg/m2, dry, since the product ^ -s generall^ legg
l.O, for many practical purposes (6-8) can be approximated as
f = YY (6-9)
r v
In this case the quantity fr/Yy (6-7) can be replaced by Y which shows
much less environmental variation than fr and Yv do separately. Note
that Yv is the total vegetative yield which can be several times the
edible portion yield for a crop. Tv, the translation factor, relates
the radionuclide concentration in the edible portion to that in the
entire plant. Baker et al. (Baa76) suggest a value of 1.0 for leafy
vegetables and fresh forage, and 0.1 for all other produce. (A value of
1.0 is used for all crops in AIRDOS-EPA.)
The value for XE is the sum of X, the radionuclide decay constant
and A the weathering rate factor. For a typical weathering half-life
of 14 days, X has € value of 5.7 KT? a~l. In general, the product
AE te >1 and (6-9) can be simplified to
6-12
-------
Radionuclides also transfer directly from the soil to vegetation
through the plant's root system. The plant concentration due to this
process can be calculated as
C® - Cs B£V (6-11)
where C is the plant concentration at harvest (Ci/kg), Cs is the soil
concentration (Ci/kg) and B£v *s c^e element specific soil to plant
transfer factor. The total concentration from both processes is
Cv - C* * Cv (6
Generally, the contribution of C to Cv is greater than that of CS for
atmospherically dispersed radionuclides.
6.4.3 Concentration in Meat and Milk
For a concentration Cv (Ci/kg) in animal feed, the concentration in
meat Cf (Ci/kg) can be calculated as
cf - Qf Ff Cv (6-13)
where Qf is the animal's feed fonsumption (kg/d) and Ff is the feed to
meat transfer factor (d/kg). if is element dependent and represents the
avsrage mean concentration at slaughter for a unit ingestion rate over
the animal's lifetime. Most systematic studies of Ff have been made for
cattle or other ruminants, although a few measurements for other species
also exist (NCRP84) . In practice, even the Ff values for beef are often
based on colateral data (Bab84).
Similarly for milk, the concentration Cju (Ci/1) can be calculated
as
Cm - Qf % Cv (6-14)
where Wm (d/L) is the equilibrium transfer factor to milk and the other
parameters are as for (6-13). Although more statistical data are avail-
able for F0 than Ff, the authors of NCRP84 note that the estimation of
transfer coefficients to animal products is a subject needing both
integration and better documentation.
6-13
-------
6.4.4 Summary
Radionuclide intake through the food chain depends upon both the
concentration in food and human usage. The concentration in food
depends upon the food source; use of foods grown in proximity to the
release location, the fraction of an individual's food that is home
produced, and other factors can strongly influence the significance of
the food pathway. Unfortunately, generally useful validation studies to
quantify the substantial uncertainties in the food chain have not been
made. References such as NCRP84, Ti83, Mi82, and Li79 cite ranges for
some parameters and make limited model uncertainty estimates but do not
make quantitative evaluations of the uncertainties for the ingestion
pathway taken as a whole.
EPA has chosen a factor of 10 as a reasonable upper bound for the
uncertainty in both the deposition rate model and the calculated intake
from eating food containing deposited radionuclides. Assuming that the
two factors are independent, uncorrelated, and correspond to the 2 sigma
values for a log normal distribution, the combined uncertainty for the
pathway (deposition and intake of radionuclides from food) is a factor
of 26.* EPA has rounded this value to 30 as an estimate of the overall
food pathway uncertainty factor.
It is useful to put this uncertainty in context, accepting the
premise that the ingestion pathway estimates should be considered
reasonable even if their uncertainty does not admit precise
quantification. Table 6.4-1 of folume II of the BID shows that for two
elemental phosphorus plants, the portion of the risk due to the
ingestion pathway was 0,7 percent for one plant and 0.5 percent for the
other. Even a factor of 30 increase in the ingestion pathway risk would
not make it a significant fraction of the total risk from all pathways.
Fortunately, the food pathway has not proved to be a significant part in
assessing the total health risks of radionuclides in air and hence the
large uncertainties associated with the food pathway do not limit the
overall uncertainty.
6.5 Calculating the Environmental Concentration of Radionuelidesj
The AIRDOS-EPA Code
6.5.1 Introduction
Environmental concentrations of radionuclides calculated by EPA may
be site specific, meaning that available data relevant for the site are
incorporated into the assessment. Or an assessment may be generic, that
is, an assessment of a hypothetical facility at a location considered an
appropriate possibility for such a facility class. Frequently, EPA
performs site-specific assessments for existing facilities, e.g., a
national laboratory. In addition, EPA often employs generic assessments
*exp[2 In2 (10)]1/2 - 26
6-14
-------
in evaluating alternative sitings for a proposed facility or assessing a
widespread class of facilities, e.g., industrial coal-burning boilers.
In any case, EPA makes both individual and collective (population)
assessments. The purpose of the individual assessment is to assess the
doses and lifetime risk to individuals living near a facility. EPA's
assumption is that these individuals reside a substantial portion of
their lives at the same location and that their exposures extend from
infancy on through adulthood. The doses and risks calculated are expec-
tation values, i.e., the estimates are intended to be typical for a
person living a long period of time under the assessed conditions.
EPA's collective (or population) assessments evaluate doses and risks
to a population that may be regional (typically up to 80 km distant),
long-range (e.g., the conterminous United States), or worldwide as
appropriate. The risk is usually expressed as the expected number of
premature deaths in the population per year of facility operation,
6.5.2 AIRDOS-EPA
EPA has used the A1RDOS-EPA code (Mo79) to calculate environmental
concentrations resulting from radionuclide emissions into air. The
results of this analysis are estimates of air and ground surface
radionuclide concentrations; intake rates via inhalation of air;
ingestion of radioactivity via meat, milk, and fresh vegetables. The
atmospheric and terrestrial transport models used in the code, their
implementation, and the applicabili.1- • of the code to different types of
emissions are described in detail in Mo79.
AIRDQS-EPA calculates atmospheric dispersion for radionuclides
released from one to six stacks or area sources. Radionuclide con-
centrations in meaf, milk, and fresh produce are estimated by coupling
the deposition rate output of the atmospheric dispersion models with
the Regulatory Guide 1,109 (NRC77) terrestrial food chain models.
Radionuciide concentrations for specified distances and directions are
calculated for the following exposure pathways; (1) iinmersion in air
containing radionuclides, (2) exposure -.o ground surfaces contaminated
by deposited radionuclides, (3) inhalation of radionuclides in air, and
(4) ingestion of food in the area. The code may be used to calculate
either annual individual exposures or annual population exposures at
each grid location. For either option, AIRDOS-EPA output tables
summarize air concentrations and surface deposition rates as well as the
intakes and exposures for each location. In addition, working level
exposures are calculated and tabulated for evaluating the inhalation of
short-lived progeny of radon-222.
Assessment Grid
AIRDOS-EPA has provision for either a rectangular or a circular
calculattonal grid. The customarily used circular grid (see Figure
6.5-1) has 16 directions proceeding counterclockwise from north to
north-northeast. The user chooses the grid distances. Generally,
successive distances are chosen with increasing spacing. It is
6-15
-------
NKW
WNW
NW
X - Assesament grid locations at up to 20 Stances
12 shown) and 16 directions (5 shown)
Figure 6.5-1. Circular grid system used by AIRDOS-EPA.
6-16
-------
important to realize that the calculational grid distances and the set
of distances associated with population and food production data are one
and the same. Hence, the concentration calculated for each grid
distance must be the appropriate average value for the corresponding
range of distances covered by the population and agricultural data.
Choosing a suitable set of grid distances may require different com-
promises of convenience for different assessments and may be different
for individual and collective assessments of the same facility.
Environmental Accumulation Time
An AIRDOS-EPA assessment is based on what can be viewed as a
snapshot of environmental concentrations after the assessed facility has
been operating for some period of time. The choice of an environmental
accumulation time affects only those pathways dependent on terrestrial
concentrations, i.e., ground surface exposure and food intakes.
Usually, the accumulation time for an individual assessment is chosen to
be consistent with the expected life of the facility (or 100 years when
a similar facility might be expected to replace the present one at the
end of its useful life), For collective assessments, 100 years is
customarily used.
Source Considerations
Point sources are characterized by their physical height and, when
desired, the parameters to calculate buoyant or momentum plume rise
using Brigg's (Br69) or Rupp's (Ru48) formulations respectively.
Alternatively, a fixed plume rise may be specified for each Pasquill-
Gifford atmospheric stability class A through G.
The area source model is similar to that of Culkowski and Patterson
(Cu76) and transforms the original source into an annular segment with
the same area. At large distances, the transformed source approaches a
point source at the origin, while at distances close to the origin it
approaches a circle with the receptor at its center.
Building wake effects and downwash are not included in the AIRDOS-
EPA models. The same type of rise calculation (buoyant, momentum, or
fixed) is used for all sources. As many as six sources may be assessed,
but for calculational purposes they are all considered to be co-located
at the origin of the assessment grid.
Radionuclide Releases
Releases for up to 36 radionuclides may be specified for AIRDOS-
EPA. Each release is characterized by the radionuclide name, effective
decay constant during dispersion, precipitation scavenging coefficient,
deposition velocity, and settling velocity as well as the annual
activity release for each source. Decay products that are significant
for the assessment of a radionuclide must be included in the list of
releases. There is no explicit method for calculating radionuclide
ingrowth during atmospheric dispersion in AIRBQS-EPA,
6-17
-------
Parameters such as particle size, respiratory clearance class, and
gastrointestinal absorption factor (fj) are passed on for use in the
DARTAB (Be81) dose and risk assessments as described in the Appendices
to Chapters 7 and 8.
The approach ORP has used for calculating a precipitation
scavenging coefficient is based on Slinn's (S177) equation (32):
J° E
(6-7)
«m
where Xsc is the scavenging coefficient, c is a constant (Slinn uses
0.5), Jo is the rainfall rate, and E is the collection efficiency for a
particle of radius a by drops of characteristic radius H^' Slinn (S177,
p. 23) considers the effects of dry deposition and interprets Dana and
Wolf's (Daa68, Wo69, Dab70) data as supporting a value for E of 0.2,
essentially independent of particle size. Adopting Slinn's typical
value of R0J for a frontal rain (0.3 mm) and selecting a long-term
average value of 1000 nnn/yr (3, 16x10"^ mm/s) for Jo, we obtain;
0.5 3.16x10-5 0.2 ,, ,
o (6"*
= 1.05x10-5 s-l
This value has been rounded to 10~5 s 1 as a working value for the
precipitation scavenging coefficient and then scaled according to the
annual precipitation at the assessment location for use in AI1DOS-EPA.
There is substantial uncertainty in interpreting environmental scav-
enging data; this estimate is clearly an order of magnitude one. The
EPA scaling procedure reflects the premise that the variation of rain-
fall from one location to another is more one of rain frequency than of
intensity during rainfall episodes.
Dispersion
Wind and stability class frequencies for each direction are the
primary data for calculating atmospheric dispersion. The required data
for AIRDOS-EPA are calculated from a joint frequency distribution of
wind speed and atmospheric stability class for each direction.
Inasmuch as the assessments require long-term average dispersion
values, the sectoi—averaged Gaussian plume option is used. The vertical
dispersion parameter (0Z) is calculated using Brigg's formulas (Gi76).
Vertical dispersion is lioited to the region between the ground and a
6-18
-------
mixing depth lid. The harmonic mean of Holzworth's (Hoa72) morning and
afternoon mixing depths is customarily employed for this value, that is,
' 2
where £a and JL, are respectively the morning and afternoon mixing depths
and hn is their harmonic mean. At large distances, the concentration is
uniform between the ground and the lid.
Deposition Rate
AIRD0S-EPA models both dry and wet deposition processes. Resuspen-
sion, the reintroduction of deposited material into the atmosphere, is
not modeled in AIRDOS-fiPA. The dry deposition rate is the product of
the deposition velocity and the near ground level air concentration
while the wet deposition rate is the product of the precipitation
scavenging coefficient and the vertically integrated air concentration.
Wet deposition decreases monotonically with distance and is independent
of the effective release height of the source, while the effect of
source height can be significant for dry deposition. For locations
close to an elevated source, wet deposition can provide the principal
source of radionuclide exposure. Concentrations are adjusted for
depletion due to deposition at each downwind distance.
Ground Surface Concentration
AIRDQS-EPA calculates the ground surface concentration from the
total (dry plus wet) deposition rate. The soil concentration is cal-
culated by dividing this value by the effective agricultural soil
surface density (kg/m^). Both concentrations are calculated for the end
of the environmental accumulation time t]j and can include the ingrowth
from deposited parent radionuclides as well as removal due to radiolog-
ical decay and environmental processes such as leaching.
Ingrowth from a parent radionuclide is calculated using a decay
product ingrowth factor. The ingrowth factor is the equivalent deposi-
tion rate for a unit deposition rate of the parent radionuclide. For
example, the ingrowth factor for lead-210 as a parent of polonium-210
would be calculated by determining the concentration of polonium-210 at
time t^ due to a unit deposition rate of lead-210 and dividing it by the
corresponding concentration for a unit deposition rate of polonium-210.
These ingrowth factors must be calculated in advance of running AIRDOS-
EPA and are dependent on both the accumulation time t{j and the soil
removal constants for the nuclides in the radionuclide chain (lead-210,
biBouth-210, and polonium-210 in this case).
6-19
-------
Concentrations in Food
Radionuclide concentrations in food are calculated using
essentially the same model as in NEC Regulatory Guide 1.109 (NRC77).
Changes from that model include consideration of environmental removal
from the root zone, and separate values for food and pasture crops of
the interception fraction, areal yield, and soil-to-plant transfer
values. Concentration calculations for meat and milk use the same
models as the Regulatory Guide model.
There are numerous parameters in the terrestrial pathways model.
Appendix A of Volume II of the BID contains tables of values used in
these assessments.
PopulationandAgricultural Data
For a collective (population) assessment, population and agricul-
tural data for each grid location must be provided. EPA uses the 1970
census enumeration district data to calculate population distributions.
AIRBQSHEPA calculates the collective assessment for agricultural
products based on consumption by the assessment area population. The
assessment can be based on agricultural production by choosing utili-
zation factors large enough to ensure that all items produced are
consumed.
Food Utilization Factors
In addition to the consumption rate for different food categories
(leafy vegetables, other produce, meat, and milk), the user may specify
the fraction of vegetables, meat, and milk that are (1) home grown,
(2) produced in the assessment area, or (3) imported from outside the
assessment area. Those in the third category are considered to contain
no radionuclides. Those from the second category have the average
concentration for that category produced within the assessment area,
while concentrations for the first category are those that would occur
at each grid location. Appendix A of Volume II of the BID provides some
typical food source fractions for urban and rural assessment areas.
Note that if the assessment considers food to be only home grown or
imported from outside the assessment area, then the actual quantity of
food produced at each location is not relevant to the assessment.
Experience has shown that the ingestion doses and risks for the nearby
individual are usually dominated by the radionuclide intake from home
grown food and hence there is generally no significant difference
between assuming that food that is not home grown is obtained from the
assessment area or is imported from outside the assessment area.
Spec i a1 Rad i onuc1i des
Special consideration is given to the radionuclides hydrogen-3
(tritium), carbon-14, and radon-222. The specific activity of tritium
in air (pCi/g of H20) is calculated for an absolute humidity of 8 mg/nH
(MC77). Etnier (EtSQ) has calculated average absolute humidities for
6-20
-------
over 200 U.S. locations. The 8 mg/ni3 value would be within a factor of
2 for most of them. The specific activity of atmospheric carbon-14
(pCi/g of C) is calculated for a C02 concentration of 330 ppm by volume
(Ki78). Concentrations of these nuclides in vegetation are calculated
on the assumption that the water and carbon content in vegetation are
from the atmosphere and have the same specific activity as in the
atmosphere. The radon-222 concentration in air is replaced by its
short-lived decay product concentration in working level units using a
fixed equilibrium fraction (typically 0.7 for calculating population
health risks).
6-21
-------
REFERENCES
Baa76 Baker D. A., Hoenes G. R., and Soldat J. K., FOOD - An
interactive code to calculate Internal radiation doses from
contaminated food products, in Proceedings of the Conference on
Environmental Modeling and Simulation, Ott W. R. editor, EPA
600/9-76-016, p. 204, Office of Research Development and Office
of Planning and Management, U.S. Environmental Protection
Agency, Washington, B.C. 20460, July 1976.
Bab84 Baes C. F. Ill, Sharp R. D., Sjoreen A. L., and Shor R. W,, A
Review and Analysis of Parameters for Assessing Transport of
Environmentally Released Radionuclides through Agriculture,
ORNL-5786, Oak Ridge National Laboratory, Oak Ridge, Tenn.,
September 1984.
Be81 Begovich C. L., Eckennan K. F., Schlatter E. C., Ohr S. Y., and
Chester R. 0., DARTAB: A program to combine airborne radio-
nuclide environmental exposure data with dosimetric and health
effects data to generate tabulation of predicted impacts,
ORNL/5692, Oak Ridge National Laboratory, Oak Ridge, Tenn.,
August 1981.
Br69 Briggs G. A., Plume Rise, TID-25075, U.S. Atomic Energy
Commission Critical Review Series, National Technical
Information Service, Springfield, Va., November 1969.
Cu76 Culkowski W. M. and Patterson M. R., A Comprehensive
Atmospheric Transport and Diffusion Model, ORNL/NSF/EATC-17,
National Oceanic and Atmospheric Administration, Atmospheric
Turbulence and Diffusion Laboratory, Oak Ridge, Tenn., 1976.
Daa68 Dana M. T. and Wolf M. A., Experimental Studies in
Precipitation Scavenging, in Pacific Northwest Laboratory
Annual Report for 1967 to the USAEC Division of Biology and
Medicine, Vol. II, Physical Sciences, Part 3, Atmospheric
Sciences, Simpson C.L. et al., USAEC Report BNWL-715-3,
pp. 128-140, Battelle Pacific Northwest Labortories, Richland,
Wa., October 1968.
6-22
-------
Dab70 Dana M. T., Wolf M. A., and duPlessis L. A., Field Experiments
in Precipitation Scavenging, in Pacific Northwest Laboratory
Annual Report for 1969 to the USAEC Division of Biology and
Medicine, Vol II, Physical Sciences, Part 1, Atmospheric
Sciences, Simpson G.L. et al., USAEC Report BNWL-1307 (Pt. 1),
pp. 77-81, Battelle Pacific Northwest Laboratories, June 1970.
Et8Q Etnier E. L., Regional and site-specific absolute humidity data
for use in tritium dose calculations, Health Phys. 39, 318-320,
1980.
Gi76 Gifford F. S. Jr., Turbulent Diffusion-Typing Schemes: A
Review, Nucl. Saf. _17_(1), 68-86, 1976.
Ha82 Hanna S. R., Briggs G. A., and Hosker R. P. Jr., Handbook on
Atmospheric Diffusion, DOE/TIC-11223, Technical Information
Center, U.S. Department of Energy, Washington, B.C., January
1982.
Hoa72 Holzworth G. C., Mixing Heights, Wind Speeds and Potential for
Urban Air Pollution Throughout the Contiguous United States,
Publication No. AP-101, U.S. Environmental Protection Agency,
Office of Air Programs, Research Triangle Park, N.C., 1972.
Hob79 Hoffman F. 0. and Baes C. F. Ill, A Statistical Analysis of
Selected Parameters for Predicting Food Chain Transport and
Internal Dose of Radionuclides, NUREG/CR-1004, Oak Ridge
National Laboratory, Oak Ridge, Tenn., 1979.
Ki78 Killough G. C. and Rohwer P. S., A new look at the dosimetry of
l^C released into the atmosphere as carbon dioxide, Health
Phys., 34, 141-159, 1978.
Li79 Little C. A. and Miller C. W., The Uncertainty Associated with
Selected Environmental Transport Models, ORNL-5528, Oak Ridge
National Laboratory, Oak Ridge Tenn., November 1979.
Mi82 Miller C. W. and Little C. A., A Review of Uncertainty
Estimates Associated with Models for Assessing the Impact of
Breeder Radioactivity Releases, ORNL-5832, Oak Ridge National
Laboratory, Oak Ridge, Tenn., August 1982.
Mo79 Moore R. E., Baes C. F. Ill, McDowell-Boyer L. M,, Watson
A. P., Hoffman F. 0., Pleasant J. C., and Miller C. W., AIRDOS-
EPA; A Computerized Methodology for Estimating Environmental
Concentrations and Dose to Man from Airborne Releases of
Radionuclides, EPA 520/1-79-009 (reprint of ORNL-5532), U.S.
Environmental Protection Agency, Office of Radiation Programs,
Washington, D.C., December 1979.
6-23
-------
NCRP84 National Council on Radiation Protection and Measurements,
Radiological Assessment: Predicting the Transport,
Bioaccumulation, and Uptake by Man of Radionuclides Released to
the Environment, NCRP Report No. 76, National Council on
Radiation Protection and Measurement, Bethesda, Md., March,
1984.
NRC77 U.S. Nuclear Regulatory Commission, Calculation of Annual Doses
to Man from Routine Releases of Reactor Effluents for the
Purpose of Evaluating Compliance with 10 CFR Part 50 Appendix I
(Revision 1), Regulatory Guide 1.109, Office of Standards
Development, Washington, B.C., October 1977.
Ru48 Rupp A. F,, Beall S. E., Bornwasser L, P., and Johnson D, H.,
Dilution of Stack Gases in Cross Winds, USA1C Report AECD-1811
(CE-1620), Clinton Laboratories, 1948.
S177 Siinn W.G.N., Precipitation Scavenging; Some Problems,
Approximate Solution, and Suggestions for Future Research, in
Precipitation Scavenging (1974), CONF-741003, Technical
Information Center, Energy Research and Development
Administration, Washington, B.C., June 1977.
Ti83 Till J. E. and Meyer H. R.» Radiological Assessment, NUREG/CR-
3332, ORNL-5968, Division of Systems Integration, Office of
Nuclear Reactor Regulation, U.S. Nuclear Regulatory Commission,
Washington, D.C., September 1983.
Wo69 Wolf M. A. and Dana M, T., Experimental Studies in
Precipitation Scavenging, in Pacific Northwest Laboratory
Annual Report for 1968 to the USAEC Division of Biology and
Medicine, Vol II, Physical Sciences, Part 1, Atmospheric
Sciences, Simpson C.L. et al.t USAEC Report BNWL-1051 (Pt. 1),
pp. 18-25, Battelle Pacific Northwest Laboratories, November
1969.
6-24
-------
Chapter 7: RADIATION DOSIMETRY
7.1 Introduction
Sadionuclides transported through the environment may eventually
reach people. This contact occurs through either external exposure to
radioactive air, water, and ground surfaces or internal exposure from
inhaling or ingesting radioactive air, water, or food. Individuals in
the population may absorb energy emitted by the decaying radionuclides.
The quantification of this absorbed energy is dosimetry. This chapter
describes the dositnetric models for internal and external exposures, the
EPA procedure for implementing the dosimetric equations associated with
the models, and the uncertainties in dositnetric calculations.
Mathematical models are used to calculate doses to specific human
body organs. The models account for the amount of radionuclides
entering the body, the movement of radionuclides through the body, and
the energy deposited in organs or tissues resulting from irradiation by
the radionuclides that reach the tissue. These models provide the basis
for the computer codes, RADRISK and DARTAB, which EPA uses to calculate
doses and dose rates, (See Addendum A.)
Uncertainties in dosimetric calculations arise from assumptions of
uniform distribution of activity in external sources and source organs
and assumptions concerning the movement of the radionuclides in the
body. The uncertainties associated with dosimetric calculations are
difficult to quantify because the data available for determining distri-
bution for the parameters used in the models are usually insufficient.
The major source of uncertainty in dosimetry is the real variation in
parameter values among individuals in the general population while doses
and dose rates are calculated for a "typical" member of the general pop-
ulation. The three sources of dosimetric uncertainty assessed by EPA
are; individual variation, age, and measurement errors. The effects of
uncertainty analysis on the dose estimates for the general population
are discussed in greater detail in Section 7.6.
7.2 Definitions
7.2.1 Activity
Radioactive decay is a process whereby the nucleus of an atom emits
excess energy. The emission of this energy is referred to as radio-
activity. The "activity" of a radioactive material is characterized by
the number of atoms that emit energy, or disintegrate, in a given period
7-1
-------
of time. The unit of activity used in this report is the picocurie
(pCi), which equals 2.22 disintegrations per minute. The excess energy
is normally emitted as charged particles moving at high velocities and
photons. Although there are many types of emitted radiations, or
particles, only three are commonly encountered in radioactive material
found in the general environments alpha radiation (nuclei of helium
atoms), beta radiation (electrons), and gamma radiation (photons).
The primary mechanism for radiation damage is the transfer of
kinetic energy from the moving alpha and beta particles and photons to
living tissue. This transfer leads to the rupture of cellular constitu-
ents resulting in electrically charged fragments (ionization). Although
the amount of energy transferred is small in absolute terms, it is
enough to disrupt the molecular structure of living tissue, and,
depending on the amount and location of the energy release, lead to the
risk of radiation damage.
7*2.2 Exposure and Dose
The term "exposure" denotes physical contact with the radioactive
material. The term "dose" refers to the amount of energy absorbed per
gram of absorbing tissue as a result of the exposure. An exposure, for
example, may be acute, i.e., occur over a short period of time, while
the dose, for some internally deposited materials, may extend over a
long period of time.
The dose is a measure of the amount of energy deposited by the
alpha and beta particles or photons and their secondary radiations in
the organ. The only units of dose used in this chapter are the rad—
defined as 100 erg (energy units) per gram (mass unit)—and the millirad
(mrad), which is one one-thousandth of a rad. The rad represents the
amount, on average, of potentially disruptive energy transferred by
ionizing radiation to each gram of tissue. Because it is necessary to
know the yearly variation in dose for the calculations described in this
report, the quantity used will be the average annual dose (or dose rate)
in rad or millirad (per year).
7.2.3 External and Internal Exposures
Radiation doses may be caused by either external or internal expo-
sures. External exposures are those caused by radioactive materials
located outside the body, such as irradiation of the body by radioactive
material lying on the ground or suspended in the air. Internal
exposures are caused by radioactive material that has entered the body
through the-inhalation or consumption of radioactive material. Having
on^.e entered the body, the contaminant may be transmitted to other
? .ernal organs and tissues.
The external exposures considered in this report are those
resulting from irradiation of the body by gamma rays only. Gamma rays
(high energy photons) are the most penetrating of those radiations con-
sidered and external gammas may normally contribute to the radiation
7-2
-------
dose affecting all organs in the body. Beta particles (electrons),
which are far less penetrating, normally deliver their dose to, or
slightly below, the unshielded surface of the skin and are not con-
sidered because their impact is small, particularly oa clothed
individuals. Alpha particles (helium nuclei), which are of major
importance internally, will not penetrate unbroken skin and so are also
excluded from the external dose calculations. The internal exposures
considered in this report originate from all three types of radiation,
7.2.4 Dose Equivalent
Different types of charged particles differ in the rate at which
their energy is transferred per unit of length traveled in tissue, a
parameter called the linear energy transfer (LET) of the particle. Beta
particles generally have a much lower LET than alpha particles. Alpha
particles are more damaging biologically, per rad, than gamma rays and
beta particles. In radiation protection, this difference is accounted
for by multiplying the absorbed dose by a modifying factor, Q, the
quality factor, to obtain a dose equivalent. The quality factor is
intended to correct for the difference in LET of the various particles.
At present, the International Commission on Radiological Protection
(ICRP77) recommends the values Q=l for gamma rays and beta particles and
Q-20 for alpha particles. The units for the dose equivalent, corre-
sponding to the rad and millirad, are rem and millirem. Thus, dose
equivalents for gamma rays and beta particles are numerically equal to
the dose since the dose equi-alent (mrem) = (Q=l) x dose (mrad) while
alpha dose equivalents are twenty times as large, dose equivalent (mrem)
= (Q=20) x dose (mrad).
7.3 DOSimetrie Models;
The radiation dose has been defined, in 7.2.2, as the amount of
energy absorbed per unit mass of tissue. Calculation of the dose
requires the use of mathematical models such as that shown later in
equation 7-2. In this equation, the amount of activity ingested, I, is
multiplied by the fraction, flt going to the blood, and the fraction,
f2, going to a specific tissue. E is the amount of energy absorbed by
the tissue for each unit of activity so that the product of all these
factors divided by the mass of the tissue is, by definition, the radia-
tion dose. The remaining term, [l-e'^j/x, indicates how the activity
deposited in the tissue changes with time. All these factors together
yield the dose rate, A more comprehensive description of the equations
used is given in Addendum A.
7.3,1 Internal Doses
Any effort at calculating dose and risk must, of necessity, involve
the use of models. In its simplest form, a model is a mathematical
representation of a physical or biological system. If, for example, the
amount of radioactive material in an organ is measured at several times
a graph of the activity in the organ, such as that in Figure 7.3-1, is
7-3
-------
o
03
O
ss
M
M
M
l-l
O
TIME
Figure 7.3-1.
Typical pattern of decline of activity of a
radionuclide in an organ, assuming an initial
activity in the organ and no additional uptake
of radionuclide by the organ (ORNL81).
7-4
-------
obtained. In the simplest case, analysis of these data may indicate
that the fraction of the initial activity, R, retained in the organ at
any time, t, is given by an equation of the form
R = e~*t (7-1)
where X is the elimination rate constant. (More generally, it may
require the sum of two or more exponential functions to properly approx-
imate the decrease of radioactivity in the organ. This may be interpre-
ted physically as indicating the existence of two or more "compartments"
in the organ from which the nuclide leaves at different rates,)
The elimination rate constant, X, is the sum of two terms, which
may be measured experimentally, one inversely proportional to the
biological clearance half-life and the other inversely proportional to
the radioactive half-life. The effective half-life, f-i/2» f°r these
processes is the time required for one-half of the material originally
present to be removed.
If radionuclides are generally found to follow this behavior, then
this equation may be used as a general ittodel for the activity in an
organ following deposition of any initial activity. In general, the
models used by EPA are those recommended by the International Commission
on Radiological Protection (ICRP79) and are documented in detail in the
cited reference. A brief description of each model is given below as an
aid to understanding the material piesented in the balance of this
chapter,
As mentioned earlier, all radiations—gamma, beta, and alpha—are
considered in assessing the doses resulting from internal exposure, that
is, exposure resulting from the inhalation or ingestion of contaminated
material. Portions of the material inhaled or ingested may not leave
the body for a considerable period of time (up to decades); therefore,
dose rates are calculated over a corresponding time interval.
The calculation of internal doses requires the use of several
models. The most important are the ICRP lung model, depicted in
Figure 7.3-2, and the gastrointestinal (GI) tract model shown in Figure
7.3-3. The lung model is comprised of three regions, the nasopharyngial
(N-P), the tracheobronchial (T-B), and the pulmonary (P) regions. A
certain portion of the radioactive material inhaled is deposited in each
of the three lung regions (N-P, T-B, and P) indicated in Figure 7.3-2.
The material is then cleared (removed) from the lung to the blood and
gastrointestinal tract, as indicated by the arrows, according to the
specified clearance parameters for the clearance class of the inhaled
material.
Deposition and clearance of inhaled materials in the lung are
controlled by the particle size and clearance class of the material.
The particle size distribution of the airborne material is specified by
7-5
-------
COMPARTMENT
N-P a
ID3 » 0.30) b
T-B c
(D4 = 0.08} d
e
P f
fd coefficie«. respectively, of a term S the appropriate re tint ion
The values shorn for D,, DA, and D* corresoond to ™
-------
INGESTION
RESPIRATORY
TRACT
I
B
L
0
0
D
*
Xab
AUL!
\ab
ALLI
SI
f Xj
ULI
\ *u
LLI
S|=6day~1
UL| - t.85day
-1
-1
Figure 7.3-3. Schematic representation of radioactivity
Movement among respiratory tract,
gastrointestinal tract, and blood.
S
SI
ULI
LLI
X
= stomach
= small intestine
= upper large intestine
= lower large intestine
= elimination rate constant
7-7
-------
giving its Activity Median Aerodynamic Diameter (AMAD) in microns (one
micron equals 10~6 meters). Where no AMAD is known, a value of 1.0
micron is assumed. Clearance classes are stated in terms of the time
required for the material to leave the lung, that is, Class D (days),
Class W (weeks), and Class Y (years).
The gastrointestinal tract model consists of four compartments, the
stomach (S), small intestine (SI), upper large intestine (ULI), and
lower large intestine (LLI). However, it is only from the small
intestine (SI) that absorption into the blood is considered to occur.
The fraction of material that is transferred into blood is denoted by
the symbol t\.
Radionuclides may be absorbed by the blood from either the lungs or
the GI tract. After absorption by the blood, the radionuclide is
distributed among body organs according to fractional uptake coeffi-
cients, denoted by the symbol f2- Since the radioactive material may be
transported through the body, dose rates are calculated for each organ
or tissue affected by using a model of the organ that mathematically
simulates the biological processes involved. The general form of the
model for each organ is relatively simple. It postulates that the
radioactive material which enters the organ is removed by both radio-
active decay and biological removal processes,
7*3.2 Ext erna1 Doses
The example just described for modeling the activity of a radio-
nuclide in an organ pertains to estimating doses from internal exposure.
In contrast, the external immersion and surface doses are calculated as
follows. First, the number of photons reaching the body is determined.
The model used here is a set of equations governing the travel of
photons (gamma radiation) in air. The simplifying assumptions used in
these calculations are that the medium (air) is an infinite half-space
and is the only material present. This makes the calculation relatively
straightforward. In the second portion of the calculation, the photons
reaching the body are followed through the body using a "Monte Carlo"
method. The "phantoms", i.e., the models of the body, are those used by
the Medical Internal Radiation Dose Committee (MIKD69). The Monte Carlo
method is a procedure in which the known properties of the radiation and
tissues are employed to trace (simulate) the paths of a large number of
photons in the body. The amount of energy released at each interaction
of the radiation with body tissues is recorded and, thus, the dose to
each organ or tissue is estimated by evaluating a large number of photon
paths.
7,3.3 Effects of Decay Products
In calculating doses from internal and external exposures, the
occurrence of radioactive decay products (or daughters) must be con-
sidered for some radionuclides. When an atom undergoes radioactive
decay, the new atom created in the process may also be radioactive and
may contribute to the radiation dose. Although these decay products may
7-8
-------
be treated as independent radionuclides in external exposures, the decay
products of each parent must be followed through the body in internal
exposures. The decay product contributions to the dose rate are
included in the dose calculations, bsssd on the metabolic properties of
the element and the organ in which they occur.
7.3.4 Dose Rate Estimates
For each external and internal exposure, dose rates to each of the
organs listed in Table 7.3-1 are calculated for each radioisotope.
These organ dose rates serve as input to the life table calculations
described in Chapter 8.
Table 7.3-1. Organs for which dose
rates are calculated
Red bone marrow Intestine
Bone Thyroid
Lung Liver
Breast Urinary tract
Stomach
Pancreas
(a'Esophagus, lymphatic system, pharynx,
larynx, salivary gland, brain.
7.4 EPA Dose Calculation
7.4.1 Dose Rates
The models described in Section 7.2 are used by EPA to calculate
radiation dose rates resulting from internal and external exposures to
radioactive materials. A more complete description of the methodology,
equations, and parameters used is given in Du84, ORNL80, and ORNL81.
EPA has adopted two refinements to the ICRP-reconmended protocol for
these calculations. The first is to track the movement of internally
produced radioactive daughters by assuming that their movement is
governed by their own metabolic properties rather than those of the
parent. Although not enough information is available to allow a
rigorously defensible choice, this appears to be more accurate for most
organs and nuclides than the ICRP assumption that daughters behave
exactly as the parent. In the second departure from ICRP recommenda-
tions, age-dependent values of the parameters governing the uptake of
transuranic nuclides have been taken from two sources, deemed appro-
priate to the general population, the National Radiological Protection
Board (MPB82) and the EPA transuranic guidance document (EPA77).
7-9
-------
The internal dose equations given by ICRP may be used to calculate
either radiation doses (rad), i.e., the total dose over a given tine
period, or radiation dose rates (rad/yr), i.e., the way in which the
dose changes with time after intake. The summation of the dose rates
is, of course, the total dose. EPA calculates dose rates rather than
doses, because SPA considers age when assessing the effects of radiation
on the population.
External irradiation does not result in any residual internal
material. Therefore, external dose rates to a given organ are constant.
That is, the dose rate caused by a given amount of radionuclide present
in air or on a ground surface becomes zero when the radionuclide is
removed.
The calculation of dose rates, rather than integrated doses, allows
the use of age-dependent metabolic parameters more appropriate to the
general population to be taken into account. In the vast majority of
cases, however, there is not now sufficient information available to
make such calculations. The major exception to this is exposure to
radon, in which EPA uses age-dependent exposure parameters. Because
most of the data available for radon are in terms of exposure, no doses
are calculated for this gaseous element. Radon assessments are
discussed in detail in Chapter 8. The effect of using age-dependent
metabolic parameters is discussed in Section 7.5.2 for some radio-
nuclides for which sufficient information is available.
7.4.2 Exposure and Usage
The ICRP dosimetric equations used by EPA are linear, i.e., an
intake of 10 picoCuries will result in dose rates ten times as large as
those from an intake of 1 picoCurie. In similar fashion, exposure to 10
times as large an air or ground surface concentration will increase the
external doses by a factor of ten. EPA uses this linearity to avoid
having to calculate radiation dose rates for a range of concentrations.
The standard EPA procedure is to use unit intakes of 1 pCi/yr and air
and ground surface concentrations of 1 pCi/cm^ and 1 pCi/crn^
respectively. The doses for other intakes and concentrations may then
be scaled up or down as required.
In most cases, it is necessary to make certain assumptions
regarding the exposure conditions in order to perform an assessment,
EPA calculates dose rates for lifetime exposure to the unit intakes and
concentrations. Chapter 8 describes the different ways in which these
rates can be applied. In addition, the exposure assessment will usually
depend on other usage conditions assumed for the exposures.
Thus, for the general population, EPA assumes a breath-ing rate,
using ICRP-recoimnended values (ICRP75), based on 8 hours of heavy
activity, 8 hours of light activity, and 8 hours of rest per day. When
required, EPA uses a drinking water intake of 2 liters per day. The
quantities of food ingested are compiled from a variety of sources.
Because there may be insufficient data for some types of food, it may be
7-10
-------
necessary to combine or substitute types in some instances. More
complete details on the values used for the ingestion of foodstuff types
are given in Appendix A of Volume 2.
?»5 Uncertainty Analysis
Uncertainty, in the dose, refers to the manner in which the calcu-
lated dose changes when the parameters used in the calculation (intakes,
metabolic factors, organ sizes, etc.) are changed. The uncertainty
associated with the dosimetric calculations is extremely difficult to
quantify because the term "uncertainty analysis" implies a knowledge of
parameter distributions that is usually lacking. Internal doses, for
example, depend on the parameters used to characterize the physiological
and metabolic properties of an individual, while external doses must
consider parameters such as organ size and geometry for a particular
individual. The data available for most of these parameters is not
sufficient to define the form of the parameter distribution. The major
source of uncertainty in calculating the dose to a distinct individual,
however, in most instances, does not result from errors in measuring the
parameters but from the real variation in parameter values among
individuals in the general population. Thus, a calculated dose is
thought to be representative of a "typical" member of the general popu-
lation and is probably reasonably precise for some large segment of that
population.
The basic physiological and metabolic data used by EPA in calcu-
lating radiation doses are taken from the ICRP Reportof the Task Group
on Reference Man (ICRP75) and from the ICRP Limits forIntakes of
Radionuclides by Workers (ICIP79). The "Reference Man" report is the
most comprehensive compilation of data available on the intake,
metabolism, internal distribution, and retention of radioisotopes in the
human body. Its major purpose, however, is to "define Reference Man, in
the first instance, as a typical occupational individual", although
differences with respect to age and sex are indicated in some instances.
The limitations inherent in defining Reference Man, and in estimat-
ing uncertainties due to variations in individuals in the general popu-
lation, are recognized by the Task Group (ICRP75):
"The Task Group agreed that it was not feasible to
define Reference Man as an "average1 or a 'median1
individual of a specified population group and that it was
not necessary that he be defined in any such precise
statistical sense. The available data certainly do not
represent a random sample of any specified population.
Whether the sample is truly representative of a particular
population group remains largely a matter of judgement
which cannot be supported on the basis of statistical
teats of the data since the sampling procedure is suspect.
Thus the Task Group has not always selected the 'average',
or the "median1, of the available measurements in making
its selection, nor has it attempted to limit the sample to
7-11
-------
some national or regional group and then seek an average
or median value. However, the fact that Reference Man is
not closely related to an existing population is not
believed to be of any great importance. If one did have
Reference Man defined precisely as having for each attri-
bute the median value of a precisely defined age group in
precisely limited locality (e.g., males 18-20 years of age
in Paris, France, on June 1, 1964), these median values
may be expected to change somewhat with time, and in a few
years may no longer be the median values for the specified
population. Moreover, the Reference Man so defined would
not have this relation to any other population group
unless by coincidence. To meet the needs for which
Reference Man is defined, this precise statistical rela-
tionship to a particular population is not necessary.
Only a very- few individuals of any population will have
characteristics which approximate closely those of
Reference Man, however he is defined. The importance of
the Reference Man concept is that his characteristics are
defined rather precisely, and thus if adjustments for
individual differences are to be made, there is a known
basis for the dose estimation procedure and for the esti-
mation of the adjustment factor needed for a specified
type of individual."
With respect to the dosimetric calculations performed by EPA to
assess the impact of radioactive pollutants on a general population,
three sources of uncertainty should be considered;
(1) that due to the variation in individual parameters among
adults in the general population
(2) that due to the variation in individual parameters with age
(3) that due to experimental error In the determination of
specific parameters
Each of these sources of uncertainty is discussed in this section.
As noted above, the data required to perform a rigorous sensitivity
analysis are lacking, and a form of uncertainty analysis called sensi-
tivity analysis is employed. The sensitivity analysis consists of sub-
stituting known ranges in the parameters for the recommended value and
observing the resulting change in the calculated dose.
7.5.1 Dose Uncertainty Resulting from Individual Variation
This section discusses the uncertainty in calculated radiation
doses occasioned by differences in physical size and metabolism among
individuals in the general population. In order to investigate the
effects of individual differences in intake, size, and metabolism, it is
necessary to consider the form of the equation used to calculate
7-12
-------
radiation dose rates. Equation 7-2 is a. simplified form of the one used
by EPA to represent the ingestion of radioactive materials.
'
where D is the dose rate
I is the intake of radioactivity
fl is the fraction of I transferred to blood after ingestion
f2 is the fraction transferred to an organ from the blood
m is the mass of the organ
X is the elimination constant, which denotes how rapidly
the activity is removed from the organ
E is the energy absorbed by the organ for each radioactive
disintegration
c is a proportionality constant.
For simplicity, we will assume that dose rates at large times, t, are to
be studied so that the term in the bracket is approximately unity.
Although the actual equations used are considerably more compli-
cated because they must describe the lung model and the GI tract, and
also treat all radioactive progeny, the essential features of the
uncertainty in dose calculation are reflected in the terms of Equation
(7.2). The sensitivity of the dose to each of the terms in the equation
may be studied by substituting observed ranges of the quantities for the
single value recommended by Reference Man. For some of these
quantities, as noted below, no range is cited because of insufficient
data.
Daily Intake. J.
As an example, postulate that the ingestion r-ode to be calculated
is for fluid intakes. The average daily fluid intake is about 1900 ml,
with an adult range of 1000 to 2400 for "normal" conditions. Unier
higher environmental temperatures, this range may be increased to 2840
to 3410 ml. Thus, a dose calculated as 1.9, for example, could range
from 1.0 to 2.4.
Transfer Fraction, fj
The value of the transfer fraction to blood depends on the chemical
form of the element under study. One of the most common naturally
occurring radionuclides is uranium, which is used here as an example.
ICRP79 cites values of fj ranging from 0.005 to 0.05 for industrial
workers, but notes that a higher value of 0.2 is indicated by dietary
data from persons not occupationally exposed. EPA has used the 0.2
value for the general population but, based on the ICRP range above, a
calculated dose determination could vary by a factor of 10.0.
7-13
-------
Organ Mass,, m
The range
investigation.
range f
grams for adult
the ortan •.„
-
nd
Liv,
"" «•" """«
the bl°°'1^" lung,
ra"8ed from M°° to 2300
" f e°al" "u., because
organ at risk.
Remaining Terms.
and the lung is usually the principal
X. E
for the L i^i^v « ab f
directly observafJthan ?
calculation can only
" ValueS tO b
f 3ntities wWch «e less
^ influence on ^e dose
7-5'2 Pose Uncertainty Resulting from
strontium Is considered
Chemical element
the uncertainty
information i
'"""" such as
"?« ff
to allow estimation of
' f"
Iodine and the Thyroid
Iodine is
7-14
-------
in Table 7.5-1. The fluid intake varies from 0.72 liters per day for a
newborn to about 2.0 liters per day for an adult.
These age-dependent parameters nay then be used in Equation (7-2)
to calculate the dose rate resulting from a constant concentration of
iodine in water and air. The resulting curves for the dose rate as a
3ge are Shown itl Fi8ures 7-5-l and 7.5-2 (note: 1 nCi -
ptij. These nay be compared to the dose rates obtained using
Reference Man parameters at all ages, indicated by the dotted lines in
the same figures. Thus, for this particular combination of organ and
isotope, the total (70 year) dose is seen to increase by about 30
percent for ingestion and 35 percent for inhalation when dependence on
age is considered.
Strontium and Bone
Because of the chemical similarities of strontium and calcium,
strontium tends to follow the calcium pathways in the body and deposits
to a large extent in the skeleton. In fact, the fraction of ingested
strontium eventually reaching the skeleton at a given age depends
largely on the skeletal needs for calcium at that age, although the body
is able to discriminate somewhat against strontium in favor of calcium
after the first few weeks of life.
The ICRP model for bone is more complicated than that for the thy-
roid because it consists of more than one compartment. For purposes
of modeling the transport of strontium by the skeleton, it suffices to
view the mineralized skeleton as consisting of two main compartments:
trabecular (cancellous, porous, spongy) and cortical (compact) bone.
Table^7.5-1. Age-dependent parameters for
iodine metabolism in the thyroid
Age
(days)
Newborn
100
365
1825
3650
5475
7300
Fractional uptake
to thyroid, f£
0.5
0.40
0.3
0.3
0.3
0.3
0.3
Thyroid mass
(g)
-
-
1.78
3.45
7.93
12.40
20.00
Biological half-time
in the thyroid
(days)
15
20
30
40
50
65
80
7-15
-------
o
a •
a
ON
O
CO
H •
O
u °
3 "'
sr
u o
UJ •
2 M
o
o
Age-dependent model
Adult model
0.0 10.0 20.0 30.0
—I 1
40.0 50.0
AGE (YEARS)
60.0 70.0 80.0 fO.O 100.0
Figure 7.5-1. Dose from chronic ingestion of iodine-131
water at a concentration of 1 HCi/1.
7-16
-------
o
Si
o
*
o _
Age-dependent model
0.0 10.0 20.0 30.0
40.0 50.0 60.0
AGE (YEARS)
70.0 80.0 90.0 100.0
Figure 7.5-2.
Dose from chronic inhalation of iodine-131 in
air at a concentration of 1
7-17
-------
Two subcompartments, surface and volume, are considered within each of
these main compartments. The four subcompartments of mineralized skele-
ton and the movement of strontium among these compartments are shown
schematically in Figure 7.5-3. The equations governing the age depen-
dence of the parameters are given in (OENL84a). Dose rate curves for
the inhalation and ingestion of constant concentrations of strontium-90
are given in Figures 7.5-4 and 7.5-5. The comparable curves for
Reference Man are again indicated by dashed lines. Thus, for this
element and organ combination, the dose rate resulting from ingestion ia
somewhat higher, while the dose rate resulting from inhalation exhibits
only minor perturbations, when the age dependence of the parameters is
considered. The lifetime (70-year) dose resulting from ingestion is
about 7 percent greater and the inhalation dose less than 1 percent
different when age dependence is considered.
Plutonium and Lung and Red Bone Marrow
Apparently plutonium and iron bear sufficient chemical resemblance
that plutonium is able to penetrate some iron transport and storage sys-
tems. It has been shown that plutonium in blood serum complexes with
transferrin, the iron-transport protein. Thus, plutonium will partially
trace the iron pathway, with the result that a substantial fraction of
systemic plutonium is carried to the bone marrow and to the liver. In
the skeleton, plutonium may be released mainly at sites of developing
red cells, Plutonium that has reached the skeleton behaves very
differently from iron1, its movement is governed by fairly complicated
processes of bone resorption and addition. Because the toful metabolic
behavior of plutonium is not closely related to that of any essential
element, any retention model for plutonium as a function of age will
involve much larger uncertainties than the analogous model for
strontium. Still, there is enough information concerning the metabolism
of plutonium by mammals to justify an examination of potential differ-
ences with age in doses to radiosensitive tissues following intake of
this radioelement.
The effect of age-dependent parameters on dose rate calculations is
most evident for the lung when the inhalation pathway is considered.
Figure 7.5-6 exhibits the variation in dose rate to the total and pul-
monary portions of the lung both for the adult and age-dependent cases.
The increased dose rate from age 0 to about 20 is typically caused by
variations in the breathing rate-lung mass ratio for infants and
juveniles. For this model, the age-dependent pulmonary lung 70-year
dose is about 9 percent greater than for the adult model.
To describe retention of plutonium in the skeleton, it is conven-
ient to view the skeleton as consisting of a cortical compartment and
trabecular compartment. Each of these is further divided into three
subcompartments: bone surface, bone volume, and a transfer compartment.
The transfer compartment, which includes the bone marrow, may receive
plutonium that is removed from bone surface or volume; plutonium may
reside in this compartment temporarily before being returned either to
7-18
-------
BLOOD
i
TRABECULAR
SURFACE
CORTICAL
SURFACE
TRABECULAR
VOLUME
CORTICAL
VOLUME
Fi :->
OoaipartfBents and pathways in model for
tttrontium in skeleton.
7-19
-------
Age-dependent model
0-0 lo.O 20.0
40.0 50.0
AGE (YEAES)
?0.0 80.0 90.0
100.0
Pi-are 7.5-4. Dose from chroaic ingestion of strontium-90 in
in water at a concentration of 1 }iCi/l.
7-20
-------
o
O-t
O
81
^ o
S s-
I -
"^ o
u *
si
S» o
i o
z
S o
« o -
asj
s p
S <
Age-dependent model
Age-dependent model
T
T
0.0 10.0 20.0 30.0
—i 1—
40.0 50.0
AGE (YEARS)
60.0
70.0 80.0 90.0 100.0
Figure 7.5-5,
Dose from chronic inhalation of strontium-90 in
air at a concentration of 1 nCi/ra-*.
7-21
-------
Age-dependent dose rates and intake rates Canary lung)
tal
Adult dose races and intake rate (pulmonary lung)
Adult dose rates and intake rate (total lung)
0.0 10.0 20.0
30.0 40,0 50.0
AGI (YEARS)
60.0
80.0 90.0 100.0
Figure 7.5-6 Dose from chronic inhalation of pultonium-239 in
air at a concentration of 1
7-22
-------
the bloodstream or to bone surfaces (Figure 7.5-7). Because of the
large amount of recycling of plutonium among the skel-?'.-u compartments,
blood, and other organs, recycling is considered explicitly in the
model. The age-dependent features of the model are described in detail
in (ORNL84a).
Red bone marrow dose rates for the age dependent model are shown in
Figure 7.5-8, for ingestion, and in Figure 7.5-9, for inhalation. The
dashed curves are the dose rates using non-age-dependent parameters. As
in the corresponding curves for strontium, the difference is more pro-
nounced for the ingestion pathway. Because of the long radiological and
biological half-lives of plutonium in the skeleton, the dose rate, for a
chronic intake, does not reach equilibrium within the one hundred year
time period of the figures. The total lifetime (70-year) dose to the
red marrow is about 25 percent greater for ingestion, and nearly
unchanged for inhalation when the age-dependent parameters are used.
In summary, it is difficult to make generalizations concerning the
uncertainty involved in neglecting age dependence in the dose calcula-
tions. Although the examples given indicate higher dose rates for the
ingestion pathway, with smaller changes for inhalation, when using age-
dependent parameters, this results from the complex interaction between
parameters in the dose equation and depends on the element/organ combin-
ation under consideration.
7.5.3 Dose Uncertainty Causedby Measurement Errors
The last potential source of uncertainty in the dose calculations
is the error involved in making measurements of fixed quantities
(ORNL84b). The radioactive half-life of an isotope, for example, may be
measured independently of any biological system, but the measurement is
subject to some error. The organ mass of a given organ may also be
measured with only a small error. Repeated determinations of these
quantities, in addition, can reduce the error. Although this source of
uncertainty may be ,of importance in other aspects of an environmental
assessment, it is of little consequence in the dosimetry, because it is
overwhelmed by the magnitude of the uncertainties resulting from age and
individual variations.
Although consideration of the factors described above implies large
uncertainties in calculated doses, the actual variation is expected to
be considerably smaller. The reason for this, and some supporting
studies on real populations, are presented in Section 7.6.
7.6 Distributionof Doses in the General Population
Although the use of extreme parameter values in a sensitivity
analysis indicates that large uncertainties in calculated doses are
possible, this uncertainty is not usually reflected in the general popu-
lation. There are several reasons for thiss the parameter values chosen
are intended to be typical of an individual in the population} it is
improbable that the "worst case" parameters would be chosen for all
7-23
-------
BLOOD
TRABECULAR
SURFACE
TRABECULAR
VOLUME
TRABECULAR
MARROW
CORTICAL
SURFACE
CORTICAL
VOLUME
CORTICAL
MARROW
Figure 7.5-7. Compartments and pathways in model for plutonium
in skeleton.
7-24
-------
Age-dependent model
0.0 10.0 20,0 30.0 40.0 50.0
AGE (YEARS)
60.0 70.0 80.0 90,0 100.0
Figure 7.5-8. Dose from chronic ingestion of plutoniuo-239 in
water at a concentration of 1 pCi/1.
7-25
-------
Age-dependent nodel
o.o
10.0 20.0 30.0
50.0
AGE (YEARS)
60.0 70.0 80.0 90.0 100.0
Figure 7.5-9. Dose from chronic inhalation of plutoniua-239 in
air at a concentration of 1
7-26
-------
terms in the equation; and not all of the terms are mutually indepen-
dent, e.g., an increased intake may be offset by more rapid excretion.
This smaller range of uncertainty in real populations is demon-
strated by studies performed on various human and animal populations.
It should be noted that there is always some variability in observed
doses that results primarily from differences in the characteristics of
individuals. The usual way of specifying the dose, or activity,
variability in an organ is in terms of the deviation from the average,
or mean, value. In the following studies, it should also be noted that,
in addition to the variability resulting from individual characteris-
tics, the exposure levels of individuals may also have varied apprec-
iably - another factor tending to increase the dose uncertainty. The
following studies are representative of those carried out on real
populations:
(1) An analysis of the thyroid from 133 jackrabbits in a nuclear
fallout area (Tu65) found that in only 2 did the iodine-131 content
exceed three times the sample mean.
(2) Measurements of the strontium-90 content of adult whole skel-
etons (Ku62) showed that only about 5 percent of the population would
exceed twice the average activity, with only about 0.1 percent exceeding
four times the average.
(3) In another study, the cesium-137 content of 878 skeletal
muscle samples (E164a, b) was measured. This radioisotope is also the
result of nuclear tests so that the muscle content depends not only on
the variation in individual parameters but also on the pathways leading
to ingestion or inhalation of the isotope. Nevertheless, analyses of
these samples indicated that only 0.2 percent exceeded three times the
mean activity at a 95 percent confidence level.
(4) A study of the variability in organ deposition among indi-
viduals exposed under relatively similar conditions to toxic substances
has also been performed (Cu79). In eleven exposure situations (Table
7.6-1), the geometric standard deviation of the apparently lognormal
organ doses ranged from 1.3 to 3.4. This means that about 68 percent of
the organ doses were between 1/6 times and 6 times the geometric mean of
the doses. From the table, for example, 68 percent of the bone doses
resulting from ingestion of strontium-90 would lie between 0.56 and 1.8
times the average.
In all but two of the situations examined, there is the compli-
cating factor that there was probably a great deal of variation in the
exposure levels experienced by members of the population. The magnitude
of geometric standard deviations of the studies listed in Table 7.6-1
may be the evidence of this variation since, except for the two beagle
studies, the exposure was not uniform. Despite these nonunifonn
exposures, however, the organ dose is not greatly affected probably
because of differences in metabolic processes. For example, there is
probably some "self-adjustment" in the amount of strontium-90 absorbed
7-27
-------
Table 7.6-1. Distributions of organ doses* from
inhalation and ingestion of metals
Population
Beagle
Humans
Humans
Iumans
iumans
Jeagles
Iumans
[umans
(smokers)
[umans
(nonsmokers)
umans
umans
Exposure
Metals
Plutonium
(fallout)
Titanium
(soil)"
Aluminum
(soil)
Vanadium
(fuel
combustion)
Strontium-90
Strontium-90
(fallout)
Cadmium
Cadmium
Lead
Lead
Principal
exposure mode
Inhalation
Inhalation
Inhalation
Inhalation
Inhalation
Ingestion
Ingestion
Inhalation and
Ingestion
Inhalation and
Ingestion
Inhalation and
Ingestion
Inhalation
Target
organ
Bone or liver
Lung
Lung
Lung
Lung
Bone
Bone
Kidney
Kidney
Bone
Lung
Geometric standard
deviation of
organ doses3
1.8
3.1b
3.4b
3.4b
3.4b
1.3
1.8b
1.8b
1.8b
2.2b
1.7b
a>The stable element organ doses used in compiling this table were generally
expressed in parts-per-million of organ mass.
^Note that exposure levels may vary considerably among individuals 5 if this
factor could be eliminated, geometric standard deviations probably would be
smaller.
aurce: (Cu79).
7-28
-------
from the small intestine to blood of different persons, since strontium-
90 tends to vary with calcium in food; if a person has a low calcium
intake, then he may absorb a higher fraction of the calcium and
strontium-90 than a person with a high calcium intake.
In the beagle studies, the geometric standard deviation is 1.8
for inhaled metals in bone or liver, but is only 1,3 for ingested
strontium-90 in bone. An important difference is that all dogs in-
gesting strontium-90 at a given level were administered the same amount,
whereas, in the inhalation studies, the exposure air concentrations were
controlled but the dogs inhaled variable amounts depending upon their
individual characteristic breathing patterns.
Thus, in real situations, the overall uncertainty in dose is seen
to be considerably smaller than would be expected solely on a basis of
the "worst case" sensitivity analyses.
7.7 Summary
This chapter presents an overview of the methods used by IPA to
estimate radiation doses. The chapter defines the basic quantities
reported by EPA and describes briefly the models employed. The chapter
also points out departures from the occupational parameters and assump-
tions employed in the basic ICRP methodology and gives the reasons for
the deviations outlined.
Many of the physiological and metabolic parameters recommended in
methods for calculating radiation doses are based on a limited number of
observations, often on atypical humans or on other species. EPA has
attempted to bound the uncertainty associated with the ranges observed
for some of the more important parameters used. In fact, some empirical
data on population doses mentioned here indicate that actual dose
uncertainties are much less than is implied by this "worst case"
analysis. For the sources of uncertainty discussed, the large dose
ranges possible because of variation in individual characteristics must
be modified by consideration of the narrower ranges indicated by studies
of real populations; the dose range resulting from age dependence
appears to be small for lifetime exposures, and the range resulting from
experimental error is negligible by comparison. Based on these observa-
tions, it is reasonable to estimate that EPA's calculated doses should
be accurate within a factor of three or four. It should be emphasized
that much of the "uncertainty" in the dose calculation is not caused by
parameter error but reflects real differences in individual character-
istics within the general population. Therefore, the uncertainty in the
dose estimates cannot be dissociated from specification of the segment
of the population to be protected.
More complete derivations and explanations for the EPA methodology
are given in the references cited in the text, and a technical descrip-
tion of the dose rate equations and their use in conjunction with the
life table risk evaluation is given in Addendum B.
7-29
-------
REFERENCES
Cu79 Cuddihy R. G., McClellan R, D., and Griffith W. C.» Variability
in Target Organ Deposition among Individuals Exposed to Toxic
Substances, Toxicol. Appl. Pharmacol. 49, 179-187, 1979.
Du84 Dunning D. E. Jr., Leggett R. W., and Sullivan R. E.» An
Assessment of Health Risk from Radiation Exposures, Health
Phys. 46 (5), 1035-1051, 1984.
E164a Ellett W. H. and Brownell G. L., Caesium-137 Fall-Out Body
Burdens, Time Variation and Frequency Distributions, Nature 203
(4940), 53-55, 1964.
E164b Ellett W. H, and Brownell G. L., The Tine Analysis and
Frequency Distribution of Caesium-137 Fall-Qut in Muscle
Samples, IAEA Proceedings Series, STI/PDB/84, Assessment of
Radioactivity in Man, Vol. II, 155-166, 1964.
IPA77 U.S. Environmental Protection Agency, Proposed Guidance on Dose
Limits for Persons Exposed to Transuranium Elements in the
General Environment, EPA 520/4-77-016, 1977.
ICRP75 International Commission on Radiological Protection, Report of
the Task Group on Reference Man, ICRP Publication No. 23,
Pergamon Press, Oxford, 1975.
ICRP77 International Commission on Radiological Protection,
Recommendations of the International Commission on Radiological
Protection, ICRP Publication No. 26, Pergamon Press, Oxford,
1977.
ICRP79 International Commission on Radiological Protection, Limits for
Intakes of Radionuclides by Workers, ICRP Publication No. 30,
Pergamon Press, Oxford, 1979.
Ku62 Kulp J. L. and Schulert A. R., Strontium-90 in Man V, Science
136 (3516), 619-632, 1962.
MIRD69 Medical Internal Radiation Dose Committee, Estimates of
Absorbed Fractions for Monoenergenetic Photon Sources Uniformly
Distributed in Various Organs of a Heterogeneous Photon, MIRD
Supplement Ho. 3, Pamphlet 5, 1969.
7-30
-------
NRPB82 National Radiological Protection Board, Gut Uptake Factors for
Plutonium, Americium, and Curium, NRPB-R129, Her Majesty's
Stationery Office, 1982.
ORNL80 Oak Ridge National Laboratory, A Combined Methodology for
Estimating Dose Rates and Health Effects for Exposure to
Radioactive Pollutants, ORNL/RM-7105, Oak Ridge, Tenn., 1980.
ORNL81 Oak Ridge National Laboratory, Estimates of Health Risk from
Exposure to Radioactive Pollutants, QRNL/RM-7745, Oak Ridge,
Tenn., 1981.
ORNL84a Oak Ridge National Laboratory, Age Dependent Estimation of
Radiation Dose, to be published.
ORNL84b Oak Ridge National Laboratory, Reliability of the Internal
Dosiraetric Models of ICRP-30 and Prospects for Improved Models,
to be published.
Tu65 Turner F. B., Uptake of Fallout Radionuclides by Mammals and a
Stochastic Simulation of the Process, in Radioactive Fallout
from Nuclear Weapons Tests, U.S. AEC, Division of Technical
Information, November 1965.
7-31
-------
Chapter 8s ESTIMATING THE RISK OF HEALTH EFFECTS
RESULTING FROM RADIONUCLIDE AIR EMISSIONS
8.1 Introduction
This chapter describes how the Environmental Protection Agency
(EPA) estimates the probability of fatal cancer, serious genetic
effects, and other detrimental health effects resulting from exposure to
ionizing radiation. Such risk estimates are complex. They are also
uncertain, even though much scientific effort has been expended to
increase the understanding of radiation effects.
Because the effects of radiation on human health are known more
quantitatively than are the effects of most other environmental pollu-
tants, it is possible to make numerical estimates of the risk that may
occur as a result of a particular source of radioactive emissions. Such
numbers may give an unwarranted aura of certainty to estimated radiation
risks. Compared to the baseline incidence of cancer and genetic
defects, radiogenic cancer and radiation-induced genetic defects do not
occur very frequently. Even among heavily irradiated populations, the
number of cancers and genetic defects resulting from radiation is not
known with either accuracy or precision simply because of sampling
variability. In addition, exposed populations have not been followed
for their full lifetime, so that information on ultimate effects is
limited. Moreover, when considered in light of information gained from
experiments with animals and from various theories of carcinogenesis and
mutagenesis, the observational data on the effects of human exposure are
subject to a number of interpretations. This in turn leads to differing
estimates of radiation risks by both individual radiation scientists and
expert groups. Readers should bear in mind that estimating radiation
risks is not a mature science and that the evaluation of radiation
hazards will change as additional information becomes available. In
this chapter a number of simple mathematical models are presented that
may describe the main features of the human response to radiation.
However, most scientists would agree that the underlying reality is
quite complicated and largely unknown, so that such models should not be
taken too literally but rather as useful approximations that will some
day be obsolete.
The risk estimates in the Draft Background Information Document
(DBID) (EPA83a) for the proposed rules on radionuclide emissions were
based on the 1972 National Academy of Science BEIR report (NAS72), To
take advantage of more recent data and analysis of radiation risks,
EPA's estimates of cancer and genetic risks in this final BID are based
8-1
-------
or Lu'- BE'h-3 report (NAS80). This report was prepared for the purpose
of a» ssing radiation risks at the low exposure levels of interest in
standard setting. As phrased by the President of the Academy, "We
believe that the report will be helpful to the EPA and other agencies as
they reassess radiation protection standards. It provides the scien-
tific bases upon which standards may be decided after nonscientifie
social values have been taken into account."
In this chapter, we outline the various assumptions made in
calculating radiation risks based on the 1980 NAS report and compare
these risk estimates with those prepared by other scientific groups such
as the 1972 NAS BEIR Committee (NAS72), the United Nations Scientific
Committee on the Effects of Atomic Radiation (UNSCEAR), and the
International Commission on Radiation Protection (ICRP). We recognize
that information on radiation risks is incomplete and do not argue that
the estimates made by the 1980 NAS BEIR Committee are highly accurate.
Rather, we discuss some of the deficiencies in the available data base
and point out possible sources of bias in current risk estimates.
nevertheless, we do believe the risk estimates made by EPA are "state-
of-the-art".
The analysis of possible health effects resulting from radionuclide
emissions in the air, EPA83a, indicated that by far the greatest risk
was radiogenic cancer, primarily lung cancer caused by inhaling
radioactive material. The risk of genetic damage was typically 10 to
100 times smaller than the risk of radiogenic cancer. Although we
include a discussion of possible genetic effects and other health
hazards due to radiation in this chapter, EPA has not included estimates
of genetic damage for the sources of radionuclide emissions described in
Chapters 11-17 of Volume II of the BID. As outlined in Section 8.7
below, the additional risk of genetic harm is so much smaller than the
uncertainty in the estimated risk of radiogenic cancer, that it has not
been a factor in this rulemaking.
In the sections below, we first consider the cancer risk resulting
from whole-body exposure to low-LET* radiation, i.e., lightly ionizing
radiation like the energetic electrons produced by X-rays or gamma rays.
Environmental contamination by radioactive materials also leads to the
ingestion or inhalation of the material and subsequent concentration of
the radionuclides in selected body organs. Therefore, the cancer risk
resulting from low-LET irradiation of specific organs is examined next.
Organ doses can also result from high-LET radiation, such as that asso-
ciated with alpha particles. Estimation of cancer risks for situations
where high-LET radiation is distributed more or less uniformly within a
body organ is the third situation considered, Section 8.3. Because
highly ionizing alpha particles have a very short range in tissue, there
are exposure situations where the dose distribution to particular organs
* Linear Energy Transfer (LET), the energy deposited per unit of
distance along the path of a charged particle.
8-2
-------
is extremely nonuniform. An example is inhaled radon progeny:
polonium-218, lead-214, and polonium-214. For these radiomiclides we
base our cancer risk estimates on the amount of radon progeny inhaled
rather than on the estimated dose, which is highly nonuniform and
cannot be well quantified. Therefore, riak estimates of radon exposure
are examined separately in Section 8.4. We review the causes of
uncertainty in the cancer risk estimates and the magnitude of this
uncertainty in Section 8.5, so that the public as well as EPA decision
makers have a proper understanding of the degree of confidence to place
in them. In Section 8.6, we review and quantify the hazard of
deleterious genetic effects due to radiation and the effects of exposur
in utero on the developing fetus. Finally, in section 8.7, we calculat
cancer and genetic risks from background radiation using the models
described in this chapter.
8.2 Cancer RiskEstimates forLow-LET Radiations
Most of the observations of radiation-induced carcinogenesis in
humans are on groups exposed to low-LET radiations. These groups
include the Japanese A-bomb survivors and medical patients treated with
X-rays for ankylosing spondylitis in England from 1935 to 1954 (Sm78).
The UNSCEAR (UNSCEAR77) and NAb Committee on the Biological Effects of
Ionizing Radiations (BEIR) (NAS80) have provided knowledgeable reviews
of these and other data on the carcinogenic effects of human exposures.
The most important epidetniological data base on radiogenic cancer
is the A-bomb survivors. The Japanese A-bomb survivors have been
studied for more than 38 years and most of them( the Life Span Study
Sample, have been followed in a carefully planned and monitored
epidemiological survey since 1950 (Kab82, Wa83). They were exposed to
a wide range of doses and are the largest group that has been studied.
Therefore, they are virtually the only group providing information on
the response pattern at various levels of exposure to low-LET radiation
Unfortunately, the dos,es received by various individuals in the Life
Span Study Sample are not yet accurately known. The 1980 BEIR
Committee's analysis o£ the A-bomb survivor data was prepared before
bias in the dose estimates for the A-bomb survivors (the tentative 1965
dose estimates, T65) became widely recognized (Lo81). It is new clear
that the T65 doses tended to be overestimated (Bo82, RERF83,84) so that
the BEIR Committee's estimates of the risk per unit dose are likely to
be too low. A detailed reevaluation of current risk estimates is
indicated when the A-bomb survivor data have been reanalyzed on t.te
basis of" new and better estimates of tha dose to individual survivors.
Uncertainties in radiation risk estimates do not result just from
the uncertainties in the Japanese data base and in other epidemiologies
studies. Analyses of these data bases require a number of assumptions
that have a considerable effect on the estimated risk. These assump-
tions are discussed below. The degree of uncertainty introduced by
choosing among these assumptions is probably greater than the uncer-
tainty of the estimated risk per unit dose among the A-bomb survivors <
other sources of risk estimates for radiogenic cancer in humans.
8-3
-------
8"2<1 As ^Ptions Needed to Make Risk Estimates
dosesshot7H0f "'V1;*1011; must be made ab°"t how observations at high
doses should be applied at low doses and low dose rates for radiation of
relllT r (LET)* TheSe assumPtio«s "duds the shape of thYdose
response function and possible dose rate effects. A dose response
hat a°LdPreSSe8 the relations^P ****** dose and the probability
occurred fofrT "T* " ^^"^ °bS"Ved eXCesS cancers have
wdUtioi /°St f "' following relatively high doses of ionizing
radiation compared to those likely to occur as a result of the
co^trollabll baCkgrTd "diation and environmental contamination from
provide a ^T6r radiation' Therefore, a dose response model
provide, a method of interpolating between the number of radiogenic
atl caLf ""t H3' HK8h dOS6S ^ the nUDber °f cance« re.ul?i£ fro.
all causes including background radiation.
ho , The"«8e 0* interpolation is not the same for all kinds of cancer
because it depends upon the radiosensitivity of a given tissue. For
IsT/'-H I Tf Pr°bable radi°8enic canc" ««r wLen is breast Leer.
not toC rfd r'KWlth,aP!r°Priate «ferences> Breast cancer appears
not to be reduced when the dose is delivered over a long period of time.
woLnX3I ' T1^ °f 6XCeSS CanCetS P6r unit dose ^ Japanese
women, who received acute dcses, is about the same per unit dose as
women exposed to small periodic doses of X-rays over many years. If
this is actually the case, background radiation is as carcinogenic for
breast tissue as the acute exposures from A-bomb gamma radiation.
Moreover the female A-bomb survivors show an excess of breast cancer at
doses below 20 rad which is linearly proportional to that observed at
several hundred rad (Tob84) . Women in their forties, the youngest age
S!7 ^njwh1ich breasc cancer is common, have received about 4 rad of
whole-body low-LET background radiation and usually some additional dose
incurred for diagnostic medical purposes. Therefore, for this cancer,
the difference between observed radiogenic cancer, less than 20 rad, and
the dose resulting from background radiation is less than a factor of
r^ "^ TVeral °rderS °f ma8nitude as « sometimes claimed. However,
it should be noted that breast tissue is a comparatively sensitive
tissue for, cancer induction and that for most cancers, a statistically
significant excess has not been observed at doses below 100 rad, low
LET. Therefore, the range of dose interpolation between observed and
calculated risk is often large.
3.2,2 Dose Response Functions
The 1980 HAS report (HAS80) examined three dose response functions
in detail: (1) linear, in which effects are directly proportional to
lose at all doses j (2) linear quadratic, in which effects are very
nearly proportional to dose at very low doses and proportional to the
iquai-e of the dose at high doses; and (3) a quadratic dose response
runction, where the risk varies as the square of the dose at all dose
eveis.
8-4
-------
We believe the first two of these functions are compatible with
most of the data on human cancer. Information that became available
only after the BEIR-3 report was published indicates that a quadratic
response function is inconsistent with the observed excess risk of solid
cancers at Nagasaki, where the estimated gamma-ray doses are not
seriously confounded by an assumed neutron dose component. The chance
that a quadratic response function underlies the excess cancer observed
in the Nagasaki incidence data has been reported as only one in ten
thousand (Wa83). Although a quadratic response function is not
incompatible with the Life Span Study Sample data on leukemia incidence
at Nagasaki, Beebe and others (Be78, Ela77) have pointed out how
unrepresentative these data are of the total observed dose response for
leukemia in that city. There is no evidence that a quadratic response
function provides a better fit to the observed leukemia excess among all
A-bomb survivors in the Life Span Study Sample than a simple linear
model (NAS80). Based on these considerations, we do not believe a
quadratic response can be used in a serious effort to estimate cancer
risks due to ionizing radiation. EPA notes that neither the NCRP, the
ICRP, nor other authorative scientific groups, e.g., NRPB and UNSCEAR,
have used a quadratic response function to estimate the risks due to
ionizing radiation.
The 1980 NAS BEIR Committee considered only the Japanese mortality
data in their analysis of possible dose response functions (NAS8Q).
Based on the T65 dose estimates, this Committee showed that the excess
incidence of solid cancer and leukemia among the A-bomb survivors is
compatible with either a linear or linear quadratic dose response to the
low-LET radiation corooonent and a linear response to the high-LET
neutron component (NA-S3D'>. Although the 1980 BEIR report indicated that
low-LET risk estimates based on a linear quadratic response were
"preferred" by most of Li,<; scientists who prepared that report, opinion
was not unanimous, ann •,: :.-2jieve the subsequent reassessment of the
A-bomb dose seriuu&l^ v*i»ak=ns the Committee's conclusion. The
Committee's analysts ft close response functions was based on the
assumption that most of the observed excess leukemia and solid cancers
among A-bomb survivors resulted from neutrons (NAS80). Current
evidence, however, i«3 conclusive that neutrons were only s minor
component of the dose in both Hiroshima and Nagasaki (Bo82, RERF83,84).
Therefore, it is likely that the linear response attributed to neutrons
was caused by the gamma dose, not the dose from neutrons. This point is
discussed further in Section 8.5.
Reanalysis of the Japanese experience after completion of the dose
reassessment may provide more definitive information on the dose
response of the A-bomb survivors, but it is unlikely to provide a
consensus on the dose response at environmental levels, i.e., about 100
mrad per year. This is because at low enough doses there will always be
sampling variations in the observed risks so that observations are
compatible, in a statistical sense, with a variety of dose response
functions. In the absence of empirical evidence or a strong theoretical
basis, a choice between dose response functions must be based on other
considerations.
8-5
-------
evidence for a nonlinear
to
i.sz-.s r ..
rule for separating nfces^rv fZ LT" " ' •' ' ViabU 8cientific
to A-bo^b survivors (s« 851? K"""* "" " the d°SeS assi^ned
well as nrudllt TK J 8'5'1>» such an approach seems reasonable, as
as prudent. Therefore, EPA has utilized the BFTR-T M»»» A
'
= fri=^r^rsirta
£»S - "" ~ "~ = SL-EL'L-S.ffi^
s
Hnear tern being indicative of this repair Us" of I lii^r ^l
dose response faction, as formulated by t^ BEIR-3 cLlittee "l. "
™ '
2.5
8-2'3 " ess
Effect, of Dose Rate on Radinc.rcinogenesi
'
rate l.. cnat to-no
-
from
"Tp-f —"--••^••••"6 iauj.uauti.ve materials, a considerable body of
NCRP Committee 40 has suggested that carcinogenic^eccTof^ow-LET
8-6
-------
radiations may be a factor of from 2 to 10 times less for small doses
and dose rates than have been observed at high doses (NCKP80).
The low dose and low dose rate effectiveness factors developed by
NCRP Committee 40 are based on their analysis of a large body of plant
and animal data that showed reduced effects at low doses for a number of
biological endpoints, including radiogenic cancer in animals, chiefly
rodents. However, no data for cancer in humans confirm these findings
as yet. A few human studies contradict them. Highly fractionated small
doses to human breast tissue are apparently as carcinogenic as large
acute doses (NAS80, LaaSO). Furthermore, small acute (less then 10 rad)
doses to the thyroid are as effective per rad as ouch larger doses in
initiating thyroid cancer (UNSCEAR77, HAS80). Moreover, the increased
breast cancer resulting from chronic low-dose occupational gamma-ray
exposures among British dial painters is comparable to, or larger, than
that expected on the basis of acute high-dose exposures (Ba81),
While none of these examples is persuasive by itself, collectively
they indicate that it may not be prudent to assume that all kinds of
cancer are reduced at low dose rates and/or low doses. However, it may
be overly conservative to estimate the risk of all cancers on the basis
of the linearity observed for breast and thyroid cancer. The
International Commission on Radiation Protection and the United Nations
Scientific Committee on Atomic Radiations have used a dose rate effec-
tiveness factor of about 2.5 to estimate the risks from occupational
(ICRP77) and environmental exposures (UNSCEAR77), Their choice of a
DREF is fully consistent with and equivalent to the reduction of risk at
low doses obtained by substituting the BEIR-3 linear-quadratic response
model for their linear model. Use of both a DREF and a linear quadratic
model for risk estimation is inappropriate (NCRP80).
The difference between risk estimates obtained with the BEIR-3
linear and linear-quadratic dose response models is by no means the full
measure of the uncertainty in the estimates of the cancer risk resulting
from ionizing radiation (Section 8.5 summarizes information on
uncertainty). The use of two dose models serves as a reminder that
there is more than one creditable response model for estimating
radiation risks and that it is not known if all radiogenic cancers have
the same dose response.
8.2,4 Ri s k Preject ion Mod els
None of the exposed groups has been observed long enough to assess
the full effects of their exposures, if, as is currently thought, most
radiogenic cancers occur throughout an exposed person's lifetime
(NAS80). Therefore, another major choice that must be made in assessing
the lifetime cancer risk resulting from radiation is to select a risk
projection model to estimate the risk for a longer period of time than
currently available observation data will allow.
To estimate the risk of radiation exposure that is beyond the years
of observation, either a relative risk or an absolute risk projection
8-7
-------
projection model projects the currently observed
.jncance: ifper unit dose int° '-^ J2r
model projects the average observed number of excess cancers Der
unit dose into the future years at risk. «*ce« cancers per
Because the underlying risk of cancer increases rapidly with aee
toward tJeVLr1 ^^ "T^' ' l^T Probability of excels canclr
model J^' f 3 Per80n S ^^time. In contrast, the absolute risk
ThJ«fP *" a T"ant incidence of *«ce.. cancer across time.
Sllo£°«n' 8"T . "?™Plete data ™ have now, less than lifetime
that e^ti™J T ""K m°del Pr°JeCtS 80mewhat *reat« "sk than
that estimated using an absolute risk model.
p The National Academy of Sciences BEIR Committee and other
scientific groups, e.g. UNSCEM, have not concluded which projection
!v?f nil - apprTiate choice for *>°st radiogenic cancers. However,
for most ^1a"Ufflulat"S that fvor. the relative risk projection modi
tor most solid cancers. As pointed out by the 1980 HAS BEIR Committee,
"If the relative-risk model applies, then the age of the
exposed groups, both at the time of exposure and as they
move through life, becomes very important. There is now
considerable evidence in nearly all the adult human
populations studied that persons irradiated at higher ages
have, in general, a greater excess risk of cancer than
those irradiated at lower ages, or at least they develop
cancer sooner. Furthermore, if they are irradiated at a
particular age, the excess risk tends to rise pari passu
I at equal pace] with the risk of the population^ Ia7g7.
In other words, the relative-risk model with respect to
cancer susceptibility at least as a function of age
evidently applies to some kinds of cancer that have been
observed to result from radiation exposure." (NAS80, p. 33)
This observation is confirmed by the Ninth A-bomb Survivor Life
Span Study, published 2 years after the 1980 Academy report. This
latest report indicates that, for solid cancers, relative risks have
continued to remain constant in recent years while absolute risks have
increased substantially (Kab82). Smith and Doll (Sm78) have reached
similar conclusions on the trend in excess cancer with time among the
irradiated spondylitic patients.
Although we believe considerable weight should be given to the
relative risk model for most solid cancers (see below), the model does
not necessarily give an accurate projection of lifetime risk. The mix
of tumor types varies with age so that the relative frequency of some
common radiogenic tumors, such as thyroid cancer, decreases for older
ages. Land has pointed out that this »ay result in overestimates of the
lifetime risk when the estimates are based on a projection model using
relative risks (Lac83). While this may turn out to be true for
8-8
-------
estimates of cancer incidence that include cancers less likely to be
fatal, e.g., thyroid, it may not be too important in estimating the
lifetime risk of fatal cancers, since the incidence of most of the
common fatal cancers, e.g., breast and lung cancers, increases with age.
Leukemia and bone cancer are exceptions to the general validity of
a lifetime expression period for radiogenic cancers. Most, if not all,
of the leukemia risk has apparently already been expressed in both the
A-bomb survivors and the spondylitics (Kab82f Sm78), Similarly, bone
sarcoma from acute exposure appears to have a limited expression period
(NAS80, Mab83). For these diseases, the BEIR-3 Committee believed that
an absolute risk projection model with a limited expression period is
appropriate for estimating lifetime risk (NAS8Q).
It should-be noted that unlike the NAS BEIR-1 report (NAS72) the
BEIR-3 Committee's relative and absolute risk models are age dependent.
That is, the risk coefficient changes depending jn the age of the
exposed persons. Observation data on how caiicer risk resulting from
radiation changes with age are sparse, particularly so in the case of
childhood exposures. Nevertheless, the explicit consideration of the
variation in radiosensitivity with age at exposure is a significant
improvement in methodology. It is important to differentiate between
age sensitivity at exposure and the age dependence of cancer expression.
In general, people are most sensitive to radiation when they are young.
In contrast, most radiogenic cancers occur late in life, ouch like
cancers resulting from other causes. In this chapter we present risk
estimates for a lifetime exposure of equal annual doses. The cancer
risk estimated is lifetime risk fron this exposure pattern. However,
age dependent analyses using BIIR-3 risk coefficients indicate that the
risk from one year of exposure varies by a factor of at least five
depending on the age of the recipient.
8.2.5 Effect of VariousAssumptions on the Numerical Risk Estimates
Differences between risk estimates made by using various
combinations of the assumptions described above were examined in the
1980 NAS report. Table 8.2-1, taken from Table V-25 (NAS80), shows the
range of cancer fatalities induced by a single 10-rad dose as estimated
using linear, linear quadratic, and quadratic dose response functions
and two projection models, relative and absolute risk.
As illustrated in Table 8.2-1, estimating the cancer risk for a
given projection model on the basis of a quadratic as compared to a
linear dose response reduces the estimated risk of fatal cancer by a
factor of nearly 20. Between the more credible linear and linear
quadratic response functions the difference is less, a factor of about
two and a half. For a given dose response model, results obtained with
the two projection models, for solid cancers, differ by about a factor
of three.
Even though the 1980 NAS analysis estimated lower risks for a
linear quadratic response, it should not be concluded that this response
8-9
-------
Table 8.2-1, Range of cancer fatalities induced by
10 rad low-LET radiation (Average value per rad
per million persons exposed)
Dose Response Lifetime Risk Projection Model
Functions Relative^-* Absolute
501
Linear Quadratic^ ) 226 77
Quadratic^) 28 10
(a>Relative risk projection for all solid cancers except
leukemia and bone cancer fatalities, which are projected
(b)by means of the absolute risk model (NAS80).
-'Response I varies as a constant times the dose, i.e.,
R=CjD. ' '
(,C>R=C2D+C3D2.
Source; NAS80, Table V-25.
function always provides smaller risk estimates. In contrast to the
1980 NAS analysis, where the proportion of risk resulting from the dose
squared term (e.g., C3 in equation c of Table 8.2-1) was constrained to
positive values, the linear quadratic function (which agrees best with
Nagasaki cancer incidence data) has a negative coefficient for the dose
squared tern (Wa83). Although this negative coefficient is small and
indeed nay not be significant, the computational result is a larger
linear term that leads to higher risk estimates at low doses than would
be estimated using a simple linear model (Wa83). Preliminarily, the
BEIR-3 analyses of mortality, which were not restricted to positive
coefficients of the dose squared terms, yielded similar results.
Differences in the estimated cancer risk introduced by the choice
of the risk projection model are also appreciable. As pointed out
above, the 1980 NAS analysis indicates that relative risk estimates
exceed absolute risk estimates by about a factor of 3, Table 8.2-1.
However, relative risk estimates are quite sensitive tc how the risk
resulting from exposure during childhood persists throughout life This
question is addressed in Section 8.2.6 below, where we compare risk
estimates made by the 1972 and 1980 NAS BEIR Committees with those of
the ICRP and UNSCEAR.
8«2.6 Comparison of Cancer Risk Estimates for Low-LET Radiation
A number of estimates of the risk, of fatal cancer following
lifetime exposure are compared in Table 8.2-2. Although all of these
8-10
-------
Table 8.2-2. A comparison of estimates of the risk of fatal cancer
from a lifetime exposure at 1 rad/year (low-LET radiation)
Cases per 10& person rad
Projection Model
BEIR-1 (NAS72)(a) §67
BEIR-1 (NAS80)(b) 568
BEIR-3 (NAS80)(b)(c) 403
BEIR-3 (NAS80)(d) 169
BEIR-3 (NASSOXb) 158
BEIR-1 (Nft'IEOKb) 115
BEIR-3 (NAS80)(d) 67
UNSCEAR (UNSC£AR77)(e) 200-300
UNSCEAR (UNSCEAR77)(e) 75-175
ICRP (ICRP77) 125
CLM (Ch83) 100-440
Relative Risk
Relative Risk
Relative Risk
Relative Risk
Absolute Risk
Absolute Risk
Absolute Risk
None—high dose > 100 rad
None—low dose/dose rate
None—occupational -
low dose/dose rate
UNSCEAR77 without A-bomb
data
-l relative risk model.
(b)Table v_4 £n NAS80, linear dose response. _
(C)L_L absolute risk model for bone cancer and leukemia; L-L relative
risk model for all other cancer.
^Table ¥-4 in NAS80, linear-quadratic dose response.
(^Paragraphs 317 and 318 in UNSCEAR77.
risk estimates assume a linear response function, they differ consider-
ably because of other assumptions. In contrast with absolute risk
estimates, which have increased since the first NAS report (BEIR-1) was
prepared in 1972 (NAS72), the 1980 NAS BEIR-3 Committee's estimates of
the relative risk, as shown in Table 8.2-2, have decreased relative to
those in the BEIR-1 report. This illustrates the sensitivity of risk
projections to changes in modeling assumptions. In the NAS80 report,
the relative risk of solid cancer observed for ages 10 to 19 was
substituted for the considerably higher relative risk observed for those
exposed during childhood, ages 0 to 9. In addition, the relative risk
coefficients used in the BEIR-3 analysis are based on excess cancer in
the Japanese A-bonb survivors compared to U.S. population cancer
mortality rates. In the 1972 NAS report this excess was compared to
cancer mortality in Japan. Moreover, the difference introduced by these
two changes, particularly the former, is somewhat greater than indicated
in the 1980 NAS report. The relative risk estimate attributed to the
8-11
-------
-1 r!Httee f\the !AS 198° reP°^ is incorrect. Therefore, two
BEIR 1 relative nsk estimates are listed in Table 8.2-2- the risk
2.3*2 thV1^ fcfruted to.the BEIR-J c«—t J'L2 U'Lt^t.
adult values (n 171 |« MAO-IT* v. «. _i_ *«rge
aoa* • i j- . NAS72}, but rather used the adult risk for all
hg re": ive^fiet^elr ^ "^T " T'M' 8'« "^ ^ ?2 i es
relative risk coefficients actually given in the BEIR-1 report.
By comparing the three relative risk estimates in Table 8 2-2 ,>
r
r -— -
cancer from childhood exposure continues throughout adult life The
^r^^^li^^
extent predicted by the NAS BEIR-1 cLnittee in 1972?
The major reason that the risk estimates in Table 8.2-2 differ is
= Jai '
population at risk nor the projection models (if any) have
e "
« M - te IGR x77).
apparently presumes the saine age distributions as occurred in
C " the Cil6d ™aily the A-b0ab --i-rs and
The last entry in Table 8.2-2 (Chb83) is of interest because it
specifically excludes the A-bOTb survivor data based on T65 dose
in SM;?^ "' 1°" ^T"^ ^ inforaati- « radiogenic cancer
in UNSCEM77 so as to exclude all data based on the Japanese experience
for hieh r" °f/f aUtieS "^ fr0ffl 10° to 4*° P« 106 P-S2
for high doses and dose rates. As indicated in Table 8.2-2. this is
e to tl»e UNSCEM estimate,
8"2"7 EPA Assumptions about Cancer Risks Resulting from LOW-LFT
Radiatio ' ~~ ~~ - — * - ==--
EPA estimates of radiation risks are based on presumed linear and
linear quadratic dose response functions. We believe these are IL
moat credible dose reaponse functions for estimating risks to exposed
8-12
-------
populations. Use of the BEIR-3 linear quadratic model is equivalent, at
low dose, to using a. dose rate effectiveness factor of 2.5.
Except for leukemia and bone cancer, where we use a 25-year
expression period for radiogenic cancer, we use a lifetime expression
period, as was done in the HAS report (NAS80). Because the most recent
Life Span Study Report (Kab82) indicates that absolute risks for solid
cancers are continuing to increase 33 years after exposure, the 1980 HAS
Committee choice of a lifetime expression period appears to be well
founded. We do not believe that limiting cancer expression to 40 years
(as has been done by the ICRP and IWSCEAE) ia compatible with the
continuing increase in solid cancers that has occurred among irradiated
populations (Kab82). Analyses of the apondylitic data have led others
to similar conclusions (Sm78).
To project the number of fatalities resulting from leukemia and
bone cancer, EPA uses an absolute risk model, a minimum induction period
of 2 years, and a 25-year expression period. To estimate the number of
fatalities resulting from other cancers, EPA uses the arithmetic average
of absolute and relative risk projection models. For these cancers, we
assume a 10-year minimum induction period and expression of radiation-
induced cancer for the talance of an exposed person's lifetime after the
minimum induction period.
8.2.8 Methodology forAssessing the Risk ofRadiogenic Cancer
EPA uses a life table analysis to estimate the number of fatal
radiogenic cancers in an exposed population of 100,000 persons. This
analysis considers not only death resulting from radiogenic cancer but
also the probabilities of other competing causes of death which are, of
course, much larger and vary considerably with age (Bu81, Co78),
Basically, it calculates for ages 0 to 110 the risk of death resulting
from all causes by applying the 1970 mortality data from the National
Center for Health Statistics (NCHS75) to a cohort of 100,000 persons.
Additional information in the details of the life table analysis are
provided in Addendum B. It should be noted that a life table analysis
is required to use the age-dependent risk coefficients in the BEIR-3
report. For relative risk estimates, we use age-specific cancer
mortality data also provided by NCHS (NCHS73). The EPA computer program
we use for the life table analysis was furnished to the HAS BEIR-3
Committee by EPA and was used by the Committee to prepare their risk
estimates. Therefore, we are sure that the population base and
calculations are the same in both the WAS and EPA analyses.
To project the observed risks of most solid radiogenic cancers
beyond the period of current observation, we use both absolute and
relative risk models, but usually present an arithmetic average based on
these projections. Use of a single estimate instead of a range of
values does not mean that our estimate is precise. As indicated in
Table 8.2-2, the range of estimated fatal cancers resulting from the
choice of a particular projection model and its internal assumptions is
about a factor of three. Although we think it is likely that the
8-13
-------
relative risk model is the best projection model for most solid cancers,
it has been tested rigorously only for lung and breast cancer (Lab78),
Until it has more empirical support, we prefer to use an average risk
based on both projection models. A second reason for this choice is to
avoid overly conservative risk estimates caused by the compounding of
multiplicative conservative assumptions.
To estimate the cancer risk from low-LET, whole-body, lifetime
exposure with the linear model, we use the arithinetic average of
relative and absolute risk projections (the BEIR-3 L-L model) for solid
cancers and an absolute risk projection for leukemia and bone cancer
(the BEIR-3 L-L model). For dose to the whole body, this yields an
estimated 280 fatalities per million person rad. For the BEIR-3 linear
quadratic model, which is equivalent to assuming a DREF of 2,5, a low-
LET whole-body dose yields an estimated life risk of about 110
fatalities per million person rad.
These risk estimates are not unduly conservative. More than 235 of
the 280 fatalities estimated with the BEIR-3 linear model result from
cancers in soft tissues for which we have used the BEIR-3 L-L model. As
explained on page 187 of that report (NAS80), the L-L model is not
derived from the observed risk of solid cancers alone but rather
includes parameters based on the Committee's analysis of the leukemia
mortality data. Therefore, as outlined in 8.5, the BEIR-3 Committee's
analysis of the Japanese leukemia data depended heavily on the assump-
tion that most of the leukemia observed at Hiroshima was caused by
neutrons. In contrast, Table V-30 in the BEIR-3 report (NAS80) esti-
mates the risk of cancer incidence in soft tissues directly, without the
additional assumptions contained in the BEIR-3 L-L model. By using the
weighted incidence mortality ratios given in Table V-15 (NAS80), the
results given in Table V-30 (HAS80) can be expressed in terms of
mortality, to yield (for lifetime exposure) an absolute risk estimate of
about 200 fatalities per 1Q& person rad and about 770 fatalities per 10&
person rad when a relative risk projection model is used to estimate
lifetime risk. The arithmetic mean of the fatalities projected by these
two models is almost 500 per 10^ person rad, more than twice as many
fatal soft tissue cancers as predicted by the BEIR-3 L-L model and about
five times as many as estimated using the BEIR-3 linear quadratic model.
By a whole-body dose, we mean a uniform dose to every organ in the
body. In practice, such exposure situations seldom occur, particularly
for ingested or inhaled radioactivity. The next section describes how
we apportion this risk estimate for whole-body exposure when considering
the risks following the exposure of specific organs.
8.2.9 Organ Risks
For most sources of environmental contamination, inhalation and
ingestion of radioactivity are more common than direct exposure. In
many cases, depending on the chemical and physical characteristics of
the radioactive material, inhalation and ingestion result in a non-
uniform distribution of radioactive materials within the body so that
8-14
-------
some organ systems receive much higher doses than others. For example,
iodine isotopes concentrate in the thyroid gland, and the dose to this
organ can be orders of magnitude larger than the average dose to the
whole body.
Fatal Cancerat Specific Sites
To determine the probability that fatal cancer occurs at a
particular site, we have performed life table analyses for each cancer
type using the information on cancer incidence and mortality in NAS80.
For cancer other than leukemia and bone cancer we used NAS80 Table V-14
(Age Weighted Cancer Incidence by Site Excluding Leukemia and Bone
Cancer) and NAS80 Table V-15, which lists the BEIR Committee's estimates
of the ratio of cancer fatality to cancer incidence for these various
organs. The proportions of leukemia and fatal bone cancer caused by
low-LET radiation were estimated using the results given in Tables V-17
and V-20 of NAS80. Normalized results, which give the proportion of
fatal cancer caused by radiogenic cancer at a particular site, are
listed in Table 8.2-3, As noted above, these proportions are assumed to
be the same for the BEIR-3 linear quadratic dose respose model.
Information on the proportion of fatal cancers resulting from
cancer at a particular organ is not precise. One reason is that the
data in NAS80 (and in Table 8.2-3) are based on whole-body exposures,
and it is possible that the incidence of radiogenic cancer varies
depending on the number of organs exposed. Except for breast and
thyroid cancer, very little information is available on radiogenic
cancer resulting from exposure of only one region in the body. Another
reason is that most epidemiology studies use mortality data from death
certificates, which often provide questionable information on the site
of the primary cancer. Moreover, when the existing data are subdivided
into specific cancer sites, the number of cases becomes small, and
sampling variability is increased. The net result of these factors is
that numerical estimates of th?. total cancer risk are more reliable than
those for most single sites.
The 1977 UNSCEAR Committee's estimated risks (UNSCEAR77) to
different organs are shown in Table 8.2-4. For all of the organs except
the breast, a. high and low estimate was made. This range varies by a
factor of two or more for most organs, Table 8.2-4. Other site-specific
estimates show a similar degree of uncertainty (Kab82), and it is clear
that any system for allocating the risk of fatal cancer on an organ-
specific basis is inexact. Table 8.2-5 compares proportional risks by
the MAS BEIR-3 Committee, UNSCEAR, and the ICRP. ICRP Report 26
provides organ-specific weights for assessing combined genetic and
cancer risks, due to occupational exposure (ICRP77). In Table 8.2-5, we
have renormalized ICRP risks so that they pertain to cancer alone.
Considering that the cancer risk for a particular site is usually
uncertain by a factor of two or more, as indicated by the range of
UNSCEAR estimates in Table 8.2-k-t we would not expect perfect agreement
8-15
-------
Table 8.2-3. Proportion of the total risk of fatal
radiogenic cancer resulting from cancer
at a oartienlar si**
_ _ »£|J Hh. «
at a particular site
~ 11 i
Site
Proportion of Total
0.21
^
Red bone marrow^) A \c.
U. io
0.099
0.009
0.085
St°mach 0.084
Intestines 0>Q39
Pancreas 0.058
Kidneys and urinary tract Q Q25
Q;
)NAS80 - Lifetime exposure and cancer expression;
results are rounded to two figures.
^"^Average for both sexes.
^'Leukemia.,
Total risk for all other organs, including the
esophagus, lymphatic system, pharynx, larynx,
salivary gland, and brain.
8-16
-------
Table 8.2-4, UNSCEAR estimates of cancer risks at specified sites
Fatalities Average Proportion
site (106/person rad) (lO^/organ rad) of Total lisk
Lung
Breast^3)
Red bone marrow(b)
Thyroid
Bone
Liver
Stomach
Intestines
Pancreas
Kidneys and
urinary tract
Other^c'
25-50
25
15-25
5-15
2-5
10-15
10-15
14-23
2-5
2-5
4-10
37.5
25.0
20.0
10.0
3.5
12.5
12.5
18.5
3.5
3.5
7,0
0.24
0.16
0.13
0.065
0.023
0.081
0.081
0.12
0.023
0.023
0.046
(a)Average for both sexes.
'Includes tj ^phagus and lymphatic tissues.
Source: (UNSCEAR77).
8-17
-------
Table 8.2-5. Comparison of proportion of the total risk
of radiogenic cancer fatalities by body organ
Site
Lung
Breast
Red bone marrow
Thyroid
Bone
Liver
Stomach
Intestines
Pancreas
Kidneys and
urinary tract
Other
H.SSOU)
.21
.13
.16
.099
.009
.085
.084
.039
.058
.025
.ll<->
UNSCEAR
(UNSCEAR77)
.24
.16
.13
.065
.023
.081
.081
.12
.023
.023
.046
XCRP77<»
.16
.20
.16
.04
.04
(.Q8)
(.08)
(.08)
(.08)
(.08)
__
(fl'Lifetime exposure and cancer expression.
^'Normalized for risk of fatal cancer (see text).
(c)pive additional organs that have the highest dose are
assigned 0.08 for a total of 0.4.
(^'Others include esophagus, lymphatic system, pharynx,
larynx, salivary gland, and brain.
8-18
-------
in apportionment of total body risks. Table 8.2-5, however, does
indicate reasonable agreement among the three sets of estimates
considered here.
The differences between the proportions of the total risk of fatal
cancer shown in Table 8.2-5 are, for the most part, small in comparison
to their uncertainty. We have used the BEIR-2 organ risks in preference
to those made by other groups such as UNSCEAR or the ICKP for several
reasons. BEIR estimates of organ risk are based on a projection of
lifetime risk using age-specific risk coefficients, rather than just
observations to date. Moreover, the 1980 BEIR Committee considered
cancer incidence data as well as mortality data. This gives added
confidence that the diagnostic basis for their estimates is correct.
And, finally, because we apply these proportional organ risk estimates
to the NAS80 cancer risk estimates for whole-body exposures, we believe
it is consistent to use a single set of related risk estimates. The way
we have used NAS80 to estimate mortality resulting from cancer at a
particular site is outlined in the next section,
8.2.10 Methodology for Calculating theProportionof Mortality
Resulting fromLeukemia
Application of NAS80 to particular problems is straightforward but
requires some familiarity with the details of that report. In this
section we provide sample calculations based on the BEIR~3 linear dose
response model for the case of fatal leukemia resulting from irradiation
of the bone narrow throughout an average person's lifetime. We then
compared this number to the average number of all fatal radiogenic
cancers to obtain the proportion caused by leukemia, as shown in Table
8.2-3.
The NAS80 estimates in Table 8.2-3 differ from the others in that
they include both a consideration of age at exposure and a full
expression of radiogenic cancer resulting from lifetime exposure. For
example, Table V—17 (NAS80) gives explicit age- and sex-dependent
mortality coefficients for leukemia and bone cancer together.
The ratio of leukemia to bone cancer fatalities is given by the
coefficient in the dose response relationship listed in Table ¥-17, i.e.
2.24/0.05. For lifetime exposure at a dose rate of one rad per year,
Table V-17 lists 3,568 leukemia (and bone) deaths per 10^ males and
2,709 deaths per 10^ females (NAS80). Using a male-female birth ratio
of 1.05 to 1.0, this averages to 3,149 fatal cancers per million persons
in the general population. The total person rad causing these excess
fatalities is the product of one rad per yr, 10" persons, and 70.7 years
(the average age of this population at death). Dividing the total
number of fatalities by this product yields 44,5 fatalities per 10"
person rad of which about 43.5 are caused by leukemia. As noted above,
for total body exposure, the average of the absolute and relative risk
projection models yielded 280 per 10" person rad. Therefore. P, the
proportion of the whole-body risk caused by the lifetime risk of a
8-19
-------
leukemia death resulting from lifetime exposure of the red boae
marrow is:
0.16 (cf. with Table 8.2-3) (8-1)
To obtain the proportional mortality for other cancers, we have used the
site-specific, age-dependent risk coefficients in Table ¥-14 (NAS80) and
the mortality ratios in Table V-15 to calculate the risk of fatal cancer
from lifetime exposure at one rad per year (for each sex) and proceeded
as in the example for leukemia outlined above.
To apply the data shown in Table 8.2-3 to a particular organ we
multiply the average of the relative and absolute lifetime risk esti-
mates for whole-body lifetime exposure for a linear dose response, 280
fatalities per 10^ person rad, and 112 fatalities per 10& person rad for
a linear nuadratic response by the proportional mortality for that
cancer. For example, using the linear model, a one rad dose (low-LET)
to the kidney (urinary tract) resulting from lifetime exposure is
estimated to cause a lifetime probability of death caused by radiogenic
cancer that is equal to (.025) x (280x10^) or 7x10"^, i.e., 7 chances in
a million.
Iodine-131 has been reported to be only 1/10 as effective as X-rays
or gamma rays in inducing thyroid cancer (HAS72, NCRP77). For this
cancer a linear dose response and a DREF of 10 is used in calculating
lifetime probability of death. For example, the risk from a one rad
dose to the thyroid from exposure to iodine-131 or iodine-129 is
calculated as follows: (0.099) x (0.10) x (280xlO~6) or 2.8xlO~6, about
3 chances in a million.
8.2.11 Cancer Risks Due toAge-Dependent Poses
As noted in Chapter 7, almost all of the dose models we have used
are based on ICRP "Reference Man". (An exception is ".he case of radon
progeny where we use an age-dependent "exposure" mode , see below.) ICRP
dosimetric models are appropriate for adult workers and do not take into
account differences resulting from the changes in physiological
parameters between children and adults, e.g., intake rates, metabolism,
and organ size. Although it is difficult to generalize for all
radionuclides, in some cases these differences tend to counterbalance
each other. Foe exampla, the ratio of minute volume to lung mass is
relatively constant with age, so that the ICRP adult model for inhaled
insoluble materials provides a reasonably good estimate of the average
annual dose throughout life.
An exception is the thyroid where the very young have a relatively
high uptake of radioiodine into a gland which is much smaller than the
adult thyroid, as noted in Table 7,5-1. This results in a larger
8-20
-------
childhood dose and an increased risk which persists throughout life.
Since this is a worse case situation, we have examined it with some
care, using the age-specific risk coefficients for thyroid cancer in
Table V-14 of the BEIR-3 report (NAS80) and the age-dependent dose model
in ORNL84, For iodine-131 ingestion, the estimated lifetime time risk
is increased by a factor of 1.56 due to the 30 percent increase in
lifetime dose over that obtained with the OKNL adult model, cf. Chapter
7. Results are about the same for inhalation of iodine-131, the
estimated lifetime risk of fatal thyroid cancer by a factor of 1.63 for
ORNL's age-dependent dose estimate.
As noted .n Chapter 7, use of an age-dependent dosimetry for other
radionuclfdes nas yielded much smaller increased doses relative to adult
models and therefore has little effect on estimates of lifetime risk.
In particular, the lung dose and risk resulting from the inhalation of
insoluble alpha particle emitters is nearly unchanged. The lifetime
dose for an age-dependent dose model is only 1.09 times greater than
that calculated using an adult model (Chapter 7); the lifetime risk of
lung cancer for this age-dependent model is a factor of 1.16 greater
than we calculate for life exposure with the adult-only model. This is
important because, as noted in Volume 11 of this BID, such radionuclides
are the major cause of increased cancer resulting from the emission of
radionuclides into air.
EPA1 s age-dependent exposure model for radon progeny outlined in
Section 8.2 yields a 12 percent greater exposure than a lifetime
exposure using just the adult intake. The lifetime risk of lung cancer
for the more realistic exposure pattern is 22 percent greater. We have
concluded that with the possible exception of some iodine isotopes,
e.g., iodine-131, the use of the ICRP dosimetry does not contribute a
significant source of uncertainty in this rulemaking. We recognize,
howevar, that good physiological data for children is not available for
many radionuclides and that there may be other exceptions. These
exceptions will not include inhaled insoluble alpha-emitting
particulates.
8.3 Fatal Cancer Risk Resulting from High-LET Radiations
In this section we explain how EPA estimates the risk of fatal
cancer resulting from exposure to high-LET radiations. In some cases,
ingestion and inhalation of alpha particle emitting radionuclides can
result in a relatively uniform exposure of the body organs by high-LET
radiations. Unlike exposures to X-rays and gamma rays where the
resultant charged particle flux results in linear energy transfers (LET)
of the order of 0.2 to 2 keV per micron in tissue, 5 MeV alpha particles
result in energy deposition at a track average rate of more than 100 keV
per micron. High-LET radiations have a larger biological effect per
unit dosa (rad) than low-LET radiations. How much greater depends on
the particular biological endpoint being considered. For cell killing
and other readily observed endpoints, the relative biological effec-
tiveness (RBE) of high-LET alpha radiations is often ten or more times
greater thsn low-LET radiations.
8-21
-------
8*3-1 Quality Factors for Alpha Particl
es
for thei^effici0168 ^^f1 "?**«* V^^ty factors, Q, to account
tor their efficiency in producing biological damage. Unlike an RBE
value, which is for a specific and well-defined eadpoint, a quality
factor is based on an average overall assessment by radiatiofprotection
experts of potential ham of a given radiation relative to Tor ga»a
rediation. In 1977, the ICRP assigned a quality factor of 20 to'a^ha
particle eradiation from radionudides (ICRP77). The reasonableness of
is Lr^n^ £aCtur "^ fatal "di°g-ic ««c.r. at a particular site
« not well known, but it is probably conservative for all sites and
highly conservative for some,
th» Jhe do" eqUiV?*entf in ^e unit rea, is the dose, in rad, time.
the c«r°T ,S qU8i?ty faCt°r f°r a 8Pecif£ed *i°d of radiation. For
the case of internally deposited alpha particle emitters the dose
equivalent from a one-rad dose is equal to 20 rem. It should be noted
10 t0 Re°rt 2
Bartic- ,— o *p*
particle irradiation was ten. That is, the biological effect from a
acute dose nf f*?^*1* W estimated to be ten times that from an
?hf TPRP ! • !ow-LET.X-rays 0^ gamma rays of the same magnitude in rad.
their Lis?onT tVnCrea" this ^ality factor to 20 followed from
their decision to estimate the risk of low-LET radiations, in
occupational situations, on the assumption that biological effects were
reduced at low dose rates for low-LET radiation. The^e is general
agreement that dose rate effects do not occur for high-LET (alpha)
radiations. The new ICRP quality factor for alpha particles of 20
largely compensates for the fact that their low-LET risks are now based
on an assumed dose rate reduction factor of 2.5. This DREF has been
0' the risk per rad for aipha
of Inhl H Polished a task group report "Biological Effects
of Inhaled Radionuciides" which compared the results of animal experi-
ments on radiocarcinogenesis following the inhalation of alpha particle
and beta particle emitters (ICRP80). The task group concluded rtat "the
experimental animal data tend to support the decision by the ICRP to
change the recommended quality factor from 10 to 20 for alpha
radiation." F
8.3.2 Dose Response Function
In the case of high-LET radiation, a linear dose response is
commonly observed in both human and animal studies and that the response
is not reduced to low dose rates (NCRP80) . Some data on human lung
cancer indicate that the carcinogenic response per unit dose of aloha
radiation is higher at low doses than higher ones (AraSl, HobSl, Wh83)-
m«J? i°n' 80me -tUdieS With animals 8how the 8ame "sponse (ChaSl,
U182>._ Me agree with the MS BEIR-3 Committee that, "For high-LET
radiation, such as from internally deposited alpha-emitting radio-
nuciides, the linear hypothesis is less likely to lead to overestimates
of the risk and may, in fact, lead to underestimates" (NAS80). However
8-22
-------
at low doses, departures from linearity are small compared to the
uncertainty in the human epidemiologieal data, and we believe a linear
response provides an adequate model for evaluating risks in the general
environnent,
A possible exception to a linear response is provided by the data
for bone sarcoma (but not sinus carcinoma) among U.S. dial painters who
have ingested alpha-emitting radium-226 (NAS80). These data are
consistent with a dose squared response (Ro78). Consequently, the MAS
BEIR-3 Committee estimated bone cancer risk on the basis of both linear
and quadratic dose response functions. However, as pointed out in
NAS80, the number of U.S. dial painters at risk who received less than
1000 rad was so small that the absence of excess bone cancer at low
doses is not statistically significant. Therefore, the consistency of
these data with a quadratic (or threshold) response is not remarkable
and, perhaps, not relevant to evaluating risks at low doses. In con-
trast to the dial painter data, the incidence of bone cancer following
radium-224 irradiation, observed in spondylitics by Mays and Spiess
(Mab83, NAS80), in a larger sample at much lower doses, is consistent
with a linear response. Therefore, for high-LIT radiations the EPA has
used a linear response function to evaluate the risk of bone cancer.
Closely related to the choice of a dose response function is what
effect the rate at which a dose of high-LET radiation is delivered has
on its carcinogenic potential. This is a very active area of current
research. There is good empirical evidence, from both human and animal
studies, that repeated exposures to radium-224 alpha particles is five
times more effective in inducing bone sarcomas than a single exposure
which delivers the same dose (Mab83, NAS80). The 1980 HAS BEIR
Committee took this into account in their estimates of bone cancer
fatalities, which EPA is using. We do not know to what extent, if any,
a similar enhancement of carcinogenicity may occur for other cancers
resulting from internally deposited alpha particle emitters.
Nevertheless, we believe that the ICRP quality factor of 20 is
conservative, even at low dose rates.
8.3.3 AssumptionsMade by EPA for Evaluating the Doae from Alpha
Particle Emitters
We have evaluated the risk to specific body organs by applying the
ICRP quality factor of 20 for alpha radiations to the risk estimates for
low dose rate low-LET radiations described in Section 8.2.9. For
some organs this quality factor may be too conservative. Several
authors have noted that estimates of leukemia based on a quality factor
of 20 for bone marrow irradiation overpredicts the observed incidence of
leukemia in persons receiving Thorotrast (thorium oxides) (Moa79) and in
the U.S. radium dial painters (Spb83). Nevertheless, in view of the
paucity of applicable human data and the uncertainties discussed above,
the ICRP quality factor provides a reasonable and prudent way of
evaluating the risk resulting from alpha emitters deposited within body
organs.
8-23
-------
a liniiJ dL f estimates for high-LET radiations are based an
the absol^r 5efP°nse Action. For bone cancer and leukemia we use
the absolute risk projection model described in the previous section.
" we use the arithraetic average °f
result in^fro,!"1 in?jcate8 EPA'« «ti-te. of the risk of fatal cancer
resulting from a uniform organ dose in various organs from internally
riralf pha,r?icles- it was prepared by -'tipiyui thr^e
dose of low T°?T % 6ar m°del f°r a uniformly Distributed whole-body
dose of low-LET radiation and, unlike the DBID (EPA83a), a dose rate
"^" f f °f 2'5) by 3 qUality fact°r of 20 a^ then Appor-
this risk by organ, as indicated in Table 8.3-1. These esti-
mates are for Hfetime doses at a congtant
Table 8.3-1. Estimated number of cancer fatalities
from a lifetime exposure to internally deposited
alpha particle emitters
„.. , . Fatalities per
5lte Proportional Risk^3) 1Q& organ
460
Red bone marrow^) _ jg
Thyroid .099
tU
Liver .085 190
Stomach -OB4 igQ
Intestine >039 gQ
Pancrea« .058 130
Kidneys and
urinary tract .Q25 55
Other-Sum (total) .n
^)proportion of whole body risk from Table 8.2-3.
'Rounded to two figures. Note that these estimates are 2 5
times smaller than those used in preparing the DBID.
^^Average for both sexes.
-"'Leukemia.
»-e'Booe endosteum as defined in ICEP-30 (ICRP79).
8-24
-------
was not followed for bone cancer. As outlined above, the risk estimate
for this cancer in the BEIR-3 report (NAS80) is based on data for high-
LET (alpha) radiation.
Some readers may note that the risk estimate in Table 8.3-1, about
20 bone cancer fatalities per 10^ person rad, is lees than the 27
fatalities listed in Table A-27 of NAS80 for alpha particles. This is
because the analysis in Appendix A of KAS80, but not Chapter V of that
report, assumes that in addition to a 2-year minimum induction period,
27 years are available for cancer expression. This is usually not the
case for doses received beyond middle age. Hence, the estimated life-
time risk is smaller when it is based on a life table analysis that
considers lifetime exposure in conjunction with death from all causes.
In the next section, we describe how we estimate the risk resulting
from inhalation of alpha-emitting radon progeny, a situation where the
organ dose is highly nonunifora.
8.4 Estimating the Risk Resulting from Lifetime Population Exposures
from Radon-222 Progeny
EPA estimates of the risk of lung cancer resulting from inhaled
radon progeny do not utilize the dosimetric approach, outlined above,
but are based on what is sometimes called an epidemiological approach.
In this approach the amount of excess human lung cancer in groups known
to have been exposed to radon progeny is determined.
When radon-222 (a radioactive atomic gas) decays, a number of short
half-life radionuclides, principally polonium-218, lead-214, bismuth-
214, and polonium-214, are formed that attach to inhalable dust par-
ticles in air. When inhaled, the dust containing the radon progeny
plates out on the surfaces of the larger bronchi of the lung. Since
two of these radionuclides decay by alpha particle emission, bronchial
epithelium is irradiated by high-LET radiation. A wealth of data
indicate that a range of exposures to the bronchial epithelium of
underground miners causes an increase in bronchial lung cancer, both in
smoking and in nonsmoking miners. Two recent reviews on the underground
miner experience are of particular interest. The 1980 HAS BEIR-3 Report
(NAS80) contains a review of the epidemiological studies on these
miners. Thomas and McNeil (Th82) reanalyze many of these epidemiologi-
cal studies in a consistent fashion so that the modeling assumptions are
the same for all of the data sets.
Although considerable progress has been mads in modeling the
deposition of particulate material in the lung (Hac82, JaaSO, JacSl), it
is not yet possible to adequately characterize the bronchial dose
delivered by alpha radiation from radon-222 progeny attached to dust
particles. This is because of the lack of knowledge concerning the
kinds of cells in which bronchial cancer is initiated (Mc78) and the
depth of these cells in the bronchial epithelium. Current estimates of
the dose actually causing radiogenic cancer resulting from inhaled
radon-222 progeny are based on average doses that may or may not be
8-25
-------
relevant. Until more reliable estimates of the bronchial dose become
available, we are following the precedents set in the 1972 and 1980 MAS
reports (NAS72, NAS80) and are estimating the risk resulting from radon-
222 progeny on the basis of exposure rather than dose per ge, This is
called the epidenioiogical approach, i.e., risk is estimated on the
basis of observed cancers following occupational exposure to radon
progeny,
8.4.1 Characterizing Exposures Co theGeneral Population vis-a-vis
UndergroundMiners
Exposures to radon under working conditions are commonly reported
in a special unit called the working level (WL). One working level is
any concentration in air of short half-life radon-222 progeny having 1.3
x 1Q5 MeV per liter of potential alpha energy (FRC67). This unit was
developed because the concentration of specific radon progeny depends on
ventilation rates and other factors, A working level month (WLM) is the
unit used to characterize a miner's exposure to one working level of
radon progeny for a working month of about 170 hours. Because the
results of epidemiological studies are expressed in units of WL and WLM,
we outline below how they can be interpreted for members of the general
population exposed to radon progeny.
For a given concentration of radon progeny, the amount of potential
alpha energy inhaled in a month by a member of the general population is
more than that received in a miner's working month. These individuals
are exposed longer, up to 24 hours per day, 7 days a week. However, the
average amount of air inhaled per minute (minute volume) by a member of
the general population is less than the amount for a working miner when
such activities as sleeping and resting are taken into account. To
compare the radon progeny exposure of a working miner to a member of the
general population, we have calculated the amount of potential alpha
energy each inhales per year.
We have assumed that (averaged over a work day) a miner inhales 30
liters per minute. This average corresponds to about 4 hours of light
activity and 4 hours of moderately heavy work per day (ICRP75). We
recognize that the new ICRP radon model assumes a 20 liter per minute
volume for miners, which corresponds to 8 hours of light activity per
day (ICRP81). Although this may be appropriate for nuclear workers,
studies of the metabolic rate of working miners clearly show that they
are not engaged only in light activity (Spa56, ICRP75, NASA73).
Therefore, we have chosen 30 liters as a more realistic estimate of
their average minute volume. A working miner with this minute volume
inhales 3.6 x 10-* cubic meters in a working year of 2000 hours (ICRP79).
One working level of radon-222 progeny is 2.08 x 1Q~5 Joules per cubic
meter. Therefore, in a working year the potential alpha energy inhaled
by a miner exposed to one working level is 7.5 x 10~^ Joules.
For adult males and females in the general population we follow the
ICRP Task Group on Reference Man (ICRP75) in assuming an inhaled air
volume of 2.3 x 10^ liters per day for males and 2.1 x 10^ liters per
8-26
-------
day for adult females, an average of 2.2 x 10^ liters per day. This
average volume results in 1.67 x 1Q~1 Joules per year of inhaled poten-
tial alpha energy from an exposure to one working level of radon-222
progeny for 365.25 days. Although it may be technically inappropriate
to quantify the amount of potential alpha particle energy inhaled by a
member of the general population in working level months, this amounts
to an annual exposure equivalent to 27 WLM (26.7) to an adult member of
the general population exposed 2A hours per day. For indoor exposure,
we have assumed an occupancy £ Jtor of 0.75 ao that an exposure to one
WL results in an annual exposure equivalent to 20 WLM (EPA78) in terms
of the amount of potential alpha energy actually inhaled.
Children have a smaller bronchial area than adults, which more than
offsets their lower minute volume, so that the dose to their bronchi,
for a given concentration of radon progeny, is greater. This problem
has been addressed by Hofmann and Steinhausler (Hoa77). Their results
indicate that exposures received during childhood are about 50 percent
greater than adult exposures. We have used the information in (Hoa77)
to prepare Table 8.4-1, which lists the age-dependent potential alpha
energy exposure we have used in the risk assessments listed below.*
Table 8.4-1. Potential alpha energy inhaled during
one year of exposure to one working level
(2.08 x 10~5 joules per cubic meter) as a
function of age by a member of the
general population'3'
Age
(years)
0-2
3-5
6-11
12-15
16-19
20-22
23 or more
Joules
0.22
0.27
0.30
0.27
0.24
0.20
0.17
«u>
35
43
49
43
38
32
27
^'Assuming a WLM corresponds to 6,2 x 10 •* Joules
of potential alpha particle energy inhaled
f *.
(see text).
Source: (Hoa77).
*The assumptions on minute volume, etc. for miners and the general
population described above are the same as those used in the preparation
of the EPA reports (EPA79,82,83a,b).
8-27
-------
The results in Table 8.4-1 have been rounded to two significant figures.
The larger exposure to children relative to adults increases the esti-
mated mortality due to lifetime exposure from birth by about 20 percent.
We have also examined the exposure model described above in terras
of the average dose delivered to bronchial tissue using the most
detailed dose model available the five-lobe lung model developed by
Harley and Pasternack (Kac82). For the breathing patterns we have
assumed for each group, the bronchial dose per WLM for working miners is
0.64 rad, and is 0.5] rad for an adult member of the general population
(Had83) . Therefore, we have concluded that the factors not included in
our simple model, such as the fraction of unattached radon progeny, are
not very important compared to other- sources of uncertainty in our risk
estimates .
8.4.2 The EPA Model
Since 1978, EPA has based risk estimates of cancer resulting from
inhaled radon-222 progeny on a linear dose response function, a relative
risk projection model, and a minimum induction period of 10 years.
Lifetime risks are projected on the assumption that exposure to 1 WLM
increases the age-specific risk of lung cancer by 3 percent over the
age-specific rate in the U.S. population as a whole. The life table
analysis described in the annex ti this chapter is used to project this
risk "over a full life span.
The EPA cGuel has been described in detail (EPA79, Elb79). In
reviewing this model in terms of the more recent information described
below, we have found that our major assumptions, linear response and
relative risk projection, have been affirmed. The A-bomb survivor data
clearly indicate that the absolute risk of radiogenic lung cancer has
continued to increase among these survivors while their relative risk
has regained reasonably constant (Kab82). The UNSCEAR, ICRP, and 1980
HAS Committee have continued to use a linear dose response to estimate
the risk of lung cancer resulting from inhaled radon progeny. Thomas
and McNeill's analysis (Th82) indicates that the use of linearity is not
unduly conservative and may, in fact', underestimate the risk at low
doses. As noted above, the 1980 NAS BE1R Committee reached a similar
conclusion.
A major limitation of the EPA model is the uncertainty in the
relative risk coefficient we have used, 3 percent increase per WLM.
This value is based on the excess mortality resulting from lung cancer
among exposed miners of various ages, many of whom smoked. Therefore,
it is an average value for a mixed population of smokers and nonsraokers,
Furthermore, the fact that smoking was more prevalent among some of the
groups of miners studied than it is among the U.S. general population
today, this may lead to an overly conservative risk estimate as
discussed below.
In a recent paper, Radford and Remard (Ra84) reported on the
results of a long-term study of Swedish iron miners who were exposed to
8-28
-------
radon progeny. This study is unique in that most of the miners were
exposed to less than 100 MLM, and the risks to smokers and nonstnokers
was considered separately. The absolute risk of the two groups was
similar, 20 fatalities per 10^ person WLM year for smokers compared to
16 for nonsmokers» The total number of lung cancer fatalities for
nonsmokers is small; so that the estimate of 16 is not too reliable.
While absolute risks were comparable for the smoking and nonsmoking
miners, relative risks were not. Nonsmokers have a much lower baseline
incidence of lung cancer mortality than smokers. This resulted in a
relative risk coefficient for nonsmoking exposed miners relative to
unexposed nonsmokers that was four times larger than the relative tisk
coefficient for exposed smokers. However, this larger relative risk
does not fully compensate for the lower base line incidence of non-
smokers. Therefores this study of Swedish iron miners indicates that a
3 percent per WLM relative risk coefficient may be too conservative when
appied to the population as a whole. Further follow-up of this and
other mining groups may provide more reliable data on the risk to non-
smckers and we expect to incorporate separate consideration of smokers
and non-smokers into EPA analyses as more data becomes available.
Although occupational exposures to pollutants other than radon-222
progeny are probably not important factors in the observed iung cancer
risk for underground miners (Elb79, Th82, Mua83, Ra84), the use of
occupational risk data to estimate the risk of a general population is
far from optimal, as it provides no information on the effect of radon
progeny exposures to children and women. Although we have continued to
assume that the risk per unit exposure during childhood is no more
effective than that occurring to adults, this assumption may not be
correct. The A-bomb survivor data indicate that, in general, the risk
from childhood exposure to low-LET radiation is greater and cortinues
throughout life (Kab82). There are no specific data for lung cancer yet
(Kab82). Another limitation of the underground miner data is the
absence of women in the studied populations. The A-bomb survivor data
indicates that women are as sensitive as men to radiogenic lung cancer
even though, on the whole, they smoke less (Pr83). These data are not
conclusive, however.
8.4.3 Comparison of Risk Estimates
Several estimates of the risk resulting from radon progeny have
been published since the EPA model was developed. One of particular
interest was expounded by the BEIR Committee (NAS80). The BEIR-3
Committee formulated an age-dependent absolute risk model with
increasing risk for older age groups. The Committee estimates of the
risk per WLM for various ages are listed on page 325 in HAS80, and their
estimated minimum induction period for lung cancer following exposure on
page 327. We have used these data, summarized in Table 8.4-2, to
calculate the lifetime risk of lung cancer mortality from lifetime
exposure to persons in the general population by raeaiis of the same life
table analysis used to calculate other EPA risk estimates.
8-29
-------
Table 8.4~2. Age-dependent risk coefficients and minimum
induction period for lung cancer resulting from
inhaling radon-222 progeny
Age
(yr)
0-14
15-34
35-49
50-64
65 or greater
Excess
(casee per 10^
WLM person years)
0
0
10
20
50
Minimum induction period
(years)
25
25-15
10
10
10
Source; NAS80.
It should be noted that the zero risk shown in Table 8.4-2 for
those under 35 years of age at exposure does r.ot mean no harm occurs but
rather that it is expressed after the person is more than 35 years old,
i.e., only after the minimum induction period. The sequence of
increasing risk with age shown in Table 8.4-2 is not unlike the increase
in lung cancer with age observed in unexposed populations, so that the
pattern of excess risk over time is similar to that found using a
relative risk projection model,
Thomas and McNeil conducted a thorough analytical investigation of
lung cancer among uranium miners for the AECB of Canada (Th82). These
investigators tested a number of risk models on all of the epidera-
iological studies that contained enough data to define a dose response
function. They concluded that, for males, a 2.3 percent increase in
lung cancer per WLM and a relative risk projection model were more
consistent with the eKcess lung cancer incidence observed in underground
miner groups than other models they tested. This is the only analysis
we are aware of that treated each data set in consistent fashion and
utilized modern epidemiological techniques, such as controlling, to the
extent possible, for age at exposure and duration of follow-up.
The AECB risk estimates for lifetime exposure to a general popula-
tion along with EPA, NAS, UNSCEAR, ICRP, and NCRP estimates of the risk
of lung cancer resulting from inhaled radon progeny are listed in Table
8,4-3. The AECB estimate for lifetime exposure to Canadian males is 830
fatalities per million person WLM (Th82), In Table 8.4-3 this estimate
has been adjusted for the U.S. 1970 population of males and females.
8-30
-------
Table 8.4-3. Risk estimate for exposures to radon progeny(a'
Organization
Fatalities per
person WLM
Exposure Period Expression Period
EPA
HAS BEIR-3^
AECB^J
ICRP
UNSCEAR
NCRpW)
760
730
600
150-450
200-450
130
Lifetime
Lifetime
Lifetime
Working Lifetime
Lifetime
Lifetime
Lifetime
Lifetime
Lifetime
30 years
40 years
Lifetime
<«>The number of fatalities per 106 person WLM listed for EPA and NAS80
in this table differs from figures we have previously published
(e.g., EPA83b) because we have now included, correctly we believe,
the increased potential alpha energy exposure during childhood in the
denominator of this ratio. Our risk estimates for various sources of
radon in the environment have not changed, because all were calcu-
lated via a life table analysis yielding deaths per 100,000 exposed,
not deaths per person WLM.
(b>Assumes increased exposure during childhood, Table 8.4-1.
(c)Adjusted for U.S. general population, see text.
(•^Assumes that risk diminishes exponentially with a 20-year half-life.
Source: EPA, EPA83b; HAS BEIR-3, NAS80; AECB, Th8?.; ICRP, ICRP81;
UNSCEAR, UUSCEAR77J NCRP, NCRP84, USRPC80.
The agreement between the EPA, BEIR-3, and the AECB estimates shown
in Table 8,4-3 is not unexpected. Each estimate is based on lifetime
exposure and lifetime expression of the incurred risk. In contrast, the
three lower risk estimates in Table 8.4-3 do not explicitly include
these factors.
The IGRP estimates are for occupational exposure to working adults.
The larger ICRP estimate is based on their epidendological approach,
that ist the exposure to miners in WLM and the risk per WLM observed_in
epidemiological studies of underground miners. The ICRP epidemiological
approach assumes an average expression period of 30 years for lung
cancer Children, who have a much longer average expression period, are
excluded from this estimate. The ICRP has not explicitly projected the
risk to miners beyond the years of observation, even though most of the
miners on whom their estimates are based are still alive and continuing
to die of lung cancer.
8-31
-------
The smaller of the two ICRP estimates listed in fable 8.4-3 is
based on their dosimetric approach. The ICRP assumes that the risk per
rad for lung tissue is 0.12 of the risk of cancer and genetic danage
following whole-body exposure (ICRP77). For the case of exposure to
radon progeny, the ICRP divided this factor of 0.12 into two equal
parts. A weighting factor of 0.06 was used to assess the risk from the
high dose to bronchial tissue, where radiogenic lung cancer is observed
in exposed underground miners. The other half of the lung weighting
factor, another 0.06 of the total body risk, was used to assess the risk
to the pulmonary region, which receives a comparatively small dose from
radon-222 progeny and where human lung cancer is seldom, if ever,
observed.
The UNSCEAR estimate is for a general population and assumes an
expression time of 40 years. Like the ICRP, UNSCEAR did not make use of
an explicit projection of risk of fatal lung cancer over a full
lifetime.
The last entry in Table 8.4-3, the NCRP risk estimate based on an
analysis by Harley and Pasternack (USRPC80, Hab82), is of particular
interest because, like the EPA and AECB risk estimates, it is based on a
life table analysis of the lifetime risk resulting from lifetime
exposure. This estimate utilizes an absolute risk projection model with
a relatively low risk coefficient, 10 cases per 10^ person WLM per year
at risk, the smallest of those listed by the HAS BEIR-3 Committee, cf.
Table 8.4-2. Moreover, they have assumed that the risk of lung cancer
following irradiation decreases exponentially with a 20-year half-life
so that exposures occurring early in life are of very little risk. The
NCRP assumption of a 20-year half-life for radiation injury reduces the
estimated lifetime risk by about a factor of 2.5, Without this
assumption, the NCRP risk estimate would be the same as the midpoint of
the UNSCEAR estimate about 325 fatalities per million person WLM. We
find this assumption particularly troublesome. If lung cancer risk
decreased over time with a 20-year half-life, the excess lung cancer
observed in Japanese A-bomb survivors would have decreased during the
period they have beetv followed, 1950 to 1982. During this period their
absolute lung cancer risk has markedly increased (Kab82).
Table 8.4-3 clearly indicates the wide divergence in risk estimates
for exposure to radon progeny. In such cases, use of a single risk
coefficient may indicate to some that this risk is well known when this
obviously not the case. The EPA and AECB estimates may be high because
they are relative risk estimates based on males, many of whom smoked.
The actual risk to a population which includes women and nonsmokers may
be smaller, but it is unlikely to be as small as estimated using the
NCRP model. Therefore, on the basis of the BEIR-3, EPA, NRPB, UNSCEAR,
and ICRP analyses, risk estimates between 700 and 300 fatalities per
million person WLM are reasonable estimates for the possible range of
effects resulting from inhaling radon progeny for a full lifetime of
exposure. These two risk estimates do not encompass the full range of
uncertainty, but do seem to illustrate the breadth of much of current
scientific opinion.
8-32
-------
8.5 Uncertainties in Risk Estimates for Radiogenic Cancer
As pointed out in the introduction of this chapter, numerical
estimates of risks resulting from radiation are neither accurate nor
precise. A numerical evaluation of radiogenic cancer risks depends both
on epidemiological observations and a number of ad hoc assumptions that
are largely external to the observed data set. These assumptions
include such factors as the expected duration of risk expression and
variations in radiosensitivity as a function of age and demographic
characteristics. A major assumption is the shape and slope of the dose
effects response curve, particularly at low doses where there are little
or no epidemiological data. In 1972, the BEIR Committee based its
estimates of cancer risk on the assumption that effects at low doses are
directly proportional to those observed at high doses, the so called
linear-nonthreshold hypothesis. As described above in 8.2, the BEII-3
Committee considered three dose response models and indicated a prefer-
ence for the linear quadratic model. The risk coefficients that the
BEIR-3 Committee derived for their linear quadratic model, and to a
lesser extent for their linear model, are subject to considerable
uncertainty primarily because of two factors: (1) systematic errors in
the estimated doses of the individual A-bomb survivors and (2) statis-
tical uncertainty because of the small number of cancers observed at
various dose levels.
8.5.1 Uncertainty in the Dose Response Models Resulting from Bias in
the A-bomb Dosiroetry
Although the BEIR-3 Committee's choice of a linear-quadratic
response has gained considerable attention, it may not be generally
realized that the BEIR-3 Cotnramittee1s numerical evaluations of dose
response functions for cancer resulting from low-LET radiation were
based exclusively on the cancer -ortality of the A-botnb survivors.
Unfortunately, the dosimetry for A-bomb survivors, on which the BEIR-3
Committee relied, has since been shown to have large systematic errors
that undermine the analyses made by the Committee. As outlined below,
the mathematical analyses made by the Committee were "constrained" to
meet certain a priori assumptions. These assumptions have since been
shown to be doubtful.
A careful state-of-the-art evaluation of the dose to A-botnb
survivors was carried out by investigators from Oak Ridge National
Laboratory in the early 1960s (Aua67, Aub77). These studies resulted in
a "T65" dose being assigned to the dose (kerma) in free air at the
location of each survivor for both gamma rays and neutrons. A major
conclusion of the ORNL study (Aua67, Aub77) was that the mix of gamaa
ray and neutron radiations was quite different in the two cities where
A-bombing occurred. These results indicated that at Hiroshima the
neutron dose was more important than the gamma dose when the greater
biological efficiency of the high-LET radiations produced by neutrons
was taken into account. Conversely, the neutron dose at Nagasaki was
shown to be negligible compared to the gamma dose for that range of
doses where there were a significant number of survivors. Therefore,
8-33
-------
the 1980 BEIR Committee evaluated the cancer risks to the survivors at
Hiroshima on the assumption that the combined effects of gamma rays and
particularly neutrons caused the observed cancer response.
Since the BEIR-3 report was published, it has become evident that
the organ doses resulting from neutrons at Hiroshima were overestimated
by about an order of magnitude, at distances where most of the irradi-
ated persons survived the bomb blast and yet received significant doses,
1000-1500 meters. In fact, the neutron doses at Hiroshima are quite
comparable to those previously assigned, at similar distances, to
Nagasaki survivors (Keb81a,b; RERF83,84). Moreover, there are now
grounds to believe that the T65 estimates of gamma ray doses in both
cities are also incorrect (RERF83,84). Although several factors need
further evaluation, reduction of the gamma dose to individual survivors
because of the local shielding provided by surrounding structures is
significant. The important point, however, is that the overestimate of
the neutron dose to the Hiroshima survivors led to the BEIR-3 Committee
attributing most of the risk to neutrons rather than gamma-rays. Hence,
they underestimated the risk for low-LET radiations by an as yet unknown
amount .
For their analysis of the A-bomb survivor data, the BEI1-3
Committee expanded the equations for low-LET radiations listed above in
8.2 to include a linear dose response function for neutrons:
p(d,D) = cjd + kjD (8-2)
?(d,D) » c2d2 + k2D (8-3)
P(d,D) = C3d + C4d2 + kjD (8-4)
where d is the gamma dose and D is that part of dose resulting from
high-LET radiations from neutron interactions. Note that in equation
(8-4) the linear-quadratic (LQ) response has two linear terms, one for
neutrons and one for gamma radiation. In analyzing approximately linear
data in terms of equation (8-4), the decision as to how much of the
observed linearity should be assigned to the neutron or the gamma
component, i»e>» &3 a**d ^3 respectively, is crucial. As shown below,
the BEIR-3 Committee attributed most of the observed radiogenic cancer
to a linear response from neutron doses that did not occur.
The BEIR-3 Committee's general plan was to examine the dose
response for leukemia and for solid cancer separately to find statis-
tically valid estimates of the coefficients cj ..... c^ and k^ ..... k3 by
means of regression analyses. The regressions were made after the data
were weighted in proportion to their statistical reliability; thus,
Hiroshima results dominate the analysis. The T65 neutron and gamma
doses to individual survivors are highly correlated because both are
8-34
-------
strongly decreasing functions of distance. This makes accurate deter-
mination of the coefficients in equation (8-4) by means of a regression
analysis extremely difficult. In addition, there is considerable
saispling variation in the A-bomb survivor data because of small sample
sige, which exacerbates the regression problem, Herbert gives a
rigorous discussion of these problems for the case of the A-bomb
survivors (He83). On account of these and other problems, agreement
between the observed response for solid cancers and that predicted by
any of the dose response functions examined by the BEIR-3 Committee is
not impressive. For example, goodness of fit, based on Chi square,
ranges from 0.20 for equation (8-3) to 0.23 for equation (8-4), to 0.30
for equation (8-2) (Table V-ll).* For leukemia, the goodness of fit
between the observed data and those predicted by the regression analysis
is better, e.g. 0.49 for equations (8-2) and (8-3) (Table V8 in NAS80).
The Committee analyzed the A-bomb survivor data in two separate
sets, i.e. first leukemia and then all cancer excluding leukemia (solid
cancers). Their treatment of these two cases was not equivalent.
Unlike the analysis of solid cancers, the Committee's analysis of
leukemia considered the Nagaski and Hiroshima data separately. Their
approach (p. 342 in NASBQ) appears to be based on an unpublished paper
by Charles Land and a published report by Ishimaru e_t al., on estimating
the RBE of neutrons by comparing leukemia mortality in Hiroshima to that
in Nagasaki (Is79). Unlike the case for solid cancers, see below, the
Committee's regression analysis of the leukemia mortality data did
provide stable values for all of the coefficients in equation (8-4), and
therefore an RBE for neutrons as & function of dose, as well as the
ratio of the linear to the dose squared terms for leukemia induction
caused by gamma rays, (cg/c^).
Estimating the linear-quadratic response coefficients for solid
cancers proved to be less straightforward. When tha BEIR-3 analysis
attempted to fit the A-bomb survivor data on solid cancers to a linear-
quadratic dose response function, they found that the linear response
coefficient, c-j in equation (8-4), varied from zero to 5.6 depending on
the dose range considered. Moreover, their best estimate of the
coefficient for the dose squared term in equation (8-4), i.e., c^ was
zero, i.e., the best fit yielded a linear response. Therefore, it was
decided that the observations on solid cancers were "not strong enough
to provide stable estimates of low-dose, low-LET radiation cancer risk
when analyzed in this fashion" (NAS80, p. 186).
As outlined in the BEIR-3 Report, the Committee decided to use a
constrained regression analysis, that is, substitute some of the
parameters for equation (8-4) found in their analysis o€ leukemia deaths
to the regression analysis of the dose response for solid cancers. That
is, both the neutron RBE at low dose (the ratio of the coefficient k3 to
*A11 references to tables with a V prefix are from Chapter V in
NAS80.
8-35
-------
C3) and the ratio of 03 ta 04 as estimated from the leukemia data were
assumed to apply to the induction of fatal solid cancers. Regression
analyses that are constrained in this isanner can yield much higher
estimates of precision than are warranted by the data, as discussed by
Land and pierce (Lac83). They can also be very misleading. Herbert has
discussed this point in detail as it applies to the BEIR-3 regression
analysis (He83). The BEIR-3 Committee's substitution of the results of
the leukemia regression for the data on solid cancers allowed then to
make stable estimates of cj, 04, spd k3- These estimates became the
basis for the "preferred" linear quadratic risk estimates for solid
cancers presented in NAS80, i.e., the LQ-L model* (NAS8Q, p. 187).
Given the information discussed above, it is possible to see, at
least qualitatively, how the high bias in the estimated T65 neutron dose
to the Japanese survivors affects the 1980 BEIE Committee's "preferred"
LQ estimates of the risk coefficients for leukemia. The Committee's
age-adjusted risk coefficients for leukemia are listed in Table ¥-8.
For the linear-quadratic response, k3, the neutron risk coefficient is
27.5. Tables A-ll and V-6 provide the estimates of neutron and gamma
doses to the bone marrow of Hiroshima survivors that were used by the
Committee. Substituting these doses in their risk equations (Table V-8)
indicates that about 70 percent of the leukemia deaths were ascribed to
the neutron dose component then thought to be present at Hiroshima. As
noted above, subsequent research indicates that the high-LET dose caused
by neutrons was actually much smaller.
It is not possible to accurately quantify what effect the
Committee's, use of these same coefficients had on their analysis of the
dose response for solid cancers. Equation V-10 for solid cancers
(NAS80, p. 187) indicates that about 60 percent of the solid tumor
response was attributed to the T65 neutron dosej but this is a minimum
estimate that ignores the effect of the assumed neutron doses on the
value of k3 and the ratio of 03 to c^.
The BEIR-3 Cotmnittee's LQ-L model assumes an SBE of 27.8 at low
doses. In the Committee's L-L linear response model, the assumed RBE is
11.3. Therefore, this linear model is considerably less sensitive to
the neutron dose component, assumed by the Committee, than their LQ-L
model. For either model, most of the A-bomb survivors' radiogenic
cancer was ascribed to the T65 neutron doses at Hiroshima.
There is no simple way of adjusting the 1980 BEIR risk estimates to
account for the risk they attributed to neutrons. Adjustment of neutron
doses alone is clearly inappropriate, because there is good reason to
believe that T65 estimates of the dose caused by gamma rays are also
subject to considerable change. Moreover, not all of the individuals in
*The response models for solid cancers that are based on the
Committee's constrained regression analysis are designated with a bar in
their 1980 report, e.g., LQ-L and L-L.
8-36
-------
a given T65 dose category will, necessarily, remain grouped together
after new estimates of neutron and gamma doses are obtained. Both the
numerator and denominator in the ratio of observed to expected cases are
subject to change, and indeed could change in opposite directions, a
fact not considered in some preliminary (and premature) analyses
(StcSl). Nevertheless, it is reasonable to conclude that bias in the
estimated neutron doses at Hiroshima has not only led to considerable
uncertainty in the BEIR-3 risk estimates but has also led to a sys-
tematic underestimation of the risk resulting from low-LET radiations.
For this reason we believe that estimates based on the more conservative
linear dose response should be given considerable weight vis-a-vis those
made using the BEIR-3 linear quadratic models.
8.5.2 Sampling Variation
In addition to the systematic bias in the BEIR-3 risk estimates for
low-LET radiation outlined above, the precision of the estimated linear
and quadratic risk coefficients in the BEIR-3 report is poor as a result
of statistical fluctuations caused by sample size. Recently Land and
Pierce (Lac83) have reevaluated the precision of the BEIR-3 linear
quadratic risk estimates to take account, at least partially, of the
Committee's use of a constrained regression analysis. This new analysis
indicates that for the BEIR-3 LQ-L model for leukemia, the standard
deviation of the linear term is nearly as large as the risk coefficient
itself (+0.93 compared to a risk coefficient of 0.99). For the LQ-L
model, solid cancer, the standard deviation is ^1.5 compared to a risk
coefficient of 1.6.
It is likely that at least part of the uncertainty attributed to
sampling variation in the BEIR-3 risk estimates is caused not by sample
size and other factors leading to random error but rather by the use of
incorrect dose estimates for the A-bomb survivors. The correlation of
neutron and gamma-ray doses has been a major underlying cause of the
uncertainty in regression analysis using the T65 doses. Analyses of
revised data with much smaller neutron doses may result in better
precision. At present, we have concluded that the BEI1-3 risk
coefficients are uncertain by at least a factor of two, see below, as
well as being biased low by an additional factor of two or more.
8.5.3 Uncertainties Arising from Model Selection
In addition to a dose response model, a "transportation model" is
needed to apply the risks from an observed irradiated group to another
population having different demographic characteristics. A typical
example is the application of the Japanese data for A-bomb survivors to
western people. Seymore Jablon (Director of the Medical Follow-up
Agency of the National Research Council, HAS) has called this the
"transportation problem," a helpful designation because it is often
confused with the risk projection problem described below. However,
there is more than a geographic aspect to demographic characteristics.
The "transportation problem" includes estimating the risks for one sex
based on data for another and a consideration of habits influencing
8-37
-------
health status such as differences between smokers and nonsmokers, as
described in 8.4 for the case of risk estimates for radon progeny.
The BEIR-3 Committee addressed this problem in their 1980 report
and concluded, based largely on the breast cancer evidence, that the
appropriate way to transport the Japanese risk to the U.S. population
was to assume that the absolute risk over a given observation period was
transferrable but that relative risk was not. Therefore, the GoBaaittee
calculated what the relative risk would be if the same number of excess
cancer deaths were observed in a U.S. population having the same age
characteristics as the A-bomb survivors. The baseline cancer rates in
the U.S. and Japan are quite different for some specific cancers, so
this is a reasonable approach. However it contains the assumption that
while the cancer initiation process is the same in the two countries,
the actual number of radiogenic cancers that actually occur is the
result of cancer promotion, the latter being a culturally dependent
variable.
An alternative approach to solving the "transportation problem" is
that of the 1972 NAS BEIR-1 Committee. This Committee assumed that
relative risks would be the same in the U.S. and Japan and transferred
the observed percentage increase directly to the U.S. population. ¥e
have compared estimates of the lifetime risk for these two treatments of
the transportation problem in order to find out how sensitive the BEIR-3
Committee risk estimates are to their assumptions. To do this, we
calculated new relative risk estimates for solid cancers based on the
age-specific cancer mortality of the Japanese population rather than the
U.S. data used by the BEIR-3 Committee, We found that this alternative
approach did not have much effect on the estimated lifetime risk of
solid radiogenic cancer, i.e. a change of 3 percent for males and 17
percent for females. We have concluded that the amount of uncertainty
introduced by transporting cancer risks observed in Japan to the U.S.
population is small compared to other sources of uncertainty in this
risk assessment. Baseline leukemia rates are about the same in the
countries, so we believe that these risks are also "transportable,"
The last of the models needed to estimate risk is a risk projection
model. As outlined in Section 8.2, such models are used to project what
future risks will be as an exposed population ages. For leukemia and
bone cancer, where the expression time is not for a full lifetime but
rather 25 years, absolute and relative risk projection models yield the
same number of radiogenic cancers, but would distribute them somewhat
differently by age. For solid cancers, other than bone, the BEIR-3
Committee assumed that radiogenic cancers would occur throughout the
lifetime. This makes the choice of projection model more critical,
because the relative risk projection yielding estimated risks about
three times larger than those obtained with an absolute risk projection,
as shown in Table 8.2-2.. Because we have used the average of these two
projections for solid cancers, we believe reduces the uncertainty
resulting from the choice of model to about a factor of two or perhaps
less, depending on the age distribution of fatal radiogenic cancer, as
outlined in 8.2 above.
8-38
-------
Similiarly, there is, as yet, insufficient information of
radiosensitivity as function of the age at exposure. The age-dependent
risk coefficients that we have used are those presented in the BEIR-3
report. As yet, there is little information on the ultimate effects of
exposure during childhood. As the A-bomb survivors1 population ages,
more information will become available on the cancer mortality of
persons irradiated when they were young. Table 8.2-2 indicates that the
conservative BEIR-1 estimates for the effect of childhood exposures
would increase BEIR-3 risk estimates by about 40 percent. As this is
probably an upper limit, the lack of more precise information is not a
major source of uncertainty in estimates of the risk caused by lifetime
exposure. Similiarly, the BEIR-3 Committee did not calculate population
risks for radiogenic cancer that included in utero radiation because
they felt the available data were unreliable. We have deferred to their
judgement in this regard. The BEIR-1 report did include in utero cancer
risk. These had little effect, I to 10 percent, on the lifetime risk of
cancer from lifetime exposure. An effect this small is not significant
relative to other sources of uncertainty in the risk assessment.
8.5.4 Summary
We can only semi-quantitatively estimate the overall uncertainty in
the risk per rad for low-LET radiations. We expect that more quantita-
tive estimates of the uncertainty will be possible only after the A—bomb
dose reassessment is completed and the A-bomb survivor data reanalyzed
on the basis of the new dose estimates. It should be noted, however,
that even if all systematic bias is removed from the new dose estimates,
there will still be considerable random error in the dose estimate for
each survivor. This random error biases the estimated slope of the dose
response curve so that it is smaller than the true dose response (Da72,
Maa59). The amount of bias introduced depends on the size of the random
error in the dose estimates, and their distribution, which are unknown
quantities at this stage of the dose reassessment.
The source of uncertainty in risk estimates for low-LET radiations
can be ranked as shown in Table 8.5-1.
The estimates of uncertainty in Table 8.5-1 are not wholly
comparable and must be interpeted carefully. However, they do have some
illustrative value, particularly when ordered in this way. The
uncertainty listed for the slope of dose response is a minimal value for
the BEIR-3 linear quadratic LQ formulation (Lac83) and is only valid
insofar as the Committee's assumptions are true. It is based on two
standard deviation errors so that the expectation of the error being
less than indicated is 95 percent. We do not believe that the
uncertainty in the BEIR-3 linear estimate, L-L, is significantly
smaller, cf. Tables V-9 and V-ll in NAS80.
The other uncertainties listed in Tsbl« 8.5-1 are quite different,
being more in the nature of informed judgements than the result of a
statistical analysis. It is doubtful that all radiogenic cancers have
the same type of response functions. However, if they were all linear,
8-39
-------
Table 8.5-1. A ranking of causes of uncertainty in
estimates of the risk of cancer
Source of uncertainty Degree of uncertainty
Choice of dose response model
Slope of dose response resulting +200%'^)
from sampling variation
Choice of an average risk
projection model
Choice of transportation model
A-bomb T65 dosimetry Plus only,
amount
(a)por choices limited to BEIR-3 linear and linear quadratic
models, see 8.2.
'b/Estimate of 2 standard deviations for
the BEIR-3 LQ model (Lac83).
(c/Average of relative and absolute projection as described
above.
'"'For the total of all cancers, not specific cancers.
as breast cancer and thyroid appear to be, the BEIR-3 linear quadratic
response model would underestimate the response by 250 percent. If most
cancers have a linear quadratic response, or equivalently, a dose rate
reduction factor equal to the difference in slope at low doses between
the BEIR~3 linear and linear quadratic models, use of a linear model
would overestimate the response by a factor of 2.5. At present, no one
knows which response model is most often appropriate. He believe that a
factor of 250 percent is a conservative estimate of the uncertainty
introduced by the lack of data at low dose rates.
As discussed above, the uncertainty resulting from the choice of an
absolute or a relative risk model is about a factor of three. Use of
the average risk for these two models reduces the uncertainty in risk
projection by more than a factor of two because it is known that a
relative risk projection is high for some kinds of cancer and an
absolute risk projection is low for others.
The uncertainties listed in Table 8.5-1 are largely independent of
each other and therefore unlikely to be correlated in sign. Their root
mean square sum is about 300 percent, indicating the expectation that
calculated risks would be within a factor of three or so of the true
value. This result is overly optimistic because it does i?ot include
8-40
-------
consideration of the uncertainty introduced by the bias in the A-bomb
dositnetry or by the constrained regression analysis used by the BEIR-3
Committee.
8.6 Other Radiation-Induced Health Effects
The earliest report of radiation induced health effects was in 1896
(Mob6V), and it dealt with acute effects in skin caused by x-ray
exposures. Within the six year period following! 170 radiation related
skin damage cases h d been reported. Such injury, like many other acute
effects, is the result of exposure to hundreds or thousands of rad.
Under normal environmental exposure situations, however, such exposure
conditions are not possible and therefore will not be considered in
assessing the risk to the general population from radionuclide
emissions.
Although radiation-induced carcinogenesis was the first delayed
health effect reported, radiation-induced genetic changes were reported
early, too. In 1927, H. J. Muller reported on X-ray-induced mutations
in animals and in 1928 L. J. Stadler reported a similar finding in
plants (Ki62), At about the same time, radiation effects on the
developing embryo were reported. Case reports in 1929 showed a high
rate of microcephaly (small head size) and nervous system disturbance
and one case of skeletal defects in children irradiated in utero
(UNSCEAR69). These effects, at unrecorded but high exposures, appeared
to be central nervous system and eye defects similar to those reported
in rats as early as 1922 (RubSO).
For purposes of assessing the risks of environmental exposure to
radionuclide emissions, the genetic effects and in utero developmental
effects are the only health hazards other than cancer that are addressed
in this BIB.
8.6.1 Types of Genetic Harm and Duration of Expression
Genetic harm or the genetic effects of radiation exposure are
those effects induced in the germ cells (eggs or sperm) of exposed
individuals, which are transmitted to and expressed in their progeny and
future generations.
Of the possible consequences of radiation exposure, the genetic
risk is more subtle than the somatic risks. Genetic risk is incurred by
fertile people when radiation damages the nucleus of the cells that
become their eggs or sperm. Damage, in the form of a mutation or a
chromosome aberration, is transmitted to, and may be expressed in, a
child conceived after the radiation exposure and subsequent generations.
However, the damage may be expressed only after many generations or,
alternately, it may never be expressed because of failure to reproduce.
The EPA treats ganetic risk as independent of somatic risk because,
although somatic risk is expressed in the person exposed, genetic risk
is expressed only in progeny and, in general, over many subsequent
8-41
-------
generations. Moreover, the types of damage incurred often differ in
kind from cancer and cancer death. Historically, research on genetic
effects and development of risk estimates has proceeded independently of
the research on carcinogenesis. Neither dose response models nor risk
estimates used for genetics are derived from data on studies of
carcinogenesis .
.Although genetic effects may vary greatly in severity, the genetic
risks considered by EPA in evaluating the hazard of radiation exposure
include only those "disorders and traits that cause a serious handicap
at some time during lifetime" (NAS80). Genetic risk may result from one
of several types of damage that ionizing radiation can cause in the DNA
within eggs and sperm. The types of damage usually considered are;
dominant and recessive mutations in autosomal chromosomes, mutations in
sex-linked (x-linked) chromosomes, chromosome aberrations (physical
rearrangement or removal of part of the genetic message on the chromo-
some or abnormal numbers of chromosomes), and irregularly inherited
disorders (genetic conditions with complex causes, constitutional and
degenerative diseases, etc.).
Estimates of the genetic risk per generation are based on a 30 year
reproductive generation. That is, the median parental age for pro-
duction of children is age 30 (one half the children are produced by
persons less than age 30, the other half by persons over age 30). Thus,
the radiation dose accumulated up to age 30 is used to estimate the
genetic risks. Using this accumulated dose and the number of live
births in the population along with the estimated genetic risk per unit
dose, it is possible to estirsate the total number of genetic effects per
year, those in the first generation and the total across all time. Most
genetic risk analyses have provided such data, EPA assessment of risks
of genetic effects includes both first generation estimates and total
genetic burden estimates.
Direct and Indirect Methods of Obtaining Risk Coefficients for Genetic
Effects
Genetic effects, as noted above, may occur in the offspring of the
exposed individuals or they may be spread across all succeeding
generations. Two methods have been used to estimate the frequency of
mutations in the offspring of exposed persons, direct and indirect. In
either case, *, .ie starting point is data from animal studies, not data
obtained froa studies of human populations.
a Direct estimate, the starting point is the frequency of a
nutation per unit exposure in some experimental animal study. The 1982
UNSCEAR (UNSCEAR82) report gave an example of the direct method for
estimating induction of balanced reciprocal translocations (a type of
chromosomal aberration) in males per rad of low level, low-LET
radiation.
8-42
-------
Induction Rate/
(1) Rate of induction in rhesus monkey
spermatogonia: cytogenetic data Q.gg x 10-4
(2) Rate of induction that relates to
recoverable translocations in the Fj
(1st filial generation) progeny
[divide (1) by 4] Q>215 x 1Q_4
(3) Rate after low dose rate X-rays:
based on mouse cytogenetic observations
[dlvide (2) b? V 0.1075 x lQ22 x 1Q_4
(5) Expected rate of unbalanced products
[multiply (3) and (4) by 2] for (3); 0.215 x 10~A
for (4): 0.043 x 10~*
(6) Expected frequency of congenitally
malformed children in the Fj, assuming
that about 6 percent of unbalanced
products [item (5) above] contribute
to this
for low dose rate X-rays 1.3
for chronic gamma radiation ^0.3
x
x
°f CEAR estiinates tha* a consequence of induction of
t 1 tr^S l°Cations in «P««» *•«*», ™ estimated 0.'
to 13 congenitally malformed children would occur in each 10* live
births for every rad of parental radiation exposure.
A complete direct estimate of genetic effects would include
estimates, derived in a manner similar to that shown above for each tvn
of genetic damage. These direct estimates could be used to calculate^
indi*ect 25
8-43
-------
(2) Estimated human spontaneous mutation
rate per gene
(3) Relative mutation risk in humans
[divide (1) by (2)]
(4) Doubling dose: the exposure needed
to double the human mutation rate
0.5 x 10-6 to
0.5 x 10~5
0.005 to 0.05
200 to 20 rad
senetlc e« r8 T "* "^ be US6d tO estifflate the
by the exfeTl8 °l «««*".*«*« » all future generations caused
P • of. Pa«nts. Since the genetic component of congenital
andl,"8 " ** P°pulati°n ca" be -tinted by epidemiological
P«nvh'- , \8 COHlponent 1S Considered to be maintained at an
bble th ^ mUtati°nS' * d°ubli<* d°- of ionizing radiation
f / ,IInetlC effects- Dividing the number of the various
J 4^ MrthS by thS doublinS *o.e yields the
of genetic pffects per rad. For example:
(1) Autosomal dominant and x-linked
diseases, current incidence
(2) Estimated doubling dose
(3) Estimate of induced autosomal
dominant and x-linked diseases
10,000 per 106 live
births
20 to 200 rad
50 to 500 per 106
live births per rad of
parental exposure.
The doubling dose estimate assunes that the total population of
and that£theS "'f^ irradlated' 8S Occurs «"» background radiatil,
and that the population exposed is large enough so that all genetic
fr^timate fTT^" fUtUre °ff8^ini- Alth-^ it is'baLc'lly
?L f M ' I -3l 8enetlC burden across a11 future generations,
the doubling dose estimate can also estimate the effects that occur in
the first generation. Usually a fraction of the total genetic burden
for each type of damage xs assigned to the first generation using
population genetics data as a basis to determine the fraction. For
example, the BEIR-3 committee geneticists estimated that one-sixth of
the total genetic burden of x-linked mutations would be expressed in the
first generation, five-sixths across all future generations. EPA
assessment of risks of genetic effects includes both first generation
estimates and total genetic burden estimates.
8.6.2 Estimates of Genetic Harm Resulting from Lo»-LET Radiations
«. »^ fir8t estifflates of genetic risk «as made in 1956 by
the NAS Committee on the Biological Effects of Atomic Radiation (BEAR
Committee). Based on Prosophila (fruit fly) data and other consid-
erations, the BEAR Genetics Committee estimated that 10 Roentgens
8-44
-------
(10 R*) per generation continued indefinitely would lead to about 5,000
new instances of "tangible inherited defects" per 1Q6 births, and about
one-tenth of them would occur in the first generation after the irradia-
tion began (NAS72), The UNSCEAR addressed genetic risk in their 1958,
1962, and 1966 reports (UNSCEAR58, UNSCEAR62, UNSCE&R6G). During this
period, they estimated that one rad of low-LIT radiation would cause a 1
to 10 percent increase in the spontaneous incidence of genetic effects.
In 1972, both the HAS BEIR Committee (NAS72) and UNSCEAR
(UNSCEAR72) reexamined the question of genetic risks. Although there
were no definitive human data, additional information was available on
the genetic effects of radiation on maniaals and insects. In 1977,
UNSCEAR reevaluated the 1972 genetics estimates (UNSCEAR77). Their new
estimates used recent information on the current incidence of various
genetic conditions, along with additional data on radiation exposure of
mice and marmosets and other considerations.
In 1980, an ICRP Task Group (ICRPTG) summarized recommendations
that formed the basis for the genetic risk estimates published in ICRP
Report 26 (Of80). These risk estimates are based on data similar to
those used by the BEIR and UNSCEAR Committees, but used slightly
different assumptions and effect categories, Table 8.6-1.
The 1980 NAS BEIR Committee revised genetic risk estimates (NAS80),
The revision considered much of the same material that was in BEI1-1
(NAS72), the newer material considered by UNSCEAR in 1977, and some
additional data. Estimates for the first generation are about a factor
of two smaller than reported in the BEIR-1 report. For all generations,
the new estimates are essentially the same, Table 8.6-2.
The most recent genetic risk estimate, in the 1982 UNSCEAR Report
(UNSCEAR82), includes some new data on cells in culture and the results
of genetic experiments using primates rather than rodents, Table 8.6-3.
Although all the reports described above used somewhat different
sources of information, there is reasonable agreement in the estimates
(see the summary in Table 8.6-4). Most of the difference is caused by
the newer information used in each report. Note that all estimates
listed above are based on the extrapolation of animal data to man.
Groups differ in their interpretation of how genetic experiments in
animals might be expressed in humans. Although there are no comparable
human data at present, information on hereditary defects among the
children of A-bomb survivors provide a degree of confidence that the
animal data do not lead to underestimates of the genetic risk following
exposure to humans. (See "Observations on Human Populations" which
follows.)
*R is the symbol for Roentgen, a unit of measurement of x-radiatioti,
equivalent to an absorbed dose in tissue of approximately 0.9 rad.
8-45
-------
Table 8.6-1. ICRP task group estimate of number of cases
of serious genetic ill health in liveborn from
parents irradiated with 1£>6 man-rem in a
population of constant sizeC^)
(Assumed Doubling Dose = 100 rad)
Category of
genetic effect First generation Equilibrium
Unbalanced translocations:
risk of malformed liveborn 23 30
Trisomics and XO 30 30
Simple dominants and sex-
linked mutations 20 100
Dominants of incomplete
penetrance and multi-factorial
disease maintained by mutation 16 160
Multifactorial disease not
maintained by mutation 0 0
Recessive disease — —
Total 89 320
(a)jn£s £s equivalent to effects per 10^ liveborn following an
average parental population exposure of 1 rem per 30-year
generation, as used by BEIR and UNSCEA1.
Source: (Of80).
8-46
-------
Table 8.6-2. BEIR-3 estimates of genetic effects of an average
population exposure of 1 tern per 30-year generation
Type of genetic
disorder
Current incidence
per 10^ liveborn
Effects per 10^ liveborn
per rem per generation
Autosomal dominant
and x- linked
Irregularly inherited
Recessive
Chromosomal aberrations
Total
First Generation
10,000 5-65
90,000 (not estimated)
1,100 Very few
6,000 Fewer than 10
107,100 5-75
Equilibrium
40-200
20-900
Very slow
increase
Increases
only
slightly
60-1100
Sources (NAS80),
8-47
-------
Table 8.6-3. UNSCEAR 1982 estimated effect of 1 rad per
generation of low dose or low dose rate, low-LET
radiation on a population of 10& liveborn
according to the doubling dose nethod
(Assumed Doubling Dose » 100 rad)
Disease classification Current incidence
First generation Equilibrium
Autosoraal dominant and
x-linked diseases 10,000 15 100
Recessive diseases 2,500 slight slow
increase
Chromosomal diseases:
Structural 400 2.4 4
NutBerical 3,000 Probably very
small
Congenital anomalies,
anomalies expressed later,
constitutional and
degenerative diseases 90,000 4.5 45
Total 105,900 22 149
Source: (UNSCEAR82).
8-48
-------
Table 8.644. Summary of genetic risk estimates per 1Q& liveborn
for an average population exposure of 1 rad of low dose or
low dose rate, low-LET radiation in a 30-year generation
Serious hereditary effects
Equilibrium
Source First generation (all generations)
BEAR, 1956 (NAS72) — 500
BEIR-I, 1972 (NAS72) 49^) (12-200) 300(a> (60-1500)
UNSCEAR, 1972 (UNSCEAR72) 9 (60-1100)
UNSCEAR, 1982 (UNSCEAR82) 22 149
Numbers in parentheses ( ) are the range of estimates
'^Geometric Mean is calculated by taking the square root of the
product of two numbers for which the mean is to be calculated.
The cube root of three numbers, etc. In general, it is the
root of the product of N numbers for which the mean is to be
calculated.
8-49
-------
It should be noted that the genetic risk estimates summarized in
Table 8.6-4 are for low-LET, low dose, and low dose rate irradiation.
Most of the data were obtained from high dose rate studies and most
authors have used a sex-averaged factor of 0.3 to correct for the change
from high dose rate, low-LET to low dose rate, low-LET exposure (NAS72,
NAS80, UNSCEAR72, UNSCEAR77). However, factors of 0.5 to 0.1 have been
used in estimates of specific types of genetic damage (UNSCEAR72,
UNSCEAR77, UNSCEAR82, Of80).
8,6.3 Estimates of GeneticHarm for High-LET Radiations
Although genetic risk estimates are made for low-LET radiation,
some radioactive elements deposited in the ovary or testis can irradiate
the germ cells with alpha particles. The ratio of the dose (rad) of
low-LET radiation to the dose of high-LET radiation producing the same
endpoint is called RBE and is a measure of the effectiveness of high-LET
compared to low-LET radiation in causing the same specific endpoint.
Studies in which the beta particle emitted isotopes carbon-14 and
tritium yielded RBEs of 1.0 and 0.7 to about 2.0, respectively
(UNSCEAR82). At the present time, the RBE for genetic endpoints
resulting from beta particles is taken as one (UNSCEAR77, UNSCEAR82).
Studies of the RBE for alpha-emitting elements in germinal tissue
have used only plutonium-239. Studies comparing cytogenetie endpoints
after chronic low dose rate gamma radiation exposure or incorporation of
plutonium-239 in the mouse testis, have yielded RBEs of 23 to 50 for the
type of genetic injury (reciprocal tr^rislocations) that might be trans-
mitted to liveborn offspring (NAS80, UNSCEAR77, UNSCEAR82). However, an
RBE of 4 for plutonium-239 compared to chronic low-LET radiation was
reported for specific locus mutations observed in neonate mice (NAS80).
Neutron RBE, determined from cytogenetic studies in mice, also ranges
from about 4 to 50 (UNSCEAR82, Gra83, Ga82). Most reports use an RBE of
20 to convert risk estimates for low dose rate, low-LET radiation to
risk estimates for high-LET radiation.
8.6.4 Uncertainty in Estimates of Radiogenetic Harm
Chromosomal damage and mutations have been demonstrated in cells in
culture, in plants, in insects, and in mammals (UNSCEAR72, UNSCEAR77,
UNSGEAR82). Chromosome studies in peripheral blood lymphocytes of
persons exposed to radiation have shown a dose-related increase in
chromosome aberrations (structural damage to chromosome) (UNSCEAR82).
In a study of nuclear dockyard workers exposed to external X-radiation
af. rates of less than 5 rad per year, Evans et al. (Ev79) found a
significant increase in the incidence of chromosome aberrations. The
increase appeared to have a linear dependence on cumulative dose. In a
study of people working and living in a high natural background area
where there was both external gamma radiation and internal alpha
radiation, Pohl-Ruling et al. (Po78) reported a complex dose response
curve. For mainly gamma radiation, exposure (less than 10 percent alpha
radiation), they reported the the increase in chromosome aberrations
8-50
-------
increased linearly from 100 to 200 mrad per year, then plateaued from
300 rarad to 2 rad per year. They concluded:
"From these data, and data in the literature, it can be
concluded that the initial part of the dose-effect curve for
chromosome aberrations is not linear or sigmoid with a
threshold at the lowest dose, but rises sharply and passes
into a complex upward form with a kind of plateau until it
meets the linear curve of the high dose."
Although chromosomal damage in peripheral blood lymphocytes cannot
be used for predicting genetic risk in progeny of an exposed person, it
is believed by some to be a direct expression of the damage analogous to
that induced in germ cells as a result of the radiation exposure. It is
at least evidence that chromosome damage can occur in vivo in humans.
Since there are no quantitative human data on genetic risks
following radiation exposure, risk estimates are based on extrapolations
from animal data. As genetic studies proceeded, emphasis has shifted
from Drosophila to mammalian species in attempts to find an experimental
system that would reasonably project what might happen in humans.
For example, Van Buul (Va80) reported the slope (b) of the linear
regression, Y = a + bD, for induction of reciprocal translocations in
spermatogonia (one of the stages of sperm development) in various
species as follows:
b x 10^ + sd x 104
Rhesus monkey
Mouse
Rabbit
Guinea pig
Marmoset
Human
0.86 + 0.04
1.29 * 0.02
1,48 + 0.13
0.91 + 0.10
7.44 + 0.95
3.40 + 0.72
to 2.90 + 0.34
These data indicate that animal based estimates for this type of genetic
effect would be within a factor of four of the true human value. In
this case most of the animal results would underestimate the risk in
humans.
However, when risk estimates such as this are used in direct
estimation of risk for the first generation, the total uncertainty in
the estimate becomes indeterminate. Even if studies have been made the
results which can be used to predict the dose response and risk
coefficient for a specific radiation-induced genetic damage for a
species, there is no certainty that this prediction will represent the
8-51
-------
response for all genetic damage of that type. In addition, as (ihown in
the example from the 1982 UNSCEAR report (UNSCEM82) shown in S«ction
8.6.1, additional assumptions based on observations, usually in other
animal species, are used to adjust the risk coefficient to what is
expected for humans. The uncertainty in these extrapolations has not
been quantified.
A rough estimate of the uncertainty can be obtained by comparing
direct estimates of risk for the first generation with doubling dose
estimates in the 1977 UNSCEAR report (UNSCEAR77). The estimates differ
by a factor of between 2 and 6 with the direct estimate usually smaller
than the doubling dose estimate.
A basic assumption in the doubling dose method of estimation is
that there is a proportionality between radiation-induced and spon-
taneous mutation rates. Some of the uncertainty was removed in the 1982
UNSCEAR report (UNSCEAR82) with the observation that in two test systems
(fruit flies and bacteria) there is a proportionality between spon-
taneous and induced mutation rates at a number of individual gene sites.
There is still some question as to whether the sites that have been
examined are representative of all sites and all gene loci or not. The
doubling dose estimate does, however, seem better supported than the
direct estimate.
Although there is still some uncertainty as to what should be
doubled, future studies on genetic conditions and diseases can only
increase the total number of such conditions. Every report, from the
1972 BUR and UNSCEAR reports to the most recent, has listed an
increased number of conditions and diseases that have a genetic
component.
Observations on Human Populations
As noted earlier, the genetic risk estimates are based on
interpretation of animal experiments as applied to data on naturally
occurring hereditary diseases and defects in man. A study of birth
cohorts was initiated in the Japanese A-bomb survivors in mid-1946.
This resulted in a detailed monograph by Heel and Schull (Nea56), which
outlined the background of the first study and made a detailed analysis
of the findings to January 1954 when the study terminated. The authors
concluded only that it was improbable that human genes were so sensitive
that exposures as low as 3 R, or even 10 R, would double the mutation
rate. Although this first study addressed morphological endpoints,
subsequent studies have addressed other endpoints. The most recent
reports on this birth cohort of 70,082 persons have attempted only to
estimate the minimum doubling dose for genetic effects in man (Sc81,
Sa82).
Data on four endpoints have beer, recorded for this birth cohort.
Frequency of stillbirths, major congenital defects, perinatal death, and
frequency of death prior to age 17, have been examined in the entire
cohort. Frequency of cytogenetic aberrations (sex chromosome
8-52
-------
aneuploidy) and frequency of biochemical variants (a variant enzyme or
protein electrophoresis pattern) have been measured on large subsets of
this cohort.
Although the updated data reported appear to suggest that radiation
effects have occurred, the numbers are small and not statistically
significant. Overall, the estimated doubling dose for low-LET radiation
at high doses and dose rates for human genetic effects is about 156 rent
(Sc81) or 250 rem
-------
dose response model is appropriate for estimating genetic effects in
humans. Until there is more consensus the linear nonthreshold model
appears to be a prudent approach that will not grossly underestimate the
risks.
The agreement in estimates made on a linear nonthreshold model, in
the various reports, is quite good. Even though the authors of the
reports used different animal models, interpreted them in different
ways, and gave different estimates of the level of human genetic
conditions in the population, the range is about an order of magnitude
(see Table 8.6-4). For the most recent more comparable estimates, the
rang? is a factor of two to four (see ICRPTG, BEIR-3 and UNSCEAI 1982 in
Table 8.6-4).
8.6.5 The EPA Genetic Risk Estimate
There is no compelling evidence for preferring any one set of the
genetic risk estimates listed in Table 8.6-4, EPA has used the esti-
mates from BEIR-3 (NAS80). These "indirect" estimates are calculated
using the normal prevalence of genetic defects and the dose that is
considered to double this risk. The HAS estimates used by EFA are based
on a doubling dose range, with a lower bound of SO rem and an upper
bound of 250 rem. We prefer these risk estimates to those made by the
ICRP task group (Of80), which used a "direct" estimate, because the
ICRPTG tabulation combines "direct" estimates for some types of genetic
damage with doubling dose estimates for others. We also prefer the
BEIR-3 risk estimates to the "direct" estimates of UNSCEAR 1982, which
tabulates genetic risk separately by the u .-^ct method and by the
doubling dose method. The risk estimated by the direct method does not
include the same types of damage estimated by doubling doses and was not
considered further. The BEIR-3 genetic risk estimate is also preferred
over the UNSCEAR 1982 and ICRPTG estimates, because BEIR-3 assigns a
range of uncertainty for multifactorial diseases (>5 percent to <50
percent) which reflects the uncertainty in the numbers better than the
other estimates do (5 percent and 10 percent, respectively).
In developing the average mutation rate for the two sexes used in
the calculation of the relative mutation risk, the BEIR-3 committee
postulated that the induced mutation rate in females was about 40
percent of that in males (NAS80). Recent studies by Dobson et al.
(Doa83, Dob83, Doc84, Dod84) suggest that the assumption was invalid and
that human occytes should have a risk equivalent to that of human
speraatogonia. This would increase the risk estimate obtained by
doubling-dose methods by a factor of 1.43.
We recognize, however, that the use of the doubling dose concept
does assume that radiation-induced genetic damage is in some way
proportional to "spontaneous" damage. As noted earlier, the recent
evidence obtained in insects (Drosophila) and bacteria (E. coli)
supports the hypothesis that, with the exception of "hot spots" for
mutation, the radiation-induced mutation rate is proportional to the
8-54
-------
spontaneous rate (UNSCEAR82). No proof that this is also true in
mammals is available yet.
The BEIR-3 estimates give a considerable range. To express the
range as a single estimate, the geometric mean of the range is used, a
method first recommended by UNSCEAE (UNSCEAR58) for purposes of
calculating genetic risk. The factor of three increase in risk for high
dose rate, low-LET radiation noted earlier is also used.
The question of PBE for high-LET radiation is more difficult. As
noted above, estimated RBEs for plutonium-239 alphas versus chronic
gamma radiation for reciprocal translocations as determined by
cytogenetic analyses is between 23 and 50 (NAS80, UNSCEM82). However,
the observed B.BE for single locus mutations in developing offspring of
male mice given plutonium-239 compared to those given X-ray irradiation
is 4 (NAS8Q). The average of RBEs for reciprocal translocations and for
specific locus mutations is 20.25. Since reported neutrons RBEs are
similar to those listed above for plutonium-239 alpha radiation, we use
an RBE of 20 to estimate genetic risks for all high-LET radiations.
This is consistent with the RBE for high-LET particles recommended for
estimated genetic risks associated with space flight (Grb83).
Genetic risk estimates used by EPA for high- and low-LET radiations
are listed in Table 8.6-5. As noted above, EPA uses the dose received
before age 30 in assessing genetic risks.
The EPA estimates (Table 8.6-5), like all other human genetic risk
estimates, are limited by the lack of confirming evidence of genetic
effects in humans. These estimates depend on a presumed resemblance of
radiation effects in animals. The magnitude of the possible error is
Table 8.6-5. EPA estimated frequency of genetic disorders in a
birth cohort due to exposure of the parents to 1 rad per generation
Cases per 10^ liveborn
First generation All generations
Type of radiation low^a) high^D)
Low dose rate, low-LET
High dose rate, low-LET
High-LET
20
60
400
30
90
600
260
780
5200
370
1110
7400
(a)Feoiale sensitivity to induction of genetic effects is 40 percent as
great as that of males.
(b)pemale sensitivity to induction of genetic effects is equal to that
of males.
8-55
-------
deriving its MM'™!/ £ "tmates. Moreover, the BEIR Committee in
cauili h esCV>ates has assuned that almost all of the risk was
b£ thf BEIR "oLutL"" O" '" ""' "°" Cation, «-
low 1FT6 "la'!Ve '"k °£ high-LIT radiation compared to low dose rate
2« ^°fl- "e^
dos^etry, i.e., the actual abjorbed
at
8. 6-. 6 Teratogenic Effects
1600 R" ^hi^tf f^ "fr^" 80me mtS t0 X~rayS 3t doses of 2°0 R to
l??? J^irtrffur of ]2° exposed females had litters and five of the
litters had anxmals with developmental defects (Moc30). He felt that
'
high radiation exposures, 25 R and above, established some
etahl relat"nshiPs- M«* imrortantly, ^ established th!
ra^iati °f/T lJiV1& °f the deVel°Pin8 'Odent embryo and fetus to
radiation effects (Ruc54, Hia53, Se69, Hic66).
Ufh' ^ ?-S reVleW °f radistion teratogenesis (Rua70), listed the
ted mammalian anomalies and the exposure causing them. The lowest
8-56
-------
reported exposure was 12.5 R for structural defects and 1 R
° "
as
--
resorption of implanted enbryos and structural abnomalities
ef,ects was linear or nearly linear, with no observable threshold
This appeared consistent with a report by Russell (Rub57) which
suted a threshold for some effects wLreas others appLred linear.
Rugh (Rua71) suggested there may be no threshold for radiation
r^^ X. -
;c£i9^:hr:^ -- -
P
»re to
chemrcals such as iodoacetimide or tetracycline (Mi 78)
o teratol°gic -tudie. in an«al» is the
of
e
ing how dose response data should be interpreted
(Ruc54) Pointed out some aspects of the problem- (1)
et atl°nK-\abSOrbed thro^h0^ the anbryo/it c" es
damage whlch ls consistently dependent on the stage of
embryonic development at the time of irradiation and (2) thfdamaged
£Lv ""Tf ' r 3/°nsistent raann-» "ithin a narrow tiioe range^
However, while low dose irradiation at a certain stage of development
produces changes only in cocoponents at their peak sensitivity? higher
other »fj «du« ""j'Lonal ^normalities «hich have peak .ewitiStJ at
other stages of development, and may further modify exoressicm of the
changes induced in parts of the embryo at peak sensitikty during he
8
' the
8-57
-------
The embryo*/fetus** starts as a single fertilized egg and divides
ind differentiates to produce the normal infant at term. The different
irgan and tissue primordia develop independently and at different rates.
lowever, they are in contact through chemical induction or evocation
Arb54). These chemical messages between cells are important in
•ringing about orderly development and the correct timing and fitting
:ogether of parts of organs or organisms. While radiation can disrupt
:his pattern, interpretation of the response may be difficult. Since
:he cells in the embryo/fetus differentiate, divide and proliferate at
lifferent times during gestation and at different rates, gestional times
rtien colls of specific organs or tissues reach maximum sensitivity to
radiation are different. Each embryo/fetus has a different timetable.
[n fact, each half (left/right) of an embryo/fetus may have a slightly
lifferent timetable.
In addition, there is a continuum of variation from the hypo-
:heticai normal to the extreme deviant, which is obviously recognizable.
there is no logical place to draw a line of separation between normal
ind abnormal. The distinction between minor variations of normal and
trank malformation, therefore, is an arbitrary one and each investigator
aust establish his own criteria and apply them to spontaneous and
induced abnormalities alike (HUC73). For example, some classify mental
retardation as IQ 80 or lower, some classify on ability to converse or
lold a job, some on the basis of the need to be institutionalized.
Because of the problems in interpretation listed above, it appears
j pragmatic approach is useful. The dose response should be given as
.he simplest function that fits the data, often linear or linear with a
threshold. No attempt should be made to develop complex dose response
nodels unless the evidence is unequivocal.
The first report of congenital abnormalities in children exposed in
itero to radiation from atomic bombs was that of Plummer (P152). Twelve
children with microcephaly of which 10 also had mental retardation had
seen identified in Hiroshima in the in utero exposed survivors. They
•rere found as part of a program started in 1950 to study children
exposed in the first trimester of gestation. In 1955 the program was
expanded to include all survivors exposed in utero.
Studies initiated during the program have shown the following
radiation-related effects: (1) growth retardation; (2) increased
aicroeephaly; (3) increased mortality, especially infant mortality;
[4) temporary suppression of antibody production against influenza; and
*The embryonic period, when organs develop, is the period from
:onception to 7 weeks gestational age.
**The fetal period, a time of in utero growth, is the period from
J weeks gestational age to birth.
8-58
-------
(5) increased frequency of chromosomal aberrations in peripheral
lymphocytes (Kaa73).
Although there have been a number of studies of Japanese A-bomb
survivors, including one showing a dose and gestational age related
increase in postnatal mortality (Kaa?3), only incidences of microcephaly
and mental retardation have been investigated in any great detail. In
the most recent report, Otake and Schull (Ot83) showed that mental
retardation was associated with exposure between 8 and 15 weeks of
gestation (10 to 17 weeks of gestation if counted from the last men-
strual period). They further found a linear dose-response relationship
for induction of mental retardation that had a slope yielding a doubling
dose for mental retardation of about 2 rad, fetal-absorbed dose (Qt83).
Classification as mentally retarded was based on "unable to perform
simple calculations, to care for himself or herself, or if he or she was
completely unmanageable or had been institutionalized". (Ot83)
Estimates of the risk of mental retardation for a rad of embryo/
fetus exposure in the U.S. population can be derived by three methods.
The first and easiest method is to use the absolute risk calculated by
Otake and Schull for the Japanese survivors (Ot84). A second method is
to use the doubling dose calculated by Otake and Schull (Ot83) times the
incidence of mental retardation per 10^ live births. Unfortunately, a
number of assumptions must be made to establish the incidence of mental
retardation per 1Q-* live births. Mental retardation may b& classified
as mild (IQ 50-70), moderate (IQ 35-49), severe (IQ 20-34) and profound
(IQ <20) (WH075). However, some investigators use only mild mental
retardation (IQ 50-70) and severe mental retardation (IQ <50) as classes
(HaaSl, Sta84). Mental retardation is not usually diagnosed at birth
but at some later time, often at school age. Since the mental
retardation may have been caused before or during gestation, at the time
of birth or at some time after birth, that fraction was caused before or
during gestation must be estimated. In like manner since mental
retardation caused before birth may be due to genetic conditions,
infections, physiologic conditions, etc.; the fraction related to
unknown causes during gestation must be estimated. This is the fraction
that might possibly be doubled by radiation exposure,
A third method to estimate the risk is indirectly, using the
relationship of microcephaly and mental retardation reported in the
Japanese survivors (Woa65, Ot83). If head size is assumed to be
normally distributed, then the fraction of the population with a head
size 2 "or 3 standard deviations smaller than average can be obtained
from statistical tables. The fraction of 10^ liveborn with microcephaly
multiplied by the proportion of mental retardation associated with that
head size yields an estimate of the incidence of mental retardation per
10^ live births; which can then be used with the doubling dose to
estimate the risk as described above.
Risk estimates for mental retardation are derived below for
comparison purposes using each of the three methods described above.
8-59
-------
A. Estimate of Incidence Per Rad Based on Direct Application of the
Slope of the Japanese Data
Otake and Sehull (Ot84) gave an estimate of 'The Relationship of
Mental Retardation to Absorbed Fetal Exposure in the "Sensitive" Period
when All "Controls" are Combined.1 The estimate of 0.416 cases of
mental retardation per 100 rad could be directly applicable to a U.S.
population. In this case the risk estimate would be about!
4 cases of mental retardation per rad per 1000 live births.
B. Estimate of Incidence Per Rad Based on the DoublingDose
The Otake and Schull report (Ot83) suggested the doubling dose for
mental retardation was about 2 rad, fetal absorbed dose or about a 50
percent increase in mental retardation per rad. It would seem reason-
able that this doubling dose would apply only to idiopathic cases of
mental retardation caused during gestation. That is those which have no
known genetic, viral, bacterial, etc. cause.
Data from studies of the prevalence of mental retardation in school
age populations in developed countries suggest a prevalence of 2.8
cases/1000 (Uppsala County, Sweden) to 7.4 cases/1000 (Amsterdam,
Holland) of severe mental retardation, with a mean of about 4.3 +_ 1.3
c.ases/1000 (Sta84). Where data is available for males and females
separately, the male rate is about 30 percent higher than the female
rate (Sta84), Historically, the prevalence of mild mental retardation
has been 6 to 10 times greater than that of severe mental retardation.
But, in recent Swedish studies, the rates of prevalence of mild and
severe mental retardation have been similar (Sta84). This was suggested
to be due to a decline in the "cultural-familial syndrome". That is,
improved nutrition, decline in infection and diseases of childhood,
increased social and intellectual stimulation, etc., combined to reduce
the proportion of nonorganic mental retardation and, therefore, the
prevalence of mild mental retardation (Sta84).
In studies of the causes of mental retardation, 23 percent to 42
percent of the mental retardation has no identified cause (Gu77, HaaSl,
St84). It is this portion of the mental retardation which may be
susceptible to increase due to radiation exposure of the embryo/fetus,
In that case, the prevalence of idiopathic mental retardation would be
0.6 to 3.1 cases per 1000 of severe mental retardation and perhaps an
equal number of cases of mild mental retardation.
For purposes of estimating the effects of radiation exposure of the
embryo/fetus a risk of spontaneous idiopathic mental retardation of 1 to
6 per 1000 will be used. If this spontaneous idiopathic mental
retardation can be increased by radiation the estimate would be:
(1 to 6 cases per 1000 live births)(0.5 increase per rad)
8-60
-------
or about 0.5 to 3 cases of mental retardation per rad per 1000 live
births.
This estimate nay be biased low. This occurs because mental
retardation induced during gestation is often associated with high
childhood death rate (Sta84). If this is generally true for idiopathic
causes of mental retardation, it would cause an underestimation of the
risk.
C. Estimate of Incidence Per Rad Based on Incidence of Microcephaly
(1) 2.275 percent of live born children will have a head
circumference 2 standard deviations or more smaller than average, 0.621
percent will have a head circumference 2.5 standard deviations or more
smaller than average and 0.135 percent will have a head circumference
3 standard deviations or more smaller than average, (statistical
estimate based on a normal distribution).
(2) There is evidence in a nonselected group of 9,379 children
that mental retardation can be estimated using the incidence of micro-
cephaly, even though head circumference in the absence of other
supporting data, e.g. height or proportion, is an uncertain indicator of
mental retardation. Based on a study of 9,379 children, Nelson and
Deutsehberger (Heb70) concluded that about half of the children with a
head circumference 2.5 standard deviations or more smaller than average
had IQs of 79 or lower. Since 0.67 percent of those studied were in
this group, the observed number is about what would be expected based on
the normal distribution of head size in a population, 0.62 percent. The
estimated incidence of mental retardation per live birth in a population
would be:
(6.7 cases of microcephaly per 1000 live births) x
In i cases of mental retardation^
case of microcephaly
or about 3.4 cases of mental retardation per 1000 live births,
(3) A first approximation of risk of mental retardation might then
be:
(3.4 cases of mental retardation per 1000 live births) x
(0.5 increase per rad)
or about 2 cases of mental retardaiion per 1000 live births per rad.
8-61
-------
Both microcephaly and mental retardation were increased In Japanese
survivors (Woa65, Wob66). About half of those with head sizes 2 or more
standard deviations smaller then average had mental retardation
rlfl^I! ' S result siailar to that observed by Nelson and Deutschberger
(Neb70). Therefore, the estimate above based on the incidence of
microcephaly in a population should be a reasonable estimate of the risk
due to radiation.
Sunnnary of the Calculated Risk of Mental Retardation
The risk of increased mental retardation per rad of embryo/fetus
exposure during the 8 to15 week gestational period estimated above
ranges from about 5 x 1CH to 4 x 10~3 cases per live birth, the largest
being a direct estimate. The geometric mean of these estimates is
1.4 x 10-J, the arithmetic mean is 2.4 x 1G~3 cases per live birth.
All the estimates derived above by any of the three methods are in
the same range as an earlier UNSCEAR (UNSCEAR??) estimate of an increase
of 1 x 10 cases of mental retardation per rad per live birth. The
UNSCEAR estimate, however, did not consider gestational age at the time
of exposure. The Otake and Schull report (Ot83) did address gestational
age and estimated a higher risk, but a narrower window of
susceptibility.
If the estimates are applicable, the 15 mrad of low-LET background
radiation delivered during the 8 to 15 week gestational age sensitive
period could induce a risk of 6 x 10~5 to 7.5 x 10~6 cases Of mental
retardation per live birth. This can be compared to an estimate of a
spontaneous occurrence of 1.5 x 10~2 to 3.4 x 10~3 cases of mental
retardation per live birth.
Japanese A-bomb survivors exposed in utero also showed a number of
structural abnormalities and, particularly in those who were micro-
cephalic, retarded growth (Woa65). No estimate has been made of the
radiation-related incidence or dose-response relationships for these
abnormalities, because of the small number of cases. UNSCEAR
(UNSCEAR77) made a very tentative estimate based on animal studies that
the increased incidence of recognizable structural abnormalities in
animals may be 5 x 10~3 cases per R per live born, but stated that
projections to humans was unwarranted. In any event, the available
human data cannot show if the risk estimates derived from high dose
animal data overestimates the risk in humans.
It should be noted that all of the above estimates are based on
high dose rate low-LET exposure. UNSCEAR in 1977 also investigated the
dose-rate question and stated!
"In conclusion, the majority of the data available for most
species indicate a decrease of the cellular and malformative
effects by lowering the dose rate or by fractionating the
dose. However, deviations from this trend have been well
documented in a few instances and are not inconsistent with
8-62
-------
the knowledge about mechanisms of the teratogenic effects.
It is therefore impossible to assume that dose rate and
fractionation factor have the same influence on all
teratological effects." (UNSCEAR77).
From this analysis, EPA has concluded that the range of risk is
4 x 10~3 to 5 x 10"^ cases of mental retardation per live birth per rad
of low-LET radiation delivered between weeks 8 and 15 of gestation, with
no threshold identified at this time.
At this tine, no attempt can be made to estimate total teratogenic
effects. However, it should be noted that the 1977 UNSCEAR estimate
from animals was 5 x 10~3 cases of structural abnormalities per R per
live birth (about the same number per rad of low-LET). This estimate
must be viewed as a minimum one since it is based, to a. large extent, on
observation of grossly visible malformations. Differences in criteria
for identifying malformations have compounded the problem, and questions
of threshold and species differences have made risk projection to humans
unwarranted.
8.6.7 Non s to chas t i c Eff ec t s
Nonstoehastic effects, those effects that increase in severity with
increasing dose and may have a threshold, have been reviewed in the 1982
UNSCEAR report (UNSCEM82). In general, acute doses of 10 rad low-LET
radiation and higher are required to induce these effects. It is
possible that some of the observed effects of in utero exposure are
nonstochastic, e.g., the risk of embryonic loss, estimated to be 10~2
per I (UNSCEAR77) following radiation exposure soon after fertilization.
However, there are not enough data to address the question. Usually, no
nonstochastic effects of radiation are expected at environmental levels
of radiation exposure.
8,7 Radiation Risk - A Perspective
To provide a perspective on the risk of fatal radiogenic cancers
and the hereditary damage due to radiation, we have calculated the risk
from background radiation to the U.S. population using the risk
coefficients presented in this chapter and the computer codes described
in Addendum B. The risk resulting from background radiation is a useful
perspective for the risks caused by emissions of radionuclides. Unlike
cigarette smoking, auto accidents, and other measures of common risks,
the risks resulting from background radiation are neither voluntary nor
the result of alcohol abuse. The risk caused by background radiation is
very largely unavoidable; therefore, it is a good benchmark for judging
the estimated risks from radionucliue emissions. Moreover, to the
degree that the estimated risk of radionuclides is biased, the same bias
is present in the risk estimates for background radiation.
Low-LET background radiation has three major components: cosmic
radiation, which averages to about 28 mrad per year in the U.S.;
terrestrial sources, such as radium in soil, which contributes an
8-63
-------
average of 26 mrad per year (NCRP75); and the low-LET dose resulting
from internal emitters. The latter differs between organs, to some
extent, but for soft tissues is about 24 mrad per year (NCRP75).
Fallout from nuclear weapons tests, naturally occurring radioactive
materials in buildings, etc., contribute something like another 10 mrem
for a total low-LET whole-body dose of about 90 mrad per year. The lung
and bone receive somewhat larger doses resulting from high-LET
radiations; see below. Although extremes do occur, the distribution of
this background annual dose to the U.S. population is relatively narrow.
A population weighted analysis indicates that 80 percent of the U.S.
population would receive annual doses that are between 75 mrad per year
and 115 mrad per year (EPA81).
As outlined in Section 8.2, the BEIR-3 linear models yield, for
life time exposure to low-LET radiation, an average life time risk of
fatal radiogenic cancer of 280 per 1Q6 person rad. Note that this
average is for a group having the age and sex specific mortality rates
of the 1970 U.S. population. We can use this datum to calculate the
average life time risk due to low-LET background radiation as follows.
The average duration of exposure in this group is 70.7 years and at
9 x 10" rad per year, the average life time dose is 6.36 rad. The risk
of fatal cancer per person in this group is:
280 fatalities . »
person rad X 6'36 rem = l'78 x 10~3
or about 0.18 percent of all deaths. The vital statistics we use in our
radiation risk analyses indicate that the probability of dying due to
cancer in the U.S. due to all causes is about 0.16, i.e. 16 percent.
Thus the 0.18 percent result for the BEIR-5 linear dose response model
indicates that about 1 percent of all U.S. cancer is due to low-LET
background radiation. The BEIR-3 linear quadratic model indicates that
about 0.07 percent of all deaths are due to low-LET background radiation
or about 0.4 percent of all cancer deaths.
The^information in Volume 2 of this BID indicates that airborne
radioactive emissions may cause additional cancer risks comparable to
those risks due to background radiation. For example, the models
described in Chapters 6 and 7 indicate that emission from the Monsanto
Plant in Idaho could result in lung doses to nearby individuals of about
30 mrad per year due to inhaled alpha particle emitters. A 30 mrad
annual dose of alpha radiation results in a dose equivalent rate to the
lung of 600 mrem per year.*
*The dose equivalent rate to other organs is 30-100 times smaller,
8-64
-------
Table 8.3-1 indicates a risk of 460 fatalities per 10^ organ rad
for alpha emitters in lung tissue. The life time cancer from this
exposure is;
460 fatalities 0.03 rad ,. _ „ nn ,„
_ x _— x 70.7y = 0.98 x 10
organ rad y
c.f. Table 6.3-13 in Volume 2 of this BID. This is twice the risk due
to low-LET background radiation calculated by means of the BEIR-3 linear
quadratic model and more than half of the risk calculated by means of
the BEIR-3 linear model.
The 1982 UNSCEAR report indicates that the average annual dose to
the endosteal surfaces of bone due to naturally occurring high-LET alpha
radiation is about 6 mrad per year or, for a quality factor 20, 120 mrem
per year (UNSCEAR82). Table 8.3-1 indicates that the life time risk of
fatal bone cancer due to this portion of the naturally occurring
radiation background is
20 cases 0.006 rad 6
Tr\f> T x —ssg-y x 70.7 years - 8.5 x 10 u
10° person rad year J
The exposure due to naturally occurring background radon-222
progeny in the indoor environment is not well known. The 1982 UNSCEAR
report lists for the U.S. an indoor concentration of about 0.004 working
levels (15 Bq m~3, (UNSCEAR82). This estimate is not based on a
national survey and is known to be exceeded by as much as a factor of
ten or more in some houses. However, as pointed out in UNSCEAR82, the
national collective exposure is not too dependent on exceptions to the
mean concentration.
Assuming 0.004 WL is a reasonable estimate for indoor exposure to
radon-222 progeny, the IPA exposure model outlined in 8.4 yields a mean
life time exposure, indoors, of 6.7 WLM. In Section 8.24 two risk
coefficients for lung cancer due to radon progeny are presented. The
largest, 700 fatalities per 10^ person WLM, yields a probability of
death of 0.0047. That is, about one-half percent of all deaths are
estimated as due to naturally occurring indoor radon progeny. We note
that this is comparable to the 1 percent fatality incidence estimated
above for low-LET background radiation. The smaller risk coefficient
listed in 8.4, 300 fatalities per 10" person WLM, implicates radon
progeny in about 0.2 percent of all deaths. The reader is cautioned,
however, that these risk estimates only apply to the U.S. population
taken as a whole, i.e. men and women, smokers and nonsmokers. While we
believe they are reasonable estimates for the U.S. 1970 population in
which the vast majority of the lung cancer mortality occurred in male
8-65
-------
smokers, we do not believe these risk estimates can be applied
indiscriminately to women or nonsmokers. As noted in Section 8.4, the
risk to these groups may not be comparable.
The spontaneous incidence of serious congenital and genetic
abnormalities has been estimated to be about 105,000 per 10^ live
births, about 10,5 percent of live births (HAS80, UNSCEAR82). The low-
LET background radiation dose of about 90 mrad/year in soft tissue
results in a genetically significant dose of 2.7 rad during the 30 year
reproductive generation. Since this dose would have occurred in a large
number of generations, the genetic effects of the radiation exposure are
thought to be an equilibrium level of expression. As noted in 8-6,
since genetic risk estimates vary by a factor of 20 or more EPA uses a
log mean of this range to obtain an average value for estimating genetic
risk. Based on this average value, the background radiation causes 700
to 1000 genetic effects per IQ& live births, depending on whether or not
the oocyte is as sensitive to radiation as the spermatogonia, see 8.6.
This result indicates that about 0.67 to 0.95 percent of the current
spontaneous incidence of serious congenital and genetic abnormalities
may be due to the low-LET background radiation.
The gonadal dose and genetic risk from airborne radionuclide
emissions is usually quite small. For example, the 30 year gonodal dose
due to the Monsanto plant, referred to above, is about 0.8 mrad, high
LET, and 0.3 mrad, low LET. From Table 8.6-5, the risk of serious
hereditary disorder from these doses, assuming equal male and female
sensitivity is:
7400
106 live births
370
106 live births
x 0.8 x 10-3 = 5.9 x 10~6 high LET
x 0.3 x 10~3 = 0.1 x 10~6 low LET
or afccut 6 cases in a million live births. This is the total for all
generations. Ten to twenty percent of these might occur in the first
generation after exposure of the parents. The total for all generations
is e. hundred times smaller than the estimated cancer risk from this
source, a result that is quite general for radionuclide air emissions of
particulates,
8-66
-------
REFERENCES
AraSl Archer V. E., Health Concerns in Uranium Mining and Milling,
J. Occup. Med. ^3, 502-505, 1981.
Arb54 Arey L. B., Developmental Anatomy, 6th ed., W. B. Sacnders,
Philadelphia, 1954.
Aua67 Auxier J. A., Cheka J. S.» Haywood F. F., Jones T. D., and
Thorngate J. H.t Free-Field Radiation Bose Distributions from
the Hiroshima and Nagasaki Bombings, Health Phys. 12, 425-
429, 1967. ~~
Aub77 Auxier J. A., Ichiban - Radiation Dosimetry for the Survivors
of the Bombings of Hiroshima and Nagasaki, TID 27080,
Technical Information Center, Energy Research and Development
Administration, National Technical Information Service,
Springfield, Va., 1977.
Ba81 Baverstock K. F., Papworth D., and Vennart J., Risks of
Radiation at Low Dose Rates, Lancet, 430-433, Feb. 21, 1981.
Be78 Beebe G. W., Kato H., and Land C. E,, Studies of the
Mortality of A-bomb Survivors, 6: Mortality and Radiation
Dose, 1950-74, Rad. Res., 25, 138-201 (RERF TR 1-77, Life
Study Report 8), 1978.
Bo82 Bond V. P. and Thiessen J. W., Reevaluations of Dosimetric
Factors, Hiroshima and Nagasaki, DOE Symposium Series 55,
CONF-810928, Technical Information Center, U.S. Department of
Energy, Washington, D.C., 1982.
Bu81 Bunger B., Cook J. R,, and Barrick M. K., Life Table
Methodology for Evaluating Radiation Risk: An Application
Based on Occupational Exposure, Health Phys. 40, 439-455,
1981.
ChaSl Chameaud J., Perraud R., Chretien J., Masse R., and Lafuma
J., Contribution of Animal Experimentation to the Interpre-
tation of Human Epidemi.ologi.cal Data, in: Proc. Int. Conf.
on Hazards in Mining: Control, Measurement, and Medical
Aspects, October 4-9, 1981, Golden, Colo., pp. 228-235,
edited by Manual Gomez, Society of Mining Engineers, New
York, 1981.
8-67
-------
Chb83
Co 78
Da72
Doa83
Dob83
Doc84
Dod84
Ela77
Elb79
SPA 78
Charles M. E., Lindop P. J., and Mill A. J., A Pragmatic
Evaluation of the Repercussions for Radiological Protection
of the Recent Revisions in Japanese A-bomb Dosiaetry, IAEA
SM-266/52, Proceedings International Symposium on the
Biological Effects of Low-Level Radiation with Special Regard
to Stochastic and Non-stochastic Effects. Venice, IAEA,
Vienna April 11-15, 1983.
Cook J. R., lunger B. M., and Barrick M. K., A Computer Code
for Cohort Analysis of Increased Risks of Death (CAIRO), ORP
Technical Report 520/4-78-012, U.S. Environmental Protection
Agency, Washington, B.C., 1978.
Davie R., Butler N», and Goldstein H., From Birth to Seven,
Longmans, London, 1972. Cited in St84.
Dobson R. L. and Felton J. S., Female Germ Cell Loss from
Radiation and Chemical Exposures, Amer. J. Ind. Med.. 4
175-190, 1983. -'
Dobson R. L., Straume J., Felton J. S., and Kwan T. C.,
Mechanism of Radiation and Chemical Oocyte Killing in Mice
and Possible Implications for Genetic Risk Estimation,
[abstract], Environ. Mutagen., 5, 498-499, 1983.
Dobson R. L., and Straume T., Hutagenesis in Primordial Mouse
Oocytes could be Masked by Cell Killing; Monte Carlo
Analysis, Environ. Mutagen. 6, 393, (1984) [Abstract].
Dobson R. L., Kwan T. C., and Straume T., Tritium Effects on
Germ Cells and Fertility, pp. 285-298, in Radiation
Protection, European Seminar on the Risks from Tritium
Exposure, EUR9065en, Commission of the European Cormmnities,
Ellett W. H. and Richardson A.C.B., Estimates of the Cancer
Risk Due to Nuclear Electric Power Generation, pp. 511-527,
in Origins of Human Cancer, Book A., H. H. Hiatt et al.,
eds., Cold Spring Harbor Laboratory, 1977.
Ellett W. H. and Nelson N. S., Environmental Hazards from
Radon Daughter Radiation, pp. 114-148, in:
Conference/Workshop on Lung Cancer Epidemiology and
Industrial Applications of Sputum Cytology, Colorado School
of Mines Press, Golden, Colorado, 1979.
Environmental Protection Agency, Response to Comments:
Guidance on Dose Limits for Persona Exposed to Transuranium
Elements in the General Environment, EPA 520/4-78-010, Office
of Radiation Programs, U.S. EPA, Washington, D.C., 1978.
8-68
-------
EPA79 Environmental Protection Agency, Indoor Radiation Exposure
Due to Radium-226 in Florida Phosphate Lands, EPA 520/4-78-
013, Office of Radiation Programs, USEPA, Washington, B.C.,
revised printing, July 1979.
EPA81 Environmental Protection Agency, Population Exposure to
External Natural Radiation Background in the United States,
Technical Note ORP/SEPD-80-12, Office of Radiation Programs,
USEAP, Washington, B.C., 1981.
EPA82 Environmental Protection Agency, Final Environmental Impact
Statement for Remedial Action Standards for Inactive Uranium
Processing Sites (40 CPR 192), Volume I, EPA 520/4-82-013-1,
Office of Radiation Programs, USEPA, Washington, D.G., 1982.
EPA83a Environmental Protection Agency, Draft Background Information
Document, Proposed Standards for Radionuclides, EPA 520/1-83-
001, Office of Radiation Programs, USEPA, Washington, D.C.
1983.
EPA83b Environmental Protection Agency, Final Environmental Impact
Statement for Standards for the Control of Byproduct
Materials from Uranium Ore Processing (40 CFR 192), Volume I,
EPA 520/1-83-008-1, Office of Radiation Programs, USEPA,
Washington, B.C., 1983.
Ev79 Evans H. J., Buckton K. E., Hamilton G. E., et al.,
Radiation-induced Chromosone Aberrations in Nuclear Dockyard
Workers, Nature, 277, 531-534, 1979.
FRC67 Federal Radiation Council, Radiation Guidance for Federal
Agencies, Memorandum for the President, July 21, 1967, Fed.
Reg., 32, 11183-84, August 1, 1967.
Ga82 Garriott M. L. and Grahn D., Neutron and Y ray Effects
Measured by the Micronucleus Test, Mut. Res. Let., 105, 157-
162, 1982.
Goa29 Goldstein L. and Murphy D. P., Etiology of Ill-health of
Children Born After Maternal Pelvic Irradiation: II,
Defective Children Born After Post Conception Pelvic
Irradiation, Amer. J. Roentgenol. Rad. Ther., 22, 322-331,
1929.
GobSO Goodhead D. T., Models of Radiation Interaction and
Mutagenesis, pp. 231-247, in Radiation Biology in Cancer
Research, R. E. Meyn and H. R. Withers, eds., Raven, New
York, 1980.
Gob82 Goodhead D. T., An Assessment of the Role of Microdosimetry
in Radiobiology, Rad. Res., 91, 45-76, 1982.
8-69
-------
Gra83
Grb83
Gu77
HaaSl
Hab82
Hac82
Had83
He83
Hia53
Hib54
Hic66
Hca77
60c
, U.S.
of Energv, Washington, D.C., January
PaSterMk B' S., Environmental
Radon
coramuni"tlon to Dr. N. Nelson, Office
"
Herbert D. E., Model or Metaphor? More Comments on the BEIR
Phys 'S B^nm3?7"390' ^ EPid-iol°gy Applied to ^Ith
Phys., COKF-830101, DE-83014383, NTIS, Springfield, Va. 1983.
Hicks S. P., Developmental Malformations Produced by
Radiation. A Timetable of Their Development, Amer J
Roentgenol. Radiat. Thera., 69, 272-293, 1953.
»5:/r '
System, J.
; sss:
., 43 (Supp!. 1). 151-178,
Hicks S P. and D'Amato C. J., Effects of Ionizing Radiations
on Mammalian Development, Adv. Teratol., I, 195-266, 1966
Hofmann W. and Steinhausler F., Dose Calculations for Infants
and Youths due to the Inhalation of Radon and it. Decay
4th Interiati^ ^^^ Envir°nment> in= Proceedings of the
n a ^ongress of the International Radiation
Protection Association, Paris, 2, 497-500, 1977.
8-70
-------
HobSl Hornung R, W. and Samuels S,, Survivorship Models for Lung
Cancer Mortality in Uranium Miners - Is Cumulative Dose an
Appropriate Measure of Exoosure?, in: Proc. Int. Conf. on
Hazards in Mining; Control, Measurement, and Medical
Aspects, October 4-9, 1981, Golden, Colorado, 363-368, edited
by Manuel Gomez, Society of Mining Engineers, New York, 1981.
HWC73 Health and Welfare Canada, The Testing of Chemicals for
Carcinoginicity, Mutagenicity and Teratogenicity, Health
Protection Branch, HWC, Ottawa, 1973.
ICRP75 International Commission on Radiological Protection,
Committee II on Permissible Dose for Internal Radiation, Task
Group on Reference Mtn, ICRP Publ. 23, Pergamon Press, 1975.
ICRP77 International Commission on Radiological Protection, Recom-
mendations of the International Commission on Radiological
Protection, ICRP Publ. 26, Aim. ICRP, _1 (1), Pergamon Press,
1977.
ICRP79 International Commission on Radiological Protection, Limits
for Intakes of Radionuclides by Workers, ICRP Publication 30,
Part 1, Ann. ICRP, 2 (3/4), Pergamon Press, New York, 1979.
ICRP81 International Commission on Radiological Protection, Limits
for Intakes of Radionuclides by Workers, ICRP Publication 32,
Part 3, Ann. ICRP, 6_ (2/3), Pergamon Press, 1981.
Is79 Ishimaru T., Otake M,, and Ishimaru M., Dose-Response
Relationship of Neutrons and Gamma-rays to Leukemia Incidence
of Cancer in 1950-1971, Based on thu Tumor Registry,
Nagasaki, Radiat. Res., _77, 377-394, 1979.
JaaSO Jacobi W. and Eisfeld K., Dose to Tissue and Effective Dose
Equivalent by Inhalation of Radon-222 and Radon-220 and Their
Short-Lived Daughters, GFS Report S-626, Gesellschaft fuer
Strahlen und Unweltforschung mbH, Munich, 1980.
Jab70 Jacobsen L., Radiation Induced Foetal Damage, Adv. Teratol.,
4, 95-124, 1970.
JacSl James A. C., et al., Respiratory Tract Dosimetry of Radon and
Thoron Daughters; The State-of-tiie-Art and Implications for
Epidemiology and Radiobiology, in; Proc. Int. Conf. on
Hazards in Mining: Control, Measurement, and Medical
Aspects, October 4-9, 1981, Golden, Colorado, 42-54, edited
by Manuel Gomez, Society of Mining Engineers, New York, 1981.
Kaa73 Kato H., LatP Effects in Children Exposed to the Atomic Bomb
While In Utero, Technical Report 18-73, Atomic Bomb Casualty
Commission, Hiroshima, 1973,
8-71
-------
382 Kato H. and Schull W. J., Studies of the Mortality of A-bomb
Survivors, 7. Mortality, 1950-1978: Part I, Cancer Mortality,
Rad. Research 90, 395-432, 1982. (Also published by the
Radiation Effect Research Foundation as: RERF TR 12-80, Life
Span Study Report 9. Part 1.)
a72 Kellerer A. M. and Rossi H. H,, The Theory of Dual Radiation
Action, Curr. Topics Rad. Res. Qaart., 8, 85-158, 1972,
a81 Kerr G. D., Review of Dosimetry for the Atomic Bomb
Survivors, in Proceedings of the Fourth Symposium on Neutron
Dosimetry, Gesellschaft fur Strahlen- und-Pmweltforsehung,
Munich-Neuherberg, Federal Republic of Germany, June 1-5, JL,
501, Office for Official Publications of the European
Communities, Luxemburg, 1981.
b81 Kerr G. D., Findings of a Recent ORNL Review of Dosimetry for
the Japanese Atomic Bomb Survivors, QRNL/TM-8078, Oak Ridge
National Laboratory, Oak Ridge, Tennessee, 1981.
62 King R. C., Genetics, Oxford University Press, Hew York, 1962.
a80 Land C. E., Boice J. D., Shore R. E., Norman J, I., and
Tokunaga M., et al., Breast Cancer Risk from Low-Dose
Exposures to Ionizing Radiation: Results of Parallel
Analysis of Three Exposed Populations of Women, J. Natl.
Cane, Inst., 65, 353-376, 1980.
b78 Land C. E. and Norman J, E., Latent Periods of Radiogenic
Cancers Occurring Among Japanese A-bomb Survivors, in: Late
Biological Effects of Ionizing Radiation, _!, 29-47, IAEA,
Vienna, 1978.
c83 Land C. E. and Pierce D, A., Some Statistical Considerations
Related to the Estimation of Cancer Risk Following Exposure
to Ionizing Radiation, pp. 67-89, in Epidemiology Applied to
Health Phys., CONF-830101, DE83014383, NTIS, Springfield,
Va., 1983.
.62 Lea D. E., Actions of Radiations on Living Cells, 2nd
edition, Cambridge University Press, 1962.
18! Loewe W. E. and Mendelsohn E., Revised Dose Estimates at
Hiroshima and Nagasaki, Health Phys., 41, 663-666, 1981.
ia59 Mandansky A., The Fitting of Straight Lines When Both
Variables are Subject to Error., J. Amer. Statis. Assoc., 54,
173-205, 1959.
8-72
-------
Mab83 Mays C. W. and Spiess H., Epidemiologies! Studies in German
Patients Injected with Ra-224, pp. 159-266, in Epidemiology
Applied to Health Physics, CONF-8301Q1, DE-83014383, NTIS,
Springfield, Va, 1983.
Mc78 McDowell E. M., Mclaughlin J. S., Merenyi D. K,t Kieffer
R. F,, Harris C. C., and Trump B. F,, The Respiratory
Epithelium V, Histogenesis of Lung Carcinomas in Humans,
J. Natl. Cancer Inst., 61, 587-606, 1978.
Mi78 Michel C. and Fritz-Niggli H., Radiation-Induced Develop-
mental Anomalies in Mammalian Embryos by Low Doses and
Interaction with Drugs, Stress and Genetic Factors, pp. 399-
408, in Late Biological Effects of Ionizing Radiation, Vol.
II, "IAEA, Vienna, 1978,
•
Moa79 Mole R. H,, Carcinogenesis by Thorotrast and Other Sources of
Irradiation, Especially Other a-Emitters, Environ. Res., 18,
192-215, 1979.
Mob67 Morgan K. Z. and Turner J. E., Principles of Radiation
Protection, John Wiley and Sons, Inc., New York, 1967.
Mua83 Muller J., Wheeler W. C., Gentleman J. F., Suranyi G., and
Kusiak R. A., Study of Mortality of Ontario Miners, 195^-
1977, Part I, Ontario Ministry of Labor, Ontario, May 1>83.
Mub29 Murphy D. P., The Outcome of 625 Pregnancies in Women Subject
to Pelvic Radium or Roentgen Irradiation, Amer. J. Obstet.
Gyn., 18, 179-187, 1929.
Muc30 Murphy D. P. and DeRenyi M., Postconception Pelvic
Irradiation of the Albino Rat (Mus Norvegieus): Its Effects
Upon the Offspring, Surg. Gynecol. Obstet., 50, 861-863,
1930,
NAS72 National Academy of Sciences - National Research Council, The
Effects of Populations of Exposures to Low Levels of Ionizing
Radiation, Report of the Committee on the Biological Effects
of Ionizing Radiations (BEIR Report), Washington, D.C., 1972.
KAS80 National Academy of Sciences - National Research Council, The
Effects of Populations of Exposure to Low Levels of Ionizing
Radiation, Committee on the Biological Effects of Ionizing
Radiation, Washington, D.C., 1980.
NASA73 National Aeronautics and Space Administration, Bioastro-
nautics Data Book, NASASP-3006, 2nd Edition, edited by J. R.
Parker and V. R. West, NASA, Washington, B.C., 1973,
8-73
-------
NCHS73 National Center for Health Statistics, Public Use Tape, Vital
Statistics - Mortality Cause of Death Summary - 1970, PB80-
133333, NTIS, Washington, D.C., 1973.
NCHS75 National Center for Health Statistics, U.S. Decennial Life
Tables for 1969-71, 1(1), DHEW Publication No. (HRA) 75-1150,
U.S. Public Health Services, NCHS, Rockville, Md., 1975.
NCRP75 National Council on Radiation Protection and Measurement,
Natural Background Radiation in the United States, NCRP
Report No, 45, NCRPM, Washington, B.C., 1975.
NCRP77 National Council on Radiation Protection and Measurements,
Protection of the Thyroid Gland in the Event of Releases of
Radioiodine, NCRP Report No. 55, NCRPM, Washington, B.C.,
1977.
NCRP80 National Council on Radiation Protection and Measurements,
Influence of Dose and Its Distribution in Time on Dose-
Response Relationships for Low-LET Radiation, NCRP Report No,
64, NCRPM, Washington, D.C., 1980.
NCRP84 National Council on Radiation Protection and Measurements,
Evaluation of Occupational and Environmental Exposures to
Radon and Recommendations, NCRP Report No. 78, NCRPM,
Washington, D.C., 1984.
Nea56 Neel J. V. and Schull W. J., The Effect of Exposure to the
Atomic Bombs on Pregnancy Termination in Hiroshima and
Nagasake, National Academy of Sciences, Publ. 461,
Washington, D.C., 1956.
Neb70 Nelson K. B. and Deutschberger J., Head Size at One Year as a
Predictor of Four-Year I.Q., Develop. Med. Child Neurol., 12,
487-495, 1970.
ORNL84 Oak Ridge National Laboratory, Age Dependent Estimation of
Radiation Dose, lin press], 1984.
OJ8Q Oftedal P. and Searle A. G., An Overall Genetic Risk
Assessment f^r Radiological Protection Purposes, J. Med,
Genetics, JJ, 15-20, 1980.
Oh65 Ohzu E., Effects of Low-Dose X-Irradiation on Early Mouse
Embryos, Rad. Res. 26, 107-113, 1965.
Ot83 Otake M. and Schull W. H., Mental Retardation in Children
Exposed In Utero to the Atomic Bombs: A Reassessment,
Technical Report RERFTR 1-83, Radiation Effects Research
Foundation, Hiroshima, 1983.
8-74
-------
Ot84 Otake M. and Schull W, J,, In Utero Exposure to A-boob Radia-
tion and Mental Retardation: A Reassessment, Brit. J.
Radiol., _57, 409-414, 1984.
P152 Plusnaer G. W., Anomalies Occurring in Children Exposed in
Utero to the Atomic Bomb in Hiroshima, Pediat., Jj), 687-692,
1952.
Po78 Pohl-Ruling J., Fischer P., and Pohl E», The Low-Level Shape
of Dose Response for Chromosome Aberration, pp. 315-326 in
Late Biological Effects of Ionizing Radiation, Volume II,
International Atomic Energy Agency, Vienna, 1978.
Pr83 Prentice R. L., Yoshimoto Y., and Mason, M. W., Relationship
of Cigarette Smoking and Radiation Exposure to Cancer
Mortality in Hiroshima and Nagasaki, J. Nat. Cancer Inst.,
W, 611-622, 1983.
Ra84 Radford E. P. and Renard K. G. St. Cl., Lung Cancer in
Swedish Iron Miners Exposed to Low Doses of Radon Daughters,
N. Engl. J. Med., 310, 1485-1494, 1984.
RERF78 Radiation Effects Research Foundation, 1 April 1975 - 31
March 1978. RERF Report 75-78, Radiation Effects Research
Foundation, Hiroshima, 1978.
RERF83 Radiation Effects Research Foundation, Reassessment of Atomic
Bomb Radiation Dosimetry in Hiroshima and Nagasaki, Proc. of
the U.S.-Japan Joint Workshop, Nagasaki, Japan, Feb. 16-17,
1982, Radiation Effects Research Foundation, Hiroshima, 730,
Japan, 1983.
RERF84 Radiation Effects Research Foundation, Second U.S.-Japan
Joint Workshop for Reassessment of Atomic Bomb Radiation
Dosimetry in Hiroshima and Nagasaki, Radiation Effects
Research Foundation, Hiroshima, 730, Japan, 1984.
Rowland R. E., Stehney A. F., and Lucas H. F,, Dose Response
Relationships for Female Radium Dial Workers, Rad. Res. 16,
368-383, 1978.
Rugh R., Vertebrate Radiobiology: Embryology, Ann. Rev.
Nucl. Sci., 3, 271-302, 1953.
Rugh R., The Effects of Ionizing Radiation on the Developing
Embryo and Fetus, Seminar Paper No. 007, Bureau of
Radiological Health Seminar Program, Public Health Service,
Washington, D.C., 1970.
Rua71 Rugh R., X-ray Induced Teratogenesis in the Mouse and Its
Possible Significance to Man, Radiol., 99, 433-443, 1971.
Ro78
Iua53
Rua70
8-75
-------
RubSO Russell L, B., X-ray Induced Developmental Abnormalities in
the Mouse and Their Use in the Analysis of Embryological
Patterns, I. External and Gross Visceral Changes, J, Exper.
Zool., 114, 545-602, 1950.
Rub56 Russell L. B., X-Ray Induced Developmental Abnormalities in
the Mouse and Their Use in the Analysis of Embryological
Patterns, II: Abnormalities of the ¥eretebral Column and
Thorax, J. Exper. Zool,, 131, 329-390, 1956.
Rub57 Russell L. B., Effects of Low Doses of X-rays on Embryonic
Development in the Mouse, Proc. Soc. Exptl. Biol. Med., 95,
174-178, 1957.
Ruc54 Russell L. B. and Russell W. L., An Analysis of the Changing
Radiation Response of the Developing Mouse Embryo, J. Cell,
Comp. Physiol., 43(Suppl. 1), 103-149, 1954.
Sa82 Satoh G., et al., Genetic Effects of Atomic Bombs, in: Human
Genetics, Part A: The Unfolding Genome, A. R. Liss, Inc.,
New York, 267-276, 1982.
Sc81 Schull W. J., Otake M., and Neel J. V., Genetic Effects of
the Atomic Bombs; A Reappraisal, Science, 213, 1220-1227,
1981.
Se69 Senyszyn, J. J. and Rugh R., Hydroeephaly Following Fetal
X-Irradiation, Radiol., 9_3, 625-634, 1969.
Sm78 Smith P. G. and Doll R., Radiation-Induced Cancers in
Patients with Ankylosing Spondylitis Following a Single
Course of X-ray Treatment, in: Proc. of the IAEA Symposium,
Late Biological Effects of Ionizing Radiation, _1, 205-214,
IAEA, Vienna, March, 1978.
Spa56 Spector W. S., editor, Handbook of Biological Data, Table
314, Energy Cost, Work: Man, W. B. Sanders Co., Philadelphia,
1956.
Spb83 Spiers F. W., Lucas H. F.. Rundo J., and Anast G. A.,
Leukemia Incidence in the U.S. Dial Workers, in: Conference
Proc. on Radiobiology of Radium and the Actinides in Man,
October 11-16, 1981, Health Phys., 44 Suppl. 1, 65-72, 1983.
Sta84 Stein Z. A. and Susser M. W., The Epidemiology of Mental
Retardation, in Epidemiology of Pediatric Neurology, B.
Schoenberg, editor, Marcel Dekker, Inc., New York, [in
press], 1984.
Stb21 Stettner E., Bin weiterer Fall einer Schadingung einer
menschichen Frucht durch Roentgen Bestrahlung., Jb.
Kinderheilk. Phys. Erzieh., 9_5_, 43-51, 1921.
8-76
-------
StcSl Straume T, and R. L. Dobson, Implications of New Hiroshima
and Nagasaki Dose Estimates: Cancer Eisks and Neutron RBE,
Health Phys. 41, 666-671, 1981.
Th82 Thomas D. C. and McNeill K. G., Risk Estimates for the Health
Effects of Alpha Radiation, Report INFO-0081. Atomic Energy
Control Board, Ottawa, 1982.
ToaSO Tobias C. A., et al., The Repair-Misrepair Model, pp. 195-
230, in R. E. Meyn and H. R. Withers, eds., Raven, New York,
1980.
Tob84 Tokunaga M., Land C. E., Yamamoto T., Asano M., Takioka S.,
Ezaki E,, and Nishimari I., Incidences of Feamle Breast
Cancer Among Atomic Bomb Survivors, Hiroshima and Nogasaki,
1950-1980, RERF TR 15-84, Radiation Effects Research
foundation, Hiroshima, 1984.
U182 Ullrich R. L., Lung Tumor Induction in Mice: Neutron RBE at
Low Doses, DE 82009642. National Technical Information
Service, Springfield, ?a., 1982.
UNSCEAR58 United Nations, Report of the United Nations Scientific
Committee on the Effects of Atomic Radiation, Official
Records: Thirteenth Session, Supplement No. 17 (A.3838),
United Nations, New York, 1958.
UNSCEAR62 United Nations, Report of the United Nations Scientific
Committee on the Effects of Atomic Radiation, Official
Records: Seventeenth Session, Supplement No. 16 (A/5216),
United Nations, New York, 1962.
UNSCEAR66 United Nations, Report of the United Nations Scientific
Committee on the Effects of Atomic Radiation, Official
Records: Twenty-First Session, Supplement No. 14 (A/6314),
United Nations, New York, 1966.
UNSCEAR69 United Nations, Report of the United Nations Scientific Com-
mittee on the Effects of Atomic Radiation, Supplement No, 13
(A/7613), United Nations, New York, 1969.
UNSCEAR72 United Nations Scientific Committee on the Effects of Atomic
Radiation, Ionizing Radiation: Levels and Effects, Volume
II: Effects, A Report to the General Assembly. Sales No. E.
72. IX.18., United Nations, New York, 1972.
UNSCEAR77 United Nations Scientific Committee on the Effects of Atomic
Radiation, Sources and Effects of Ionizing Radiation, Report
to the General Assembly, with Annexes, UN publication E.77
IX.1., United Nations, New York, 1977.
8-77
-------
UNSCEAR82 United Nations Scientific Committee on the Effects of Atomic
Radiation, Ionizing Radiation: Sources and Biological
Effects, 1982 Report to the General Assembly, Sales No.
B.82.IX.8, United Nations, New York, 1982.
USRPC80
Va80
Wa83
Wh83
WH075
Wi54
Woa65
Wob66
U.S. Radiation Policy Council, Report of the Task Force on
Radon in Structure, USRPC-80-0Q2, Washington, D.C., 1980.
Van Buul P.P.W., Dose-response Relationship for X-ray Induced
Reciprocal Translocations in Stem Cell Sperraatogonia of the
Rhesus Monkey (Macaca mulatta), Mutat. Res., 73, 363-375
1980. (Cited in ONSCEAR82.) — *
Wakabayashi T., Kato H. , Ikeda T., and Schull W. J., Studies
of the Mortality of A-bomb Survivors, Report 7, Part III,
Incidence of Cancer in 1959-78 Based on the Tumor Registry,
Nagasaki, Radiat. Res., j?3, 112-142, 1983.
Whittemore A. S. and McMillan A., A Lung Cancer Mortality
Among U.S. Uranium Miners? A Reappraisal, Technical Report
No. 68, SIAM Inst. Math. Soc., Stanford University, Stanford,
1983.
World Health Organization, International Statistical
Classification of Diseases, Injuries, and Causes of Death.
9th Revision, WHO, Geneva, 1975.
Wilson J. G., Differentiation and the Reaction of Rat Embryos
to Radiation, J. Cell. Camp. Physiol., 43 (Suppl 1). 11-37,
Wood J- W., Johnson K. G., and Omari Y., In Utero Exposure to
the Hiroshima Atomic Bomb: Follow-up at Twenty Years
Technical Report 9-65, Atomic Bomb Casualty Commission,
Hiroshima, 1965.
Wood J. W., Johnson K. G., Omari Y., Kawamoto S., and Keehn
R. J., Mental Retardation in Children Exposed in Utero to the
Atomic Bomb— Hiroshima and Nagasaki, Technical Report 10-66,
Atomic Bomb Casualty Commission, Hiroshima, 1966.
8-78
-------
Chapter 9; SUMMARY OF DOSE AND RISK ESTIMATES
9.1 Introduction
This chapter summarizes the calculated doses and risks for the
facilities analyzed in Chapters 2 through 7 of Volume II. Also, overall
uncertainties in these estimates are discussed.
Four separate steps are involved in estimating the health impact of
a specific source of radioactivity; (1) measurement of emissions of
radionuclides to air from the source, (2) estimation of the radionuclide
concentration and annual intake of radionuclides at various locations,
(3) calculation of the estimated dose and risk resulting from a unit
intake or unit concentration of radioactivity in the environment, and
(4) a means of scaling the risk estimates to match the specific source.
In EPA s analysis, each step is associated with a computer code that
performs the necessary calculations; the relationship of these codes is
illustrated in Figure A-l (Addendum A).
EPA uses the AI1DOS-EPA code (Mo79, Ba81) to analyze radionuclide
emissions into air from a specific source. The results of this analysis
are estimates of air and ground surface radionclide concentrations,
intake rates via inhalation of air, and ingestion of radioactivity via
meat, milk, and fresh vegetables. Chapter 6 presents a description of
the techniques used and their limitations. The atmospheric and
terrestrial transport models used in the code, their implementation, and
the applicability of the code to different types of emissions are
described in detail in Mo79.
The computer code used to calculate dose and risk is RADRISK
(Dub84, Su81, DuaBO). RADRISK calculates the radiation dose and risk
resuiting from an annual unit, e.g., 1 pCi/y, intake of a given
radionuclide or the risk resulting from external exposure to a unit,
e.g., 1 pCi/m3, 1 pCi/m2, concentration of radionuclide in air or on
ground surface. Since both dose and risk models are linear, the unit
dose and risk results can then be scaled to reflect the conditions
associated with a specific source. The assessment of radiation doses is
discussed in Chapter 7j Chapter 8 discusses estimating the risk of
health effects.
Once the radionuclide intakes and concentrations are calculated for
a specific source by means of the environmental transport code, it is
necessary to scale the dose and risk values resulting from a unit intake
or concentration to the intake and concentration values predicted by the
9-1
-------
transport code. As shown in Figure A-l (Addendum A), the DARTAB
computer code (BeSl) performs this step using RADRISK unit doses and
risks and AIRDOS-EPA concentrations and intakes. DARTAB is independent
of both the environmental transport code, e.g., AIRDOS-EPA, and the
dosimetric and health effects code, e.g., RADRISK. This eliminates
redundant dose/risk calculations and the need for extraneous coding to
calculate doses and health impacts in each environmental transport code.
9.2 Doses and Risks for Specific Facilities
Tables 9.2-1 and 9.2-2 are summaries of the doses and risks to
critical groups of individuals and populations in the vicinity of
facilities that discharge radioactive emissions. Data for selected
facilities from each category are presented in the order they are
presented in Chapters 2 through 7 of Volume II.
These dose and risk values were estimated using the environmental
transport codes of AIRDOS-EPA, the dose and risk tables of DARfAB and
the risk estimates that compose the RADRISK code. More detailed infor-
mation, including a description of the facility, the processes causing
the emissions, estimates of rates of emission, and estimates of doses
and risks that result to individuals and populations are found in the
respective chapters of Volume II.
9.3 Overall Uncertainties
Although the doses and risks presented in Tables 9.2-1 and 9.2-2
seem well defined and sometimes given to more than one significant
figure, there are considerable uncertainties that persist when trying to
fix their exact value. The individual uncertainties in the components
which lead to the results in Tables 9.2-1 and 9.2-2 have been previously
discussed. Source term measurement errors were discussed in Chapter 4,
possible errors introduced in evaluating movement from the source
through various pathways were discussed in Chapter 6; variations which
could be introduced in the calculation of doses and dose rates were
evaluated in Chapter 7} finally, Chapter 8 discussed the potential
errors that could be introduced in the risk calculations.
9.3.1 Emission and Pathway Uncertainties
Measurement of emissions from sources have been estimated in
Chapter 4 to be valid within a factor of 1.4.
In the evaluation of pathways, the uncertainties in results
predicted by the atmospheric dispersion models make the most significant
contribution. As discussed in Section 6.2.3, the studies by Little
(Li79) and Miller (Mi82) indicate that for average annual concentra-
tions, an uncertainty of approximately a factor of 2 for locations
within 10 km of the release could be expected. Inasmuch as nearby
locations to releases are of greatest concern, this uncertainty value is
the most appropriate.
9-2
-------
Table 9.2-1. Doses and risks to nearby individuals
facility
DOt facilities
Peed Materials
Production Center
Oak Ridge leaervation
Portsmouth Gaseous
Diffusion Plant
Savannah River Plant
HtC faeilitiea
Research and test
reactor'*'
Accelerator '^)
•adiopharmaceutical
supplier*'*'
Afgltl(l)
U.S. Army facility
U.S. Havy facility'*)
TisaueU)
luag
lung
bone surface
thyroid
average all organs
average all organs
thyroid
average all organs
apleen
average all organs
Doae rate
(airaai/ year)
as
SO
11
4.9
1
0.0001
0.3
0.005
0.03
0.02
!4feti»s(l>,c,d) tisfc
(deaths/10*6 persons)
100 (100)
100 (100}
20 (20)
40 (20)
20 (8)
0.002 (0.0008)
0.2 (0.1)
0.09 (0.04)
0.4 (0.2)
0.3 (0.1)
•adiation aeurce
manufacturerI*'
Coal Fired toilers^*)
Utility boilers (rural)
Industrial boilers
Uranium Mint'e)
Ground level release
(at 2000 meters)
PIuse rise release
(at 2000 meters)
Phosphate
average all organi
bone surface
bone surface
lung
lung
0.2
5
0.4
4 (2)
30 (10)
0.6 (0.5)
Ho doie available 10,000 (5.000>
Ho date available 1,000 (500){h'
Drying and grinding^6*
Element si phosphorus
Pocstello, Idiho
Soda Springe, Idaho
Mineral Extraction Induatryt*)
Aliaainum reduction plant
Copper saielter
Zinc BBelter
Lead smelter
bone surface
lung
lung
kidney
lucg
bone lurface
lung
15
290
610
1.2
0.2
0.02
4.8
10
500
1000
0.8
O.J
0.02
8
'•'organ with highcit annual doae.
l^'liak ii that due to the total expoiure not juet that due to highlit organ. Thii value
represent! the exceaa caocert in a lifatuie for orgaa dote ritei ehown ac offtit*
pointa of higheat riak.
'c'lhe riak tttimatei in parantheaea include a doat rate reduction factor of 2.5 for low-LET
radiation, ai deacribed in Chapter 8 (Volume 1) of thia report.
'^'Jiak» arc (zpreated per Billion population; for individual rieki aulciplj each value by
10-*.
'•'lefennci facility.
M* Van de Craaff.
nJ force* ladiobiology Bueirch Initicute.
«luei in the firat coltmn are baaed on UIK-3 (NA.S80), KRPS (KUPBB2), and EPA BOdel*
(CoTg, «184, Mo79); the valuei io pareothaaee are baaed an UMSCEAt (UMSCEAK82) and 1C IP
(OflO) riik eeti«atea (lac Chapter 8, Voltmt I).
9-3
-------
Table 9.2-2. Doses and risks to regional population
Facility
DOE Facilities
Feed Materials
Production Center
Oak Ridge Reaervaticm
Portsaoush Giieouf
Savannah liver Flint
MIC Facilities
•esearch and ceit
reactor
Radiapharmaceutical
auppliera(d)
ATUtttO
tl.S, Army facility
0.5. IJavy facility^'
•anufacturer^)
Coal Fired Boiler*^)
Utility boiler* (rural)
Industrial boiler*
Uranium Mioe^d'
Phosphate Industry
Drying and grinding*"*
Met process fertiliier
Sleoeatal phocphoruB
?oc«Cfllo, Idiho
Soda Springs, Idaho
Mineral Exctictiou Induitry
Alunineia reduce ion plaot
Capper faclcer
Zinc taelter
Lead iaelter
Organ<«)
lung
lung
thyroid
average all or gins
. ^
thyroid
average all organ*
•pleen
average all orfaaa
average all organs
bane aurface
bone aurface
lung
bone surface
booe surface
lung
lung
CdJ
kidaey
lung
bone surface
lung
Collective
doae rate
(pera-rew/year )
4AO
212
35
120
34.0
Q.DQO&
3
0.002
0.09
0.09
8
UC
90
Ho doae available
110
41
1170
150
4.1
0.95
2.5
69
gi.ktb,c)
(fatal cancers /year)
0.01 (0.01)
O.OOi (0,OQ&>
<0.001
0.03 (0.01)
0.1 (0.04)
<0.001
<0.001
O.OOI
<0.001
0.001
0.0005
0.03
0.02
(v V(a Af Graaff.
^^Arwed Farce* Ksdiabialogy Eeaearch Institute.
C|)fhe values in the fine coluan are baaed on BElk-3 (HAS90), NRPB (H8PS82), and EPA
•odeIs (Co78, £184. Mo7
-------
9.3.2 Dose Uncertainties
As discussed in Chapter 7 and summarized in Section 7.7, dose
uncertainties are much less than would be implied by sensitivity
analyses of maximum ranges of variables. The large dose ranges possible
because of variation in individual characteristics must be modified by
consideration of the narrower ranges indicated by studies of real
populations; the dose range resulting from age dependence appears to be
small for lifetime exposures, and the range resulting from experimental
error is negligible by comparison. Based on these observations, it is
reasonable to estimate tha?: EPA's doses calculated on the basis of unit
intakes or unit concentrations should be accurate within a factor of
three or four. Much of the "uncertainty" in the dose calculation is not
caused by parameter error but reflects real differences in individual
characteristics within the general population. Therefore, the
uncertainty in the dose estimates cannot be dissociated from
specification of the segment of the population to be protected.
9.3.3 Risk Uncertainties
The uncertainties in estimating risk have been discussed in Chapter
8. Table 8.5-1 ranks and estimates the degree of uncertainty introduced
by various sources in estimating the risk of cancer. The uncertainties
listed in Table 8.5-1 are largely independent of each other and
therefore unlikely to be correlated in sign. Their root mean square sum
is about 300 percent, indicating the expectation that calculated risks
would be within a factor of three or so of the true value. (This result
is likely to be somewhat low because it does not include consideration
of the uncertainty introduced by the bias in the A-bomb dosimetry or by
the constrained regression analysis used by the BEIR-3 Committee.)
9.3.4 Overall Uncertainty
As indicated in the previous discussion, the individual uncer-
tainties which combine to provide a basis for the overall uncertainty in
risk evaluation are the following:
» Emission estimates are valid within a factor of 1.4*
• Air concentration estimates are valid within a factor of 2
* Dose calculations should be valid within a factor of 3 or 4
• Risk calculations should be valid within a factor of 3.
If these uncertainty estimates are independent and uncorrelated and can
reasonably be considered to estimate the 20 fractile of a log normal
*If the nominal value is multiplied or divided by the factor to
give a range, the true value is expected to be within that range.
9-5
-------
distribution, then the overall uncertainty in EPA's risk estimation can
be estimated as a factor of about 7*. That is the maximum expected
variation would range from about 15 percent to 700 percent of the
nominal value.
The various uncertainties, however, may not be uncorrelated or
independent. In this case, the overall uncertainty is likely to be less
than predicted by the above procedure.
EPA concludes that risk estimates in this Background Information
Document are accurate within a factor of 10, This estimate of
uncertainty is believed representative of state-of-the-art procedures
for estimating risks due to airborne radionuclide emissions.
-------
REFERENCES
Ba81 Baes C. F. Ill and Sharp R. D., A Directory of Parameters
Used in a Series of Assessment Applications ol the AIRDOS-EPA
and DARTAP. Computer Codes, ORNL-5720, Oak Ridge National
Laboratory, Oak Ridge, Tetm., March 1981,
Be81 Begoyich C. L., Eckerman K. F,, Schlatter E. C., Ohr S. Y.,
and Chester R. 0., DARTAB: A Program to Combine Airborne
Radionuclide Environmental Exposure Data with Dosimetric and
Health Effects Data to Generate Tabulation of Predicted
Impacts, ORKL/5692, Oak Ridge National Laboratory, Tenn.,
August 1981.
Co78 Cook J. R., Bunger B., and Barrick M. K., A Computer Code for
Cohort Analysis of Increased Risks of Death (CAIRO),
EPA 520/4-78-012, 1978,
DuaSO Dunning D. E, Jr., Leggett R. W.» and Yalcintas M, G., A
Combined Methodology for Estimating Dose Rates and Health
Effects from Exposure to Radioactive Pollutants, ORNL/TN-
7105. 1980.
Dub84 Dunning D. E. Jr., Leggett R. V., and Sullivan R. E., An
Assessment of Health Risk from Radiation Exposures, Health
Phys., 46(5)51035-1051, 1984.
E184 Ellett W. H., RABRISK/BEIR-3, Part I; Basis for EPA
Radiation Risk Estimates, to be published, 1984.
Li79 Little C. A. and Miller C. W., The Uncertainty Associated
with Selected Environmental Transport Models, ORNL-5528, Oak
Ridge National Laboratory, Oak Ridge Tenn., November 1979.
Mi82 Miller C. W. and Little C. A., A Review of Uncertainty
Estimates Associated with Models for Assessing the Impact of
Breeder Radioactivity Releases, ORNL-5832, Oak Ridge National
Laboratory, Oak Ridge, Tenn., August 1982.
9-7
-------
Mo79 Moore R. E., Baes C. F. Ill, McDowell-Boyer L. K.» Watson
A. P., Hoffman F. 0., Pleasant J. C., and Miller C. W,,
AIRDOS-EPA: A Computerized Methodology for Estimating
Environmental Concentrations and Dose to Man from Airborne
Releases of Radionuclides, EPA 520/1-79-009, EPA Office of
Radiation Programs, Washington, D.G., December 1979.
NAS80 National Academy of Sciences - National Research Council, The
Effects on Population of Exposure to Low Levels of Ionizing
Radiation, Committee on rh<: Biological Effects of Ionizing
Radiation, Washington, D.'-.., 1980,
NRPB82 National Radiological Protection Board, Gut Uptake Factors
for Plutonium, Asnericium and Curium, NRPB-R129, 1982.
Of80 Oftedal P. and Searle A. G., An Overall Genetic Risk
Assessment for Radiological Protection Purposes, J. Med,
Genetics, _T7, 15-20, 1980.
SuSl Sullivan R, 1., Nelson N. S., Ellett W. H., Dunning D. E,
Jr., Leggett R, W., Yalcintas M. G., and Eckerman K. F.,
Estimates of Health Risk from Exposure to Radioactive
Pollutants, ORNL/TM-7745, 1981.
UNSCEAR82 United Nations Scientific Committee on the Effects of Atomic
Radiation, Ionizing Radiation: Sources and Biological
Effects, 1982 Report to the General Assembly, Sales No.
E,82.IX.8, United Nations, New York, 1982.
9-8
-------
ADDENDUM f
COMPUTER CODES USED BY EPA TO ASSESS
DOSES FROM RADIATION EXPOSURE
A-l
-------
ADDENDUM A: COMPUTER CODES USED BY EPA TO ASSESS
DOSES FROM RADIATION EXPOSURE
CONTENTS
Page
A.I Introduction A-5
A.2 Overview of the EPA Analysis A-5
A.3 Dose Rates froa Internal Exposures , 4-7
A.4 Dose Rates from External Exposures A-12
TABLES
A-l Snail intestine to blood transfer fractiors, fj, for
transuranic elements » A-ll
FIGURES
A-l Assessment of radiological health impacts ...*.....*....... A-6
Preceding page blank
A-3
-------
ADDENDUM A: COMPUTER CODES USED BY EPA TO
ASSESS DOSES FROM RADIATION EXPOSURE
At 1 Introduction
This addendum (to Chapter 7) provides a brief overview of the
computer codes used by the Environmental Protection Agency (EPA) to
assess the health risk from radiation exposures. It describes how the
basic dose calculations are performed. Comprehensive descriptions of
the various parts of this methodology have been published in a series of
reports by the Oak Ridge National Laboratory and the Environmental
Protection Agency (Dub84, Be81, Ba81, Moa79). The risk estimates in
current use are described in Chapter 8 and reflect the change from the
BIIR-1 report (NAS72) to the BEIR-3 report (NAS80).
Three separate steps are required to estimate the health impact of
a specific source of radioactivity: (1) estimate at various locations
the radionuclide concentration and annual intake of radionuclides
resulting from specific sources of radioactivity in the environment,
(2) calculate the estimated dose and risk resulting from a unit intake
or unit concentration of radioactivity in the environment, and (3) use a
means of scaling the risk estimates to match the specific source. In
EPA's analysis, each step is associated with a computer code that
performs the necessary calculations, as illustrated in Figure A-l.
A. 2 Overview of theEPA Analysis
The computer code used to calculate dose and risk is RAD1ISK
(Dub84, Su81, DuaSO). RADRISK calculates the radiation dose and risk
resulting from an annual unit intake of a given radionuclide or the risk
resulting from external exposure to a unit concentration of radionuclide
in air or on ground surface. Since both dose and risk models are
linear, the unit dose and risk results can then be scaled to reflect the
exposure associated with a specific source.
As outlined in Chapter 7, estimates of the annual dose rate to
organs and tissues of interest are calculated using, primarily, models
recommended by the International Commission on Radiological Protection
(ICRP79, ICRP80). Because EPA usually considers lifetime exposures to a
general population, these dose rates are used in conjunction with a life
table analysis of the i—reased risk of cancer resulting froa radiation
(Co78). This analysis, described in Addendum B, takes account of both
competing risks and the age of the population at risk.
Preceding page blank
-------
AIRDOS-EPA
NUCLIDE
TRANSPORT
DATA
ENVIRONMENTAL
EXPOSURE AND
INTAKE RATE
n
DARTAB
IT
TABULATIONS
HEALTH
IMPACTS
DISOMETRIC AND
HEALTH EFFECTS
DATA
T__
/—x T s—s
RADRISK
HEALTH
EFFECTS
DATA
Figure A-l. Assessment of radiological health impacts,
A-6
-------
Various computer codes are available to predict how radionuclides
are transported through environmental pathways. As noted in Figure A-l,
IPA uses the AIRDOS-IPA code (Moa79, Ba81) to analyze the transport of
radionuclide emissions into air from a specific source. The results of
this analysis are estimates at various distances from the source of air
and ground surface radionuclide concentrations! intake rates via
inhalation of air, and ingestion of radioactivity via meat, milk, and
fresh vegetables. The atmospheric and terrestrial transport models used
in the code, their implementation, and the applicability of the code to
different types of emissions are described in Chapter 6.
A. 3 Dose Rates from Internal Exposures
Internal exposures occur when radioactive material is inhaled or
ingested. The RADRISK code implements contemporary dosimetric models to
estimate the dose rates at various times to specified reference organs
in the body from inhaled or ingested radionuclides. The dosimetric
methods in RADRISK are adapted from those of the INREM II code (Ki78),
based primarily on models recommended by the International Commission on
Radiological Protection (ICRP79). The principal qualitative difference
is that RADRISK computes dose rates to specified organs separately for
high and low linear energy transfer (LET) radiation^ . Aereas IHREM II
calculates the committed dose equivalent to specifi "• > »ans. The time-
dependent dose rates are used in the life table calcuj jcions of RADRISK.
In RADRISK, the direct intake of each nuclide is treated as a
separate case. For chains, the ingrowth and dynamics of daughters in
the body after intake of a parent radionuclide are considered explicitly
in the calculation of dose rate. Consideration is also taken of
different metabolic properties of the various radionuclides in a decay
chain .
•
The dose rate D£(X,t) to target organ X at time t due to
radionuclide i (l£i£N) residing in organs Yj, ¥2, ..., Ym is a measure
of the energy deposited annually in a given mass of tissue as a result
of radioactive decay, and is computed as:
m
D(X-*-Yjt) (A-l)
k=l
where
, (A-2)
activity of radionuclide i in organ Y^ at
time t measured from the initial intake of i
into the body,
A-7
-------
average dose rate to target organ X per unit
activity of the radionuclide i uniformly
distributed in source organ Y^ (Sn74, DuaSO).
The summation is taken over all source organs Y. Implicit in the
definitions is the assumption of uniform distribution of activity of
radionuclide i in each source organ, as is the assumption of averaging
the dose rate over the mass of the target organ. Although estimates of
dose to an organ include contributions from activity distributed
throughout the body (for penetrating radiations), activity within that
organ generally contributes the principal component of dose [i.e.,
D£(X*-Xjt) is the principal component of D£(Xjt)] .
The time rate of change of activity in the body is modeled by a
system of ordinary differential equations, with each equation describing
the rate of change of activity in a conceptual compartment of the body.
For radionuclides that are part of a decay series there may be formation
of radioactive daughters in a given compartment that have different
chemical and physical properties from those of the parent. Unlike the
models given in ICRP80, the specific metabolic properties of the
daughter are evaluated when they differ from those of the parent. This
refinement is under active consideration by ICRP experts. In almost all
cases, doses to soft tissues calculated on this basis differ only
slightly, if at all, from ICRP80 dose estimates, but the difference is
large for some radionuclides when the parent is incorporated into bone,
for example lead-210. For this radionuclide the 1CRP80 model has been
used without any modifications.
The pathways in the body by which activity is assumed to move were
illustrated in Chapter 7. Except for radon daughters, which are
considered separately, inhaled activity is assumed to be originally
deposited in the lungs (distributed among the nasal-pharyngeal,
tracheobronchial, and pulmonary regions), whereas ingested activity is
originally deposited in the stomach. From the lungs, activity may be
absorbed by the bloodstream or migrate to the stomach. Activity in the
stomach may proceed through the small intestine, upper large intestine,
and lower large intestine; activity may be absorbed by the bloodstream
from any of these four segments, although only absorption from the small
intestine is considered in this study.
The activity, Ajjc(t), of nuclide i in organ k may be divided among
several "pools" or "compartments", denoted here by the subscript SI.
Each differential equation describing the rate of change of activity
within a compartment is a special case of the equation:
i E1 Bi £ Ar*V« i-1--' %k (A'3)
A-8
-------
where
•
^iJtk = activity of radionuclide i in compartment I of organ k»
L£j,fc ~ number of exponential terms in the retention function for
nuclide i in organ k,
Bjj = branching ratio of nuclide j to nuclide i,
a
X. = rate coefficient (time'*) for radiological decay of
nuclide i,
•a
^i&k = rate coefficient (time~*) for biological removal of
nuclide i from compartment $, of organ k,
ci£k = fractional coefficient for nuclide i in the £-th
compartment of organ k,
I*ik = inflow rate of nuclide i into organ k.
If the inflow rate Pijj remains constant, the equations may be
solved explicitly for Af^Ct) as described by Killough, Dunning, and
Pleasant (Ki78). In many cases the inflow into a compartment will not
be a constant rate over a long period of time. To handle this problem,
the time interval over which solution of the activity equation is
desired (e.g., 110 years) is divided into 1-year subintervals. The
inflow rate on each subinterval is then taken to be that constant value
which would yield the total activity flowing out of the preceding
compartment (s) during the same subinterval.
The model used in RADRISK for particulate deposition and retention
in the respiratory tract is the ICRP task group lung model (Mob66,
ICRP72). In this model, shown in Chapter 7, there are four major
regions! the naso-pharyngeal, tracheobronchial, pulmonary, and
lymphatic tissues. A fraction of the inhaled activity is initially
deposited in each of the naso-pharyngeal, tracheobronchial, and
pulmonary regions. The material clears from the lung to the blood and
the gastrointestinal tract, also as shown in Chapter 7. Deposition and
clearance of inspired particulates in the lung are controlled by the
particle size and solubility classification.
The size distribution of the particles is specified by the activity
median aerodynamic diameter (AMAD)j in this document, all particulates
are assumed to have an AMAD equal to 1.0 micron unless otherwise stated.
The model employs three solubility classes, based on the chemical
properties of the nuclide; classes D, W, and Y correspond to rapid
(days), intermediate (weeks), and slow (years) clearance, respectively,
of material deposited in the respiratory passages. Inhaled nonreactive,
i.e., noble,, gases are handled as a special case.
A-9
-------
Movement of activity through the gastrointestinal (Gl) tract is
simulated with a catenary model, consisting of four segments: stomach,
small intestine, upper large intestine, and lower large intestine.
Exponential outflow of activity from each segment into the next or out
of the system is assumed. Outflow rate constants are calculated from
the transit times of Eve (Ev66). Although absorption may occur from any
combination of the four segments, only activity absorbed into the blood
from the small intestine is normally considered! the fractional
absorption from the small intestine into the blood is traditionally
denoted fj.
Activity absorbed by the blood from the GI or respiratory tract is
assumed to be distributed immediately to systemic organs. The distri-
bution of activity to these organs is specified by fractional uptake
coefficients. The list of organs in which activity is explicitly
distributed (herein termed source organs) is element-dependent, and may
include such organs as bone or liver where sufficient metabolic data are
available. This list is complemented by an additional source region
denoted as OTHER, which accounts for that systemic activity not dis-
tributed among the explicit source organs; uniform distribution of this
remaining activity within OTHER is assumed.
Radioactive material that enters an organ may be removed by both
radioactive decay and biological removal processes. For each source
organ, the fraction of the initial activity remaining at any time after
intake is described by a retention function consisting of one or more
exponentially decaying terms»
The metabolic models and parameters employed in the present study
have been described by Sullivan et al. (Su81). In most cases, the
models are similar or identical to those recently recommended by the
ICRP (ICRP79, ICRP80, ICRP81). However, some of this work was performed
prior to the publication of these documents, so that differences in
model parameters do exist for some radionuclides (Su81). In particular,
parameter values that are thought to be more representative of
metabolism following low-level environmental exposures, rather than
occupational exposures, have been used in this analysis [e.g., fj=0.2
for uranium in the environnent (ICRP79, NAS83)]. For transuranic
isotopes, metabolic parameters from the Proposed Guidance on Dose Limits
for Persons ExposedtoTransuranium Elementain the General Environment
(EPA77), related comments (EPA78), and from the National Radiological
Protection Board (NEPB82), have been used rather than those from ICRP80.
These parameters are listed in Table A—I.
The EPA Values were recommended by U.S. experts en. transuranic
element metabolism at lattelle Pacific Northwest Laboratory (EPA78).
The recently adopted National Radiation Protection Board fj valuCS for
transuranics in the general environment are closer to those values
proposed by EPA in 1977 than those currently advocated by ICRP for
occupational exposures. Use of these larger f] values increases the
estimated dose and risk from ingestioa. of transuranic materials b\it-
little effect On 35SeS foll&wirtg inhalation.
A-10
-------
Table A-l. Small intestine to blood transfer fractions, fj,
for transuranic elements
EPA
Element
Isotope
238
Pu2*1
Oxide form
Nonoxide form
Bio. inc.'*)
239
„ 240
Pu
Oxide form
Nonoxide form
Bio. inc.
Am
Oxide form
Nonoxide form
Bio. inc.
Cm
Oxide form
Nonoxide form
-Bio. inc.
N|>
Child
0-12 mo
10-2
10-2
5x10-2
10-3
10-2
5x10-2
10-2
10-2
5xlO"2
10-2
10-2
5x10-2
-
Adult
>12 mo
10-3
10-3
5x10-3
10~*
10-3
5x10*3
10-3
10-3
5x10-3
10-3
10-3
5x10-3
10-3
Adult
10-5(b)
5x10-*
5x10-*
10-5(b)
5x10**
5,10-4
5x10-*
5xlO~*
5x10-*
5x10-*
5x10-*
5x10-*
10-3
NRPB
Child
0-12 mo
5X10-*
-------
A.4 Dose Rates from External Exposures
As a result of the penetrating nature of photons, radioactivity
need not be taken into the body to deliver a dose to body organs.
Energy absorbed from photons emitted by radionuclides residing in the
air or on the ground surface may also contribute to the overall risk.
Indeed, natural background radiation is an example of an important
external exposure, ordinarily contributing the largest component of dose
to mankind.
Dose rates to organs of an individual immersed in contaminated air
or standing on a contaminated ground surface are computed by the
DOSFACTER computer code of Kocher (Ko81), These calculations assume
that the radionuclide concentration is uniform throughout an infinite
volume of air or area of ground surface, and that the exposed individual
is standing on the ground surface. Only photons penetrate the body
sufficiently to deliver a significant dose to internal organs, and only
doses from photon radiation are considered in this analysis. Beta
radiation is far less penetrating and delivers a dose only to the body
surface; because skin is not a target tissue of concern in this
analysis, no consideration of beta contributions to dose is required.
Alpha particles have even less penetration ability, and are also
excluded from consideration here.
.y
The photon dose rate factor D^ (X) for a given target org*in, X, of
an individual immersed in a unit concentration of contaminated air at
any time may be expressed as:
D!(X) = c K -
pm p., /—r n n
r a n
(Ji/p),
GX (A-4)
where
pa = density of air,
Kpm = 0.5 = particle-medium correction factor,
fY = intensity of nfch discrete photon (number/disintegration),
n
energy of n1-*1 photon,
n
= photon mass energy absorption coefficient, with
subscripts "t" and "a" denoting tissue and air,
respectively for photons of energy En,
gX = ratio of absorbed dose in organ X to absorbed dose at
the body surface,
c = unit conversion proportionality constant.
A- 12
-------
The tetms u/p and G^ are functions of photon energy,
.
The photon dose rate factor D: (x) to organ X of an individual at
a distance z above a unit concentration contaminated ground surface may
be computed as:
(X) = 0.5 c Kpffl £ fj
I1/r
Gx (A-5)
dr-[Can/(Dan-l)]
where
Kpm = 1.0 = particle-material correction factor,
Han = mass attenuation coefficient for the n*-h discrete photon,
z = height of reference position above ground surface (taken
to be 1 meter in this study),
c = unit conversion proportionality constant.
The coefficients Can and Dan are functions of the photon energy. A
detailed discussion of the derivation of these equations as well as an
extensive tabulation of dose rate factors for various radionuclides is
presented by Kocher (Ko79, Ko81).
In the analysis here, the dose rate factors described by these
equations are scaled to achieve a continuous exposure of 1 pCi/em^ for
air immersion and 1 pCi/cm^ f0j ground surface exposure. Risk estimates
for these exposure pathways are based on continuous lifetime exposure to
these levels.
Once the radionuclide intakes and concentrations are calculated for
a specific source by means of the environmental transport code, it is
Jiecessary to scale the dose and risk values due to a unit intake or
concentration to the intake and concentration values predicted by the
transport code. As shown in Figure A-l, the DARTAB computer code (Be81)
performs this step using RADRISK unit doses and risks and AIRDOS-EPA
concentrations and intakes. DARTAB is independent of both the environ-
mental transport code, e.g., AIRDOS-EPA, and the dosimetric and health
effects code, e.g., RADRISK. This eliminates redundant dose/risk
calculations and the need for extraneous coding to calculate doses and
health impacts in each environmental transport code.
A-13
-------
REFERENCES
Ba81
Be81
Co?8
DuaSO
Dub84
EPA77
EPA78
Ev66
ICRP72
Baes C. F, III and Sharp R. D., A Directory of Parameters Used
in a Series of Assessment Applications of the AIRDOS-EPA and
DARTAB Computer codes, ORNL-S72Q, Oak Ridge National
Laboratory, Oak Ridge, Tenn., March 1981,
Begovich G. L., Eckerman K. F., Schlatter E. C., Ohr S. Y.f and
Chester R. 0., DARTAB: A program to combine airborne radio-
nuclide environmental exposure data with dosimetric and health
effects data to generate tabulation of predicted impacts,
ORNL/5692, Oak Ridge National Laboratory, Oak Ridge, Tenn.,
Aurusf 1981.
Cook J. R., Hunger B., and Barrick M. K., A Computer Cede for
Cohort Analysis of Increased Risks of Death (CAIRD), EPA 520/4-
78-012, 1978.
Dunning D. E, Jr., Leggett R. W., and Yalcintas M. G., A
Combined Methodology for Estimating Dose Rates and Health
Effects from Exposure to Radioactive Pollutants, ORNL-7105,
1980,
Dunning D. E. Jr., Leggett R. W., and Sullivan R. E.,
Assessment of Health Risk from Radiation Exposures, Health
1031-1035, 1984.
U.S. Environmental Protection Agency, Proposed Guidance on Dose
Limits for Persons Exposed to Transuranium Elements in the
General Environment, EPA 520/4-77-016, 1977.
U.S. Environmental Protection Agency, Response to Comments:
Guidance on Dose Limits for Persons Exposed to Transuranium
Elements in the General Environment, EPA 520/4-78-010, 1978.
Eve I. S., A Review of the Physiology of the Gastrointestinal
Tract in Relation to Radiation Doses from Radioactive
Materials, Health Phys., JJ, 131-162, 1966.
International Commission on Radiological Protection, The
Metabolism of Compounds of Plutonium and Other Actinides, ICRP
Publication 19, Pergaraon Press, 1972.
A-14
-------
ICRP79 International Commission on Radiological Protection, Limits for
Intakes of Radionuclides by Workers, ICRP Publication 30,
Part 1, Annals of the ICRP, 2 (3/4), Pergamon Press, 1979.
ICRPSO International Commission on Radiological Protection, Limits for
Intakes of Radionuclides by Workers, ICRP Publication 30,
Part 2, Annals of the ICRP, 4 (3/4), Pergamon Press, 1980.
ICRP81 International Commission on Radiological Protection, Limits for
Intakes of Radionuclides by Workers, ICRP Publication 30,
Part 3, Annals of the IURP, j> (2/3), Pergamon Press, 1981.
Ki78 Killough G. G., Dunning D. E. Jr., and Pleasant J. C., IN1EM
II: A Computer Implementation of Recent Models for Estimating
the Dose Equivalent to Organs of Man from an Inhaled or
Ingested Radionuclide, ORNL/NUREG/TM-84, 1978.
Ko79 Kocher D. C., Dose-Rate Conversion Factors for External
Exposure to Photon and Electron Radiation from Radionuclides
Occurring in Routine Releases from Nuclear Fuel-Cycle
Facilities, ORNL/Nl/REG/TM-283, 1979.
Ko81 Kocher D. C., Dose-Rate Conversion Factors for External
Exposure to Photon and Electron Radiatim from Radionuclides
Occurring in Routine Releases from Nuclear Fuel-Cycle
Facilities, Health Phys., J38, 543-621, 1981.
Moa79 Moore R. E., Baes C. F. Ill, McDowell-Boyer L. M., Watson
A. P., Hoffman F. 0., Pleasant J. C., and Miller C. W., AIRDOS-
EPA; A Computerised Methodology for Estimating Environmental
Concewtrations and Dose to Man from Airborne Releases of
Radionuclides, EPA 520/1-79-009, EPA Office of Radiation
Programs, Washington, B.C., December 1979.
Mob66 Morrow P. E., Bates D. V., Fish B. R., Hatch T. F., and Mercer
T. T., Deposition and Retention Models for Internal Dosimetry
of the Human Respiratory Tract, Health Phys., 12, 173-207,
1966,
!»A?72 National Academy of Sciences - National Research Council, The
Effects on Populations of Exposures to Low Levels of Ionizing
Radiation, Report of the Committee on the Biological Effects of
Ionizing Radiations, Washington, D.C., 1972.
NAS80 National Academy of Sciences - National Research Council, The
Effects on Populations of Exposure to Low Levels of Ionizing
Radiation, Committee on the Biological Effects of Ionizing
Radiation, Washington, B.C., 1980.
NAS83 National Academy of Sciences - National Research Council,
Drinking Water and Health, Vol. 5, Safe Drinking Water
Committee, Washington, B.C., 1983.
A-15
-------
NRPB82 Harrison J. D., Gut Uptake Factors for Plutonium, Aoericium and
Curium, NRPB-R129, January 1982.
Su81 Sullivan R. E., Nelson N. S., Ellett W. H., Dunning D. E. Jr.,
Leggett R. W., Yalcintas M. G., and Eckeraan K. F., Estimates
of Health Risk from Exposure to Radioactive Pollutants
ORNL/TM-7745, 1981. '
A-16
-------
ADDENDUM B
MECHANICS OF THE LIFE TABLE
IMPLEMENTATION OF THE RISK ESTIMATES
B-l
-------
ADDENDUM B: MECHANICS OF THE LIFE TABLE
IMPLEMENTATION OF THE RISK ESTIMATES
CONTENTS
Page
B.I Introduction B-5
B,2 Life Table Analysis to Estimate the Risk of Excess Cancer.... B-5
B.3 Risk Analysis Methodology. B-7
Preceding page blank
B-3
-------
ADDENDUM B: MECHANICS OF THE LIFE TABLE
IMPLEMENTATION OF THE RISK ESTIMATES
B.I Introduction
This addendum describes the mechanics of the life table
implementation of the risk estimates derived in Chapter 8. The
calculation is performed as an integral part of the RADRISK code,
described in Chapter 7, since time dependent organ dose rates are used.
B.2 Life Table Analysis to Estimate the Riskof ExcessCancer
Radiation effects can be classified as stochastic or nonstochastic
(NAS8Q, ICRP77), For stochastic effects, the probability of occurrence
of the effect, as opposed to the severity, is a function of dose; induc-
tion of cancer, for example, is considered a stochastic effect. Non-
stochastic effects are those health effects for which the severity of
the effect is a function of dose; examples of nonstochastic effects
include cell killing, suppression of cell division, cataracts, and
noiuaalignant skin damage.
At the low levels of radiation exposure attributed to radionuclides
in the environment, the principal health detriment is the induction of
cancers (solid tumors and leukemia), and the expression, in later gener-
ations, of genetic effects. In order to estimate these effects, instan-
taneous dose rates for each organ at specified times are sent to a
subroutine adaptation of CAIRO (Co78> contained in the RADRISK code.
This subroutine uses annual doses derived from the transmitted dose
rates to estimate the number of incremental fatalities in the cohort due
to radiation-induced cancer in the reference organ. The calculation of
incremental fatalities is based on estimated annual incremental risks,
computed from annual doses to the organ, together with radiation risk
factors such as those given in the 1980 HAS report BEIR-3 (NAS80).
Derivation of the risk factors in current use is discussed in Chapter 8.
An important feature of this methodology is the use of actuarial
life tables to account for the time dependence of the radiation insult
and to allow for competing risks of death in the estimation of risk due
to radiation exposure. A life table consists of data describing age-
specific mortality rates from all causes of death for & given popula-
tion. This information is derived from data obtained on actual mortal-
ity rates in a real population; mortality data for the U.S. population
during the years 1969-1971 (HEW75) are used throughout this study.
'receding page Wank
B-5
-------
The use of life tables in studies of risk due to low-level radia-
tion exposure is important because of the time delay inherent in radia-
tion risk. After a radiation dose is received, there is a minimum
induction period of several years (latency period) before a cancer is
clinically observed. Following the latency period, the probability of
occurrence of a cancer during a given year is assumed to be constant for
a specified period, called & plateau period. The length of both the
latency and plateau periods depends upon the type of cancer.
During or after radiation exposure, a potential cancer victim may
experience years of life in which he is continually exposed to risk of
death from causes other than incremental radiation exposure . Hence,
some individuals will be lost from the population due to competing
causes of death, and are not potential victims of incremental radiation-
induced cancer.
It is assumed that each member of the hypothetical cohort is
exposed to a specified activity of a given radionuclide. In this analy-
sis each member of the cohort annually inhales or ingests 1 pCi of the
nuclide, or is exposed to a constant external concentration of 1 pCi/cm-*
in air or 1 pCi/cm^ on ground surfaces. Since the models used is
RAD1ISK are linear, these results may be scaled to evaluate other
exposure conditions. The cohort consists of an initial population of
100,000 persons, all of whom are simultaneously liveborn. In the sce-
nario employed here, the radiation exposure is assumed to begin at birth
and continue throughout the entire lifetime of each individual.
No member of the cohort lives more than 110 years. The span from
0 to 110 years is divided into nine age intervals, and dose rates to
specified organs at the midpoints of the age intervals are used as esti-
mates of the annual dose during the age interval. For a given organ,
the incremental probability of death due to radiation-induced cancer is
estimated for each year using radiation risk factors and the calculated
doses during that year and relevant preceding years. The incremental
probabilities of death are used in conjunction with the actuarial life
tables to estimate the incremental number of radiation-induced deaths
each year.
The estimation of the number of premature deaths proceeds in the
following, manner. At the beginning of each year, a, there is a proba-
bility P^ of dying during that year from nonradiological causes, as
calculated from the life table data, and an estimated incremental
probability PR of dying during that year due to radiation-induced cancer
of the given organ. In general, for the m-th year, the calculations
are:
M(m) = total number of deaths in cohort during year mf
> + PR(m)] x N(m)
Q(m) = incremental number of deaths during year m due to
radiation-induced cancer of a given organ,
= PR(ni) x N{m)
B-6
-------
N(ia-t-l) = number of survivors at the beginning of year m + 1
= N(m) - M(m)
(N(1)=1QO,000).
P° is assumed to be small relative to P^, an assumption which is reason-
able only for low-level exposures (BuSl), such as those considered here.
The total number of incremental deaths for the cohort is then obtained
by summing Q(m) over all organs for 110 years.
In addition to providing an estimate of the incremental number of
deaths, the life table methodology can be used to estimate the total
number of years of life lost to those dying of radiation-induced cancer,
the average number of years of life lost per incremental mortality, and
the decrease in the population's life expectancy. The total number of
years of life lost to those dying of radiation-induced cancer is com-
puted as the difference between the total number of years of life lived
by the cohort assuming no incremental radiation risk, and the total num-
ber of years of life lived by the same cohort assuming the incremental
risk from radiation. The decrease in the population's life expectancy
can be calculated as the total years of life lost divided by the
original cohort size (N(1)=10Q,QOQ).
Either absolute or relative risk factors can be used. Absolute
risk factorsj given in terms of deaths per unit dose, are based on the
assumption that there is some absolute number of deaths in a population
exposed at a given age per unit of dose. Relative risk factors, the
percentage increase in the ambient cancer death rate per unit dose, are
based on the assumption that the annual rate of radiation—induced excess
cancer deaths, due to a specific type of cancer, is proportional to the
ambient rate of occurrence of fatal cancers of that type. Either the
absolute or the relative risk factor is assumed to apply uniformly
during a plateau period, beginning at the end of the latent period.
The estimates of incremental deaths in the cohort from chronic
exposure are identically those which are obtained if a corresponding
stationary population (i.e., a population in which equal numbers of per-
sons are born and die in each year) is subjected to an acute radiation
dose of the same magnitude. Since the total persons years lived by the
cohort in this study is approximately 7,07 million, the estimates of
incremental mortality in the cohort from chronic irradiation also apply
to a one year dose of the same magnitude to a population of this size,
age distribution, and age-specific mortality rates. More precise life
table estimates for a specific population can be obtained by altering
the structure of the cohort to reflect the age distribution of a
particular population at risk.
B.3 Risk Analysis Methodology
Risk estimates in current use at EPA are based on the 1980 report
(BEIR-3) of the National Academy of Sciences Advisory Committee on the
Biological Effects of Ionizing Radiation (NAS80). The form of these
risk estimates is, to some extent, dictated by practical considerations,
B-7
-------
e.g., a desire to limit Che number of cases which must be processed for
each environmental analysis and a need to conform to limitations of the
computer codes in use. For example, rather than analyze sale and female
populations separately, the risk estimates have been merged for use with
the general population; rather than perform both an absolute and a
relative risk calculation, average values have been used.
The derivation of the risk estimates from the BEIR-3 report is pre-
sented in Chapter 8. A brief outline of the general procedure is sum-
marized below. Tables referenced from Chapter ¥ of NAS80 are designated
by a V prefix.
(1) The total number of premature cancer fatalities from lifetime
exposure to 1 rad per year of low LET radiation is constrained to be
equal to the arithmetic average (280 per million person rad) of the
absolute and relative risk values (158 and403) given in Table V-25 of
the BEIR-3 report (NAS80) for the L-L and L-L models for leukemia and
solid cancers respectively.
(2) For cancers other than leukemia and bone cancer, the age and
sex specific incidence estimates given in Table V-14 were multiplied by
the mortality/incidence ratios of Table V-15 and processed through the
life table code at constant, lifetime dose rates of 1 rad per year. The
resulting deaths are averaged, using the male/female birth ratio, and
proportioned for deaths due to cancer in a specific organ as described
in Chapter 8. These proportional risks are then used to allocate the
organ risks among the (235.5) deaths per million person rad remaining
after the 44.5 leukemia and bone cancer fatalities (Table V-17) are
subtracted from the arithmetic average of 280 given in Table V-25.
(3) The RADRISK code calculates dose rates for high- and low-LET
radiations independently. A quality factor of 20 has been applied to
all alpha doses (ICRP77) to obtain the organ dose equivalent rates in
rent per year. The derivation of the proportional organ risks and mor-
tality coefficients for alpha particles are, however, based on the dose
in rad as described in Chapter 8, Table 8-6.
.A typical environmental analysis requires that a large number of
radionuclides and multiple exposure modes be considered. The RADRISK
code has been used to obtain estimates of cancer risk for intakes of
approximately 200 radionuclides and external exposures by approximately
500 radionuclides. For each radionuclide and exposure mode we assume
that each member of a cohort of 100,000 persons is exposed to a constant
radionuclide intake of 1 pCi/year, or a concentraton of 1 pCi/cc-year
for air immersion, or of 1 pCi/cm^-year from the ground surface, until
they die or are 110 years old, the maximum cohort. The mean life span
of the cohort population is 70.7 years, a result obtained from 1970 age
specific mortality rates. The calculated dose rates and mortality
coefficients described in the preceding sections are then processed
through the life table subroutine of the RADRISK code to obtain lifetime
risk estimates. At the low levels of contamination normally encountered
in the environment, the life table population is not appreciably
B-8
-------
perturbed by the excess radiation deaths calculated and, since both the
dose and risk models are linear, these unit exposure results may be
scaled to reflect excess cancers due to the radionuclide concentrations
predicted in the analysis of a specific source.
As noted in the discussion of the life table analysis, risk esti-
mates for chronic irradiation of the cohort may also be applied to a
stationary population having the same age-specific mortality rates as
the 1970 U.S. population. That is, since the stationary population is
formed by superposition of all age groups in the cohort, each age group
corresponds to a segment of the stationary population with the total
population equal to the sum of all the age groups. Therefore, the num-
ber of excess fatal cancers calculated for lifetime exposure of the
cohort at a constant dose rate would be numerically equal to that cal-
culated for the stationary population exposed to an annual dose of the
same magnitude. Thus, the risk estimates may be reported as a lifetime
risk (the cohort interpretation) or as the risk ensuing from an annual
exposure to the stationary population. This equivalence is particularly
useful i i analyzing acute population exposures. For example, estimates
for a stationary population exposed to annual doses which vary from year
to year may be obtained by summing the results of a series of cohort
calculations at various annual dose rates.
B-9
-------
REFERENCES
Bu81 Bunger B. M., Cook J. R., and Barrick M. K., Life Table
Methodology for Evaluating Radiation Risk; An Application
Based on Occupational Exposures, Health Ph;/s. 40, 439-455.
Co78 Cook J. R., Bunger B., and Barrick M. K., A Computer Code for
Cohort Analysis of Increased Risks of Death (CAIRD),
EPA 520/4-78-012, 1978.
HEW75 U.S. Department of Health Education and Welfare, 1975, U.S.
Decennial Life Tables for 1969-1971, Vol. 1., No. 1., DREW
Publication No. (HRA) 75-1150, Public Health Service, Health
Resources Administration, National Center for Health
Statistics, Rockville, Maryland.
ICRP77 International Commission on Radiological Protection, 1977,
Recommendations of the International Commission on Radiological
Protection, Ann. ICRP, Vol. 1, No. 1, Pergamon Press, 1977.
NAS80 National Academy of Sciences - National Research Council, 1980,
The Effects on Population of Exposure to Low Levels of Ionizing
Radiation, Committee on the Biological Effects of Ionizing
Radiation, Washington, D.C.
B-10
------- |